E. Letournel R.

Judett

Fractures of the
Acetabulum
Second Edition
Entirely Revised and Enlarged

With 458 Figures in 1754 Separate Illustrations,

Some in Color

Springer-Verlag
Berlin Heidelberg New York
London Paris Tokyo Hong Kong
Barcelona Budapest
EMILE LETOURNEL

Professeur d'Orthopectie et de Traumatologie, Universite de Paris VI

Chirurgien des H6pitaux de Paris
Chef du Service d'Orthopectie et de Traumatologie
Centre Mectico-Chirurgical de la Porte de Choisy
6, Place de Port-au-Prince
F-75634 Paris Cedex 13

ROBERT JUDETt

Professeur honoraire d'Orthopectie et de Traumatologie

a la Facultede Medecine de Paris
Chirurgien honoraire des H6pitaux de Paris

Translator and Editor:

REGINALD A. ELSON

The Northern General Hospital

Sheffield 5, UK

Title of the original French edition:

Les fractures du Coty/e. © Masson et Cie., Paris, 1974

Library of Congress Cataloging-in-Publication Data

Letournel, Emile. [Fractures du cotyle. English] Fractures of the acetabulum/E. Letournel, R. Judet; [translator
and editor, Reginald A. Elson]. - 2nd ed. entirely rev. and en!. p. cm. Includes bibliographical references and
index.
ISBN-13: 978-3-642-75437-1 e-ISBN-13: 978-3-642-75435-7
DOl: 10.1007/ 978-3-642-75435-7
1. Acetabulum (Anatomy) - Frac-
tures. 2. Hip joint - Surgery. I. Judet, Robert. II. Elson, Reginald. III. Title. [DNLM: 1. Acetabulum - in-
juries. 2. Fractures. WE 750 L649f] RD549.L4713 1993, 617.1'58 - dc20. DNLM/DLC

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned,
specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is
permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version,
and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution
under the German Copyright Law.
© Springer-Verlag Berlin Heidelberg 1981, 1993
Softcover reprint of the hardcover 2nd edition 1993
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific
statement, that such names are exempt from the relevant protective laws and regulations and therefore free for
general use.
Product liability: The publishers cannot guarantee the accuracy of any information about dosage and applica-
tion contained in this book. In every individual case the user must check such information by consulting the
relevant literature.
Reproduction of the figures: Gustav Dreher GmbH, Stuttgart
"TYpesetting: K + V Fotosatz GmbH, Beerfelden

24/3130 - 5 4 3 2 1 0 - Printed on acid-free paper

Dedication
It is a great pleasure for me to honour the members
of the "Acetabular Fracture Club". The promotion
of acetabular fracture surgery has now been passed
on to them. They already know this speciality as
well as me and practice it with very similar en-
thusiasm. All the members came to Choisy many
times, either to improve their knowledge in this
field or to discuss with me the most difficult prob-
lems these fractures may present. These pupils and
friends are, in the order of their first stay at Choisy:
Claude Martimbeau, MD, from Fort Smith,
Arkansas, USA
Joel Matta, MD, from Los Angeles, California,
USA
Top right: Claude Martimbeau.
Bot/om, from left to right: Joel Matta, Jeffrey Mast, Keith
Mayo, Eric D. Johnson.
Jeffrey Mast, MD, from Detroit, Michigan,
USA
Keith Mayo, MD, from Seattle, Washington,
USA
Eric D. Johnson, MD, from Los Angeles,
California, USA
With all of them, over the last 12 years we have
organised hands-on courses in France or in the
USA to teach younger surgeons, without forgetting
to take the opportunity at these meetings to com-
pare our results and techniques, always with one
goal in mind: the improvement of acetabular frac-
ture surgery. And out of our mutual interest in the
acetabulum we have become close friends.
Preface to the Second Edition

At the request of our publishers, I accepted the task of preparing this second
edition. I felt this was necessary for several reasons: new imaging technologies
such as CT scanning and 3-D reconstructions are now used routinely, the in-
dications for employing improved approaches are clearer, and reconstructions
are facilitated by new internal fixation devices. Above all, I thought it was time
to report the long-term results of the 940 acetabular fractures, 90070 of which
were treated surgically - a unique series.
In spite of the experience acquired from the three previous reviews of cases
(1966, 1971, and 1978), I failed to foresee the amount of time this revision
would need. In fact, it took more than 3 years to follow up the larger number
of cases, and 159 patients (out of 800, i.e. 22.7%) were not included as they
had moved since their last review and simply could not be located. At a time
when it is in fashion to evaluate the cost of health care, it is strange to see how
public administrators, so keen on evaluating the immediate cost of our opera-
tions, do not care about the quality of their long-term results, which appears
to us, however, to be the best basis for the choice of the initial treatment.
Of 849 patients operated upon, 28 had less than 1 year follow-up, 74 had
died, and 49 did not undergo true reconstructive surgery, so 698 patients were
theoretically available for evaluation. We were able to examine personally 539
of them, that is 77.2%. Patients who died of a cause other than one having
to do with acetabular fracture surgery (61 patients) and patients who were
impossible to evaluate for this review but who had been followed up for more
than 1 year (105 patients) are included in our statistics. So, on the whole,
705 patients, i.e. 88.12% of the series, are considered as having been evaluat-
ed.
On a much larger scale than for the previous reviews, patient data were com-
puterised. We took particular care in attempting to evaluate the durability of
the results, and in Chap. 26 we analyse the evolution of the results of patients
who were evaluated in our successive reviews.
As far as the results are concerned, operated patients were divided into three
groups according to the time of operation with respect to the time of injury.
The largest group comprises patients operated on within 3 weeks from injury,
when surgery is not always easy but the original fracture lines can be found.
Patients operated on between 3 weeks and 4 months from injury form the sec-
ond group. This group is characterised by the fact that during the healing pro-
cess, which sometimes occurs rapidly, callus has formed, which must be totally
removed, and the fracture lines have been remodelled, both factors increasing
the difficulties of surgery. The third group comprises patients operated on
more than 4 months after the injury, i.e. beyond the average healing time, and
we have to deal with non-unions, mal-unions, or non-unions/mal-unions; then
attempts at surgical reconstruction are attempts to salvage these hips from an
early total hip replacement.
VIII Preface to the Second Edition

We continue to believe that total hip replacement is not a treatment for an

acetabular fracture. Patients so treated because the condition of the femoral
head precluded any attempt at reconstruction are not regularly included in our
statistics.
The classification we established between 1961 and 1965 has stood the test
of time and is widely used. Even if some other classifications are suggested,
we regularly find our groups, but in a different order. The regular use of 3-D
reconstruction confirms the description of the ten categories we proposed,
with all the possible transitional forms. The new radiological technologies im-
prove regularly, but are not yet reliable enough to be able to dispense with the
systematic analysis of the three plain views we have promoted. That is why the
teaching of these three standard views has been pursued and even increased.
We have done our best to analyse the results as precisely as possible - most
importantly, the quality of the immediate reconstruction which was evaluated
on the three views. The increasingly frequent use of the CT scan post-
operatively demonstrated that what appears perfect on the three views may
show some imperfections on CT sections, such as remaining step-offs of
1 - 2 mm. These hidden imperfections may explain the slightly increasing rate
of osteoarthritis over the years, following cases we initially considered as
perfectly reduced. Osteoarthritis developing 20 - 30 years after surgery is also
attributed to the acetabular fractures, and this, of course, is very questionable.
What remains difficult is the evaluation of an imperfect reconstruction.
Faults in reduction imply an intra-articular incongruence, which is most proba-
bly the dominating factor for long-term prognosis. The precise measurement
of post-surgical and post-traumatic incongruence is still impossible today,
which is why it is not possible to accurately establish the relationship between
these incongruences and their long-term results either. However, it would be
extremely helpful to know how much of the incongruence is compatible with
a very good long-term result in determining indications for surgical treatment.
We still feel the need for a precise method of evaluation of these imperfections
and for some intra-operative means of control of the quality of the reduction
before the time of wound closure.
Over the years we have been pleased to note that in many countries surgeons
are showing an increasing interest in the field of acetabular fracture surgery.
Our frequent participation in symposia or lectures, in hands-on courses
with practical exercises on pre-fractured plastic bones, and in X-ray lecture
teaching in small groups demonstrates the regular improvement in the quality
of the surgery achieved and in the knowledge of the surgeons who make special
effort to learn about this difficult surgery.
Because of the difficulties involved in the surgery of acetabular fractures,
an intensive period of learning is required, which, of course, will benefit both
patients and surgeons.
We do hope that this second English edition will help to spread interest in
acetabular fracture surgery and help in choosing the best type of treatment for
each patient.

EMILE LETOURNEL
December 1992
Acknowledgements

However much he has done himself, an author does not forget that his book
could not have come about without the contributions and encouragement of
many others; when the time comes to express his gratitude, however, he is anx-
ious not to forget anyone. When speaking and reporting on a surgical activity
of over 30 years, one cannot forget that the surgical work, which forms the
basis of this book, cannot be accomplished without the assistance of the
nurses in the operating theatre, on the ward and in the out-patient clinic, the
orthopaedic staff residents, assistants, medical doctors, and other colleagues
of the clinic who gave their assistance at various times. Also not to be forgotten
are those doctors and surgeons from all over France and many other countries,
too, who referred the patients who were treated.
The Administrative Council of the Fondation de l'Avenir and its president,
Jean Pierre Davant, and the scientific council have intensively supported the
review of the patients by providing us with all the computer means that such
a study needed.
The Choisy Medical Surgical Centre and its director, Jean Jacques Monteil,
have greatly contributed to this work. Most of the patients were operated on
there, and all the patients were reviewed there, benefitting from all the financial
and human facilities that such a centre has to offer - in the orthopaedic
department as well as in the radiology department and the out-patient
clinic.
Robert Palau, head of the radiology department at "Choisy", not only
helped out in the radiological evaluation of the patients, but, with his extensive
knowledge in the communication field, has developed the software which
allowed us to computerise our data.
Jean Pierre Moulinie, the internist and anaesthesiologist in the orthopaedic
department of Choisy, handled all the patients pre- and post-operatively. He
also wrote the chapter on the prevention of D. V. T. and infections.
My secretaries, and particularly Michele Rosec, did a great job in using all
resources available to try and get in touch with the patients and in trying to
convince them of the usefulness of the follow-up study.
John Lyttle, from Little Rock, Arkansas, USA, spent 1 year of residency
with me and contributed widely to the evaluation of the clinical and radiolog-
ical results. Bruce Buhr, from Kansas City, Missouri, USA, began his residency
at Choisy when I had just started to correct the proofs of this book. He helped
me in pursuing the typing errors, in checking the tables, and in correcting my
English. Remy Serr has been responsible for the preparation of most of the x-
ray illustrations and intra-operative pictures. He is also the man who
manipulates the Judet table so wonderfully and positions the patients so well.
The Osteo Company, Switzerland, and its President, Mr. Beat Leu, trusted
me enough, more than 10 years ago, to consider manufacturing the first com-
plete armamentarium for acetabular fracture surgery.
x Acknowledgements

Our publisher, Springer-Verlag, deserves all our grateful thanks. They ac-
cepted all the changes we felt necessary to make to the first edition. The in-
credible improvement in their editing techniques over the last 10 years made
our work and communication with them very easy. They paid regular and very
close attention to the quality of the illustrations and did their best to have this
book published as quickly as possible. I want to thank personally Ms. M.
Aryan, Mr. A. Gosling, the production editor and Ms. K. Wagstaff, the copy
editor who had the task of correcting my English.
Finally, this book owes a lot to Francine. She was responsible for inputting
all the information into the computer, and for the computerised analysis of the
cases. She also typed all the changes and additions we decided to make in revis-
ing the first edition. Nobody knows, except for me, the number of hours, even-
ings and week-ends she has spent to give birth to this second edition. She
deserves my very special and "affectionate" thanks.
Preface to the First Edition

It has been a pleasure to comply with requests to publish this book in English.
During the intervening years, there has been little to add to our views as to the
best management of acetabular fractures, but an additional chapter has been
incorporated comprising recent findings in our patients and slight changes in
emphasis on the indications for operations.
Additionally, having recognised that one of the greatest difficulties in this
method of treatment lies in the pre-operative assessment of the standard
radiographs, we have prepared a short series of radiographs which the reader
may find advantageous for study.
We are grateful to Mr. REGINALD ELSON who has translated and revised
the French edition. Considerable alteration of the text and the general presen-
tation was necessary in order to make the material palatable in English.
Our thanks are due to our new publishers, Springer-Verlag, for their keen
interest and skill.

E. LETOURNEL
R. JUDET
Editor's Preface to the First English Edition

It is a privilege to have been entrusted with the preparations of the English edi-
tion of this book. I have been aware of the authors' work for some years and
personal observation has inspired enormous respect for their experience and
expertise in this field. Undoubtedly, the information they present forms a
unique collection.
As a practising orthopaedic surgeon, I feel bound to warn colleagues not
to approach the subject lightly. Even now, having translated every sentence, I
find aspects here and there difficult to comprehend. The authors have lived
with the work for years and this shows abundantly in their ability to interpret
radiographs and to operate. It is necessary to learn some new terminology, not
difficult in itself but requiring effort nevertheless; thereafter, having grasped
the classification and basis underlying the spectrum of fractures, their inter-
pretation from radiographs is more straightforward. Even here, the average or-
thopaedic surgeon will be surprised how much effort is necessary to appreciate
the fractures in three dimensions. The operations are logical but always of con-
siderable magnitude. Undoubtedly, the anterior approaches should be prac-
tised on a cadaver. The number of fractured acetabula likely to be encountered
by one surgeon is going to be small and practical experience hard to acquire.
If open operation is to be encouraged, it is yet another procedure which should
be centralised; after resuscitation and treatment of other injuries, operation on
the acetabulum is not urgent (apart from reduction of a posterior dislocation);
after a week, the majority of patients can be transported. The authors
themselves warn against operating unless the necessary background study has
been performed. Chapter 2 may be read to advantage after study of radiology
and the classification in Chapter 3. .
A superficial appraisal of post-operative radiographs gives the impression
that the surgeon has scattered metal irresponsibly. In fact, when exposed at
operation, the application of the plates and screws is so obviously correct; one
of the most remarkable features is the immediate solidity obtained when all
goes to plan, even in the most complex fractures.
In the translation, I have tried to render the French presentation into a style
acceptable to English readers and yet preserve some of the original flavour.
Much of the French text is in the first person and sounds foreign to us; never-
theless, while putting into the third person passive the authors' recommended
instructions and teaching, I have left in the first person their discussion when
this related to what they did in treating their patients or when they debate on
opinion.
It is remarkable how many terms cannot be translated directly. For example,
the French call the root of the superior pubic ramus, Ie corps (body) du pubis,
while our body of the pubis becomes la lame quadrilatere (the quadrilateral
plate) du pubis.
XIV Editor's Preface to the First English Edition

I wish to thank VALERIE BARCLAY and ANN JOHN who typed the manu-
script.
Finally, there are a number of features on the innominate bone of which
nomenclature may cause problems. A glossary of these is appended in the hope
that the defined meaning will aid the reader who decides to pursue a study of
the subject.
REGINALD A. ELSON
Contents

Introduction: History and Development of Our Methods

of Classification and 'freatment of Acetabular Fractures

1 Anatomy of the Acetabulum ................................ 17

1.1 Columns of the Acetabulum ................................ 17
1.2 Posterior Column . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.3 Anterior Column .......................................... 18
1.3.1 Iliac Segment ........................................ 18
1.3.2 Acetabular Segment. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . 19
1.3.3 Pubic Segment ....................................... 19
1.4 Structure of the Innominate Bone in Relation to Load-Bearing .. 20
1.5 Vascular Supply ........................................... 20
1.5.1 Internal Surface ...................................... 21
1.5.2 External Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.5.3 Acetabulum .......................................... 22

2 Mechanics of Acetabular Fractures . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2.1 Force Applied to the Greater Trochanter in the Axis
of the Femoral Neck ....................................... 24
2.1.1 Neutral Abduction-Adduction .......................... 24
2.1.2 Abduction and Adduction ............................. 24
2.2 Force Applied to the Flexed Knee in the Axis
of the Femoral Shaft ....................................... 25
2.2.1 Hip Joint Flexed 90° .................................. 26
2.2.2 Different Degrees of Hip Flexion ....................... 26
2.3 Force Applied to Foot with Knee Extended ................... 26
2.3.1 Hip Flexed .......................................... 26
2.3.2 Hip Extended ........................................ 26
2.4 Force Applied to Lumbo-sacral Region ....................... 27
2.5 Comment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.6 Clinical Correlation ........................................ 27
2.6.1 Blow on Knee or Dashboard Injuries ................... 27
2.6.2 Blow on Greater Trochanter ........................... 28
2.6.3 Blow Under Foot ..................................... 28
2.6.4 Blow on Sacro-iliac Region ............................ 28
2.6.5 Antero-posterior Compression .......................... 28

3 Radiology of the Normal Acetabulum ........................ 29

3.1 Standard Radiography. .. . . . . . . .. . . . . . . . . . . .. . . . . . .. . . . . . . . . 29
3.1.1 Anterior-posterior Radiograph of Pelvis ................. 31
XVI Contents

3.1.2 Antero-posterior Radiograph of Acetabulum ............ 31

3.1.3 Obturator-oblique Radiograph. . . . . . . . . . . . . . . . . . . . . . . . . 37
3.1.4 Iliac-oblique Radiograph ............................. 43
3.2 Computed Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.2.1 CT of a Normal Acetabulum ......................... 45
3.2.2 Special Advantages of CT ............................ 47
3.2.3 Disadvantages of CT ............................ . . . . . 49
3.3 Tomography ............................................. 49
3.4 Stereo-radiography........................................ 49
3.5 Interpreting the Radiographs ............................... 49
3.5.1 Interpreting the Standard Views ....................... 49
3.5.2 Interpreting the CT Sections to Aid
or Complete the Diagnosis ...................... . . . . . . 59
4 Classification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5 Posterior Wall Fractures .................................. . 67
5.1 Typical Posterior Wall Fractures ........................... . 68
5.1.1 Morphology ....................................... . 68
5.1.2 Radiology ......................................... . 68
5.2 Postero-superior Fractures ................................ . 75
5.2.1 Morphology ....................................... . 75
5.2.2 Radiology ......................................... . 75
5.3 Postero-inferior Fractures ................................. . 79
5.3.1 Morphology ....................................... . 79
5.3.2 Radiology ......................................... . 79
5.4 Special Forms of Posterior Wall Fractures .................. . 79
5.4.1 Extended Posterior Wall Fractures .................... . 79
5.4.2 Horizontal Extension of Fracture Line ................ . 79
5.4.3 Massive Posterior Wall Fractures ..................... . 79
5.4.4 Posterior Wall and Incomplete Transverse Fractures ..... . 82
5.5 CT Study of Posterior Wall Fractures ...................... . 85
6 Fractures of the Posterior Column .......................... 89
6.1 Typical Posterior Column Fractures ......................... 89
6.1.1 Morphology ........................................ 89
6.1.2 Radiology .......................................... 90
6.2 Extended Posterior Column Fractures ....................... 93
6.2.1 Morphology ........................................ 93
6.2.2 Radiology .......................................... 94
6.3 Atypical Posterior Column Fractures ........................ 94
6.3.1 Other Associated Pelvic Ring Fractures. . . . . . . . . . . . . . . . . 94
6.3.2 Epiphyseal Injury .................................... 94
6.4 Transitional Posterior Column Fractures ..................... 99
6.4.1 Partial Superior Fractures ............................ 99
6.4.2 Partial Inferior Fractures ............................. 101
6.5 CT Study of Posterior Column Fractures .................... 101
7 Anterior Wall Fractures ................................... 103
7.1 Morphology ............................................. 103
7.2 Radiology ............................................... 104
7.2.1 Antero-posterior View ................................ 104
7.2.2 Obturator-oblique View .............................. 107
Contents XVII

7.2.3 Iliac-oblique View ................................. 107

7.3 Atypical Examples ....................................... 107
7.4 CT Study of Anterior Wall Fractures . . . . . . . . . . . . . . . . . . . . . . . 111

8 Fractures of the Anterior Column ......................... 115

8.1 Morphology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115
8.1.1 Very Low Fractures ................................ 115
8.1.2 Low Fractures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115
8.1.3 Intermediate Fractures ............................. 116
8.1.4 High Fractures .................................... 117
8.1.5 Atypical Examples ................................. 119
8.2 Radiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 119
8.2.1 Very Low Fractures ................................ 121
8.2.2 Low Fractures ..................................... 121
8.2.3 Intermediate Fractures ............................. 123
8.2.4 High Fractures .................................... 124
8.2.5 Atypical Examples ................................. 127
8.3 CT Study of Anterior Column Fractures ................... 128

9 Pure Transverse Fractures ................................. 141

9.1 Morphology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 141
9.1.1 Orientation of Fracture. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 141
9.1.2 Displacement in Transverse Fractures. . . . . . . . . . . . . . . .. 142
9.2 Radiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.2.1 Antero-posterior View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.2.2 Obturator-oblique View ............................ 150
9.2.3 Iliac-oblique View ................................. 152
9.3 Atypical Cases .......................................... 152
9.4 CT Scan Study of Transverse Fractures ..................... 155

10 T-shaped Fractures ....................................... 163

10.1 Morphology ............................................ 163
10.1.1 Transverse Component ............................. 164
10.1.2 Stem Component .................................. 164
10.1.3 Displacement ..................................... 167
10.2 Radiology .............................................. 167
10.2.1 Transverse Component ............................. 167
10.2.2 Stem Component .................................. 168
10.3 Atypical Examples ....................................... 170
10.3.1 Additional Vertical Fracture of Obturator Ring. . . . . ... 170
10.3.2 Additional Fracture Line in Cotyloid Fossa ........... 170
10.3.3 Association of a Posterior Column
and an Anterior Hemitransverse Fracture ............. 179
10.4 CT Study of T-Shaped Fractures. . . . . . . . . . . . . . . . . . . . . . . . . .. 179

11 Associated Posterior Column and Posterior Wall Fractures .... 189

11.1 Morphology ............................................ 189
11.1.1 Posterior Wall Component .......................... 189
11.1.2 Posterior Column Component . . . . . . . . . . . . . . . . . . . . . .. 189
11.2 Radiology .............................................. 189
11.2.1 Antero-posterior View .............................. 189
XVIII Contents

11.2.2 Obturator-oblique View ............................ 190

11.2.3 Iliac-oblique View ................................. 190
11.3 Atypical Examples ....................................... 190
11.4 Comment............................................... 191
11.5 CT Study of Associated Posterior Column
and Posterior Wall Fractures .............................. 194

12 Associated Transverse and Posterior Wall Fractures .......... 201

12.1 Cases with Posterior Dislocation .......................... 202
12.1.1 Morphology ...................................... 202
12.1.2 Radiology ........................................ 203
12.1.3 Atypical Examples ................................. 208
12.2 Cases with Central Dislocation ............................ 208
12.2.1 Morphology ...................................... 208
12.2.2 Radiology ........................................ 211
12.3 Comment............................................... 213
12.4 Very Large Postero-superior Fragment
Extending to the Iliac Crest ............................... 213
12.5 CT Study of Associated Transverse
and Posterior Wall Fractures .............................. 221

13 Associated and Posterior Hemitransverse Fractures ........... 231

13.1 Morphology ............................................ 231
13.1.1 Anterior Fractures with Associated Complete Posterior
Hemitransverse Fracture ............................ 231
13.1.2 Anterior Fractures with Associated Incomplete Posterior
Hemitransverse Fractures ........................... 232
13.1.3 Important Remarks ................................ 232
13.2 Radiology .............................................. 234
13.2.1 Anterior Fracture .................................. 234
13.2.2 Posterior Column Fracture .... . . . . . . . . . . . . . . . . . . . . .. 237
13.2.3 A Special Feature of this Group ..................... 239
13.3 Atypical Examples ....................................... 241
13.4 Radiological Differential Diagnosis ........................ 245
13.5 CT Study of Associated Anterior
and Posterior Hemitransverse Fractures ..................... 246

14 Associated Both-Column Fractures. . . . . . . . . . . . . . . . . . . . . . . .. 253

14.1 Morphology ............................................ 253
14.1.1 Posterior Column Components ...................... 254
14.1.2 Additional Posterior Components ................... 255
14.1.3 Anterior Column Component ....................... 256
14.1.4 Result of Both-Column Fracture. . . . . . . . . . . . . . . . . . . .. 257
14.1.5 Displacement of the Fragments and the Femoral Head . 258
14.1.6 Atypical Examples ................................. 259
14.1.7 The Key to Reconstruction .......................... 259
14.2 Radiology .............................................. 261
14.2.1 Antero-posterior View. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 261
14.2.2 Obturator-oblique View ............................ 263
14.2.3 Iliac-oblique View ................................. 264
14.3 Summary............................................... 265
Contents XIX

14.4 Atypical Examples ....................................... 266

14.5 Differential Radiological Diagnosis ........................ 267
14.6 CT Study of Associated Both-Column Fractures. . . . . . . . . . . .. 267

15 Transitional and Extra-articular Forms ..................... 315

15.1 Transitional Forms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 315
15.2 Extra-articular Forms .................................... 316

16 Associated Injuries ...................................... 323

16.1 Injury of the Femoral Head .............................. 323
16.1.1 Macroscopic Injury ................................ 323
16.1.2 Vascular Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 324
16.1.3 Molecular Injury .................................. 324
16.2 Capsular Injury ......................................... 324
16.3 Vascular Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 324
16.3.1 Acetabular Wall ................................... 324
16.3.2 Pelvic Vessels ..................................... 324
16.3.3 Retro-peritoneal Haematoma ........................ 325
16.4 Other Pelvic Injuries ..................................... 325
16.5 Associated Hip Injuries .................................. 326
16.6 Other Skeletal Injuries ................................... 328
16.7 Urinary Tract Injuries. . . . . .. . . . .. . . . . .. . . . . . . .. .. .. . . . ... 328
16.8 Other Visceral Injuries ................................... 328
16.9 Associated Skull Trauma ................................. 328
16.10 Sciatic Nerve Injuries .................................... 328

17 Distribution of the Clinical Series ......................... 329

17.1 Distribution According to Age ............................ 329
17.2 Distribution According to Sex ............................. 329
17.3 Distribution According to Time After Injury ................ 329

18 Clinical Presentation ..................................... 333

18.1 Clinical Findings ........................................ 333
18.1.1 Posterior Dislocation ............................... 333
18.1.2 Central Dislocation ................................ 333
18.2 Early Complications ..................................... 333
18.2.1 Traumatic Shock .................................. 333
18.2.2 Retro-peritoneal Haematoma ........................ 333
18.2.3 Pre-operative Sciatic Nerve Injury ................... 334
18.2.4 Morel-Lavalle Lesion ............................... 337
18.2.5 Intra-articular Incarceration of Bone Fragments ....... 337
18.2.6 Other lYpes of Palsies ............................. 341
18.3 Special Cases ........................................... 343
18.3.1 Children.......................................... 343
18.3.2 Elderly Patients ................................... 343
18.3.3 Pathological Fractures. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 343

19 General Principles of Management of Acetabular Fractures ... 347

19.1 Conservative Treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 347
19.1.1 Indications ....................................... 347
xx Contents

19.1.2 Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 348

19.1.3 Results ........................................... 349
19.2 Justification for Operative Treatment .... . . . . . . . . . . . . . . . . . .. 358
19.3 Indications for Operative Treatment ........................ 359
19.4 Timing of Surgery ....................................... 359

20 Surgical Approaches to the Acetabulum .................... 363

20.1 Classical Approaches ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 363
20.2 Kocher-Langenbeck Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 364
20.2.1 Technique ........................................ 365
20.2.2 Application ....................................... 369
20.2.3 Closure .......................................... 370
20.2.4 Dangers .......................................... 371
20.2.5 Complications............ . . . . . . . . . . . . . . . . . . . . . . . .. 371
20.3 Ilio-femoral Approach ................................... 373
20.3.1 Technique ........................................ 373
20.3.2 Application ....................................... 374
20.3.3 Closure .......................................... 375
20.3.4 Dangers .......................................... 375
20.3.5 Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 375
20.4 Ilio-inguinal Approach ................................... 375
20.4.1 Technique ........................................ 375
20.4.2 Application ....................................... 381
20.4.3 Closure .......................................... 382
20.4.4 Dangers .......................................... 382
20.4.5 Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 383
20.5 Combined Anterior and Posterior Approaches .............. 385
20.6 Extended Ilio-femoral Approach. . . . . . . . . . . . . . . . . . . . . . . . . .. 386
20.6.1 Technique ........................................ 386
20.6.2 Application ....................................... 391
20.6.3 Closure .......................................... 391
20.6.4 Dangers .......................................... 392
20.6.5 Complications ..................................... 393
20.7 Post-operative Care ...................................... 395
20.8 Summary of the Use of Different Surgical Approaches ....... 395
20.9 Addendum: The Kocher-Langenbeck Extended
to a Triradiate Approach ................................. 395

21 Operative Treatment of Displaced Fractures

Within Three Weeks of Injury ............................ 399
21.1 Pre-operative Care ....................................... 399
21.2 Choice of Surgical Approach ............................. 399
21.2.1 Kocher-Langenbeck Approach. . . . . . . . . . . . . . . . . . . . . .. 400
21.2.2 Ilio-femoral Approach ............................. 400
21.2.3 Ilio-inguinal Approach ............................. 400
21.2.4 Fracture Types for Which There Is a Choice
of Approach ...................................... 400
21.3 Operative Details ........................................ 402
21.3.1 Where and How to Insert Screws .................... 402
21.3.2 Special Instruments ................................ 403
21.3.3 Implants for Osteosynthesis ......................... 407
21.3.4 Method of Internal Fixation ........................ 410
Contents XXI

21.3.5 Reduction of Dislocation .......................... 411

21.3.6 Reduction of Fracture ............................. 411
21.4 Post-opertive Care ....................................... 412
21.4.1 Local Care ... , . . . . .. . . . . . . . . . . ... . . . . . . .. . . . . . . .. 412
21.4.2 Physiotherapy .................................... 412
21.4.3 Medical Treatment. JEAN-PIERRE MOULINIE ......... 412

22 Operative Treatment of Specific Types of Fracture .......... . 417

22.1 Posterior Wall Fractures ................................. . 417
22.1.1 Postero-superior Fractures ........................ . 420
22.1.2 Postero-inferior Fractures ......................... . 421
22.1.3 Special Features ................................. . 421
22.2 Posterior Column Fractures .............................. . 421
22.2.1 Special Features ................................. . 428
22.3 Anterior Wall Fractures ................................. . 428
22.4 Anterior Column Fractures .............................. . 431
22.4.1 Middle and Low Fractures ........................ . 431
22.4.2 High Fractures .................................. . 431
22.4.3 Special Features ................................. . 436
22.4.4 Insertion of Screws Along the Pelvic Brim .......... . 436
22.5 Pure Transverse Fractures ................................ . 442
22.5.1 Pure Juxta-tectal or Infra-tectal Transverse Fractures .. 442
22.5.2 Pure Trans-tectal Transverse Fractures .............. . 442
22.5.3 Special Features ................................. . 447
22.6 Associated Posterior Column and Posterior Wall Fractures ... . 447
22.7 Associated Transverse and Posterior Wall Fractures ......... . 455
22.7.1 Kocher-Langenbeck Approach ..................... . 455
22.7.2 Extended Ilio-femoral Approach ................... . 456
22.7.3 Special Features ................................. . 461
22.8 T-shaped Fractures ...................................... . 461
22.8.1 Special Features ................................. . 475
22.9 Associated Anterior and Hemitransverse Posterior Fractures .. 475
22.10 Both-Column Fractures .................................. . 484
22.10.1 Approach ...................................... . 484
22.10.2 Reduction and Fixation Through Posterior Approach .. 484
22.10.3 Reduction and Fixation Through Ilio-inguinal
Approach ...................................... . 491
22.10.4 Reduction Necessitating Both Approaches .......... . 502
22.10.5 Reduction and Fixation Through Extended
Ilio-femoral Approach ........................... . 503
22.10.6 A Particular Both-Column Fracture ................ . 505
22.11 Special Examples ....................................... . 507
22.11.1 Incarcerated Intra-articular Fragments .............. . 507
22.11.2 Bilateral Acetabular Fractures ..................... . 507
22.11.3 Fractures of Paralysed Hips 507

23 Anatomical Results of Operation

Within Three Weeks After Injury 521

23.1 Analysis of the Immediate Radiological Results ............. 522

23.2 Analysis of Imperfect Radiological Reductions .............. 526
23.3 The Learning Curve ..................................... 528
XXII Contents

Appendix: CLAUDE MARTIMBEAU'S Method of Assessing

Displacement in Acetabular Fractures ...................... 533

24 Early Complications of Operative Treatment

Within Three Weeks of Injury ............................ 535
24.1 Death.................................................. 535
24.2 Infection ............................................... 535
24.2.1 Analysis of Post-operative Infections ................. 535
24.2.2 Cause of Infection ...... . . . . . . . . . . . . . . . . . . . . . . . . . .. 536
24.2.3 Prophylaxis ....................................... 536
24.2.4 Treatment ........................................ 537
24.3 Nerve Damage .......................................... 537
24.3.1 Sciatic Nerve Damage .............................. 537
24.3.2 Other Nerve Damage .............................. 539
24.4 Secondary Displacement of Fracture Site ................... 539
24.5 Thrombo-embolism ...................................... 540
24.6 Wound Complications .................................... 540
24.7 Miscellaneous Complications .............................. 540

25 Late Complications of Operative Treatment

Within Three Weeks of Injury ............................ 541
25.1 Pseudarthrosis .......................................... 541
25.2 Cartilage Necrosis ....................................... 543
25.3 Avascular Bone Necrosis ................................. 545
25.3.1 Aetiology......................................... 549
25.3.2 Time of Presentation .............................. 550
25.3.3 Clinical and Radiological Course .................... 550
25.3.4 Clinical and Radiological Results .................... 551
25.3.5 Conclusion ....................................... 551
25.4 Post-traumatic Osteoarthritis .............................. 551
25.4.1 Osteophytes ....................................... 551
25.4.2 Osteoarthritis ..................................... 553
25.5 Post-operative Ectopic Ossification. . . . . . . . . . . . . . . . . . . . . . . .. 558
25.5.1 Clinical and Radiological Presentation ............... 558
25.5.2 Aetiology ......................................... 559
25.5.3 Treatment ........................................ 559
25.5.4 Prevention ........................................ 560
25.5.5 Results of Surgical Excision of Ectopic Bone .......... 562
25.5.6 Ectopic Ossification and Cranio-cerebral Trauma ...... 562
25.5.7 Ectopic Ossification and Type of Fracture ............ 562

26 Clinical and Radiological Results of Operation

Within Three Weeks of Injury ............................ 565
26.1 Clinical Results .......................................... 565
26.1.1 Type of Fracture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 568
26.1.2 Age of Patient .................................... 569
26.2 Radiological Results ..................................... 569
26.3 Late Overall Clinical Results and Quality of Reduction ....... 573
26.3.1 Perfect Reductions ................................. 574
26.3.2 Imperfect Reductions .............................. 574
26.3.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 579
26.4 Summary of Results ..................................... 580
Contents XXIII

26.4.1 Early Results 580

26.4.2 Late Results 580
26.5 Conclusions ............................................ . 581
26.6 Comment .............................................. . 581

27 Reassessment of Patients Treated Operatively

Within Three Weeks of Injury ............................ 583
27.1 Evolution in Patients Operated on Before 1966 .............. 583
27.2 Evolution in Patients Operated on 1966-1971 ............... 586
27.3 Evolution in Patients Operated on 1971-1978 ............... 588
27.4 Assessment of Patients Operated on 1978 -1990 ............. 588
27.5 Longitudinal Assessment of All Excellent
or Very Good Results .................................... 588

28 Operative Treatment Between Three Weeks and Four Months

After Injury ............................................ 591
28.1 Condition of Fracture Healing ............................ 591
28.2 Surgical Approach ....................................... 591
28.3 Surgical Technique ....................................... 593
28.3.1 Cases with Visible Fracture Lines .................... 593
28.3.2 Mal-union ........................................ 593
28.3.3 Non-union/Mal-union .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 593
28.3.4 Neglected Posterior Dislocations of the Femoral Head .. 593
28.3.5 Incarcerated Fragments ............................. 607
28.3.6 Review of Surgical Techniques. . . . . . . . . . . . . . . . . . . . . .. 619
28.4 Intra-operative Complications ............................. 619
28.5 Early Post-operative Complications ........................ 623
28.6 Late Post-operative Complications ......................... 626
28.7 Results ................................................. 630
28.8 Conclusion ............................................. 633

29 Operative Treatment More Than Four Months After Injury ... 635
29.1 General Considerations and Condition of Fracture Healing ... 635
29.2 Preconditions for Surgery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 635
29.3 Time of Operation After Injury ........................... 636
29.4 Choice of Surgical Approach ............................. 636
29.5 Surgical Techniques Employed ............................. 636
29.5.1 Cases in Which Reconstruction Was Impossible ....... 636
29.5.2 Cases in Which Reconstruction Was Possible .......... 637
29.6 Overview of the 123 Cases Treated More Than Four Months
After Injury ............................................ 665
29.6.1 Reconstruction Impossible (49 Cases) ................ 665
29.6.2 Repositioning of Posteriorly Dislocated Femoral Head
(11 Cases) ........................................ 682
29.6.3 Missed Incarcerated Fragment (16 Cases) ............. 682
29.6.4 Mal-unions, Non-unions, Mal-union/Non-unions ...... 682
29.7 Conclusion ............................................. 684

30 Exercises in Radiographic Diagnosis ....................... 685

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 723

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 731

Glossary

Acetabular notches both anterior and posterior lips of acetabular margin present with
well defined indentations in their mid-parts
Acetabulo-obturator radiological landmark described in Sect. 3.2.1
line
Anterior pillar thickened area of iliac wing which supports gluteus medius tubercle
Anterior tubercle of tubercle located on margin of obturator foramen at junction of in-
obturator foramen ferior pubic ramus and ischial ramus
Gluteal surface external iliac fossa
Gluteus medius tubercle tubercle of iliac crest
Innominate bone used throughout text for os innominatum or pelvic bone
Interspinous notch area of anterior border of innominate bone between anterior
superior and anterior inferior iliac spines
Ischio-pubic notch recess at junction of body of ischium with root of superior pubic
ramus
Marginal impaction (French: fracture mixte) - the impaction and incarceration into the
underlying cancellous bone of small osteochondral fragments from
the shattered margin of the acetabulum (further explanation on
page 35)
Obturator ring convenient description of the whole bone structure enclosing the
obturator foramen
Osteosynthesis widely used and convenient term which embodies reduction and
fixation of fractures
Pelvic brim refers to brim of true pelvis: an anatomically complex formation
which could be variously called ilio-pectineal, innominate or
(posteriorly) arcuate lines
Posterior tubercle of tubercle located on margin of obturator foramen at junction of
obturator foramen body of ischium and pubic ramus
Psoas gutter ilio-pectineal surface of innominate bone in which the psoas lies at
its departure from the pelvis
Root of superior pubic that part of the superior pubic ramus which contributes to the
ramus acetabulum (in French: body of the pubis)
Sciatic buttress condensation of trabecular bone which related to the angle of the
greater sciatic foramen and formed by system described in Sect. 1.4
Sciatic nerve its two large branches are named in this book: lateral popliteal
( = Common peroneal nerve) and Medial popliteal (Tibial nerve)

Sub-cotyloid groove groove under overhanging lip of postero-inferior aspect of inferior

horn of articular surface
Teardrop commonly used radiological feature which the French call (more ac-
curately) a "U"
Introduction: History and Development of Our Methods
of Classification and Treatment of Acetabular Fractures
Many readers will already know, and the rest must
be told, that a significant part of the work pre-
sented here was carried out in collaboration be-
tween myself and my Boss, the late Professor
Robert JUDET (Fig. 1.1). I refer to him constantly,
so great are the respect, the admiration and the af-
fection that I feel towards him. I am pleased to
acknowledge the amount of theoretical and prac-
tical things I learned from him; above all, I feel that
in the more than 20 years that I worked with him,
he conveyed to me some ways of thinking and prac-
tising surgery that I will use with benefit until my
last day in the operating theatre.
Robert JUDET died suddenly on 20 December
1980, so this new edition of this book will include
new sections in various chapters to which he was
not able to contribute, relating to such matters as
new devices, new tools and the use of computed
tomographic (CT) scanning.
In the past few years I have been able to follow
up some of the first acetabular fractures on which
he operated, and I and the patients paid tribute
together to this most exceptional man and most
Fig. 1.1. Robert JUDET in his boat
gifted surgeon. In this new Introduction, to pay
homage to Robert JUDET'S memory, it gives me
great pleasure to tell readers the story of the devel-
opment of our classification and management of
acetabular fractures as I experienced it.
I became Robert JUDET'S Resident at the Ray-
mond Poincare Hospital, near Paris, in May 1957.
My initial appointment was for 6 months; he kept
me for a year. In October 1961, I returned to Robert
JUDET'S department and stayed with him until his
retirement in 1978, being successively Senior
Registrar and Assistant and Associate Professor in
Orthopaedics. We thus worked together for 23
years. In 1963, in addition, I was offered the oppor-
tunity to take over headship of the Orthopaedic
Department of the Porte de Choisy Surgical Cen-
tre, still keeping the advantage of working four
mornings a week at the Raymond Poincare Hospi-
tal.
In 1957, Robert JUDET, extremely disappointed
with the results of conservative treatment of ace-
tabular fractures, had already operated upon about
12 patients through the Kocher-Gibson or Smith-
2 IntroductIOn
Petersen approaches. Figure 1.2 shows a diagram
drawn by Robert JUDET of a T-shaped and
posterior wall fracture, operated on through the
Oilier approach. For the first and the last time in
his life, he had used a cerclage wire around the
pelvic bone.
During my residency, I had the luck to assist
Robert JUDET in a few cases and was able to ap-
preciate the difficulties he met in complex cases
despite his unusual surgical skill. As he saw it, the
problem was that the approaches were inadequate,
and he asked me to go to the anatomy laboratory
to try to improve the approaches to the broken
acetabuli. So I did. But whatever the modifications
we introduced into the approaches already men-
tioned, whatever the new approaches by which we
tried to improve access to the pelvic bone, the dif-
ficulties remained practically the same in some
cases. Finally, I realised, and told the Boss, that in
my opinion the difficulties we met intra-operatively
were due in part to inadequate approaches, but also
to the fact that in most cases we were completely
unable to grasp the precise outline of the fracture
we had to treat from the traditional antero-
posterior and lateral radiographs, and that
therefore the choice of approach was often more or
less a toss-up. Consequently, I asked Robert JUDET
r
Fig. 1.2. Robert JUDET'S diagram of a T-shaped and posterior
wall fracture he operated on in 1956 through an Oilier approach.
He used a cerclage around the pelvic bone
to allow me to collect data on all the cases he had
treated in both private practice and the national
health system, in order to carry out a detailed
radiographic study. In 1959 I was able to collect
material on 75 cases, 30 of which had been
operated upon. These were all analysed in my thesis
published in 1961.
What, then, was the state of the art in acetabular
fractures in the late 1950s? From the literature of
that period, it can be seen that four types of frac-
tures had already been identified:
posterior wall fractures
transverse fractures
T- or Y-shaped fractures
the association of a transverse and a posterior
wall fracture
All others were grouped under the name of "central
fracture dislocations", "stove-in hips", bursting
fractures, etc.
As for treatment, conservative methods were
used almost universally; attempts to treat these
fractures surgically were very infrequent. By 1960,
a short series of open reduction and internal fixa-
tion of posterior wall fractures had been published:
- JUDET: 9 cases through the
Kocher-Langenbeck
approach
WALLER: 27 cases through the
Oilier approach
STEWART and MILFORD: 15 cases
However, the treatment of central dislocations re-
mained almost exclusively conservative. In 1951,
however, CAUCHOIX mentioned occasional at-
tempts to reduce such fractures surgically, per-
formed by LAMBOTTE, VAUGHAN, and LERICHE,
but he did not say with what results. In 1960, I was
able to collect from the literature only 20 cases of
central dislocations treated surgically by other
authors:
LEVINE (1943): 1 case
URIST (1949): 1 case
OKELBERRY (1955): 7 cases
ELIOT (1956): 3 cases
KNIGHT and SMITH (1958): 8 cases
In addition to these, Robert JUDET had operated
upon 13 cases.
When I began my study of Robert JUDET's 75
cases, I did not know that that would be the pro-
gramme of my week-ends and holidays for several
 Introduction
years! The classification was in fact completed 6
years later, in 1965.
We began our work by trying better to under-
stand the architecture of the innominate bone. Per-
forming horizontal cuts (Fig. 1.3) 0.5 cm thick and
other cuts parallel to the acetabular rim, which
were X-rayed, we finally reached the conclusion
that we should consider the acetabulum as being
contained within the open arms of an inverted V
(Fig. 1.4A) formed by a posterior column or the
ilio-ischial component, so called in reference to its
double embryological origin, the ilium above and
the ischium inferiorly; and an anterior column or
ilio-pubic column, embryologically formed by a
part of the ilium above and the pubic bone inferior-
ly. The anatomical roof of the acetabulum formed
the keystone of the arch. This was the way in which
the acetabulum was described in the classical article
in the American Journal of Bone and Joint
Surgery, December 1964 issue (JUDET and
LETOURNEL 1964).
A few years later, the study of many new cases
changed our minds and prompted us to regard the
acetabulum as being located within the open arms
of an inverted Y (Fig. I.4B), still formed by two col-
umns of bone. Of these, the posterior or ilio-ischial
Fig. 1.3. A Radiographs of vertical cuts of the innominate bone. B Radio-
graphs of horizontal cuts of the innominate bone. These radiographs help
one to grasp the concept of the columnar architecture of the innominate
A bone
3
column remains unchanged, but the concept of the
anterior column is completely different: it is now
seen as extending from the anterior part of the iliac
crest to the pubic symphysis, the upper end of the
posterior column being attached to the posterior
aspect of the anterior one just a little above its mid-
level, forming an angle of about 60 o. As in the pre-
vious description, the summit of the angle is filled
with a fillet of compact bone which constitutes the
anatomical roof of the acetabulum and forms the
keystone of the arch. The two columns are linked to
the auricular articular surface of the sacro-iliac
joint by the "sciatic buttress", which forms an
angle of about 70 ° with the acetabular rim
(Fig. 1.4 C).
To study the radiographs carefully, I found no
other way than to follow the radiological land-
marks; very quickly, tracing paper proved to be
most helpful and I began to "draw the lines". At
the beginning (Fig. 1.5), I superimposed all the
lines. I rapidly realised that most of them were
either undescribed or improperly described and
that a careful study of some of them sufficed to
make a correct diagnosis. However, I must confess
that I waited until 1965 before undertaking a com-
plete, systematic radiological study of the dry
4 Introduction

c
Fig. 1.4. A The columns of the acetabulum as we understood them since 1966. C The two columns are linked to the sacral
them initially (JUDET et al. 1964). B Diagram of the two col- bone by the "sciatic buttress"
umns supporting the acetabulum such as we have described

pelvic bone, superimposing or covering the ar-
ticular surfaces with thin sheets of lead, underlay-
ing pelvic edges with lead wires, and underlaying
with image-intensifier guidance the lines due to
tangencies to bone surfaces such as the ilio-ischial
line. At last we acquired a complete knowledge of
the relationships between the radiological land-
marks and the anatomical surfaces related to them.
It also quickly appeared that the traditional
antero-posterior and lateral views taken of most hips
were inadequate for an understanding of the frac-
ture lines. Then we tried to use stereo radiographs,

A
Fig. 1.5. A Antero-posterior view of the first pure posterior column fracture ever described. B Diagram from November 1959, show-
ing the way we used initially to study the radiographs
Introduction
but they soon proved to be of little help, as they are
not easy to read and it is basically impossible for
several people to look at them at the same time. We
also tried all the different oblique incidences de-
scribed by Ake WALLER, Merle D'AuBIGNE and
others, but those were essentially designed to show
the posterior wall. Finally, I noticed that in the long
series of radiographs that some patients had, it was
the two 45 0 oblique views of the L 5-S 1 disc which
best showed the two major parts of the innominate
bone, the iliac wing and the obturator ring, which
are anatomically disposed at approximately right
angles. Centring the X-ray beam a little further
downwards and on the femoral head, we obtained
the oblique views which became known as JUDET'S
views. One of these displays the iliac wing and
superimposes onto each other the anterior and
posterior limits of the obturator frame; we later
called it the "iliac-wing-oblique view" (or IOV). On
the other, the obturator frame is perfectly displayed
and the iliac wing is seen in profile; this is the "ob-
turator-oblique view" (or OOV). From 1959, all our
patients underwent radiography with an antero-
posterior view of the pelvis and an antero-posterior
and the two 45 0 oblique views of the injured hip.
The improvement in tomography by 1963 was
such that we used it for most patients. Things re-
mained stable at this until the arrival of the stan-
dard CT scan, which for us was in 1984. Routine
use of the 3-D CT scan, promoted by Dana MEARS
in 1980, which so greatly simplifies the understand-
ing of fractures, began for us in 1989.
But I have not yet explained how we found the
basis of our classification. The whole story starts
with one particular case (Fig. 1.6). Mr. P. had been
operated on by Robert and Jean JUDET on 21 June
1955, 29 days after injury, through a Kocher-Gib-
son approach. The operative report is short: "Very
difficult reduction of the posterior fragment which
is not intrapelvic but displaced posteriorly. The
posterior fragment is fixed with 2 screws to the
anterior part of the bone:' Post-operatively, a
plaster cast was applied for 3 months. I had already
many times traced the lines on the radiographs of
this case, and on a Sunday morning in autumn 1959
that I well remember, it suddenly became obvious
to Robert JUDET and me that the ilio-pectinealline
was entirely intact and not involved, meaning that
the main radiological landmark of the anterior col-
umn was intact, and so, therefore, should be the
anterior column itself. More careful analysis
(Fig.1.6A) showed that, in the same way, the
anterior lip, the roof and the teardrop were intact
5
and at their proper positions, so none of the land-
marks of the anterior column was involved in this
injury. On the other hand, the main fragment ap-
parently pushed inwards by the femoral head
(Fig. 6B) was obviously limited posteriorly by the
posterior margin of the innominate bone with the
two characteristic sciatic notches, and on this frag-
ment the entire ilio-ischialline was visible and had
lost, in an inwards direction, its normal relation-
ship with the teardrop, its faithful companion. Not
less obviously, the upper part of the posterior lip of
the acetabulum was broken. So all the anatomical
landmarks of the posterior column belonged to the
displaced fragment, which in fact represented the
entire posterior column. Thus, Robert JUDET,
without realising it, had operated 4 years previously
on a pure posterior column fracture (Fig. 1.6 C).
Among our 75 cases we found two other similar
ones, and we were then able to make a distinction
between the posterior wall fracture, which is in fact
a partial fracture of the posterior column, and
complete fracture of the posterior column.
If trauma was able to detach the posterior col-
umn partially or totally, and if our concept of the
columnar architecture was correct, partial and total
fractures should also occur at the level of the
anterior column. Among our first 75 cases, we
found this case: Mr. D., 81 years old, hit at hip level
by a car in January 1956 and treated conservatively
with a plaster cast for 60 days (Fig.1.7)! As I de-
scribed before in relation to the posterior column
fracture, I had drawn and redrawn the radiological
lines before realising one day that it was obvious,
beyond any possible argument, that the posterior
acetabular rim was intact, and the ilio-ischial line
was intact, together with the major part of the
roof. That meant that all the landmarks of the
posterior column, and consequently the posterior
column itself, were not involved in that injury. On
the other hand, the pelvic brim, the main landmark
of the anterior column, was broken in several
places, and a part of the superior pubic ramus was
obviously detached and verticalised. The teardrop
was invisible. Thus, all the landmarks of the
anterior column were involved, and therefore, I
thought, we had found the counterpart of the pure
posterior column fracture: namely, a pure anterior
column fracture. In fact this was not completely
true, as only the middle segment of the anterior
column was involved and the iliac wing and the
pubic angle were intact. Robert JUDET made the
drawing (Fig. 1.7 C) of this case, which is a pure
anterior wall fracture, the counterpart of the
 6 Introduction

Fig. I.6A-C Same case as in Fig. 1.5. A The intact lines of the anterior column.
B The broken lines of the posterior column. C Post-operative appearance at 3
C months
 Introduction 7

Fig. I. 7 A - C. The first anterior wall fracture described.

A Antero-posterior view, 1956. B The diagram I did in 1959.
C Robert JUDET'S diagram of the case (December 1959)

•
~I
/ ~~

B
C

posterior wall fracture, but it was classified for Pure transverse fractures 19 cases
several years as an anterior column fracture. To T-shaped fractures 3 cases
recognise and see intra-operatively a true anterior Associated transverse + anterior col- 3 cases
column fracture, we had to wait another 3 years. umn fractures
With these elements, I was able to propose in Anterior column fractures 5 cases
1961, in my thesis, a new classification of ace- Partial fractures (iliac wing extending 2 cases
tabular fractures, to be overlaid on the columnar to the roof)
concept of the architecture of the acetabulum. The 75 cases
75 cases were grouped as follows:
Posterior wall fractures 24 cases
Posterior column fractures 6 cases This classification included three newly described
Associated transverse + posterior 6 cases types of fracture: two elementary patterns (the pure
wall fractures + posterior dislocation posterior and anterior column fractures) and one
Associated transverse + posterior 7 cases association (the anterior column and posterior
wall fractures + central dislocation hemitransverse fracture).
8 Introduction

c D
 Introduction
E segment of the pelvic brim. From the back it was
Fig. I.SA-E. The first anterior column fracture operated upon.
A Pre-operative antero-posterior view. B Diagram of the antero-
posterior view. C Pre-operative obturator oblique view. D The
lines drawn in over the view in C; the tracing was done post-
operatively. E Post-operative view after an inadequate Kocher-
Langenbeck approach
Some elements of the final classification were
still missing, but anyway we felt already able to
claim that it was impossible to divide acetabular
fractures into strict categories: in reality, they form
one continuous, complete spectrum. The usual
separation between the posterior wall fractures, so
easy to identify, and the other types then appeared
illogical, as all the fractures are produced by the
same mechanism (see Chap. 2). A force applied at
the level of the femur is transmitted to the
acetabulum by the femoral head, the type and ex-
tent of the fracture depending on the localisation
of the impact, which is directly related to the posi-
tion of the femur at the time of trauma and, of
course, on the magnitude of the fracturing force.
However, the work was still not finished. As
time passed and the number of cases increased, the
classification was improved. On the afternoon of
30 October 1961, we encountered and operated
upon our first anterior column fracture. The pa-
tient had been admitted to the clinic the previous
9
evening, and due to the usual hurry preceding the
Boss's holidays, she did not have a complete set of
radiographs taken, only antero-posterior and ob-
turator-oblique views (Fig. 1.8). These do in fact
suffice to make the right diagnosis, but at that time,
after only 4 years (!) of work, we were still insuffi-
ciently experienced to make it. Robert JUDET
diagnosed a transverse and posterior wall fracture.
In fact, the fracture does look like a transverse one
at first sight (if the uninjured iIio-ischial line is
overlooked), and the posterior lip seems disturbed
on the antero-posterior but not on the obturator-
oblique view. A posterior approach was decided
upon.
I assisted Jean JUDET in this case. A typical and
complete Kocher-Langenbeck approach was per-
formed. The only thing we found posteriorly was a
discrete haematoma of the gluteus minimus (the
only sign that we were not on the wrong side!), and
the posterior column appeared totally uninjured. It
was finally the index finger introduced through the
greater sciatic notch that felt the fracture line
through the quadrilateral surface and the displaced
difficult to mobilize the fragment; we achieved a
partial reduction which was fixed with two screws
(Fig. 1.8 E) and a plaster cast was applied for 60
days. In spite of this error and the poor reduction,
it is difficult to describe the pleasure I had in hav-
ing finally identified, felt, and palpated, if not
seen, my first anterior column fracture.
Returning to the reading of the radiographs dur-
ing the operation, however, I must confess that our
mistakes were obvious: all the landmarks of the
posterior column (the posterior lip and the iIio-
ischial line) were intact, whereas all the landmarks
of the anterior column (the pelvic brim, the
anterior lip of the acetabulum) were involved. The
teardrop had kept its normal relationship with the
pelvic brim but was considerably displaced inwards
with respect to the iIio-ischialline (the opposite to
what occurs in posterior column fracture). The
roof was intact, but medial to it there was a clear
triangular surface corresponding to the area of
detachment of the anterior column.
From that day on, the identification of anterior
column fractures was clear, and in 1962 we pub-
lished a description of the pure fractures involving
the acetabular columns in the French Revue de
Chirurgie Orthopedique. This clear understanding
of anterior column fractures enabled us to distin-
guish them from pure anterior wall fractures, which
are partial fractures of the column, involving only
 Introduction

'1 'i1"'t t-.

7t . (O.f~
, ,

-----
.. --- --_.

c D
\
 Introduction
E
Fig.I.9A-E. Unrecognised both-column fracture operated on
by a Kocher-Langenbeck approach in November 1959. A Pre-
operative antero-posterior view. B Diagram of A. C Pre-
operative iliac wing oblique view. D Diagram of C. E Post-
operati ve appearance at 4 years
the medial part of the anterior column and leaving
the iliac wing and the angle of the pubis intact.
It is also true to say that this initial classification
did not satisfy us completely, as some complex
cases were still difficult to interpret, such as one
that I operated upon in November 1959 and was
classified as transverse and posterior wall and roof
fracture (Fig. 1.9). The temptation was great to
identify this as a transverse fracture which had
caused such a central dislocation of the hip. As in
a pure or associated transverse fracture, the case
was operated on through the Kocher-Langenbeck
approach. In fact, what we failed to recognise
was the fracture line through the iliac wing and,
most importantly, on the iliac-oblique view
(Fig.1.9C, D), the fact that the fragment obviously
carrying the whole roof was actually a large
anterior part of the iliac wing, i.e. the upper part of
the anterior column. Through the Kocher-
Langenbeck approach we achieved only an in-
complete reduction (Fig.1.9E).
Actually, it was the systematic posterior
approach to comparable cases, leading to further,
11
similar intra-operative findings, which finally
awoke us:
the posterior column was detached exactly as it
was in cases of pure posterior column fracture;
from any point of the fracture line dividing the
posterior column between the greater sciatic
notch and the posterior lip of the acetabulum,
a fracture line started which extended through
the iliac wing up to the iliac crest or the anterior
border of the pelvic bone, and it was the frag-
ment delineated above by this line which took
with it the whole roof.
The fracture line freeing the posterior column and
the fracture line through the iliac wing were suffi-
cient to free the anterior column component, which
nevertheless was in most cases further divided by
other, secondary fracture lines.
Thus we identified the "both-column" fractures
that detach from the posterior part of the iliac wing
(which remains undisturbed) the whole architecture
of the two columns and consequently, in two or
more pieces, the whole acetabular articular surface
(see Chap. 14).
The radiological features of these fractures were
published in La Presse Medicate in 1963. On the
three views all the typical landmarks are broken
and displaced. Not one is intact and in its normal
place, and on the obturator-oblique view the
pathognomonic sign of the both-column fracture is
visible, the "spur" sign, which corresponds to the
radiographic profile of the intact part of the wing
(see Chap. 14, Fig. 14.22). As a matter of fact, in
my thesis, five of these fractures were misinter-
preted and assigned to other categories.
Another wrong choice of approach led a step
further in the process of classification. For in-
stance, one patient operated upon in March 1963
was considered as presenting a both-column frac-
ture and, as was usual at that time, was approached
from the back, lying prone (see Chap. 30, case 33).
Through this approach the only thing we were able
to see was a horizontal fracture line, dividing
horizontally the ischial spine and involving the last
centimetre of the posterior wall. How and why did
we not notice that from studying the iliac-oblique
view (Fig. 1.10B, C)? Through the greater sciatic
notch we felt the displaced fragments of the
anterior column, and by elevating the external iliac
fossa and dividing the glutei tendons close to the
greater trochanter we were able to visualise a frac-
ture line which started from the apex of the
12
Fig.I.I0A-C Unrecognised anterior column and posterior
hemitransverse fracture operated on from the back in 1963
acetabular roof and reached the iliac crest
(Fig. 1.10A). Consequently, between these two frac-
ture lines, a segment of the acetabular articular sur-
face remained in place, and this is the critical dif-
ference between the both-column fracture, which
detaches the whole articular surface, and this case,
in which an anterior column fracture was associat-
ed with a posterior transverse fracture line dividing
the posterior column (Fig. 1.10A), which we called
for this reason a hemitransverse fracture.
Our classification was then completed and was
published in the December 1964 issue of the
American Journal of Bone and Joint Surgery. The
detailed description appeared in the original
French edition of the present book in 1974 and, in
the excellent English translation of my friend
Reginald ELSON, in the first English edition in
1981.
Finally, we found it practical to divide the
acetabular fractures into two categories:
the five elementary fracture patterns: posterior
wall, posterior column, anterior wall, anterior
column, transverse;
the five associated fractures, each of which
combines at least two of the five elementary
patterns.
We know of course that the possible fractures in-
volving the two columns are symmetrical, so strict-
ly speaking 13 varieties of fractures should be de-
Introduction
/'
I "-
I
/
/
/
/ /
./
I /
J /
/
/
/
c
B
(Chap. 30, case 33). A, B Diagram of the fracture lines. C
Diagram of the iliac-oblique view
scribed. However, to keep the classification as sim-
ple as possible, we have grouped three of them with
others, since they require the same surgical
approach:
anterior wall and posterior hemitransverse frac-
tures and anterior column and posterior hemi-
transverse fractures are included in the same
group under the name "anterior and posterior
hemitransverse fractures";
posterior column and anterior hemitransverse
fractures were grouped with the T-shaped frac-
tures, as they were very difficult to differentiate
before CT scanning and, incidentally, require
the same surgical approach (see Chap. 10,
Fig. 10.3);
associated anterior wall and anterior column
fractures are included in anterior wall fractures,
as their treatment does not differ.
We know also that every transition between these
ten categories of fracture may be encountered: the
possible fractures of the acetabulum form a com-
plete fan. Nevertheless, for the last 25 years we have
always succeeded in assigning a new case to one of
these ten categories, and the fantastic images ob-
tainable today from 3-D reconstruction confirm the
description we have given of these fractures, which
is now accepted nearly worldwide.
To be able to classify a fracture appeared essen-
tial and helped us a great deal in choosing ap-
propriately between the two approaches we were
Introduction
then using (the Kocher-Langenbeck and the Smith-
Petersen). These approaches were not in all cases
adequate, and sometimes surgery remained dif-
ficult because the approaches were deficient even
though we now had a better understanding of the
fracture lines. So we were back to square one and
had to start improving the approaches to our
favourite bone.
In March 1962, for one particular case of
anterior column fracture (see Figs. 8.8, 8.9), we
modified the routine anterior Smith-Petersen
approach. Always following the anterior part of the
iliac crest, we continued downwards along the belly
of the sartorius muscle for about 15 cm. Deeply, we
freed the outer border of this muscle, which was
elevated in continuity with the iliac muscle up to
the pelvic brim. This approach, later called the ilio-
femoral, allowed us to reach the upper two-thirds
of the anterior column no further down than the
ilio-pectineal eminence, even if we cut the psoas
tendon. This approach quickly proved insufficient
for the treatment of complex anterior column or
anterior wall fractures and is now strictly reserved
for treating high anterior column fractures when
the fragment is in one piece. It is true to say, how-
ever, that subsequent use of the Kocher-Langen-
beck and the ilio-femoral approach during the
same operation enabled us to reduce correctly some
difficult both-column fractures (see Figs. 14.23,
22.80).
So, our main problem in the early 1960s re-
mained to find a method of gaining complete access
to the anterior column. In my thesis (1961), I pro-
posed an approach to the anterior column based on
cadaver studies which was attractive but which we
had never strictly applied in humans. With some
modif~cations it was used to fix one particular case
(see Fig. 12.11) in 1965, and as we were opening the
inguinal canal completely, it was called the ilio-
inguinal approach. In this, the incision follows the
anterior two-thirds of the iliac crest and continues to
the mid-line, slightly concave from above. The in-
guinal canal is divided through its anterior and
posterior walls, and all the elements passing beneath
the inguinal ligament are put under three rubber
slings: the ilio-psoas and the femoral nerve, the ex-
ternal iliac vessels, and the spermatic cord or the
round ligament. This approach, if we accept work-
ing within three windows - lateral, medial and in-
ternal - gives complete access to the anterior col-
umn and beyond the pelvic brim to the quadrilateral
surface, but this is only possible if total division of
the ilio-psoas sheath has been completed.
13
With these two approaches the Kocher-Langen-
beck and ilio-inguinal, used in succession in some
complex cases, we were happy for several years and
performed some nice reconstructions, such as that
of the T-shaped fracture shown in Fig. 22.55. How-
ever, more and more frequently we had complex or
very complex cases referred to us, and in addition
we began to attempt delayed reconstruction weeks
or months after trauma, in cases in which conser-
vative methods had failed. For them, we felt the
need for simultaneous access to the two columns,
or the need to have simultaneous control over both
extremities of a transverse fracture line. So, in 1974,
we went back to the cadaver again and developed,
through an extra-long Smith-Petersen incision, a
very extensile approach to the bone. With the pa-
tient lying in the lateral position on the orthopaedic
table, a transcondylar Steinmann pin is inserted,
allowing traction with the knee flexed. The
approach comprises division of the two gluteal
muscles (medius and minimus) at both their ex-
tremities, i.e. their elevation from the outer aspect
of the iliac wing (together with the tensor fasciae
latae), and section of their tendons close to the
greater trochanter. Further on, the external rotators
are divided as they are through the Kocher-Langen-
beck approach. This approach allows access to the
whole external aspect of the bone, except for the
part of the anterior column situated below the ilio-
pectineal eminence, and to the joint surface after
capsulotomy along the acetabular lips. We called
this the Extended ilio-femoral approach.
This appeared to be still not enough in some
cases, and after many hesitations, when I was
operating one day with Monsieur JUDET, we decid-
ed to elevate the abdominal muscles from the iliac
crest in continuity with the iliacus from the iliac
fossa. Thus we were able to gain access to both
aspects of the wing and to the posterior half of the
pelvic brim. We do this now in about 40070 of the
cases treated through this extensile approach with-
out inconvenience.
Since 1975, we have been satisfied with these
three approaches. We use the ilio-inguinal as often
as we can, since it is followed by practically 0%
ectopic bone formation. We remain faithful to
the easy Kocher-Langenbeck which can be enlarged
in a tri-radiate Dana MEARS approach in case of
difficulties. The extended ilio-femoral approach is
suited to the most complex cases, but unfortuna-
tely it is associated, like the Kocher-Langenbeck,
with a significant rate of ectopic bone forma-
tion.
14
Conjointly with the development of the classifi-
cation and surgical approaches, we had to try to
solve other problems, some of which are still pend-
ing: the post-operative complications, the devices,
the special tools.
As regards post-operative complications, pro-
phylactic administration of antibiotics has reduced
the incidence of post-operative infections to a very
low level (below 1070), and anticoagulants have
reduced the massive pulmonary embolisms, but un-
fortunately nothing has yet proved totally reliable
in preventing the ectopic bone formations, al-
though indomethacin and radiotherapy seem very
promising.
Post-operative sciatic palsies were at first very
frequent: they occurred at an incidence of 18%. For
years, we "fought" intra-operatively against a nerve
stretched as tight as a violin string, up to the day
we had the idea of applying traction with the knee
flexed at approximately 40°, using a transcondylar
Steinmann pin. We still see about a 3.3% incidence
of palsies, probably due to the side effects of our
retractors, but they are less extensive than before.
As regards the devices: during the first years, we
used Vitallium screws exclusively, but the fixation
was not reliable enough and a plaster cast had to be
applied for 30-60 days. The first straight Vitallium
plate was applied in 1958 for a pure posterior col-
umn fracture, but we did not use it routinely
straightaway (see Figs. 6.3, 22.11).
From 1960 onwards, we practically gave up the
use of isolated screws in favour of the combination
of lag screws and plating. With one or two lag
screws precariously fixing the reduction, we had all
the time we needed to perfectly contour the buttress
plate applied along the column concerned, which
Fig. 1.11. Some of the different clamps I have been using
Introduction
brought about a perfect and very reliable fixation,
allowing immediate mobilisation and the use of
continuous passive motion.
The use of the ilio-inguinal approach led us fre-
quently to apply a plate along the superior aspect
of the pelvic brim. Straight plates appeared inade-
quate and Vitallium plates were difficult to bend on
the flat; furthermore, this induced needless stresses
into the plate. For this purpose, we designed a
curved plate specially adapted to the curvature of
the pelvic brim. Cadaver studies showed that there
were two main types of curvature of the pelvic
brim, with a radius of 88 mm and 108 mm, respec-
tively. That is why we have two series of curved
plates which are now available in stainless steel. We
routinely use 3.5-mm screws, but 4.5-mm and
6.5-mm stainless steel screws appear useful in some
cases when inserted:
between the two tables of the iliac wing (first
case in 1968); or
along the axis of the anterior column (per-
formed for the first time in 1978).
I will finish with a brief commentary about tools.
Each surgeon has his or her own habits and
favourite tools. We favour the curved chisel to act
as a lever, the ball spike instrument to push the
fragment without splitting it, the sciatic nerve
retractor, the AO pelvic reduction clamp which is,
however, too heavy. The AO distractor can be
useful even though we routinely use JUDET'S or-
thopaedic table.
However, a main problem, still incompletely
solved, is that of the clamps. Figure 1.11 shows the
results of quite a lot of work that I have put in this
field without complete success, because some
Fig. 1.12. Joel MATTA'S clamps
Introduction
clamps are useful in some cases, useless in others;
still, on the whole they have enabled me to perform
good reductions over many years.
Joel MATTA has designed some new clamps
which appear very useful (Fig. 1.12). A group of
surgeons particularly involved in acetabular frac-
ture surgery (Joel MATTA, Jeffrey MAST, Keith
MAYO, Tim POHLEMANN and myself) are working
15
on the clamp problem and hope in the near future
to be able to propose a set of clamps very conve-
nient for the reconstruction of the broken in-
nominate bone.
Such is the story which has brought to me the
most enjoyable working hours of my life as a
surgeon. Thank you, Monsieur JUDET.
1 Anatomy of the Acetabulum
ROUVIERE (1940) has given us a particularly clear
description of the acetabulum, which we have in no
way altered. The fractures with which we are con-
cerned comprise significant areas of the walls of
the acetabular articular surface together with the
bony segments which support them.
It is common to distinguish between the anterior
wall, posterior wall, and the superior wall or roof
of the acetabulum, but for reasons which will
become clear, we shall also distinguish the postero-
superior segment (which forms the bridge between
the roof and the posterior wall), and the postero-
inferior segment (which comprises the lower part
of the posterior wall together with the posterior
horn of the acetabular articular surface).
For the purposes of surgery, a clear understand-
ing of the underlying bone supporting the acetabu-
lum is more important than any descriptive anatomy
of the socket itself: it is by a perfect reconstruction
of the foundation that an adequate osteosynthesis
of the surface will be achieved. It is equally impor-
tant to learn the relationships of certain key anatom-
icallandmarks, for the correct realignment of these
will restore the innominate bone and enable us to fix
the parts effectively. The innominate bone is a com-
plex and irregular structure; although it is difficult
to learn, we cannot over-emphasise the necessity for
detailed anatomical clarity. For example, for a long
time the integrity of the concavity of the iliac wing
was not regarded as particularly important to
restore. We now appreciate how important is the
seemingly small detail that, in a patient lying recum-
bent, the anterior superior iliac spine should lie
perpendicularly above the posterior nutrient for-
amen of the iliac wing.
1.1 Columns of the Acetabulum
At first sight the acetabulum appears to be con-
tained within an arch. The limbs of the arch are
posterior (or ilio-ischial), and anterior (or ilio-
pubic). For a better understanding of the patholog-
ical anatomy of the fractures, we must alter
somewhat this basic concept of the architecture. It
is better to regard the acetabulum as being contain-
ed within the open arms of an inverted Y formed
by a posterior column, the ilio-ischial component,
and an anterior column which is much longer and
extends from the anterior end of the iliac crest to
the pubic symphysis; the upper end of the posterior
column is attached to the posterior aspect of the an-
terior column, a little above its mid-level (Fig. 1.1).
1.2 Posterior Column
We have called this the ilio-ischial column in order
to indicate its components: above, iliac, and below,
..
ischial. It is thick, the surfaces are easily recog-
nised, and it offers solid material for internal fixa-
tion. It is triangular in section, and presents inter-
nal, posterior and antero-Iateral surfaces.
(a) The internal surface comprises the quadrilateral
surface on the inner aspect of the body of the
ischium. This is continuous at the middle part
of its posterior margin with the inner surface of
the spine of the ischium, being orientated here
more obliquely internally and posteriorly.
(b) The posterior surface comprises (from above
downwards) a convex area forming part of the
posterior wall of the acetabulum, called the
retro-cotyloid surface, the sub-cotyloid groove,
in which runs the tendon of obturator externus,
and the ischial tuberosity. Note that the spine of
the ischium is situated at a slightly higher level
than the inferior margin of the posterior horn
of the acetabular articular surface.
(c) The antero-lateral surface includes above, the
posterior part of the acetabular articular sur-
face. This is bounded inferiorly by the project-
ing inferior horn which forms the edge of the
18 Anatomy of the Acetabulum

A B c

Fig. 1.1A-D. Columns of the acetabulum (dried bone). A Lateral aspect, B obturator-oblique
view, C iliac-oblique view, D scheme of endopelvic aspect. White, Anterior column; Red,
Posterior column; Blue, The beam uniting inferior ends of the columns (ischio-pubic ramus)

sub-cotyloid groove for the obturator externus 1.3.1 Iliac Segment

tendon. Below, the anterolateral surface is
formed by the body of the ischium.
The posterior border of the ilio-ischial column is
formed by the posterior edge of the innominate
bone with the greater and lesser sciatic notches
separated by the spine of the ischium.
1.3 Anterior Column
This, the ilio-pubic column, extends from the ante-
rior end of the iliac crest to the pubic symphysis. In
general form it is concave both anteriorly and
medially, its arc being bridged by the inguinalliga-
ment. One can distinguish from above to below
three segments: iliac, acetabular, and pubic.
This forms the anterior part of the iliac wing and
presents two surfaces.
(a) The pelvic surface is concave from above to
below and extends as far as the ilio-pectineal
line.
(b) The external surface is markedly roughened and
forms a large anterior part of the gluteal sur-
face of the ilium including the anterior pillar
which extends upwards to the gluteus medius
tubercle.
Its anterior border is marked by the anterior
superior and anterior inferior iliac spines separated
by the interspinous notch. The inferior portion of
the anterior inferior iliac spine is immediately adja-
cent to the acetabular margin.
Anterior Colum 19

1.3.2 Acetabular Segment 1.3.3 Pubic Segment

This is triangular prismatic in shape and presents This is the superior pubic ramus and constitutes the
three surfaces. slenderest piece of the column as well as its most
forward and medial part. It also is triangular in sec-
(a) The postero-lateral surface supports the
tion and presents three surfaces.
anterior articular segment of the acetabulum
and the front part of the cotyloid fossa. Note (a) The antero-superior surface affords insertion
that the anterior horn is located about 1 cm for the pectineus muscle. It is distinctly spiral in
above the level of the upper border of the ob- configuration and in order to apply a plate for
turator foramen. internal fixation, it is always necessary to twist
(b) The internal surface is generally concave and is this to fit the shape of the segment. The area is
formed by the anterior part of the quadrilateral bounded posteriorly by the medial part of the
surface. It extends as far forward as the ob- anatomical brim of the true pelvis. The most
turator canal and is limited superiorly and medial part of the surface is horizontally
anteriorly by the ilio-pectineal line. disposed and represents the upper aspect of the
(c) The antero-superior surface presents from body of the pubis, giving insertion to the rectus
above downwards, the gutter of the ilio-psoas abdominis muscle.
tendon just below the anterior inferior iliac (b) The internal surface is generally concave poste-
spine, and the ilio-pectineal eminence. At the riorly and above, it faces almost medially.
level of the latter the bone of this surface forms Following its curve, it comes to be continuous
the anterior lamella of the anterior wall of the with the pelvic surface of the body of the pubis
acetabulum, and is roughly parallel to the ante- which looks postero-superiorly.
rior acetabular articular surface; at that level (c) The inferior surface forms the bony roof of the
the bone is 6 - 10 mm thick. It is limited inter- obturator canal and faces mainly inferiorly and
nally by the ilio-pectineal line, which is always increasingly anteriorly as it approaches the
interrupted in fractures of the anterior wall of body of the pubis. Its anterior limit, which it
the acetabulum. It may be useful to remember has in common with the antero-superior sur-
that the inferior limit of the ilio-pectineal face, ends medially at the pubic spine, where the
eminence is situated at the same horizontal level inguinal ligament has its lower insertion.
as the anterior horn of the acetabulum, or a lit-
The anterior column is complicated and the main
tle higher (Fig. 1.2).
guide when assessing its continuity is the ilio-pec-
tineal line which constitutes an unbroken arc and
which appears to reinforce the antero-superior part
of the acetabulum; as stated already, fractures of
this line always indicate a fracture of the anterior
column.
Iliac spine Outline of anterior wall Both anterior and posterior columns unite a lit-
of acetabulum
I
/ tle above the level of the mid-point of the anterior
I
I
column and form an angle of approximately 60°.
I
I Within this angle is located the acetabulum itself.
I
I
I
The summit of the angle is filled with a fillet of
I
I compact bone which constitutes the roof of the
\
\
,_/
I
I
,/
;
'
acetabulum and forms the keystone of the arch. So,
I this anatomical roof corresponds to a segment of
I
I articular surface which sub tends to an angle of 45°
I
I to 60° and is located between the anterior inferior
I
\ iliac spine and the ilio-ischial notch of the ace-
\-,
---- ; ....
tabular margin (not always clearly visible) poste-
riorly. Medially, the anatomical roof does not reach
the edge of the cotyloid fossa; on the contrary, it is
Fig. 1.2. Landmarks of the middle segment of the anterior col- joined to the superior border of the cotyloid fossa
umn by a distinct and thinner plate of compact bone
20
which is often easily visible in a coronal section of
the area and on the antero-posterior view.
Joined together in the manner described, the
two columns are linked with the auricular surface
of the sacro-iliac joint by the sciatic buttress de-
scribed by ROUVIERE (1940).
1.4 Structure of the Innominate Bone
in Relation to Load-Bearing
The shape and internal structure of the innominate
bone is related to the forces which must be trans-
ferred from the head of the femur to the vertebral
column and vice versa.
ROUVIERE (1940) describes how these forces
pass firstly through thick condensed areas of bone
which arise tangentially from the auricular surface
of the ilium (i.e. its articular surface at the sac-
roiliac joint) and pass adjacent to the upper border
of the greater sciatic notch. He calls the particular
condensation of bone in this region the sciatic but-
tress (Fig. 1.3).
From this point two systems of bony trabeculae
arise: the first is relevant to the erect posture and
comprises elements related to the postero-superior
part of the acetabulum and which continue thence
into the anterior column; and the second is of func-
tional significance during sitting, in which the bony
thickenings pass almost vertically as far as the
ischial tuberosity (LATARGET, ROUVIERE, DEL-
MAS, 1940) and involve essentially the posterior
column.
Anatomy of the Acetabulum
CAMPANACCI (1967), discussing fractures of the
pelvis, presented a beautiful analysis of our under-
standing of the architecture of the innominate
bone. It confirmed the above account and cor-
responded exactly with our radiological studies of
the various laminae and thickenings of the bone.
He distinguished three fundamental trabecular
systems within the bone structure: sacro-acetab-
ular, sacro-pubic, and sacro-ischial. We can relate
the trabecular architecture in a logical fashion to
the anterior and posterior columns (Fig. 1.3).
The posterior column contains the lower or
ischial members of the sacro-acetabular group of
thickenings, together with the sacro-ischial com-
ponents. The anterior column corresponds to the
remaining upper members of the sacro-acetabular
group and the sacro-pubic components, which join
another system in the ilium - the ilio-acetabular
group.
The massive thickening of bone trabeculae in
the region of the sciatic buttress explains why this
region is only exceptionally involved in fractures of
the acetabulum; fractures involving either column
are confined below this particularly strong area.
1.5 Vascular Supply
The blood supply is rich, largely on account of the
broad areas of muscular attachment. A study made
in 1933 by J. LAPART was repeated in 1960 and
1961 by pupils of Professor SALMON (L. LOUIS
and M. BERGOUIN). It is from these authors that
Fig. 1.3. Internal structure of the innominate bone
(CAMPANACCI 1967). Trabeculae of the anterior
column: 1 sacro-acetabular, 4 sacro-pubic, 5 ilio-
acetabular. Trabeculae of the posterior column:
2 sacro-acetabular, 3 sacro-ischial. There is overlap
between these groups
Vascular Supply
External antenor ~oJ:?"7f-~Ic---++-- Ilia-lumbar
artery
Iliac artery
A B
Fig.l.4A,B. Vascular supply of the innominate bone. A Internal aspect, B external aspect (LOUIS and BERGOUIN 1960)
we have extracted the following short account
(Fig. 1.4).
1.5.1 Internal Surface
The largest nutrient foramina, one artery of which
we are most frequently obliged to divide when us-
ing the iIio-inguinal approach to fractures of the
acetabulum, is situated in the iliac fossa 1 cm in
front of the auricular surface of the sacro-iliac
joint and 1 cm above the iIio-pectineal line. It
receives a branch of the iIio-lumbar artery.
Other smaller nutrient foramina are located:
below the iIio-pectineal line in front of the
greater sciatic notch, and also in the roof of the
obturator canal where branches of the obtura-
tory artery supply the bone;
on the internal surface of the ramus of the
ischium, supplied by the internal pudendal
artery;
above the anterior inferior iliac spine and along
the iliac crest, by branches of the external
anterior iliac artery which anastomose with
branches of the fourth lumbar artery;
and posteriorly, by branches of the ilio-Iumbar
artery which penetrate the rough surface behind
the auricular articular surface of the sacro-iliac
joint.
21
-c~_ ~-:_:~_~~-~~
~-~
'-"-"
Artery of the roof
of the acetabulum
(follows reflected head of
rectus femons)
Pubic branch
Artery to sCiatic nerve of obturator artery
Acetabular branch of
obturator artery
,-,,"--'i"----+- Obturator artery
1.5.2 External Surface
There is one particularly large nutrient foramina in
the middle of the gluteal area of the wing of the
ilium, just in front of the anterior gluteal line. It
receives a branch of the superior gluteal artery.
Other supplies are:
multiple nutrient vessels around the margins of
the acetabulum which form a complete vascular
circle, from the obturator artery, the inferior
gluteal artery and other local anastomotic
branches including the artery of the roof, a con-
stant branch of the superior gluteal artery;
the cotyloid fossa itself which is perforated by a
number of small vessels from the acetabular
branch of the obturator artery;
the body of the pubis which is supplied by
branches of the obturator artery;
and the region of the sciatic buttress which
receives several branches of the superior gluteal
artery.
The intra-osseous distribution of the arterial sup-
ply has also been studied by the same authors.
From the periphery of the innominate bone, small
arteries are directed towards the central areas. They
are approximately parallel and form a palisade
which can be observed on arteriography.
22
The two principal nutrient arteries entering the
gluteal surface of the ilium and the surface of the
iliac fossa respectively are located at approximately
the same level. They divide in the spongy bone into
many vessels which anastomose and form a star-
shaped configuration.
1.5.3 Acetabulum
Around the periphery of the acetabulum there ex-
ists a vascular palisade of radiating vessels (the
peri-acetabular circle), while at the level of the
cotyloid fossa branches from the obturator artery
develop another star-shaped group of anastomoses.
Anatomy of the Acetabulum
We have noticed, when treating fractures of the
acetabulum involving the posterior wall, often
associated with a dislocation of the head of the
femur, that we do not meet branches of the periace-
tabular circle. For practical purposes it seems that
in all fractures involving the posterior part of the
acetabulum, these arteries are torn and subsequent-
ly thrombose. On the contrary, when there is a cen-
tral dislocation of the hip and the posterior wall of
the acetabulum remains intact, detachment of the
soft tissues of the area is accompanied by bleeding
from these vessels.
Despite the rich blood supply with so many
anastomoses, extensive periosteal stripping of frac-
ture fragments can lead to avascular necrosis and
must be restricted to a minimum.
2 Mechanics of Acetabular Fractures

Fractures of the acetabulum occur as a result of
force acting between this part and the head of the
femur, the last link of a chain of transmission from
the greater trochanter, the knee or the foot. Alter-
natively, a blow on the back of the pelvis can have
the same effect.
In general, there are four points of application
for the energy necessary to produce a fracture of
the acetabulum: the greater trochanter, the knee
(when this joint is in the flexed posture), the foot
(the knee being extended), and the posterior aspect
of the pelvis. Before considering each of these, cer-
tain general aspects should be understood.
The resultant force acting at any time between
the head and the acetabulum may be regarded as
passing through the centre of the head, which is,
for practical purposes, identical with that of the
centre of curvature of the acetabulum.
It is seen in Fig. 2.1 that a force F applied at a
point A on the femur may be resolved into two
components: F' is that component of force trans-
mitted to the acetabulum and is represented by a
projection from the point of application of force F
and passing through the centre of curvature of the
head. The line of this force passes through the
point of impact I on the acetabulum. The other
component, j, is perpendicular to F' and has the ef-
fect of tending to rotate the upper shaft of the
femur in the direction shown.
In all cases where a fracture has occurred, the
direction of force F must have been near to that of
the neck axis AC, therefore F' must have
approached it in magnitude; correspondingly, force
f must have been negligible.
If the hip were a perfect ball and socket joint,
the two opposing surfaces meeting exactly and be-
ing made of homogeneous material, the forces
applied to the femur and transmitted by the head
to the acetabulum would tend to be distributed
evenly to the whole acetabular articular surface
(Fig. 2.1 A). However, the interposition between the
two articular surfaces of double compliant carti-
laginous layers considerably modifies the transmis-
sion of the forces. Of the various theoretical anal-
yses, we favour the following:
Referring to Fig. 2.1 B, it can be seen that the
force F' is applied to the acetabulum at point I and
is attenuated by the compound cartilage. It remains
maximal at point I but becomes distributed over a
circular zone seen shaded in section. The mag-
nitude of the force at each point of the circular area
may be represented vectorially by the arrows on
either side of F' and these form with respect to
magnitude an elliptical area. The sum of these im-
aginary forces amounts to F'. The fracture line and
subsequent displacement of the fragments can be
related to the force pattern displayed in this
diagram. The position and magnitude of the force
pattern is determined at the moment of impact by
the attitude of the femur and the direction, mag-
nitude and point of application of the force itself.
The different strain-rate sensitivities of the tissues
involved render an exact analysis of the situation
very difficult, but clearly the rate of application of
the force is highly significant.
F'
Fig.2.lA,B. Analysis of force acting on the acetabulum
through the greater trochanter
24 Mechanics of Acetabular Fractures

2.1 Force Applied to the Greater Trochanter buttress afforded by the anterior column and
in the Axis of the Femoral Neck involves this part predominantly.
(c) If the external rotation is even more extreme at
40° to 50°, the force is exerted entirely on the
The point of impact in the acetabulum is deter-
anterior wall of the acetabulum.
mined by the degree of abduction or rotation of the (d) With internal rotation of varying degrees, the
femur. For practical purposes, flexion of the femur central zone of the acetabulum is involved and
plays little part.
the anterior column progressively less so. At
about 20° of internal rotation, the zone of im-
pact is to some extent shared by both columns.
2.1.1 Neutral Abduction-Adduction
The lesion produced will be variable and, de-
pending upon the exact description of the force,
Throughout the range of external/internal rotation
it may comprise a simple transverse fracture of
of the hip the site of impact lies on a line of latitude
the acetabulum, a T-shaped fracture, or in ex-
of 30° to 40°, the axis of which is determined by the
treme cases a fracture of both columns.
inclination of the neck of the femur. Referring to
(e) In extreme internal rotation approaching 50°,
Fig. 2.2: the point of impact involves the junction be-
(a) In neutral rotation, because of the normal tween the posterior articular horn and the
anteversion of the femoral neck, a blow over the cotyloid fossa. This area is supported by the
trochanter is transmitted approximately to the posterior column, which is likely to be fractured
cotyloid border of the anterior horn of the ace- and is always associated with a complete or in-
tabular articular surface. The diagram shows complete transverse component.
that the fracture will involve the centre of the
acetabulum and the anterior column, produc-
ing an anterior column + posterior hemitrans- 2.1.2 Abduction and Adduction
verse fracture.
(b) In external rotation, when this approaches For any given degree of rotation, the main point of
about 25°, the impact operates against the bony impact in the acetabulum will be altered according

(b) 25° external rotatio n

anterior column fract ure \
\
\
+ (c) Maximum 50° external rotation
I .... Ant erior wall fract ure
\ 1
1
1
(al Neutral rotat ion
.... anterior co lumn and ..........
posterior hemi-transverse fracture ............
......
......
.....
......

(d) 20° internal rotation

.... Transverse or I
both-cofum n fractures I
... \ I
"' ... \ I
-- ... \ I

-- ---
---~~

(e) Maximum 50° internal rotation _ Fig. 2.2. Hor izontal section
.... Transverse and posterior through hip joint showing
wall fract ure
sites of application of force
as influenced by internal and
external rotation
Force Applied to the Flexed Knee in the Axis of the Femoral Shaft 25

CD ,
,,
@ ,
(c) 50° abduction of neck ,,
--. ,,

...
, , \
, ,,
\ '
\ '.
" ,, \\ \\
,, ,
CD ,, '.
(a) Neutral abduction, ,~~' . JI ,, '''
, ''
",,;,'
,>
: ,'
I \
0 1 neck o r 60°
abduction of lower limb , .,,'~"' --" .. ", ,\\
20 ~dduction of
0
,..- / " '\
~~::... / / / / \:~'"
(b) 30 0 maximum lower limb
Neutral abduction of
, ' ' ' '.,./'
' ",",\
.3 ~ .. ) ' . . /' /
,, \ ,

\ \ ..... , ......
.. .,
of neck o r 30 0
abduction of lower limb lower limb (b) 50° abduction /
,, ,
, ,
- transverse fracture , \
,
and central d islocation \
\
'- -- - '
. , "'-{~:~)
\ ;,'

(c) 15 0 abduction':·-:.: ':;"! '--:::>:1--- (d) 25 0 adduction

.... Iract ure of posterior column - -+ posterior dislocation
A
(a) Neutral abd uction and fracture of lip
-+ posterior dislocation
of posterior wall
Fig. 2.4. Horizontal section through the hip Joint showing force C> ' and fracture of posterior
acting through the knee wall

to any abduction or adduction posture at the mo-
ment of injury.
If we take as an example 20° of internal rota-
tion, the main line of impact corresponding to dif-
ferent degrees of abduction will approximate to a
coronal section through the centre of the hip joint.
Referring to Fig. 2.3:
(a) In neutral abduction-adduction the centre of
the area of impact occurs at the inner margin of
the roof of the acetabulum. A transverse frac-
ture at this level or, alternatively, aT-shaped
fracture or a both-column fracture can all
result. The force considered here results from a
blow over the trochanter but a similar pattern
would prevail if the force were transferred along
the axis of the shaft of the femur, the hip being
abducted to 60° at the moment of impact (see
Fig. 2.6).
(b) With significant degrees of adduction, the im-
pact affects increasingly the roof of the ace-
tabulum and a transverse fracture through this
part is the result.
(c) In abduction, a much more common state of
affairs, the main point of impact shifts pro-
gressively inferiorly, leading to transverse frac-
tures below the articular margin of the roof and
becoming progressively horizontally disposed.
In summary, it can be seen that any combination of
rotation and abduction-adduction can occur and
correspondingly, an infinite number of fracture
patterns can result. Nevertheless, each can be ra-
tionalised according to the above analysis.
2.2 Force Applied to the Flexed Knee
in the Axis of the Femoral Shaft
In principle, if the hip is flexed at 90° and a blow
is sustained on the knee, provided the neck of the
femur is not fractured, the acetabulum will break if
the force is sufficient. In Fig. 2.4 the component of
the fracture F which is directed towards the centre
of rotation of the femoral head is responsible for
the direction and magnitude of the zone of impact
sustained by the acetabular surface. The degree of
rotation of the femur does not play a significant
part but combinations of flexion-extension and ab-
duction-adduction together produce another range
of possibilities.
26
2.2.1 Hip Joint Flexed 90°
From Fig. 2.5 a it can be seen that the main impact
received by the acetabulum is in the horizontal
plane corresponding to an arc, part of a great circle.
Referring also to Fig. 2.4:
(a) In neutral abduction, the posterior wall of the
acetabulum receives the impact and a pure frac-
ture of this part may result.
(b) In maximum abduction approaching 50°, it is
the postero-medial segment which is damaged,
resulting in posterior column fractures often as-
sociated with transverse fractures.
(c) In about 15° abduction most of the force is
exerted on the posterior column, from which a
pure fracture results.
In these situations with the hip flexed at 90°, it is
the posterior column which is overwhelmingly in-
volved, with the possible participation of the floor
of the acetabulum in maximum abduction; the
anterior column can be involved only secondarily
by a transverse-type fracture line.
(d) If the femur is adducted, the main site of im-
pact approaches the posterior margin of the
acetabulum and it is under these circumstances
that posterior dislocation with or without frac-
tures of the margin of the acetabulum is pro-
duced.
2.2.2 Different Degrees of Hip Flexion
Referring to Fig. 2.5:
(a) With increasing flexion (Fig.2.5b), the lowest
part of the posterior wall becomes exposed to
force transmitted along the shaft of the femur.
(b) 115 0 flexion
/ / -+ fracture of posterior horn
F'1
'-.. (cl 60 0 flexion
- posterior-superior fracture
of posterior wall
Fig. 2.5. External aspect of hip showing sites of application of
force acting through the knee with the hip flexed
Mechanics of Acetabular Fractures
In extreme flexion, the lowest part of the poste-
rior wall can be fractured, the line extending to
the upper pole of the ischial tuberosity.
(b) As flexion of the hip becomes less than 90°
(Fig. 2.5 c), it is the postero-superior segment of
the acetabulum that is increasingly prejudiced.
This is precisely the situation seen in dashboard
injuries when the passenger in a car having his
legs crossed is flung forward at the moment of
collision, striking his knee. A dislocation with
or without fracture of the acetabular margin
may be produced. Alternatively, if the hip is ab-
ducted and in less than 90° flexion, a posterior
fracture, perhaps with posterior dislocation of
the hip, may be associated with a transverse
fracture.
As in previous examples, the spectrum of possi-
bilities combining abduction-adduction and flex-
ion-extension is infinite but at the same time
capable of definition and predictable in effect.
2.3 Force Applied to Foot
with Knee Extended
2.3.1 Hip Flexed
These are the circumstances operating when, dur-
ing a frontal collision in a car, force is transmitted
through the brake pedal to the foot and thence
through the extended knee and the hip joint. If the
hip is neutral or tending to lie in abduction at the
time, the postero-superior zone of the acetabulum
is involved and the upper wall fracture is commonly
associated with a transverse component. The pat-
tern is similar of course to that prevailing when the
blow is received on the flexed knee, with the hip in
the same amount of flexion.
2.3.2 Hip Extended
This occurs typically when a subject falling from a
height lands on his feet in a vertical posture. If the
limb is a little abducted, the main point of impact
with respect to the acetabulum is on the inner seg-
ment of the roof; the degree of rotation does not in-
fluence the result, a trans-tectal transverse fracture.
Clinical Correlation 27

2.4 Force Applied to Lumbo-sacral Region
If in the stooping posture the hip is flexed at 90°
and a blow is received on the back, it is easy to see
that the posterior wall of the acetabulum can be
fractured. Similarly, with different degrees of flex-
ion precisely the same range of fractures can be
seen as applied to circumstances in which the force
is introduced along the femur. The injury has oc-
curred in miners stooping at the time of a roof fall;
we have six examples, one in a man stooping at the
bottom of a trench who was injured by a fall of
stone and the others from similar accidents.
2.5 Comment
From the above considerations two points should
be emphasised: Firstly, the force which produces a
fracture of the acetabulum will be responsible for
the degree and direction of displacement of the
fragments and also for the direction of a disloca-
tion of the head of the femur, should this occur.
Secondly, we have explained how, from a de-
scription of the force, the site of potential fracture
of the acetabulum can be deduced. It does not mat-
ter how this force is transmitted to the acetabulum,
as is demonstrated in Fig. 2.6. In position 1 (flexion
90°, abduction 0° and internal rotation 20°) the
angle of abduction with respect to the neck of the
YI
femur is at 45°. In position 2 (flexion 90°, abduc-
tion 45°, internal rotation 0°) the angle of abduc-
tion with respect to the shaft of the femur approx-
imates to 45°. It matters not whether the force is
applied to the trochanter in position 1 or to the
knee in position 2; the site of impact with reference
to the acetabulum will be similar in both cases.
2.6 Clinical Correlation
In our series of 940 fractures of the acetabulum, we
know the cause of the accident in 918 cases
(Table 2.1). From Table 2.2 it is seen that it has been
impossible, despite careful enquiry, to establish the
site of impact in 560/0 (530) of the accidents ana-
lysed. The descriptions refer to cases in which the
site of impact and mechanism of injury was fairly
certain, and could be related to the type of fracture
which resulted.
2.6.1 Blow on Knee or Dashboard Injuries
Posterior wall fractures of all types 76
Posterior column fractures 3
Transverse fractures 4
T-shaped fractures 2
Associated transverse and posterior wall 18
fractures
Posterior column and posterior wall 5
fractures
108

Somewhat uncertain with respect to our theoretical

analysis are the four transverse fractures and the
two T-shaped fractures.

Table 2.1

Activity at time of accident Number

Car occupant 621

Motorcyclist 81
Pedestrian 119
Fall from a height 66
Simple fall 25
Blow on the back 6
Unknown 22
Fig. 2.6. Diagram showing how the same force can act on the
acetabulum through the knee when the hip is abducted or Total 940
through the greater trochanter with the hip in neutral abduction
28 Mechanics of Acetabular Fractures

Table 2.2

Activity at time of accident Blow on knee Blow on trochanter Blow on foot Blow on back Antero-post. Unknown
compression

Car occupant 103 67 14 437

Motorcyclist 5 39 37
Pedestrian 103 4 11
Fall from a height 45 2 19
Simple fall 21 4
Mining accident or similar 5
Unknown 22

Total 108 275 16 6 5 530

We feel that the configuration of the fractures are interpreting the site of impact must of course be
explicable only if the blow occurred on the knee recognised.
while the hip was approximately fully extended and
being in a small degree of abduction. We believe
that this posture and the site of the blow apply in
each of the accidents mentioned. 2.6.3 Blow Under Foot

Various types of posterior fracture 10

2.6.2 Blow on Greater Trochanter
Posterior wall fractures
Posterior column fractures
Transverse fractures
Anterior wall fractures
Anterior column fractures
T-shaped fractures
Associated transverse and posterior
fractures
Associated posterior column and posterior
wall fractures
Associated anterior and posterior
hemitransverse fractures
Both-column fractures
It is remarkable that 216 of these 275 fractures were
both-column fractures, pure anterior fractures, or
mixed anterior and posterior hemitransverse frac-
tures. The only difficult examples to explain me-
chanically comprise the four posterior wall frac-
tures, the four posterior column fractures, and the
five associated posterior column and posterior wall
fractures. Even in full internal rotation, it is dif-
ficult to explain how such a posteriorly directed
force could be transmitted. The possibility of mis-
Associated transverse and posterior wall 5
fractures
4 Epiphyseal displacement of the
4 posterior column
11 16
14 All of these fractures corresponded to our mechan-
22 ical theory.
18
17
2.6.4 Blow on Sacro-i1iac Region
5
Six blows on the back resulted in two posterior
49
wall, one anterior column, one transverse, one T-
shaped and one associated transverse and posterior
131 wall fracture. The anterior column fracture is im-
possible to explain. Posterior wall or column frac-
275
tures are easy to understand; transverse, T-shaped
or transverse associated posterior wall fractures can
be explained by the hip's being in some abduction
and slight flexion at the time of injury.
2.6.5 Antero-posterior Compression
Five accidents of this type appear to have been the
cause of one anterior column fracture, three
transverse fractures and one T-shaped fracture.
3 Radiology of the Normal Acetabulum
Since the first edition of this book, the use of CT
scanning has become widespread and is now almost
routine. At present only a few centres have access to
3-D CT reconstruction; when this is available
everywhere, the understanding of acetabular frac-
tures will be much easier. With good 3-D analysis,
one just needs to know the classification to be able
to link the case being treated to one of the fracture
types described. It made us very happy to see that
the 3-D images fully confirm the descriptions we
gave of the different types of acetabular fracture.
In the meantime, while waiting for the universal
advent of 3-D, nearly all centres have a 2-D CT
scanner, and it must be stressed here that, whatever
the advantages of a 2-D CT image, it should never
be read in isolation. Combined interpretation of
the standard radiographic views and the 2-D CT
images is essential and can provide all the necessary
pre-operative information about the case under
treatment.
This chapter wil cover first standard radiogra-
phy of the innominate bone, then CT scanning and
other techniques.
3.1 Standard Radiography
The reader studying this section is advised to have
at hand a dried innominate bone or pelvis. In order
to interpret accurately the radiological features of
a fractured acetabulum, it is necessary to pursue in
a disciplined fashion an orderly study of each stan-
dard view, then to put these together in three
dimensions, comparing when necessary with the
appearance of the opposite side, and finally to
relate the findings to the radiological features of
the columns of the acetabulum. Each landmark
must be followed in turn, and scrutinised for possi-
ble traversing fracture lines; it is important to note
whether at the site of such a break, there is frank
displacement or not. Treatment should not be com-
menced until a full understanding is achieved and,
most important, during the consequent well-
planned operation the surgeon will know what to
expect and not have to try to analyse the fracture
during the procedure. A systematic and detailed
study of the pre-operative radiographs with the
unashamed aid of a dried bone will enable this goal
to be achieved. As experience grows, more rapid ap-
praisal becomes possible.
During our studies we believe that we have
assessed and tried to use most of the special radio-
graphs which have been advocated. The orientation
of the acetabulum is such that the usual antero-
posterior and lateral radiographic views do not give
sufficient information and we have attempted to
find others which would rectify the deficit.
The elementary fact should be recalled that a ra-
diological line is produced by rays tangential to a
bony surface or crossing a border, and represents,
truly, a line of tangency; it must not be interpreted
as a surface, of which it is only an "optical" cut
(Fig. 3.1). A bone surface yields the same radio-
logical line provided the angle of incidence of the
ray remains constant. Provided the slight inevitable
variation in the position of a patient at the time of
radiological examination is within reasonable
limits, the appearance of most landmarks will be
constant. It is then possible to define a zone of a
surface which will offer a constant radiological
outline (Fig. 3.8A). Outside the limits of this zone,
the configuration of the adjacent surfaces will offer
a markedly altered radiological image. Neverthe-
less, it can be difficult to find pairs of films which
can be accurately superimposed, and yet on casual
inspection, appear similar.
We have come to the firm conclusion that in
order to study the masses of bone which limit and
enclose the acetabulum and which may be involved
in fractures thereof, four radiographic views suf-
fice:
a standard antero-posterior view of the whole
pelvis,
30 Radiology of the Normal Acetabulum

lila-Ischial Line of X-ray tangency lila-Ischial

line producing the Ilia-Ischial line line

A Sagittal diagram B Coronal diagram

Fig. 3.2..Innominate bone seen from above, showing the ob-
Fig.3.1A,B. Generation of a radiological line here the ilio- turator nng at an angle of approximately 90 0 to the main part
ischial line. A Sagittal, B coronal view. I f the ~osition of the of the iliac wing
pelvis varies even a little, the line of tangency will be a little dif-
ferent, but it is always situated along the ilio-ischial surface and
will appear with practically the same configuration

a standard antero-posterior view centred on the ner and outer surfaces of the bony ring enclosing
affected hip, and the obturator foramen (the obturator ring). We
two oblique views taken accurately at 45 o. have called this the iliac-oblique view. The other
displays perfectly the outline of the obturator ring
The antero-posterior view of the pelvis must be
around the obturator foramen, while showing in
taken on a large plate (36 x 43 cm) and is centred on
section the iliac wing and superimposing the anteri-
the pubic symphysis. The antero-posterior view of
or and posterior iliac spines; this we have called the
the injured hip is centred on the head of the femur
and must include the whole of the innominate obturator-oblique view.
The two views are obtained by rolling the injured
bone.
patient carefully from one side to the other and
Of all the possible oblique views, the two at 45 0
supporting him during exposure of the films on
to the coronal plane are by far the best. These views
suitable cushions. The transverse axis of the pelvis,
are justified by the fact that the obturator ring is at
45 0 to the coronal plane and is also approximately which must in each case be at 45 0 relative to the X-
ray table, is easy to assess.
perpendicular to the plane of the anterior two-
For the iliac-oblique view, it is the uninjured hip
thirds of the iliac wing. One or the other is seen in
which is elevated, the injured part resting on the
profile on each oblique view (Fig. 3.2).
table. For the obturator-oblique view, the injured
The obturator-oblique view is very close to that
hip is raised. For both oblique views the centre
described by WALLER (1955) and which approx-
beam of the vertically disposed tube is aimed above
imated that described by TEUFEL (1930). It ap-
the femoral head; under fluoroscopy a good centr-
proaches that used by D'AuBIGNE (1968) called the
ing is the level of the superior border of the greater
postero-internal view. These authors used a beam
coincident at 40 0 (and directed superiorly at 10 0 in sciatic notch.
It is essential that on each view the whole of the
the case of WALLER). Oblique views have received
iliac bone should be seen both vertically and trans-
many names, a state of affairs which has produced
versely.
confusion.
In the obturator-oblique view, the perfectly ex-
?ne of the oblique views exposes perfectly the posed picture superimposes accurately the anterior
entIre surface of the iliac wing and projects the in- and posterior iliac spines, the iliac wing is seen in
Standard Radiography
section as narrow as possible, and correspondingly
the outline of the obturator foramen is as large as
possible, being parallel with the plane of the film.
If the iliac wing appears widened, it is due to the
rotation of the patient being insufficient; unless at-
tempts are made to correct this, accurate appraisal
will be difficult.
In a correctly positioned iliac-oblique view, it is
the iliac wing which is seen widely spread out and
the bony outline of the obturator ring as thin as
possible in section, the obturator foramen being in-
visible.
The two oblique views can be taken in the vast
majority of cases, even following recent injuries,
without recourse to general anaesthesia. Should the
presence of other injuries make radiography dif-
ficult, there are two possibilities: the X-ray exami-
nation can be conducted under general anaesthesia,
or, without moving the patient, the tube can be
tilted to 45 0 , the injured person resting in neutral
decubitus. Unfortunately the latter technique re-
sults in varying distortions according to the build
of the patient, this applying especially to the iliac
bone. We advise that following a study of these
distorted views, which will give some idea of the
fracture configurations, the radiographs should be
repeated at the time of operation under anaesthesia
in the approved fashion. One further possibility to-
day is to obtain the 45 0 oblique views from CT
scanning. They are not as good as standard radio-
graphs but present no distortion at all.
31
3.1.1 Antero-posterior Radiograph of Pelvis
This must always be taken and may reveal:
the uncommon bilateral acetabular fracture
(Fig. 3.3) which has occurred only 12 times in
940 cases:
another fracture of the pelvic ring independent
of the fractured acetabulum and passing
through perhaps the obturator ring, the iliac
wing or even the sacrum;
dislocation through one or more of the joints of
the pelvic ring.
3.1.2 Antero-posterior Radiograph
of Acetabulum
Because of the more laterally disposed central ray
there is some modification in this view (Fig. 3.4A)
compared with the previous one. The difference,
however, is slight and perhaps one could dispense
with it.
Six fundamental radiological landmarks of the
acetabulum are seen in this view:
borders of the anterior and posterior walls of
the acetabulum;
roof;
teardrop;
ilio-ischial line;
pelvic inlet (brim of the true pelvis): the innomi-
nate or ilio-pectineal line.
Fig. 3.3. Antero-posterior radiograph
of pelvis showing T-shaped fracture on
the right and associated T-shaped and
posterior wall fracture on the left
32
3.1.2.1 Anterior Border of Acetabulum
This is visible only on films of good quality (2 in
Fig.3.4a). It begins at the external border of the
roof but it is distinctly more horizontally disposed
than the posterior border and is superimposed on
the shadow of the posterior wall. Its mid-point is
marked by a change of direction at a notch of char-
acteristic angle, below which it descends further
almost parallel to the outline of the posterior
border. Finally, the line follows a curved path to
become continuous with the superior border of
the radiological obturator foramen. We call the
whole line, the acetabulo-obturator line (Figs. 3.5,
3.7 A, B). It is seen to comprise three segments,
separated by two distinct changes of curvature:
from above downwards and medially these are the
upper part of the anterior lip of the acetabulum,
the inferior part of the anterior lip and the root of
the pubis and then the anterior lip of the bony roof
of the obturator canal which becomes continuous
with the inner border of the obturator foramen.
The inferior margin of the anterior horn of the
acetabular articular surface is located at the mid-
point of the middle segment (C in Fig. 3.5A); this
A a
Radiology of the Normal Acetabulum
is important to keep in mind, as all fracture lines
dividing the anterior lip above this point will in-
volve the acetabular articular surface.
3.1.2.2 Posterior Border of Acetabulum
Clearly visible as an approximately straight line (1
in Fig. 3.4a) this is continuous inferiorly with the
beginning of the outline of the posterior horn of
the articular surface. Here the line forms a sharp
curve convex inferiorly which terminates medially
and overlies the outline of the upper margin of the
ischial tuberosity.
3.1.2.3 Roof
The radiological roof, the classical "dome" (3 in
Fig. 3.4 a), is in fact a radiological line and nothing
else, so it has the width of a line! It is obviously
produced by the rays tangential to the most cranial
point of the acetabular crescent.
When we underline the roof of a dry bone with
a sheet of lead under the image intensifier, we apply
a strip of lead not more than 2-3 mm in width
(Fig. 3.7 B, D), and this could define the "roof sur-
Fig. 3.4. Antero-posterior radiograph (A) and diagram (a) show-
ing radiological landmarks: 1 posterior border of acetabulum,
2 anterior border of acetabulum, 3 roof, 4 teardrop, 5 ilio-ischial
line, 6 brim of true pelvis
.."._, i 1·
I /i_..... 2--
•.4. ..
\ __/;r· ~ 1\
Superior pole ( • \
01 ischial Posterior' horn \ . .
tuberOSity of articular Tangent to s urface prolectlng
surface and supporting postenor horn
Standard Radiography
A bone showing lead marker on the acetabulo-obturator line
face" to which the rays are tangential in the posi-
tion of the pelvis close to the ideal one. All possible
ray tangencies through this area, following slight
variations in the position of the pelvis, will produce
comparable images.
So, the outline of the roof represents a narrow
segment of bone corresponding only to the width
of the articular crescent. It does not give any in-
dication of the overall integrity of the so-called an-
atomical roof. This must be assessed from the con-
tinuities of the anterior and posterior lips which
remain attached to the roof (Fig. 3.6).
When traced inferiorly and medially, the dense
sharp line of the roof pursues a bayonet course
which corresponds to the rays tangential to the up-
per part of the cotyloid fossa and then forms the
external limb of the radiological teardrop.
3.1.2.4 Teardrop
From Fig. 3.7 it can be seen that the outline of the
radiological teardrop (4 in Fig. 3.4 a) is produced by
a V-shaped continuous surface of bone; the lower
border is located in the ischio-pubic notch which at
the same time forms the superior border of the ob-
turator foramen at its highest point.
33
Fig.3.5A,B. Source of acetabulo-obturator line. A Outline on
antero-posterior radiograph of dried bone (lead foil has been
applied to the anterior part of the articular cartilage and the
outline of the anterior horn is indicated by the arrow C), B dried
The external limb of the teardrop is tangential to
part of the surface of the outer aspect of the
cotyloid fossa; the curvature is regular and slightly
concave externally. It is the middle third of the
cotyloid fossa surface which becomes tangential to
the X-ray beam producing the image.
The internal limb of the teardrop is formed by
the outer wall of the obturator canal which merges
posteriorly with the outline of the quadrilateral
surface of the ischium. Horizontally, this surface is
slightly concave internally, which contrasts with the
externally concave surface of the cotyloid fossa
(Fig.3.7C).
If the dry specimen is carefully examined it will
be seen that the zones of the surfaces responsible
for the two limbs of the teardrop do not lie in the
same coronal plane and for this reason, in different
subjects, marked variations in the relative disposi-
tion of the two limbs can result - in some in-
stances they can actually cross. When the two limbs
merge a true teardrop appearance is seen. In most
instance, however, the two limbs are more or less
parallel, and a radiological "V" is usually a more
exact description than teardrop.
As stated above, the bottom of the teardrop is a
tangency to the acetabular notch which forms the
34
Alea of X-ray
tangency
producing
the radiological
roof---+--:L-~-
superior border of the obturator foramen. It has a
variable relationship with the acetabulo-obturator
line and is projected in the neighbourhood of its
third segment. Depending on the exact location of
the central ray, the lowest part of the U passes
sometimes above or sometimes below the line; this
Radiology of the Normal Acetabulum
Fig. 3.6. A The difference between the radiological roof and the
anatomical roof. B Antero-posterior view of a typical transverse
fracture. C To assess the status of arc CAD, we must regard the
lengths of segment AC of the anterior lip and segment AD of
the posterior lip that have remained undamaged
is easily explained when it is appreciated that the
two elements are not in the same coronal plane (see
Fig.3.7B).
The shape of the radiological U varies con-
siderably from even a slight modification of the
orientation. Thus external rotation of that side of
Standard Radiography
A B
W: :
c o
Fig. 3.7 A-E. The teardrop. A Radiograph of dried bone (lead
wire accentuates the outline of the teardrop and the acetabulo-
obturator line, and lead foil outlines the roof), B corresponding
dried bone. C-E Diagrams. C Horizontal section showing the
the pelvis results in a U produced by the anterior
moiety of the cotyloid fossa and a more anterior
part of the external wall of the obturator canal. At
this level the bone is thicker and the resulting U is
generally larger at its base; further, its internal limb
is often shorter. That is what we call an "anterior
teardrop" (see Fig. 3.70).
3.1.2.5 Ilio-ischial Line
The ilio-ischialline (5 in Fig. 3.4 a) was described by
DUVERNAy-PARENT and in our opinion has a fun-
35
Roof of obturator canal
E
limbs of the teardrop formed by tangential incidence at sites not
in the same coronal plane, 0 possible sources of the teardrop,
E zone of bone responsible for the internal limb of the teardrop
damental importance which has not been recog-
nised sufficiently in the past.
The line is a result of the incidence of the X-ray
beam tangent across a segment of the quadrilateral
surface of the ischium, the axis of which is parallel
to the X-ray cassette (Figs. 3.1, 3.8). Its anterior
limit coincides with the posterior border of the ob-
turator foramen and its posterior part lies about a
fingerbreadth in front of the tip of the ischial spine;
superiorly it ends 1 cm below the top of the greater
sciatic notch but the inferior limit is always dif-
ficult to locate precisely because it is superimposed
on the external border of the obturator foramen.
36 Radiology of the Normal Acetabulum

The line usually begins above in common with

the ilio-pectineal line, with which it appears to
blend. Occasionally, however, the two can be seen
as separate entities being separated by a clear space
on the radiograph. Traced downwards it pursues a
straight or more often slightly curved course and
terminates as the outer border of the obturator
foramen. On the way it crosses the teardrop or is
superimposed on one of its limbs. The exact rela-
tionship beween these two, of which we shall see the
importance later, depends again on the orientation
and individual shape of the pelvis as well as its rela-
tion to the X-ray beam (Fig. 3.8 B).
The respective sources of the teardrop and the
ilio-ischial line are distinct entities although some-
times radiographically superimposed. The relation-
ship of these landmarks assists in the exact location
of certain fractures of the acetabulum.

3.1.2.6 Pelvic Brim, or Ilio-pectineal Line

Radiologically, the line indicating the pelvic brim

(i.e. brim of the true pelvis; 6 in Fig. 3.4a) does not
correspond throughout with the arcuate line, the
anatomical boundary. Between the superior border
of the symphysis pubis and the beginning of the
ilio-ischialline the radiological pelvic brim and the
anatomical pelvic brim correspond exactly; the seg-
ment constitutes about the anterior three-quarters
of the brim of the true pelvis as seen on the radio-
graph.
The posterior quarter of the radiological pelvic
brim is formed by a surface situated 1 - 2 cm below
the arcuate line and which corresponds roughly to
the lower half of the internal face of the sciatic
buttress and then by the internal part of the roof of
the greater sciatic notch which is related to the
gluteal vessels (Fig. 3.9B, C).
The radiological outline of the arcuate line itself
would extend to the projection of the anterior sur-
face of the second sacral vertebra.
3.1.2.7 Other Radiological Features
(a) The posterior border of the ilium (J in
Fig. 3.lOa). This is not visible except where it con-
stitutes the most internal part of the superior
border of the greater sciatic notch which shows
below the arcuate line (3 in Fig. 3.10a). The ischial
spine is sometimes visible a little above the tear-
drop, between the ilio-ischial line and the pelvic
brim (2 in Fig. 3.10a).
A
Fig.3.8A-D. Source of iIio-ischial line. A Dried bone with
main source of the line removed from the quadrilateral surface.
B-D Antero-posterior radiographs of dried bone. B Ilio-ischial
line crossing the teardrop, C ilio-ischialline absent in a specimen
prepared as in A, D ilio-ischial line reappears when the area of
cortical bone removed in A is replaced with lead foil
(b) The articular surface (4 in Fig. 3.10a). A clear
indication of the breadth of the acetabular surface
at the level of the roof is given because it cor-
responds exactly to the lamina of compact bone
which supports it in this region. The posterior horn
is outlined by the posterior border and inferiorly by
the medially directed curve with which it is con-
tinuous. The anterior horn, narrow and slender,
descends to a level about a fingerbreadth above the
outline of the posterior horn and stops at the mid-
point of the middle segment of the acetabulo-
obturator line (see Sect. 3.1.2.1).
(c) The spine of the pubis. The line produced by the
roof of the obturator canal is usually visible cross-
ing the body of the pubis (5 in Fig. 3.10a); the sur-
face of bone at the medial end of this line cor-
responds to the condensation of bone supporting
the pubic spine inferiorly but the spine itself is
Standard Radiography
B c
poorly seen, if at all (5 in Fig. 3.10a). Laterally, the
roof of the obturator canal appears to be con-
tinuous with the acetabulo-obturator line.
(d) The iUo-pectineal eminence. Although not visi-
ble in the antero-posterior radiograph, its site
should be known (Fig.3.10a).
(e) Inferior border of the ischio-pubic ramus. This
is simply and anatomically displayed (6 in
Fig.3.10a).
(f) Iliac wing. A thickening, the anterior pillar,
forms the main support of the iliac wing and passes
from the roof of the acetabulum to a point at the
junction of the anterior and middle thirds of the
crest (7 in Fig.3.10a). The highest part of this
pillar, which forms the gluteus medius tubercle, is
not the most external point of the convexity of the
crest seen on the antero-posterior radiograph. The
37
D
latter arises from the wing, anterior and inferior to
the gluteus medius tubercle.
3.1.3 Obturator-oblique Radiograph
When taken perfectly, this view shows the tip of the
coccyx approximately above the centre of the co-
tyloid fossa (Fig. 3.11 a). The following features
should be studied:
pelvic brim (the fundamental line of the anteri-
or column) or ilio-pectineal line;
certain elements of the articular surface, espe-
cially the posterior border;
obturator foramen and surrounding ring of
bone;
iliac wing as seen in section;
anterior lip and wall.
38 Radiology of the Normal Acetabulum

Fig. 3.9A-C. Source of radiological pelvic brim, which differs

in its posterior part from the anatomical pelvic brim (arcuate
line). A Antero-posterior radiograph of pelvis with lead wire
applied to the anatomical pelvic brim, B dried bone on which
the zone of quadrilateral surface which is the source of the
Wi thin thiszone,
ilio-ischial line has been excised and where distinct triangular anatomical and
surfaces have been covered with lead foil in order to outline radiological pelvic
completely the posterior part of the radiological pelvic brim. brims correspo nd Lead wire marker
e Diagram, explaining B exact ly along pelvic brim

B
Standard Radiography
/'
fillD
Ante rior superior
ili ac spine
Anterior inferior
iliac spine
Fig. 3.10. Antero-posterior view.
A Radiograph of dry specimen in
which as many as possible of the linear
radiological landmarks have been ac-
centuated with lead wire. The iliac
spines, the ilio-pectineal eminence and
the articular surfaces have been covered a 6~ine
with thin lead sheet, a diagram
3.1.3.1 Pelvic Brim
In the obturator-oblique view of a dry bone
perfectly positioned at 45 0 the pelvic brim does not
appear as regular in outline as in the antero-
posterior view. It is seen as a line composed of
several segments (Fig. 3.11 B, b). It corresponds to
the ilio-pectinealline from the angle of the pubis as
far laterally as a point situated a little above the
roof of the acetabulum at the level of the antero-
inferior iliac spine; this first segment is perfectly
clear and well defined. It then continues as a short
segment measuring 3 - 4 cm (Fig. 3.11 b, between
points B and C) relating to the internal margin of
the ilium tangential with respect to the X-ray beam,
39
Posterior inferior
iliac spine
(rip
Ilia-pectineal
eminence
\
.----
surface supporting
just above the pelvic brim (l in Fig. 3.11 D and
Sect. 3.1.3.4). The terminal portion of the pelvic
brim changes direction to become continuous with
the outline of the incidence on the wing of the first
sacral vertebra, making an obtuse angle with the
brim. Between this terminal portion of the radio-
logical pelvic brim and the internal limit of the iliac
fossa (B, C segment on Fig. 3.11 b) is a further short
segment about 1 cm long, which breaks the smooth
curvature twice and corresponds to the anatomical
arcuate or innominate line (Fig. 3.11 b, between
points C and D).
For practical purposes one must remember that
this exact description of the composition of the
radiological pelvic brim on the obturator-oblique
40 Radiology of the Normal Acetabulum

A B

it
Supra-acetabular suriac:.)
(site of spur sign when I
fractured and junction of I
anterior and posterior ~ Fig.3.llA-D. Obturator-oblique view. A Standard radiograph
colum ns) (perhaps centred slightly too low but with tip of coccyx above
\, centre of head), a diagram of radiological lines to be sought;
/" '- ._ ._.1 B radiological outlines marked on dried bone with lead wire,
b diagram; C dried bone with external landmarks of the iliac
i wing outlined with lead wire (see text), D dried bone with source
\ of the inner radiological outline of the iliac wing marked with
\
a \ lead wire

view of a dry bone perfectly oriented at 45 ° will
not always apply, because of changes in orientation
and projection. In some instances, the outline may
appear as a perfectly regular curve, and then cor-
responds exactly to the tangency of the X-ray
beams to the upper aspect of the anatomical pelvic
brim, but this will be the case only if the rotation
of the pelvis is a little below 45 o. In humans
generally the ilio-pectineal line on this view is a
regular curve and then corresponds to a tangency to
the upper aspect of the pelvic brim.
Between the pelvic brim and the projection of
the articular surface of the acetabulum a somewhat
dense vertically disposed line is visible. It corre-
sponds to the psoas gutter. It is slightly curved, be-
ing concave externally, and the point where it
Standard Radiography 41

....
i \
i . \
~ .
Gluteus medius tubercle
!/
""'{;:-.,.
~~
\ Projection of internal iliac fossa
.' I (1 on Fig. 3.11 0)
"';..., \\ . . The anatomical pelvic brim
Posterior superior iliac spine , j..' ~ \\ I. is in dotted line when it differs in
... ) :), \ \ \ position from radiological outline
Anterior pillar of iliac wing '\ , .. ;"1"#
(1 on Fig. 3.1 1 C) \ ... \ \ \ i, ----~ Projection of ala of sacrum
Gluteus maximus tubercle ~ A. \ 0 Short segment of anatomical pelv ic bri m
Projection of external iliac fossa A .~
(2 on Fig.3.11 C)
Anterior superior iliac spine i~ PrOjection of iliac w ing immediately
Exfernal surface of sciatic buttress ~i \ above pelvic brim
"~-- Anterior inferior iliac spine
(3 on Fig. 3 11 C) .
k~_- External margin of psoas gutter
Roof of acetabulum
Retro-acetabular surtace - - -_'fJ
Projection of anterior articular surface
superimposed on anterior lip
of acetabulum

\
Proj~ction of anterior part of
the Ischlo·publc notch :::><\
'-' . \ \.
//
...~ Projection of wall supporting
pubiC spine
Projection'of surtace which supports the ' . ............. ../
tip of the posterior horn of the ._ ._ . >- ...... Roof of obturator canal
articular surtace
Posterior border of obturator canal
Site of posterior border of innominate bone Anterior border of obturator canal
which is not visible in this view
b

c D
42
rejoins the radiological pelvic brim marks the ter-
mination of the projection due to the anatomical
pelvic brim (B in Fig. 3.11 b).
3.1.3.2 Posterior Border of Acetabulum
This is easily visible in its entirety delineating the
outer aspect of the posterior horn and the posterior
wall of the acetabulum.
A study, in which a lead marker is applied to the
articular surface, proves that the image which ap-
pears below the posterior horn is formed by the
outline of the wall which supports the latter part
and forms also the upper wall of the subcotyloid
groove.
3.1.3.3 Obturator Foramen
The outline is complete and the symphyseal surface
of the pubis opposite to the side being examined is
superimposed thereon. The line we have described
previously on the antero-posterior radiograph,
formed in its outer part by the roof of the obturator
canal and on its inner part by the condensation of
bone supporting the pubic spine, is seen as if ex-
tending from the superior border of the obturator
foramen. The whole bony frame of the obturator
foramen (the obturator ring) is practically undis-
torted in this projection.
3.1.3.4 Iliac Wing
There are several vertically disposed outlines, often
superimposed, being projected from various sur-
faces (Fig. 3.11 C, D). The outer limit of the iliac
wing, as seen in this view, is a line externally convex
in its upper two-thirds, and externally concave in its
inferior third. It is produced, from above down-
wards (see Fig. 3.11 B, b), by the hindmost part of
the iliac crest (the posterior superior iliac spine is
its most lateral projecting point), then by a tangen-
cy to the gluteus maximus tubercle - the thickest
part of the posterior gluteal line - and then to the
lateral aspect of the sciatic buttress (Fig. 3.11 band
3 on Fig. 3.11 C). Finally, just above the roof of the
acetabulum, it merges with the line we shall now
describe. The beam tangential to the pillar of bone
which extends from the tubercle of the iliac crest to
the roof of the acetabulum generates a line
(Fig. 3.11 band 1 in Fig. 3.11 C) which describes a
smooth curve concave externally. (This bony thick-
ening in the iliac wing we shall refer to as the
anterior pillar.) Below, it is often continuous with
Radiology of the Normal Acetabulum
a second curve convex externally and formed by the
upper outer segment of the wall of the acetabulum
posteriorly. It terminates at a point corresponding
to the margin of the upper third of the articular
surface. The exact point will correspond to the
degree of pelvic tilt relative to the X-ray beam of in-
cidence, appearing higher or lower corresponding-
ly. Similarly if the degree of rotation of the pelvis
is less than 45 0, the external limit of the iliac wing
appears to terminate externally at the outer border
of the acetabulum, level with the roof. Above, this
line protrudes where it meets the outline of the
tubercle of the iliac crest.
The internal surface of the iliac fossa projects a
line apparently continuous with the arcuate line; it
is seen on a dried bone that it is produced by the
deepest part of the internal iliac fossa (1 in
Fig. 3.11 D).
Between the concavities of the two lines just de-
scribed, which although approaching each other
never actually cross, there exists a third line, almost
vertical, which corresponds to the most concave
part of the outer (gluteal) surface of the iliac wing
(2 in Fig. 3.11 C). The location of this line on the
dried specimen is seen to coincide almost exactly
with the outline of the segment of bone producing
the image of the inner limit of the iliac wing.
In summary, the shadow of section of the iliac
wing is marked by three principal lines. These are:
(a) the tangency to the anterior pillar of bone
supporting the iliac crest;
(b) the internal limit which appears to be a pro-
longation of the pelvic brim and corresponds to the
most concave part of the iliac fossa; and
(c) an intermediate line with respect to (a) and
(b), tangential to them or cutting them, and which
represents the most concave part of the projection
considered here of the gluteal surface.
3.1.3.5 Junction of Anterior
and Posterior Columns
This area, which may be disrupted in certain frac-
ture types, is seen as a line just above the roof in
this view (Fig. 3.11 a).
3.1.3.6 Anterior Lip and Wall of the Acetabulum
Note that (1) only the upper part of the anterior
border of the acetabulum is visible because, below
its notch, the outline is superimposed on the
Standard Radiography
shadow of the anterior wall seen in section, and (2)
the anterior wall, seen largely in section, terminates
about 1 cm above the upper limit of the obturator
foramen. The vertical line which appears to con-
tinue the outline inferiorly and reaches the upper
limit of the obturator foramen is the floor of the
cotyloid fossa (Fig. 3.11 b).
3.1.4 Iliac-oblique Radiograph
On this view (Fig. 3.12) the following features are
displayed:
posterior border of the iliac bone;
anterior border of the acetabulum;
iliac wing.
3.1.4.1 Posterior Border of the Innominate Bone
This is strictly anatomical in outline and easy to
define (Fig. 3.12 A, B).
3.1.4.2 Outer Limit of Acetabulum
An approximately straight line is produced in its
upper two-thirds by the anterior border of the
acetabulum and in its lower part by rays tangential
to the root of the superior ramus of the pubis
(Fig. 3.12a, b). In vivo, it often appears that this
outer limit of the acetabulum is not straight, its in-
ferior half being convex outwardly (Fig.3.12C).
This convex part corresponds to rays tangential to
the external limit of the root of the superior pubic
ramus.
3.1.4.3 Iliac Wing
The complete profile of the iliac wing is displayed
together with both anterior iliac spines separated
by the interspinous notch, behind which there is a
slightly more dense area of bone extending from
the roof of the acetabulum to the crest - the ante-
rior pillar. The whole iliac crest is seen.
3.1.4.4 Posterior Border of Acetabulum
This appears regularly concave externally and me-
dial to the outline of the anterior border
(Fig. 3.12B, b). It terminates at a distinct bony pro-
jection formed by the posterior horn of the ar-
ticular surface which is supported by a short curved
section, concave infero-Iaterally; the latter is the
43
supporting wall of the posterior horn and runs ad-
jacent to the outline of the pubic spine.
3.1.4.5 Other Radiological Features
The following features should be observed:
(a) The roof of the obturator canal appears as a
line concave supero-externally and usually distin-
guishable as being more acutely concave than the
outline of the posterior articular horn to which it
is roughly parallel and superiorly disposed
(Fig. 3.12A, C, D). The line relates to two further
dense lines emanating from the pubis. The upper
(or inner) of these is the internal surface of the
superior pubic ramus; the lower or external is from
the anterior surface of the superior pubic ramus -
it terminates below at the pubic spine, but is con-
tinued upwardly by the outline of the anterior sur-
face of the root of the superior pubic ramus
(Fig. 3.12A, C).
(b) The inner limit of the acetabulum is marked
in its lower part by two lines, roughly concentric,
the outer being the internal border of the posterior
wall, and the inner being the cotyloid fossa
(Fig. 3.12 B, b, C).
(c) The outer limit of the lower part of the whole
radiological outline is formed by the ischial tuber-
osity, below the posterior horn of the acetabular
articular surface.
(d) The ischio-pubic ramus appears within the
outline of the body of the ischium as a well delin-
eated "hanging drop" (Fig. 3.12B, b).
(e) The pelvic brim is never visible on this view.
It is useful to know that it is located about a finger-
breadth higher than the angle formed by the greater
sciatic notch (Fig.3.12B).
(f) The section of the roof seen on both this
oblique view and the other must be interpreted in
the knowledge that, in comparison with the antero-
posterior radiograph, it is an oblique articular sec-
tion at 45 0 to the latter.
3.2 Computed Tomography
Technology moves fast, and what we get today from
the CT scanner is probably already nearly old-
fashioned. We have been CT scanning in acetabular
fractures since 1984, and over this short period of
44 Radiology of the Normal Acetabulum

A B

Posterior superior
Pelvic brim (invisible a iliac spine
on this view) ~ ......-- '-'-
Anterior /
superior /
Poste rior border of iliac spine i
acetabulum i Projection of floor of
Anterior inferior cotyloid fossa
Projection of posterior lip of iliac spine J~
Internal margin
obturator cana l Articu lar surface -~~.\\\'~-----'''' of posterior wall of
Proj ection of acetabulum
root of superior Roof of obturator canal
pubic ramus \ I" " "
\ I ',\
\
,,
\
I
\ " .,
\
, '(
~
Out li ne of ischia-pubic ramus
Outline of pubic
symphysis

a seen in section b Ischia-pubic ramus

Computed Tomography
c
time techniques have improved and have already
been modified several times.
Today, working with a GE 8000 scanner, we pro-
ceed this way:
through the iliac wing, lO-mm-thick sections
every 10mm;
through the acetabulum (in fact from the supe-
rior border of the greater sciatic notch) 5-mm-
thick sections every 3 mm, so they overlap each
other;
through the obturator foramen, 10-mm-thick
sections every 10 mm again.
This procedure appears to give better images for
3-D reconstruction. Naturally, the whole pelvis is
visualised, as comparisons with the intact side are
very often useful.
CT scanning shows us true or nearly true hori-
zontal sections of the whole pelvis. These sections
involve the true anatomical landmarks of the in-
nominate bones. On the other hand, the con-
structed landmarks, due to X-rays tangential to
bony surfaces such as the ilio-ischial line or the
45
Fig. 3.12A-D. Iliac-oblique view. A Standard radiograph, a dia-
gram of radiological outlines to be sought, B radiological
outlines marked with lead wire, the acetabular articular surface
being covered with lead, b diagram, C Radiograph of dried bone
with lead wire marking the roof of the obturator canal and the
two dense lines relating to it and emanating from the pubis,
D dried bone as in C with lead wires in situ. The more lateral
lead wire is the inner branch of the teardrop
teardrop, do not appear on these sections. For this
reason, proceeding from the selected landmarks of
the standard radiographic views, CT scanning al-
lows perfect analysis only of both lips of the ace-
tabulum, the pelvic brim and the posterior edge of
the innominate bone.
It appears useful in some cases to take thinner
sections, say, 1 mm thick, for instance, to study in-
carcerated fragments or impaction of the femoral
head.
3.2.1 CT of a Normal Acetabulum
We shall analyse 15 sections of a normal innomi-
nate bone which include the most useful by which
to characterise a fracture through that bone
(Fig. 3.13). It must be mentioned that the sections
vary slightly in shape from one case to another, as
they depend on the thickness of bone analysed and
also on the position of the pelvis, which may be
more or less flexed or tilted. It is essential and very
46 Radiology of the Normal Acetabulum

4 10

rJ
5 8 11 14
'\D-BOdY of pubis

~
4
~Ischlum

i/6
o
3 9 12 15
5?2
Cotyloid
fossa r)0 1

"SacrO-lllac Ischial spine

JOint ramus
o
B
Fig.3.13A,B. CT sections of an intact and dry innominate the anatomical roof, 4 posterior border of the innominate bone,
bone: A scans, B diagrams explaining A. 1 Posterior lip of ace- 5 pelvic brim (brim of the true pelvis or ilio-pectineal line);
tabulum, 2 anterior lip of acetabulum, 3 sub-chondral bone of AIlS, anterior inferior iliac spine; GSN, greater sciatic notch
Computed Tomography
helpful to have true horizontal sections involving
both innominate bones at the same level.
Section 1, above the sacro-iliac joint, shows the il-
iac wing, its shape, thickness and direction.
Sections 2 and 3 show the iliac wing and the adja-
cent sacro-iliac joint; if necessary, several sections
are made through the sacro-iliac joint, demonstrat-
ing the condition and the direction of the joint
space, and a possible compression fracture of the
ala of S 1.
Section 4 passes immediately below the roof of the
greater sciatic notch; posteriorly adjacent to the
section of the wing is a segment of this roof.
Section 5 involves at the front the upper part of the
antero-inferior iliac spine and passes above the
acetabulum to finally reach the greater sciatic
notch.
Section 6 demonstrates the sub-chondral bone of
the dome of the acetabulum, which has a circular
cross-section and corresponds to the anatomical
roof. It is an essential section to analyse fracture
lines passing through the anatomical roof. This
allows a much better study of the anatomical dome
than is possible with standard views, because we see
the entire dome and not just an X-ray transection.
At the front the anterior inferior iliac spine forms
the limit of the section.
In order to interpret the CT section accurately, we
have also studied with CT a dry bone with the land-
marks of the standard views overlaid with lead wire
or sheeting. The section passing through the ana-
tomical dome shows perfectly that the structure
that on the antero-posterior view represents the
radiological roof, called by some the "dome", and
which was overlaid with a narrow lead strip, is only
a narrow strip of the sub-chondral bone con-
stituting the anatomical dome (Fig. 3.14).
Section 7 is taken a little below this latter; as the an-
atomical dome is spherical, we still see the sub-
chondral bone like a corona surrounding the sec-
tion of the femoral head, from which it is separated
by the radiolucency of the two layers of articular
cartilage.
Sections 8 and 9 transect the acetabular region
where the femoral head opposes the anterior and
posterior columns separated by the cotyloid fossa.
These sections divide the fossa where the ligamen-
tum teres inserts on the femoral head. Section 9
also transects the ischial spine.
47
On section 10, the posterior border appears wider
at the level of the lesser sciatic notch. The anterior
wall articular surface is obvious.
On sections 11 and 12, the anterior wall is just com-
ing to an end or (section 12) has already disap-
peared.
On section 13, the posterior wall, which goes more
than 1 cm further down than the anterior wall, is
still well seen. The section divides in front of the
posterior wall the ischio-pubic notch, the obturator
canal and the superior pubic ramus.
Section 14 transects the neck of the femur and the
trochanteric region, and, on the pelvic side, the
pubic ramus anteriorly and the ischium posteriorly.
Section 15, the most inferior, reveals the ischio-
pubic ramus and the ischial tuberosity. It shows the
condition of the symphysis pubis.
The anatomical landmarks of the innominate bone
that we studied on the standard views and which
are perfectly displayed by CT are the posterior lip
of the acetabulum (J in Fig. 3.13B), the anterior lip
of the acetabulum (2 in Fig. 3.13 B), the posterior
border of the innominate bone (4 in Fig.3.13B)
and the brim of the true pelvis (5 in Fig. 3.13 B).
The last can be studied here in its entirety: we must
remember that on the antero-posterior radiograph-
ic view only the anterior two-thirds of the radio-
logical ilio-pectineal line corresponded to the true
pelvic brim - its posterior third was a constructed
image. We have not shown on the sections the sur-
faces responsible for the teardrop and ilio-ischial
line, as these are landmarks proper to the standard
views and are produced by X-ray beam tangen-
cies.
3.2.2 Special Advantages of CT
Throughout this book we shall see what advances
CT scanning has brought and how it has comple-
mented the interpretation of the standard
radiographs.
We learnt from surgery that not all fracture lines
are visible on standard views. They cannot - theo-
retically - escape the CT scanner, whatever they
are, displaced or undisplaced. However, to achieve
this the CT studies have to be complete and the sec-
tions numerous and preferably slightly overlapping
each other, since a horizontal fracture line may be
 48
invisible even on CT if it is located between two sec-
tions that are too far apart. Possible associated
sacral fractures or alterations of the symphysis
pubis or of the sacro-iliac joint are now very easy
to recognise and assess.
CT is also the ideal means for evaluating the
marginal impactions of the articular surface which
are so frequent in acetabular fractures. Marginal
impaction can well accompany all types of frac-
tures. In its displacement by the fracturing force, in
whatever direction, the femoral head breaks the
acetabular articular surface. Regularly the head
breaks off segments of the articular surface,
pushing them in front of itself. Frequently, in addi-
tion to these fragments delineated by a fracture line
running through the articular surface, the femoral
8 place
Radiology of the Normal Acetabulum
head also breaks the inner part, the "marginal
part", of these articular segments, but instead of
pushing them in front of it, it subjects them to a
rotational movement, resulting in their impaction
into the underlying cancellous bone. These
marginal impactions have different features on CT
according to their situations.
At the level of the roof, marginal impactions oc-
cur around a roughly horizontal axis, and these ap-
pear, on a CT section passing through the roof, as
polygonal sectors of sub-chondral bone which are
not at their normal place, often appearing more
dense than the dome itself (see Figs. 7.7E, 13.15F).
At the level of the walls of the acetabulum the im-
pacted fragments rotate around a roughly vertical
axis, so they appear as circular segments of ar-
Fig. 3.14. A Antero-posterior
radiograph of a dry pelvis with the
main landmarks overlaid with lead
wire; especially, the radiological roof
has been emphasised with a strip of
lead. 8 CT section of the same bone
through the anatomical roof with the
strip of lead marking the radiological
roof of the antero-posterior view still in
Tomography
ticular surface which have lost their normal orien-
tation and look outwards (see Fig. 5.19E). Medial
to these impacted fragments, the CT scan shows
condensation of the underlying cancellous bone,
thus revealing the impaction or the crushing of the
spongy bone.
It is well known that CT scanning is an excellent
method, and extremely reliable (provided the sec-
tions are numerous) in detecting incarceration of
bony fragments in the joint space or the cotyloid
fossa.
A reminder: if it is impossible to obtain the ob-
lique views because of pain, they can be taken as
"scout views" with the scanner tube steadily in-
clined to 45 ° on each side.
3.2.3 Disadvantages of CT
Has CT scanning any disadvantages? Well, as men-
tioned by Dana MEARS, "Conventional computer-
ized scans are hampered by their obliquity of sec-
tion through the pelvis of a recumbent patient:'
Furthermore, the pelvic tilt varies from one patient
to another. Careful study of fractures through the
acetabulum or the sacral bone would be made
easier by using sections taken with different obli-
quities. MEARS and others developed angled com-
puting techniques in which both the gantry of the
scanning unit and the radiographic table were tilted
by about 20 o.
CT sections showing all the fracture lines and all
the small fragments may be frightening from a
surgical point of view: fractures appear generally
much more complex on CT than on the standard
views. We have succeeded in perfect reconstruction
in some cases which from the CT scan could have
been regarded as irreparable (Fig. 3.15).
Finally, it should be said that the improvement
brought by CT reconstruction will greatly facilitate
comprehension of these fractures, but it is still very
expensive and an experienced acetabular surgeon
must be able to do without it, at least in fresh cases.
Moreover, it must be remembered that the present
imperfections in the reconstruction process and the
smoothing of the CT sections may hide some frac-
ture lines partly or completely.
3.3 Tomography
Tomography proved useful for the precise location
of incarcerated fragments of bone during the first
49
25 years of our practice, but in this particular point
it is not superior to CT.
In cases seen some time after injury it has been
superior to plain radiographs in the assessment of
fracture union before a decision is made about any
active operative procedure. I still use it for this pur-
pose, and I also find this technique superior to CT
for an assessment of the healing process. It also
provides the best way to evaluate the congruency
between the anatomical roof of the acetabulum and
the femoral head. If we want to be certain that the
conditions of secondary congruence are fulfilled
we need tomograms; CT is not so good for assess-
ing congruence at the level of the roof (see
Fig. 19.3). Unfortunately, the traditional tomo-
graphs are getting old and disappearing one by one,
while digitalised tomography is far from being as
good for this purpose.
3.4 Stereo-radiography
We tried this technique in our early days and found
it difficult. It is of no further interest in the age of
3-D.
3.5 Interpreting the Radiographs
When 3-D images are obtainable for every case, the
diagnosis of the type of an acetabular fracture will
perhaps be easier. Comparison of these views to a
series of typical cases reproduced on dry bones
should suffice. However, while waiting for such a
time, interpretation of the standard views and the
2-D CT scan will have to suffice.
3.5.1 Interpreting the Standard Views
3.5.1.1 Preliminary Remarks
Radiological magnification and other distortions
must be taken into account. There is greater en-
largement of the anterior structures of the pelvis
when the views are taken with the patient lying
supine, and more enlargement of the posterior
structures when the patient is prone. It must be
remembered that the displacement of a fracture
50 Radiology of the Normal Acetabulum

A B c

D E F

G H

Fig. 3.15 A-I. Example of a fracture which on CT appeared to ture, D- F three alarming CT sections, G- I post-operative
be very difficult to reconstruct. A-C Standard radiographic follow-up at 1 year
views showing an infra-tectal transverse and posterior wall frac-
Interpreting the Radiographs
seen on a radiograph is the projection of the real
linear displacement. This displacement may be in
any direction, and the imaginary line that joins two
points formerly contiguous and forming the same
anatomical structure will appear very differently in
our three standard views. When a fragment rotates
on a vertical axis with respect to the X-ray beam, its
appearance in the antero-posterior view approx-
imates increasingly to that which it would normally
project in one or the other oblique view.
These simple observations need to be kept in
mind; a conscious awareness of them will con-
tribute enormously to a disciplined understanding
of the standard views recommended. A knowledge
of the classification provides the background
which facilitates understanding.
In radiographs showing the post-operative ap-
pearance, plates and screws used for internal fixa-
tion often appear very large. This is in part due to
simple radiographic magnification of the implants
and gives a false appearance, but, additionally, the
area of projection of the innominate bone on a
radiograph is relatively less magnified, especially in
the obturator-oblique view, and these two factors
combine to enhance the false impression that the
method of osteosynthesis is disproportionately
gross. These facts can be proved simply by placing
a Shermann plate on a dried bone and observing
the radiographic appearance thereof.
When the linear image of a structure is the result
of its presenting a surface tangential to the X-ray
beam - for example, the ilio-ischialline seen in the
antero-posterior view of the acetabulum - the im-
age so produced will appear only in that single
view; the appearance is considerably modified or
disappears completely once sufficient rotation of
the part represented has occurred. Further, a cons-
tant and therefore a comparable appearance is
preserved only if the displacement remains parallel
to its original orientation and thus perpendicular to
the X-ray beam.
A fracture line cannot be assumed to be un-
displaced unless this is confirmed in at least two of
the views; the image of a displacement may be in-
visible in the antero-posterior view but consider-
able in an oblique view (Fig.3.16A). Similarly, a
maximal displacement on one oblique view,
because it is parallel to the X-ray cassette, may ap-
pear undisplaced on the other oblique view as a
result of the rotation through the 90 0 difference
between these views (Fig.3.16B). On the antero-
posterior view, such a displacement will appear
moderate.
51
3.5.2.1 Interpretation
From experience we know that the best way to
recognize a fracture of the acetabulum is to analyse
the main landmarks of the two columns on the
antero-posterior and the two oblique views. Fig-
ure 3.17 shows these landmarks overtraced in white.
The landmarks of the anterior column are:
On the antero-posterior radiograph:
the anterior lip of the acetabulum, not always
clearly visible;
the ilio-pectinealline, the main landmark of the
column;
the teardrop, representing the cotyloid fossa (as
it is usually not involved in a posterior column
fracture, we may regard it as belonging to the
anterior column);
the anterior part of the obturator ring.
On the obturator-oblique radiograph:
the pelvic brim or ilio-pectineal line, always
perfectly displayed (the X-ray beam is in fact a
tangency to its superior aspect);
the anterior part of the obturator ring;
the external limit of the bone along 4 - 6 cm
above the roof, which may be the site of a spur
sign (see Sect. 14.2.2e).
On the iliac-oblique radiograph:
the anterior part of the iliac crest and the
anterior border of the iliac wing, with the two
anterior iliac spines and the notch separating
them;
the anterior lip of the acetabulum, limiting on
that incidence the acetabular area.
The landmarks of the posterior column are:
On the antero-posterior radiograph:
the posterior lip of the acetabulum, always
clearly visible;
the ilio-ischial line, the main landmark of the
posterior column on this view;
the ischial tuberosity.
On the obturator-oblique radiograph:
the posterior lip of the acetabulum, here
regularly convex, in continuity downwards with
the contour of the posterior horn of the aceta-
bular articular surface;
52 Radiology of the Normal Acetabulum

7J most cranial part of the acetabular articular sur-

face, the optical section of this part on the oblique
views being at 45 0 in either direction to the roof on
~
A'
the antero-posterior view.
~ Table 3.1 summarises all the above information.
! To study these three standard views properly, you
Pi AzA, need sheets of transparent material, felt-tip pens of
A Antero-posterlor Iliac-oblique
different colours and a dry bone.
On the viewing box, place transparencies in
front of the three views. On each of the three views,
each set of landmarks must be followed in turn, in
any order you like - for instance, those of the
anterior column, then those of the posterior col-
umn - but it may also be easier to trace the ap-
parently intact landmarks or those appearing less
disturbed first. The essential thing is that all the
Iliac-oblique main landmarks shown in Fig. 3.17 are checked.
B
Choosing one colour for the landmarks of the
anterior column, and another for the landmarks of
the posterior column, scrutinise each landmark in
its entirety for a possible traversing fracture line.
Sometimes several fracture lines may divide a land-
mark. The important thing is to locate as precisely
as possible the fracture line or lines, and then to
transfer onto the dry bone the point or points of
rupture of the landmark in question. Mark it with
, chalk or a felt-tip pen. At the same time, note
"-
"- whether there is a frank displacement or not at the
"-
lOso/- fracture site. If this is done for all the landmarks of
------~~----~-~ Pi.B'
the three views, the final result on the dry bone is
a series of points that only have to be joined up to
c Antero-postenor Iliac-oblique
give a near-accurate representation of the outline of
Fig. 3.16 A - C Diagrams of fracture-dislocation. A No dis-
the fracture to treat.
placement is seen on the antero-posterior view but marked dis-
placement on the iliac-oblique view. B Marked displacement on Let us take an example. Figure 3.18A shows the
one oblique view may not be apparent on the other. C Displace- three standard views of the case concerned. First we
ment along an axis 45 0 to the horizontal may be apparent on the check the landmarks of the anterior column.
antero-posterior view but absent on one oblique view
On the antero-posterior view:
the posterior part of the obturator ring and the
the anterior lip (Fig. 3.18 B) is broken at the level
ischial tuberosity.
of its notch; we mark a red point on the dry
bone at the top of the notch (Fig. 3.18b);
On the iliac-oblique radiograph:
the ilio-pectinealline (Fig. 3.18 C) is interrupted
the posterior border of the innominate bone is at the junction of the anterior two-thirds and
the most remarkable landmark with the outline the posterior third; we mark a red point at
of the two sciatic notches separated by the about that level on the dry bone (Fig.3.18c);
ischial spine; the teardrop is not disturbed, except for being
the quadrilateral surface (the hatched area in displaced inwards;
Fig. 3.17 C, bottom) is exposed, parallel to the the anterior part of the obturator ring is intact.
film, but fracture lines traversing it are not
always clearly visible. On the iliac-oblique view:
As for the radiological roof, it is seen on all three the anterior lip is broken at one point, of course
views, and results from the rays tangential to the still the same;
Interpreting the Radiographs 53

the anterior part of the iliac wing and its anteri- the antero-posterior view; no new mark on the
or limits are intact; no mark on the dry bone. dry bone;
the same goes for the rupture of the ilio-pec-
tineal line;
On the obturator-oblique view:
the anterior part of the obturator is intact; no
- the anterior lip is broken at the same point as on mark on the dry bone.

A B C

I)
z;! z;.

Anlero-poslerior

a ~ Iliac-oblique

b ~
Fig. 3.17 A-C. The main radiological landmarks of the anterior
Ob,~",,·06liq",· ~
c
....
,

B obturator-oblique radiograph, b diagram, C iliac-oblique

column (above) and posterior column (below) as seen in the radiograph, c diagram
three standard views. A Antero-posterior radiograph, a diagram,
 54 Radiology of the Normal Acetabulum

Table 3.1

Antero-posterior view Obturator-oblique view Iliac-oblique view

Landmarks of Ilio-pectineal line ++ Pelvic brim ++ Anterior border ++

anterior column (ilio-pectineal line) acetabulum
Anterior border ± Anterior part of obturator ++ Anterior border iliac wing ++
ring and crest
Teardrop and relationships + Iliac wing ++
with ilio-ischial line

Landmarks of Posterior border of + Posterior border of ++ Quadrilateral surface of +

posterior column acetabulum acetabulum ischium
Ilio-ischial line ++ Posterior part of ++ Posterior border of ++
obturator ring innominate bone

Roof Top segment of the Nearly similar Nearly similar

articular surface

+ + Landmark especially well displayed; + Landmark usually well seen; ± Landmark may not be seen

Now we check the landmarks of the posterior On the iliac-oblique view:

column.
On the antero-posterior view:
the posterior lip is broken a little above its mid-
part (Fig. 3.18D); a corresponding red point is
marked on the dry bone (Fig. 3.18d);
the ilio-ischialline is detached and displaced in
its entirety (Fig. 3.18E), so a red point is marked
at the upper tip of the ilio-ischial surface on the
bone (Fig.3.18e).
the posterior border of the innominate bone is
interrupted at the level of the angle of the
greater sciatic notch (Fig. 3.18 F); a red point is
marked at that level on the dry bone
(Fig. 3.18 f).
As to the roof, on the antero-posterior view
(Fig. 3.180), the fracture line divides the extreme
upper part of the cotyloid fossa; a red point is
marked on the dry bone at the upper part of the
fossa (Fig. 3.18 d).

Fig. 3.1SA-I. Radiological study of a

pure transverse fracture. A The three
standard views, B-G, b-f analysis of
the landmarks, H, I the fracture lines
drawn on the dry bone, inner and outer
A aspects
B b

c c
 56 Radiology of the Normal Acetabulum

d
Interpreting the Radiographs 57

e
 58 Radiology of the Normal Acetabulum

G
 Interpreting the Radiographs
Some landmarks appear on two of the three
views; that helps to locate the fractured points more
precisely.
Joining the six points we have marked on the dry
bone (Fig. 3.18H,I), we now have the outline of
our fracture. It appears as a single fracture line
dividing the innominate bone into two parts; it is
thus a pure transverse fracture. Since it passes at the
junction of the roof and the cotyloid fossa, it is a
juxta-tectal transverse fracture.
Figure 3.19 shows a more complex case which
was analysed in the same way. We marked the posi-
tions of the fractured point on the iliac crest, the
fractured point on the superior pubic ramus, and
the spur of bone which was detached from the
pillar of the wing and took with it a part of the
acetabular roof. However, despite the apparent
complexity of the case, from this study we knew
that the posterior border of the innominate bone
was intact, so it could not be other than an anterior
column fracture, with multiple additional fracture
59
lines, but still involving essentially the anterior col-
umn structure, leaving the posterior column un-
disturbed.
3.5.2 Interpreting the CT Sections
to Aid or Complete the Diagnosis
Once again a reminder: interpretation of the CT
sections must be done conjointly with that of the
standard views.
The CT sections will allow more precisely loca-
tion of the points of rupture of the true anatomical
landmarks: iliac crest, iliac wing at different levels,
acetabular dome, posterior border, cotyloid fossa,
pubic and ischial ramus, anterior and posterior
wall. The position of these fracture points can also
be transferred onto the dry bone in order to gain a
more accurate representation of the fracture lines.
CT sections also allow one to assess the orienta-
tion of a fracture line; we have learned that on CT:
 60 Radiology of the Normal Acetabulum

Fig. 3.19A-C A complex anterior col-

umn fracture analysed in the same way
as in Fig. 3.18. A The three standard
views, 8, C the fracture lines drawn on
the dry bone

8 c
Interpreting the Radiographs 61

a transverse fracture line has an antero-posteri-

A B C antero-posteriorly oriented and more or less exter-
Fracture of 1 or Transverse
of 2 columns fracture
Fig. 3.20 A - C. Orientation of fracture lines as seen on CT.
A Fracture detaching a whole column: coronal orientation;
B transverse fracture line: antero-posterior orientation; C frac-
ture detaching a posterior wall: outward/frontal orientation;
fracture detaching an anterior wall: outward/posterior orienta-
tion
a fracture detaching the whole architecture of a
column is frontally or coronally disposed or
nearly so (Fig.3.20A);
or orientation (Fig.3.20B);
a fracture detaching a wall is oriented outwardly
and frontally for a posterior wall, posteriorly
and outwardly for an anterior wall (Fig. 3.20C).
So, on the section involving the anatomical roof we
can easily distinguish a transverse fracture, which is
nally situated depending on whether or not it in-
volves the roof, from a pure column fracture, which
divides the roof coronally (Fig.3.20A). CT scan-
ning also permits an assessment of the displace-
ment of the fractured fragments and, especially,
their rotation, which is often difficult to evaluate
on the plain X-rays.
In regard to evaluation of the healing of frac-
tures, I have so far found CT not very reliable.
4 Classification
Before surgical intervention in a fractured
acetabulum can be accepted as a means of treat-
ment, accurate diagnosis based on radiology is
essential. As in other conditions, classification aids
accurate understanding of these sometimes com-
plex fractures. Recognition of the existence of a
fracture does not in general present a great difficul-
ty although a vertical crack in the coronal plane or
in the anterior wall of the acetabulum may not be
so easy to confirm. Once a fracture is recognised,
its anatomico-pathological type should be defined
with precision, together with an assessment of dis-
placement of the fracture fragments. It is essential-
lyon this appraisal that the fundamental decision
with regard to choice of operative approach will be
made.
For a long time the classical typing of fractures
of the acetabulum led to two broad categories:
central dislocation; and
- posterior dislocation of the hip with a fracture
of the acetabulum.
CAUCHOIX and TRUCHET (1951) subscribed to this
basic classification. However, these authors felt
that it did not encompass all aspects and they were
obliged to recognise the existence of intermediary
forms:
fractures of the posterior wall of the acetabulum
associated with a central dislocation (this group
corresponds exactly with our fracture complex
which associates a transverse and a posterior
element with central dislocation); and
trans-acetabular fractures of the pelvis with
posterior dislocation (again, this necessitates an
association between two basic fracture types,
namely a transverse element with a posterior
wall fractu~e).
We believe that this distinction between a central
dislocation of the hip and a posterior fracture-
dislocation is too crude and must be abandoned, if
only for one reason: a consideration of the mecha-
nisms whereby a fracture of the acetabulum is pro-
duced.
There is no doubt that these fractures result
from a force acting between the femoral head and
the acetabulum itself. The force can be transmitted
from the knee or the foot or alternatively from the
pelvis onto the femoral head as a result of a blow
on the back. These forces can drive the femoral
head into any aspect of the acetabulum, producing
a central or a posterior dislocation at any level, and
clearly, the number of fractures and dislocations
which can be produced is infinite.
To say that an injury of the acetabulum is simply
a central dislocation and to leave it at that is totally
inadequate if a serious attempt to reduce the frac-
ture accurately is envisaged. Until the late 1970s,
almost without exception, American, English and
German authors used this crude distinction and we
cannot accept their arguments.
CREYSSEL and SCHNEPP (1961) have tried to
categorise fractures using the above broad division,
and distinguishing principal and accessory fracture
lines. We do not agree with this approach for it
seems that all fracture lines traversing the
acetabulum are of similar significance; while cer-
tain fracture complexes are commoner because of
the modes of injury responsible, until now, no one
has proved that any particular zone of the
acetabulum is less useful than another, as far as the
long-term prognosis is concerned. Further, their
classification used the term "trans-acetabular" to
describe transverse fractures alone. We reject this
concept, for by definition, all fractures traversing
the acetabulum must be trans-acetabular.
ROWE and LOWELL published their classifica-
tion and results in the American Journal of Bone
and Joint Surgery in 1961. The main or essential
element they used for classifying the fractures was
the radiological dome. We saw in the previous
chapter what that radiological dome corresponds
to, and thus why it cannot be of any help in the
assessment of the extent of the post-traumatic con-
64 Classification

of the acetabulum, of which 839 have been

Fractures involving Fractures involving Fractures involving
the posterior col- both columns the anterior col-
operated upon. We remain of the belief that open
umn only umn only reduction and internal fixation is the method of
choice. We propose the division of these fractures
Posterior column Anterior column into two large groups (elementary fractures and as-
Posterior wall
Posterior wall and
sociated fractures), each of which will be broken
posterior column down as shown below.
Anterior wall
Anterior wall
and anterior Elementary fractures (Fig. 4.2) comprise those
column in which a part or all of one column supporting the
acetabulum has been detached. We include with
these, by virtue of its purity, the transverse fracture.
Transverse
Transverse and There are five elementary forms:
posterior wall fractures of the posterior wall of the acetabu-
Both columns
lum;
fractures of the posterior column;
T-shaped fractures of the anterior wall of the acetabulum;
Posterior column fractures of the anterior column;
and anterior
hemitransverse
transverse fractures.

Associated fractures (Fig. 4.3) all include at least

Anterior wall and two of the elementary forms above. There are five
posterior
hemitransverse principal associations:
Anterior column T-shaped fractures;
and posterior
hemitransverse fractures of the posterior column and posterior
wall;
transverse and posterior fractures (with a dislo-
Fig. 4.1. System of classification of acetabular fractures: final cation of the femoral head either posteriorly or
form for practical use
centrally);
fractures of the anterior column or anterior wall
associated with a hemitransverse fracture poste-
gruency or incongruency between the femoral head riorly;
and the intact part of the acetabular articular sur- both-column fractures.
face.
In actual fact, the range of fractures that may in-
Our classification, first published in my thesis in
volve the two columns is symmetrical. We should
1961, underwent some modifications before 1965.
thus have enumerated 13 types of fracture instead
Since then it has remained unchanged and is now
widely accepted. We readily recognise a number of of 10, these being:
fractures already described elsewhere and include Posterior column
these in our classification based on what we regard Posterior wall
as the fundamental grouping of lesions of the two Posterior wall and posterior column
columns which support the acetabulum, not taking Posterior column and anterior hemitransverse
into account the direction of displacement of the Anterior column
femoral head. The latter aspect is important, of Anterior wall
course, but not from the point of view of classify- Anterior wall and anterior column
ing the fractures. In effect, we shall demonstrate a Anterior wall and posterior hemitransverse
single family of fractures, each of which can be de- Anterior column and posterior hemitransverse
scribed individually yet forming together a con- Transverse
tinuous spectrum of possibilities. T-shaped
At the date of publication of the second English Transverse and posterior wall
edition of this book, we have treated 940 fractures Both-column
Classification
Fig. 4.2. The five elementary fracture
patterns reproduced on dry bones
To keep the classification as simple as possible,
however, we group some types together, for various
reasons:
1. On plain radiographs, it is difficult to tell a T-
shaped fracture from a posterior column and
anterior hemitransverse fracture (see Figs. 10.3
and 10.19). Although, with CT scanning and
3-D imaging, distinction between them is now
possible, we still group these fractures together
as "T-shaped", because posterior column and
Fig. 4.3. The five most frequent asso-
ciated fracture patterns reproduced on
dry bones
65
anterior hemitransverse fractures are rare and
anyhow they always need the same surgical
approach as T-shaped ones.
2. Anterior wall fractures and anterior wall and
anterior column fractures are grouped together
because the second merely adds to the first a
fracture through the ischio-pubic ramus that is
never approached and fixed. The same anterior
approach is required for both fracture types (see
Fig. 7.5).
66
3. Anterior wall and posterior hemitransverse frac-
tures and anterior column and posterior hemi-
transverse fractures are grouped together under
the name "anterior fracture and posterior hemi-
transverse fracture" since they require the same
approach and the fracture through the ischio-
pubic ramus, if present, is not fixed (see
Fig. 13.7).
The counterpart of the anterior wall and posterior
hemitransverse fracture, which would be a pos-
Classification
terior wall and anterior hemitransverse fracture,
does not exist, as a posterior wall fracture leaves
the anterior border of the greater sciatic notch
undisturbed, and when it is accompanied by a
transverse fracture, the fracture line involves the
greater sciatic notch and thus the posterior and
anterior columns; such a fracture cannot be quali-
fied as other than transverse. The classification of
fractures that emerges after all the considerations
have been taken into account is thus as shown in
Fig. 4.1.
5 Posterior Wall Fractures
Fractures of the posterior wall of the acetabulum
involve separation of a segment of the posterior ar-
ticular surface; the fracture line leaves undisturbed
the major portion of the posterior column. A pos-
terior dislocation is usually associated.
Strictly, these are partial fractures of the posteri-
or column and one could include them under this
heading. However, they have been recognised as a
particular entity for a long time and concern a
clearly defined anatomical area. Their clinical
presentation, especially with a dislocation, is so
typical that it is reasonable to describe them in a
special group.
These fractures fall into the following distinct
sub-groups:
typical fractures of the posterior wall confined
totally below the roof;
postero-superior fractures in which part of the
adjacent roof becomes separated;
postero-inferior fractures in which the detached
fragment includes the inferior horn of the artic-
ular surface, the sub-cotyloid groove and often
the superior portion of the ischium.
These three main types may result in detachment of
one or more fragments. Less commonly the follow-
ing are seen:
extended fractures which may detach in several
fragments, a segment extending from the roof to
the superior portion of the ischial tuberosity;
transitional forms almost amounting to com-
plete fractures of the posterior column detach-
ing in one or several fragments the posterior
wall, and a portion of the posterior cortex adja-
cent to the acetabulum including the anterior
border of the greater sciatic notch and a seg-
ment of variable size of the quadrilateral sur-
face of the ischium - massive posterior wall
fractures;
Table 5.1
Fractures of the posterior wall 223 23.720/0
Pure posterior, one fragment 64 6.80%
Pure posterior, multiple fragments 47 5.00%
Pure posterior with marginal impaction 39 4.14%
Postero-superior, one fragment 7 0.74%
Postero-superior, several fragments 10 1.06%
Postero-superior with marginal impaction 9 0.95%
Postero-inferior 5 0.53%
Extended posterior wall 13 1.38%
Posterior wall with incomplete transverse 29 3.08%
transitional forms in which there are associated
incomplete transverse fractures.
In all varieties of posterior wall fractures, when-
ever a segment of the posterior wall becomes sepa-
rated, allowing a posterior dislocation to occur, the
detached fragment may remain attached by its cap-
sule to the femur or the capsule may rupture
(Fig. 5.1). The significance of the capsular injury is
that it facilitates the escape of the femoral head. In
contrast, when the capsule remains intact, the head
dislocates after fragmenting the inner edge of the
fracture margin, the osteochondral segments be-
coming incarcerated and impacted into the under-
lying cancellous bone. We have distinguished these
two mechanisms and call the results simple fracture-
dislocations (by far the commoner and accounting
for 175 cases out of 223 - 78.5070), and fracture-
dislocations with marginal impaction (less common
and accounting for 48 cases of the 223 - 21.5%).
It is important to recognise the impacted fragments
due to marginal impaction, perhaps radiologically,
but failing this, during operation or by CT scanning
(which shows the marginal impaction perfectly), for
only by freeing these and restoring them to proper
alignment can the complete articular surface be re-
constituted. After effective support with cancellous
bone, redisplacement of these fragments of articular
surface need not occur.
68
Fig. S.IA,B. Diagrams of posterior fracture. A Pure fracture-
dislocation, B fracture-dislocation with marginal impaction (see
text)
5.1 Typical Posterior Wall Fractures
Incidence: 150 out of 223 cases, of which 39 ex-
hibited marginal impaction.
5.1.1 Morphology
There is a detachment of a single or several
fragments involving a segment of the posterior ar-
ticular surface but not involving the posterior horn
or the roof (Fig. 5.2).
The detached fragment or fragments vary in
size, in site, and in the amount of articular surface
Posterior Wall Fractures
involved. Thus all possibilities may occur, from a
mere abrasion of the posterior lip of the acetabu-
lum detaching fragments only 2 - 3 mm in breadth
and therefore impossible to reposition, to gross
fragments comprising the whole posterior wall.
These fractures result more usually in the detach-
ment of one large fragment.
Referring to Fig. 5.2 B, the possibilities with
regard to the soft tissue attachment include total
loss of capsular and muscular attachment (Ba),
retention of capsular attachment (Bb) which may
lead to marginal impaction (see above), retention of
retro-acetabular soft tissue attachment (Bc), or im-
paction of fragments which have lost soft tissue at-
tachment following marginal impaction (Bd). Loss
of soft tissue attachment leads to avascularity; dur-
ing operative intervention, while the vascular sup-
ply is rich, it must not be abused by careless man-
agement of the soft tissue pedicles.
When the posterior wall segment is shattered,
polyhedral and sometimes small fragments bearing
only cortical bone of the retro-acetabular surface
may be encountered; they are of value in providing
mechanical stability. Fragments bearing articular
cartilage must be pieced together accurately.
5.1.2 Radiology
This is straightforward but a detailed description
offers a useful introduction into the understanding
of the more complicated fractures which are en-
countered later.
(a) Antero-posterior view. Before reduction of the
dislocation, the detached bone of the posterior
wall looks like a cap adjacent to the femoral
head (Fig. 5.3 A). The reduction of the disloca-
Fig. S.2 A, B. Posterior wall fracture. A Scheme of
pure posterIor wall fracture. B possible attachments
of posterior wall fragment (see text)
Typical Posterior Wall Fractures 69

A B

/
.-._....j
./ ~_.Jl
Fig. S.3A-D. Posterior wall fracture. A Antero-posterior radio- .I "." I.
graph, B antero-posterior radiograph after reduction, b dia- ;I
'-.\
gram, C iliac-oblique radiograph, c diagram, D obturator-
oblique radiograph, d diagram. Note incarcerated fragment on
\
Band D, and irregular area on infero-internal aspect of femoral
\
head on A and D caused by damage at the moment of disloca- \
tion b \

tion (Fig. 5.3 B), usually performed as a matter
of urgency, may well be stable and then the pro-
blem is to recognise the nature and magnitude
of the posterior fragment. From this view alone
it is not possible to assess its size or its displaced
position.
Five of the six fundamental radiological landmarks
are not disturbed, namely the roof, the pelvic brim,
the ilio-ischial line, the teardrop, and the anterior
border of the acetabulum. Only a segment of the
posterior lip is missing, its extremities being united
by an abnormal notch created by the loss of the
detached fragment. The notch is variable in size
and can be obscured by the intact anterior wall.
(b) Iliac-oblique view. This confirms (Fig. 5.3 C)
that the posterior border of the innominate
bone, the anterior border of the acetabulum,
and the iliac wing are all intact. The detached
posterior wall fragment is superimposed on the
iliac wing and often difficult to see.
(c) Obturator-oblique view. Most information with
regard to the posterior wall fracture is here
(Fig. 5.3 D). The innominate line and the ob-
70 Posterior Wall Fractures

c D

;"
(
/ .-i;-._-,
,-
i . j ./
,.....-." I
r ~' - -/I i
/ -.~-"""
/I
I ·
i ! I i ':'.,\
-\ ". i \ /
i \ \ /
! ) \ f
! i \ \
c see Fig. 3.1 2 d ./
Fig.S.3C-D
 Typical Posterior Wall Fractures 71

..
.............- - ... , J
/ \ I
I ~~+
I \
I \
I \
I \
i \
\
\
\
\ Fig.S.4A,B. Detached fragment. A Antero-posterior radio-
\
\ graph on which the fragment detached from the posterior wall
\
\ appears small but with care the large defect in the posterior wall
\ can be perceived, a diagram, B obturator-oblique radiograph on
a \ which the detached fragment is seen to be large
72
a seen markedly displaced
turator ring are intact. The following three
features should be sought:
The head of the femur may be fully reduced
or there may be a persistent subluxation not
apparent on the antero-posterior view.
The notch created by the fracture and which
interrupts the posterior border of the
acetabulum creates a deep defect in the
posterior wall.
The exact size of the detached fragment is
displayed because in this view it is exposed
Posterior Wall Fractures
Fig. 5.5A,B. Detached posterior wall fragment. A Antero-poste-
rior view, on which the fragment appears a little displaced,
a diagram, B obturator-oblique view, on which the fragment is
(being unsuperimposed), and its largest sec-
tion is perpendicular to the X-ray beam.
The comparison between an antero-posterior view
and an obturator-oblique view in which the true
magnitude of a detached fragment is seen is shown
strikingly in Fig. 5.4. The obturator-oblique view
also shows the displacement of the posterior frag-
ment and Fig.5.5 illustrates how deceptive the
antero-posterior view can be in suggesting that an
adequate reduction has been achieved.
Typical Posterior Wall Fractures 73

Fig. 5.6. Intra-operative appearance of

a marginal impaction

It remains to attempt an exact evaluation of the

nature of the detached fragment. Most examples
comprise a large single piece of bone similar to that
just described. Simple comminution of the de-
tached bone is usually visible and easily detected.
Marginal impaction may be more difficult to detect
but in light of its presence in 26070 of our posterior
wall fractures, this should be sought carefully. In a
few instances, impacted fragments of articular sur-
face have been found at operation (Fig. 5.6) and
even retrospectively these cannot be detected on the
radiographs. In general, however, the evidence is
available:

Accompanying a perfect reduction, segments of

Fig. 5.7. Posterior wall fracture with marginal impaction. In this
antero-posterior view the fragment is adjacent to the roof out-
line after reduction of the dislocation
varying length which create a dense curved line
and which should be concentric with the femo-
ral head are seen to be displaced. In Fig. 5.7 the
displaced line is visible above the roof of the ac-
etabulum, and in Fig. 5.8 two impacted frag-
ments are seen, in addition to the main detached
posterior wall fragment, these being driven
posteriorly into the cancellous bone of the
posterior column and appearing superimposed
on the femoral head.
Sometimes reduction of the femoral head is in-
complete, so that it still lies slightly posterior,
74 Posterior Wall Fractures

Fig. 5.8A-C Posterior wall fracture with margmal impactIOn.

A Antero-posterior view, a diagram, B iliac-oblique view, C ob-
turator-oblique view

,.-.- .- ...... i
1" ).
i '- ' -'j
i \
.
I
\.
I
\
\ \
\ i
i
a "' /
A f

Fig. 5.9. Scheme of pos-

C tero-superior fracture
Postero-superior Fractures
and it is seen to be related to a dense line of vari-
able length which is perfectly concentric with its
surface and separated from it by a clear space.
The clear space is created by the radiolucent ar-
ticular cartilage lying on a detached fragment
which has become impacted. Its bony margins
are not visible (Fig. 5.10).
5.2 Postero-superior Fractures
Incidence: 26 out of 223 fractures (11.60;0), of
which nine exhibited marginal impaction.
5.2.1 Morphology
The detached fragment comprises the postero-
superior sector of the acetabulum (Fig. 5.9). The
lower part of the posterior wall always remains in
place, and sometimes nearly the whole. Varying
amounts of the roof are included in the detached
fragment, this being the characteristic feature of
75
Detached
fragments
1
i
\
'\.,
".-._.">.,
i
i
a i
Fig. 5.10. Postero-superior fracture with marginal impaction.
The roof has been almost totally detached in several fragments
postero-superior fractures. Also, the detached frag-
ment takes part of the lower outer surface of the
ilium. The cavity created by the fracture as seen
from the outer aspect can vary in size and position.
The detached fragment may be in one piece
(seven cases), in two or more pieces (ten cases), or
include marginal impaction (nine cases).
5.2.2 Radiology
Distinct differences are seen in comparison with the
previous typical posterior wall fracture.
(a) Antero-posterior view. Only four landmarks are
intact: the ilio-ischial line, the teardrop, the
pelvic brim line and the anterior border of the
acetabulum (Fig. 5.11). It is the roof of the ace-
tabulum which is seen to be involved, but the
detached fragment appears as before like a cap
over the femoral head. The amount of the roof
contained by the fragment varies according to
how far forward the fracture line reaches; usual-
ly part of the roof remains visible towards the
76 Posterior Wall Fractures

Fig.S.llA-C. Postero-superior fracture. A Antero-posterior

radiograph, a diagram (note the thin outline of the roof), B ob-
turator-oblique view, C iliac-oblique view

Outline of roof +.~~~

still visible
but faint

'\
\
a \

C
B
Postero-superior Fractures
'\ .- ...
.1(,\ \ •
•...! .I ' •
; J
i
i
\ Fig, S.12A,B. Pure superior fracture taking the anterior part of the iliac wing.
b A Scheme, B amero-posterior radiograph, b diagram
medial side of the acetabulum. Only the upper
part of the posterior border of the acetabulum
is seen to be involved.
In some cases, the line of the roof of the
acetabulum remains in its proper place but ap-
pears very thin, continuing the curve of the
most internal portion of the roof which has its
normal density (Fig. 5.11 A). This effect is the
result of a fracture line which extends very far
anteriorly and leaves in place only the most for-
ward portion of the roof.
An extreme example, which we have seen on
only two occasions, is that in which the roof is
totally detached, this amounting to a pure
superior fracture completely above the bound-
ary of the posterior border and posterior wall
but involving the upper part of the anterior lip
of the acetabulum (Fig. 5.12). It can be seen
77
from the diagram that the detached roof takes
with it the anterior part of the iliac wing but the
pelvic brim remains intact. We believe that this
rare fracture must be included with the postero-
superior group because, in contradistinction to
typical fractures of the anterior column, so
much of the roof is detached and the pelvic
brim is undisturbed. It is reasonable to regard
this interesting example as a transition between
postero-superior fractures and fractures of the
anterior column.
A postero-superior fracture can also entail
marginal impaction (Fig. 5.10).
(b) Oblique views. These confirm the injury to the
roof but do not supply any further notable
detail. It is always on the obturator-oblique
view that the size of the detached fragment can
be assessed (Fig. 5.11 B).
78 Postenor Wall Fractures

Fig.S.13A-c' Postero-infenor fracture. A Scheme, B antero-

posterior radiograph, C IlIac-oblIque radiograph

B C
Special Forms of Posterior Wall Fractures
5.3 Postero-inferior Fractures
Incidence: 5 out of 223 fractures (three single frag-
ment, two multi fragment).
These fractures are rare and their anatomical
reduction sometimes offers considerable difficulty
because of their low position on the edge of the
acetabulum. We isolated them as a special group in
1962.
5.3.1 Morphology
With respect to the articular surface, the detached
fragment includes the lower part of the posterior
wall and the posterior horn. As a result of the con-
struction of the bone in this region, the split in the
innominate bone separates a fragment which in-
cludes the sub-cotyloid groove, the upper part of
the ischial tuberosity, and the ischial spine.
Tracing the surface marking of the fracture
(Fig. 5.13A), this passes along the shallow groove
under the posterior horn of the articular surface
descending towards but not reaching the obturator
foramen. It then curves to reach the ischial tuberos-
ity at a variable level and returns on the inner
aspect of the bone to reach the greater sciatic
notch. Finally it crosses the retro-acetabular sur-
face to reach the posterior lip of the acetabulum.
In a single case, we have observed an isolated
detachment of the posterior horn of the acetabu-
lum articular surface taking with it a portion of the
upward-sloping surface of the sub-cotyloid groove.
5.3.2 Radiology
(a) Antero-posterior view. In typical cases the pos-
teriorly dislocated femoral head is accompanied
by a large fragment of bone which remains con-
gruent with its infero-medial aspect and carries
a portion of the articular surface. The fragment
includes the sub-cotyloid groove and the upper
pole of the ischial tuberosity, both being
recognised by their typical configurations
(Fig. 5.13 B). The roof, the teardrop and the
pelvic brim are intact as in all posterior frac-
tures, but conspicuous within the outline of the
true pelvis are seen the ischial spine together
with a segment of the anterior border of the
greater sciatic notch, which are parts of the
detached fragment. Sometimes the fracture line
can be seen passing under the ischial spine
79
where it cuts the posterior border of the ischi-
um (Fig.5.13B).
(b) Obturator-oblique view. The fracture of the
ischial tuberosity is confirmed and also the fact
that the obturator ring, while damaged, re-
mains in continuity.
(c) Iliac-oblique view. The points are displayed at
which the fracture line crosses the posterior
border of the innominate bone, that is, at the
levels of the greater sciatic notch above and the
ischial tuberosity below.
5.4 Special Forms
of Posterior Wall Fractures
5.4.1 Extended Posterior Wall Fractures
In this variant, the detached fragment with artic-
ular surface has been in several pieces; it included
the most posterior part of the roof, the posterior
wall, and a segment from the upper pole of the
ischial tuberosity. This has been seen in 13 in-
stances and in each the anterior border of the
greater sciatic notch was either not disturbed or
merely cracked (Fig. 5.14).
5.4.2 Horizontal Extension of Fracture Line
The fracture extends horizontally and takes with it
the whole retro-acetabular surface and the anterior
border of the greater sciatic notch. Sometimes a
fragment from the greater sciatic notch margin
becomes isolated (Fig. 5.15).
5.4.3 Massive Posterior Wall Fractures
In two remarkable cases there was an enormous
posterior fragment comprising in one piece the
posterior wall, the whole retro-acetabular surface,
the angle and the anterior border of the greater
sciatic notch, the ischial spine itself, and in one
case the superior pole of the ischium. These large
fragments amount to a significant part of the
posterior column and could be considered amongst
this latter group of fractures, namely partial
superior fractures of the posterior column. We have
preferred to include them amongst special fractures
of the posterior wall in order not to disturb the
homogeneity of fractures of the posterior column.
80 Posterior Wall Fractures

/ ~\'
I ..\
I \
\.
\.
\
a \
Fig. 5.14A-C. Extended posterior fracture. A Antero-posterior
radiograph, a diagram, B iliac-oblique radiograph, C obturator-
oblique radiograph c

Fig.5.15A-C. Extended posterior fracture taking the anterior a diagram, B obturator-oblique radiograph, C iliac-oblique
border of the sciatic notch. A Antero-posterior radiograph, radiograph, c diagram
Special Forms of Posterior Wall Fractures 81

I.'-. .~
._ ./( \
) i
i
j
I
/
Fragment which
corresponds to site of
break in outlin e of
g reater sciatic

) . . . . .,/z:u.

' .\
\
\
c c
82
b posterior wall fragment
Further, the most massive posterior wall fracture
which we have seen (Fig. 5.16) spared the greater
part of the ilio-ischialline, except perhaps in a very
small degree in its uppermost part, and its relation-
ship with the teardrop was not disturbed, in con-
trast with what is seen in true posterior column
fractures. In addition, the femoral head dislocated
posteriorly in these two cases, and not centrally, as
it does in posterior column fractures. These mas-
sive posterior wall fractures are transitional with
respect to posterior column fractures.
5.4.4 Posterior Wall
and Incomplete Transverse Fractures
Incidence: 29 out of 223 posterior wall fractures
(13070 ).
In this fracture type, in addition to a typical
posterior wall fracture with all its characteristics on
the three views, there is a fracture line starting from
the anterior border of the notch created in the
Posterior Wall Fractures
Fig. 5.16A, B. Very extended posterior fracture. A Scheme,
B antero-posterior radiograph, b diagram. The femoral head is
posteriorly dislocated and has lost all its relationship to the
posterior wall by the posterior wall fracture. This
fracture extends anteriorly like a transverse frac-
ture, transgressing the ilio-ischial line on the ante-
ro-posterior view and the anterior border of the
greater sciatic notch on the iliac-oblique view.
However, the ilio-pectineal line (the pelvic brim)
appears intact on the antero-posterior and obtura-
tor-oblique views; the anterior column is therefore
undisturbed (Fig. 5.17). We call this additional
fracture line an "incomplete transverse" fracture,
which is what it is. Usually the fracture line is un-
displaced or only very slightly displaced. It may
show a little displacement which is an internal rota-
tion of the inferior fragment, which has to be re-
duced in order to restore perfect congruency to the
posterior wall.
We have observed the same additional incom-
plete transverse fracture line in a case of postero-
inferior fracture (Fig. 5.18). However, in this case
the fracture line extended up to the pelvic brim, and
due to its slight displacement produced at that level
a small deformity rather like a greenstick fracture.
 Special Forms of Posterior Wall Fractures 83

Fig. 5.17 A-C. Posterior wall and incomplete transverse frac-

ture. A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph. The hemitransverse compo-
nent is clearly seen on the iliac-oblique view, where it cuts the
mid-part of the greater sciatic notch, and on the antero-posteri-
or view, where it interrupts the ilio-ischial line. This fracture
quite often escapes CT scanning unless the cuts are thin

B c
84 Posterior Wall Fractures

,..- ,_ ./
/'---.
/
i
i
\
".

\
\
\
c b

Fig. S.18A-C. Extended postero-inferior fracture with trans- brim. A Scheme, B antero-posterior radiograph, b diagram,
verse element crossing and kinking the outline of the pelvic C iliac-oblique view
CT Study of Posterior Wall Fractures
5.5 CT Study of Posterior Wall Fractures
CT allows a comprehensive study of the bone frag-
ment, as it helps one to assess:
its size, which increases from the uppermost sec-
tion in which it appears, up to a maximum, and
then decreases again until it disappears;
its degree of posterior displacement, cranial dis-
placement (it often appears on sections which
still show an intact posterior wall), and rotation
around any axis;
the direction of the fracture line separating the
fragment, which is always oblique forwards and
outwards, nearly always at 45°;
the amount of articular surface it brings with it;
the exact site of origin: CT demonstrates the up-
per and lower limits of the gap created by the
fracture in the posterior wall;
the position of the femoral head; CT can show
perfect reduction, or persistent posterior dislo-
cation or subluxation, or an eccentric position
of the head due to an incarcerated fragment;
the presence of a marginal impaction, which
appears as a segment of the articular surface
85
of the posterior wall that has rotated to a greater
or lesser extent externally, sometimes so far as
to attain an antero-posterior orientation
(Fig. 5.19). This marginal impaction may be
fragmented into several pieces. Sometimes a
densification of the bone beneath the fragment
can be seen, indicating the crushing of the un-
derlying cancellous bone.
the presence of an intra-articular osteochondral
fragment, which may be completely free and
located within the cotyloid fossa (Fig. 5.20), or
may be attached to the capsule, in which case it
remains in the upper joint space and does not
reach the cotyloid fossa;
an avulsion fracture from either end of the liga-
mentum teres insertions.
An associated incomplete transverse fracture
may escape detection by CT if the sections are not
very thin, as this fracture line is often horizontal.
The displacement and origin of a posterior wall
fracture are perfectly shown in 3-D reconstruction
(Fig. 5.21).
 86 Posterior Wall Fractures

A B

c E

Fig. S.19A-E. A postero-superior wall fracture, showing a tions. D shows that part of the subchondral bone of the roof
marginal impaction in A and B. A Antero-posterior, B ob- belongs to the fragment; E exhibits a marginal impaction with
turator-oblique and C iliac-oblique radiographs. D, E CT sec- an antero-posterior orientation
CT Study of Posterior Wall Fractures 87

A B

D,E

C F,G
Fig.S.20A-G. Posterior wall fracture with incarcerated frag- fragment which is, moreover, included within the fracture line of
ment. A-C Multiple fragmentary fractures ofthe posterior wall the femoral head. F,G Magnetic resonance images showing sat-
associated with an infero-medial segmental fracture of the head. isfactory vascularization of the segmental fracture of the head,
D, E CT sections demonstrating the incarceration of an articular which is shown very clearly
88 Posterior Wall Fractures

A B

c E

Fig. 5.21 A-E. A mal-united posterior wall fragment, 4 months after injury. A - C Radiographs, D, E the 3-D reconstruction
6 Fractures of the Posterior Column

It was in 1959 that we identified, from the amor- Table 6.1

phous general diagnosis of central dislocations of
the hip, the fractures of the posterior column; this Fractures of the posterior column 30 3.190/0
was the earliest development of the column classifi- Typical total 15 1.60070
cation which we promoted. Other writers had rec- Extended (detaching teardrop) 7 0.740/0
ognised that there were some with special features. Partial superior 2 0.210/0
Thus KNIGHT and SMITH (1958) likened the left Partial inferior 3 0.310/0
Epiphyseal separation 3 0.310/0
acetabulum, seen from the outer aspect, to a clock
and described fracture lines disposed vertically and
from two to six o'clock. These included fractures of
the posterior column but they did not emphasise As seen from the inner aspect, the fracture line
their individual character nor describe their radio- leaves the angle of the greater sciatic notch and
logical characteristics. TANTON (1916) reported an descends across the quadrilateral surface at first
experimental fracture of the ischium, performed by obliquely downwards and forwards, and then ver-
WALTHER (1891), which corresponds to a pure tically. It follows a path separate from the innomi-
fracture of the posterior column (Fig. 6.1). nate line which is not transgressed and reaches the
obturator foramen behind the anterior wall of the
obturator canal. The separated fragment of the
posterior column is limited by the posterior border
of the innominate bone with the two sciatic notches
6.1 Typical Posterior Column Fractures and the ischial spine.

6.1.1 Morphology

The whole posterior column is detached in one

fragment (Fig. 6.2).
On the outer surface of the innominate bone the
fracture line begins near the summit of the angle of
the greater sciatic notch, detaching more or less of
the curved dense trabeculae present here. It de-
scends obliquely downwards and outwards across
the retro-acetabular surface, reaching and splitting
the lip of the acetabulum behind the roof sector.
From this point, the line crosses the articular sur-
face at the junction of the roof with the posterior
wall segment and then crosses the cotyloid fossa in
its most posterior part so that the components
forming the teardrop are not disturbed. The op-
posite side of the obturator ring is fractured at a
variable point along the ischio-pubic ramus, most
usually in its middle part; we have one example in Fig.6.1. Experimental fracture of the posterior column pro-
which the ischio-pubic ramus was not fractured. duced by WALTHER (1896), reported by TANTON (1916)
90 Fractures of the Posterior Column

Fig. 6.2. Scheme of typical posterior column fracture

In addition to the main fracture line, in the up-

per part there may be some comminution and sepa-
ration of tiny bone fragments not involving the ar-
ticular surface.
In typical fractures the thick part of the roof is
left intact, but occasionally there is some impaction
of one or two fragments from the inner thin margin
of the roof which seem to have been displaced by
the head during its postero-medial displacement.
A
The posterior column in these fractures is driven
inwards and posteriorly. In most cases the head
follows it and remains congruent with the posterior
wall and the posterior horn of the articular surface;
27 examples of the 30 posterior column fractures
could be regarded as centrally dislocated. Of opera-
tive importance is that the posterior capsule is nor- ____ Posterior border
mally attached to the posterior column fragment. of aceta butum
.'<"::=... . ..".
\.
6.1.2 Radiology \
" \. i
(a) Antero-posterior view (Figs. 6.3, 6.4, 6.5, 6.6A). \ /
The femoral head appears displaced inwards I"'
I
i I
and often encroaches upon the outline of the
pelvic inlet. It gives the clear notion of having a \. \
driven medially before it the massive bony frag- Fig.6.3. Typical fracture of the posterior column. A Antero-
ment, which carries the posterior horn of the posterior radiograph, a diagram. The other standard radio-
articular surface, and on which one sees more graphs are not l1vailable for this early case (1956)
or less clearly the ilioischial line. The posterior
fragment is limited above by a fracture line
crossing the posterior wall, sometimes very
the ischiopubic ramus is variable. Usually, at
clear. On other occasions the fracture line is in-
the innermost limit of the displaced fragment,
dicated only by an interruption of the posterior
the outline of the ischial spine can be seen.
border of the acetabulum, and the superior
limit of the fragment cannot be detected Alongside these abnormal appearances the
precisely. Below, the position of the fracture in following features should be noted:
Typical Posterior Column Fractures 91

A
B

Posterior border

'-',\
of acetabulum
lIio-ischi al line Anterior border
. - .{ r . \ 01 acetabulum
.
1~ . -'t,
i/> \
\ \
./ \
( j
I I
I I
I b
I
a

Fig. 6.4A,B. Fracture of the posterior column. A Antero-posterior radiograph, a diagram, B iliac-oblique radiograph, b diagram

The roof of the acetabulum maintains its
normal density. It is intact and in its proper
position.
The integrity of the ilio-pectineal line from
the sacro-iliac joint to the pubis is funda-
mental in establishing that the anterior col-
umn is intact.
The anterior border of the acetabulum is in-
tact and its outline with the characteristic
notch is abnormally clear due to the inward
displacement of the posterior column which
it normally overshadows.
The teardrop is intact and it maintains its
normal relationship with the ilio-pectineal
line. In contrast the ilio-ischial line is dis-
placed inwards with respect to the teardrop.
 92 Fractures of the Posterior Column

A B c
Fig. 6.SA-C. Typical fracture of the posterior column. A An- in red. A small additional fragment is visible on the obturator-
tero-posterior view, B obturator-oblique view, C iliac-oblique oblique view
view. Intact landmarks are outlined in black, broken landmarks

Fig.6.6A,B. Epiphyseal separation of

the posterior column with posterior
dislocation: A immediate radiograph
which is between a true antero-
posterior and an obturator-oblique
view; B antero-posterior view after
A B reduction of the posterior dislocation
Extended Posterior Column Fractures
Thus, all the radiological landmarks of the
anterior column are seen to be undisturbed.
Like the roof, the anterior column is completely
spared in the typical posterior column fracture.
In summary, one recognises the typical pos-
terior column fracture on the antero-posterior
view, from the massive fragment which includes
the ilio-ischialline pushed inwards by the head,
the integrity of the ilio-pectineal line and its
normal relationship to the teardrop, while the
ilio-ischialline is displaced inwards with respect
to both these structures.
In the majority of instances (27 out of 30
cases), the displacement of the dislocated
femoral head is predominantly inwards, i.e. the
appearance is one of central dislocation. In
three cases the femoral head lay slightly dis-
placed with respect to the iliac wing and
acetabular outline, in an outward and upward
direction; in these examples it was always adja-
cent to the posterior horn of the articular sur-
face of the acetabulum; the obturator-oblique
view confirmed that the displacement was
essentially posterior. We do not speak of
posterior subluxation, for at operation the
posterior capsule is not detached from the
posterior column.
The antero-posterior view may also show:
one or several tiny fragments detached from
the upper part of the fracture line;
possible impaction of the inner thin part of
the roof;
the site of fracture often visible at the upper
part of the posterior column. This is
brought about by the inward displacement
of the column which subtracts from the den-
sity of this part of the innominate bone and
causes it to appear much more transparent
in an area more or less triangular, situated
just outside the ilio-pectinealline (Figs. 6.3,
6.8);
the outline of the cotyloid fossa which
sometimes appears duplicated because its
two segments, from the anterior and posteri-
or columns, respectively, are only slightly
displaced or tilted with respect to each other.
(b) Obturator-oblique view. Occasional posterior
displacement of the head is demonstrated
(Fig. 6.6A). Above all, this view confirms with
great clarity the integrity of the ilio-pectineal
line and of the anterior column.
93
The exact position of the fracture of the
ischiopubic ramus can be seen clearly as also
the level of the break in the posterior border of
the acetabulum. Most frequently the latter is
high but exceptionally it is low, the direction of
the fracture line in the back of the acetabulum
then being very oblique; we have seen this in
only one case. The roof is confirmed as being
intact.
(c) Iliac-oblique view. There are two important
features (Figs. 6.4B and 6.5 C): Firstly, the inter-
nal and superior boundaries of the displaced
fragment are delineated clearly. Internally, it is
limited by the posterior border of the pelvic
bone extending to a variable point along the
greater sciatic notch; frequently, the fragment
includes the angle of the greater sciatic notch
but occasionally the angle is not detached with
it. In the latter instance the angle remains in its
proper place and the upper part of the fracture
line, instead of being directed obliquely up-
wards and inwards, becomes more horizontally
disposed in its inner part and cuts the anterior
border of the greater sciatic notch at a variable
level.
Secondly, the iliac-oblique view demon-
strates that the anterior border of the acetabu-
lum, of which the outline is fairly clear, is in-
tact.
Note that the roof is intact, only its internal
part occasionally exhibiting impaction, and
that the iliac wing, the iliac crest and the anteri-
or border of the ilium are uninjured.
6.2 Extended Posterior Column Fractures
In seven of our 30 posterior column fractures the
appearance is modified because the teardrop re-
mains attached to the posterior column, being
displaced with it and with the ilio-ischial line
(Fig. 6.7).
6.2.1 Morphology
These fractures approach the limit of the posterior
column group and an enormous fragment of bone
is detached. Viewed from the outer side the fracture
line always includes the angle of the greater sciatic
notch; in one example it actually included the
94
Fig. 6.7. Scheme of extended fracture of the posterior column
whole superior border of the notch. It descends
towards the acetabulum sometimes as in the typical
form but tends to pass higher on the articular sur-
face and detach the posterior part of the roof with
the posterior column. The essential modification
of the fracture line which creates this subgroup oc-
curs at the level of the cotyloid fossa: instead of
dividing the area vertically in its more posterior
part, the fracture line descends in the anterior part.
On the medial surface of the innominate bone,
the fracture line is situated very far forward, and it
skirts the pelvic brim (to which it is almost tangen-
tial) and then descends to the obturator foramen in
such a manner that the posterior part of the exter-
nal wall of the obturator canal is taken with the
posterior column fragment.
Thus all of the elements which form the radio-
logical teardrop are detached with the posterior
column and the remaining innominate bone is very
thin at approximately the mid-part of the ilio-
pectineal line. The ischiopubic ramus is broken at
a variable site.
6.2.2 Radiology
(a) Antero-posterior view. It is a little difficult to
recognise the fracture in this view (Fig. 6.8 A)
because one sees both the teardrop and the ilio-
ischial line on the fragment which has been
driven inwards by the femoral head; they may
preserve their normal relationship or exhibit
slight separation. In contrast, the loss of the
relationship of the teardrop with the ilio-pec-
tineal line is very manifest. The pelvic brim
seems to be intact, for its curve is regular; how-
Fractures of the Posterior Column
ever, it has lost radiological density in its middle
part and this can make its integrity difficult to
confirm. The roof is always a little impacted in
its inner part. In some very extreme examples of
this group, the column takes with it the hind
part of the roof and this segment, having lost its
normal relationship with the segment remain-
ing on the anterior column, forms with it an im-
age like a gull in flight (Fig. 6.9 A, a). (This ap-
pearance may be seen in the oblique views as
well.) The rupture of the posterior border of the
acetabulum is often very high. The ischio-pubic
ramus is broken, as before, at a variable site.
The oblique views are indispensable in diag-
nosing this fracture.
(b) Obturator-oblique view. This confirms
(Fig. 6.8 B) the integrity of the ilio-pectinealline
and the anterior border of the acetabulum, thus
demonstrating the preservation of the anterior
column. The thinning in the region of the pelvic
brim visible on the antero-posterior view com-
pletely disappears on the obturator-oblique as
this view is especially valuable in displaying the
sloping upper segment of the pelvic brim (Figs.
6.8 B, 6.9 B), which is always intact in these
cases.
(c) Iliac-oblique view. This confirms (Fig. 6.8 C)
the extended character of the upper limits of the
fracture which sometimes includes the superior
border of the greater sciatic notch.
6.3 Atypical Posterior Column Fractures
6.3.1 Other Associated Pelvic Ring Fractures
In three patients, there has been an anterior, ver-
tical fracture of the innominate bone on the same
side (one case) or both sides (two cases). The frac-
ture of the superior pubic ramus (Fig. 6.9) may ap-
pear difficult to explain but it should be observed
that this lesion was totally extra-articular, did not
constitute an element of the fractured acetabulum
itself, and is probably linked to an associated pelvic
injury.
6.3.2 Epiphyseal Injury
We have seen three examples of an epiphyseal frac-
ture-separation of the posterior column, two in
boys of 12 and 14 years (Fig. 6.10) and the other in
a girl of 10 years.
lransitional Posterior Column Fractures 95

c
Fig. 6.8A-C. Extended fracture of the posterior column. A An-
tero-posterior radiograph, a diagram, B obturator-oblique
radiograph, b diagram, C iliac-oblique radiograph. The teardrop
a is detached with the posterior column. The middle segment of
the pelvic brim is thinned, and almost appears to be broken, but
it is shown in B to be intact (this view best demonstrates this
landmark)

B b
96 Fractures of the Posterior Column

...- ..........
("- ......
I
.i \.\
.
.
I
\.
. )
f / c
i
i
a
i
i

Fig.6.9A-C. Extended fracture of the posterior column.

A Antero-posterior radiograph, a diagram, B obturator oblique
radiograph, C iliac-oblique radiograph, c diagram. In this case
there is an associated fracture of the superior pubic ramus which
could cause difficulty in interpretation; with the fracture of the
inferior ramus, an essential component of the posterior column
fracture, 1l resembles a vertical fracture through the obturator
ring
 Transitional Posterior Column Fractures 97

B
Fig. 6.10A-C Epiphy eal separation of the posterior column.
A Amero-posterior radiograph of pelvis, 8 obturator-oblique
radiograph, C iliac-oblique radiograph c
98 Fractures of the Posterior Column

Reduplication of
ilio-isc hial line

Fig. 6.11A-C. Partial superior fracture of the posterior column b diagram, C iliac-oblique radiograph (taken after post-traumat-
(transitional form). A Scheme, B Antero-posterior radiograph, ic osteoarthrosis had been established)
Transitional Posterior Column Fractures
Fig. 6.12A,B. Pseudarthrosis of ischium at the base of the pos·
terior column. A Iliac·oblique radiograph, B tomograph
A
6.4 Transitional Posterior Column Fractures
6.4.1 Partial Superior Fractures
We have seen two high partial fractures of the pos-
terior column, transitional with respect to posterior
wall fractures and therefore meriting special
description.
6.4.1.1 Morphology
The fracture line detaches the upper part of the
posterior column, taking the angle of the greater
(b) Oblique views. We have only the iliac-oblique
sciatic notch, and descends across the hind part of
the quadrilateral surface so as to detach a large
fragment of the ischial body not involving the tu-
berosity but including the ischial spine
(Fig. 6.11 A).
6.4.1.2 Radiology
(a) Antero-posterior view. A reduplication of the
ilio-ischialline is created by the vertical fracture
line crossing and displacing the quadrilateral
surface, approximately along the vertical axis of
99
the ilio-ischial surface (delimited by the dotted
lines in Fig. 6.11 A). The two separate parts of
the ilio-ischial surface as displaced in a parallel
fashion produce two ilio-ischial lines on the
antero-posterior view. The reduplication of the
outline of the roof, of which the posterior seg-
ment has accompanied the displaced fragment
and has hinged inwards, creates with the un-
disturbed segment an image like a gull in flight
(Fig. 6.11 B). Below, the inferior angle of the
detached fragment appears in the superoexter-
nal quadrant of the obturator foramen.
view, this having been taken when an osteoar-
throsis of the hip had already developed.
Degenerative changes occurred in spite of slight
displacement of the original fracture, which,
however, created a real roof-head incongruency.
This case is an example of a transition be-
tween the fractures of the posterior column and
the extended fracture of the posterior wall,
which takes with it the anterior border of the
greater sciatic notch but does not disturb the
ilio-ischial line or the obturator foramen.
 100 Fractures of the Posterior Column

A B

Fig. 6.13A-F. Extended right posterior column fracture detach-

ing with it the teardrop associated with a fracture of the superior
pubic ramus. A-C The three standard views. D-F Three CT
sections: D above the roof, E through the roof, F through the
cotyloid fossa. The upper part of the fracture line is at an angle
of about 50° inwards and forwards. The posterior column is in-
ternally rotated around a vertical axis. A T-shaped fracture was
c present on the other side
 CT Study of Posterior Column Fractures
-
I
""-
..... fj-:- .. "
~...
o ~
E f
6.4.2 Partial Inferior Fractures
We have operated on a pseudarthrosis of the
ischium (Fig. 6.12) which was associated with no
displacement of the femoral head. This fracture
followed a direct blow to the ischium which was
clearly the site of Paget's disease. The fracture line
crossing the ischium in its upper part was of the
same character as those separating the posterior
column and cutting the greater sciatic notch in its
middle part; it displaced the posterior part of the
articular surface together with the lower fragment.
CAMPANACCI (1967) has reported three similar
cases but these were not treated operatively. This
fracture is in fact a partial fracture of the posterior
column, involving only its inferior part, the ischio-
pubic ramus being fractured in its middle part and
the upper part of the column remaining undis-
turbed. Since the first edition of this book we have
operated on two more instances of these partial in-
ferior fractures of the posterior column, making a
total of three.
6.5 CT Study of Posterior Column Fractures
In the typical cases of this fracture type, the frac-
ture appears on the section involving the roof of
101
the greater sciatic notch or just below it. The frac-
ture line is then followed, in continuity, up to the
ischio-pubic ramus, which is broken in its mid-part.
It is striking that, in the typical case, the orienta-
tion of the fracture line is roughly coronal or fron-
tal, all along its track. From its starting point the
fracture line is more and more anteriorly situated,
dividing the roof section just behind the sub-chon-
dral compact bone area and then crossing the
cotyloid fossa in its mid-part, always with a frontal
orientation.
Displacement of the posterior column, which is
most often followed by the femoral head, is very
easy to analyse: the fragment displaces backwards
but also rotates internally along a vertical axis.
In the extended type of these fractures
(Fig. 6.13), the fracture line passes through the
compact sub-chondral bone area of the roof, tak-
ing with it a good part of the latter.
Above the roof and through it the fracture line
is not coronal but oblique forwards and inwards at
about a 45°- 50° angle (notice it is the opposite
direction to the fracture line detaching the posterior
wall fractures), becoming, however, coronally ori-
ented at the level of the cotyloid fossa. Through the
fossa the fracture line skirts very closely the inner
border of the anterior wall, so the posterior frag-
ment takes with it all the area responsible for the
teardrop on the antero-posterior view.
7 Anterior Wall Fractures
These are the counterparts of the posterior wall
fracture and to our knowledge had not been de-
scribed as an entity until we separated them in
1968. They are uncommon (Thble 7.1) but their ex-
istence is significant in that they complete the range
of fractures which can occur, disposed like a fan
around the boundaries of the acetabulum.
Fractures of the anterior wall of the acetabulum
result in a separation of the anterior part of the ar-
ticular surface together with a large part of the
middle third of the anterior column.
7.1 Morphology
Seen from the outer side (Fig. 7.1), the fracture line
begins at the anterior border of the acetabulum a
little below the anterior inferior iliac spine. It
crosses the articular surface detaching the anterior
wall with a small, variable amount of the anterior
part of the anatomical roof of the acetabulum. It
then descends across the cotyloid fossa to reach the
ischio-pubic notch a little behind the anterior horn
of the articular surface. Traversing the obturator
foramen the fracture line cuts the superior pubic
ramus obliquely forwards and inwards.
On the internal surface the fracture line, leaving
the anterior border of the acetabulum, is directed
backwards and inwards, reaching the ilio-pectineal
line 3 - 4 cm in front of the sacro-iliac joint. It
crosses the ilio-pectineal line and then descends
nearly vertically to reach the upper border of the
obturator foramen at the level of the ischio-pubic
notch, rejoining the fracture line dividing the outer
aspect of the bone.
The detached fragment is seen to be trapezoidal,
involving much of the upper surface of the superior
pubic ramus. The longer parallel side is from the
pelvic brim and the shorter is the anterior border of
the acetabulum. The detached fragment does not
include the medial part of the superior pubic
Table 7.1.
Fractures of anterior wall 18 1.910/0
Pure anterior wall 12 1.27%
Anterior wall and anterior column 6 0.63%
ramus, the pubic angle, or the anterior inferior iliac
spine, so leaving undisturbed a large part of the
anterior column.
Sometimes the typical form described is modi-
fied: (1) The detached fragment may be split along
a line which, with respect to the longitudinal axis
of the superior face of the fragment, may be direct-
ed transversely or longitudinally. (2) In 16 examples
of the 18, there was a bony plate of varying size as-
sociated with the typical detached fragment and
separated from the cotyloid fossa and the
quadrilateral surface. It was limited anteriorly by
the principal fracture line below the pelvic brim,
and above and below by splits running horizontally
and more or less parallel across the quadrilateral
surface. The segment, which comprises the floor of
the cotyloid fossa and part of the quadrilateral sur-
face, was seen to be deflected and driven in by the
femoral head. The posterior column was most
often not disturbed. The segment in all cases main-
tained an osseous hinge situated posteriorly at a
variable distance from the anterior border of the
greater sciatic notch. We thought for a long time
that this bony plate did not bear any articular sur-
face, but CT has shown that the posterior hinge
may involve the innermost part of the posterior
wall (see Fig.7.6E,F); in these instances it is of
course useful to achieve perfect reduction of this
plate of bone.
It is easy to understand that the displacement of
the anterior wall itself does not suffice to allow the
head to penetrate the pelvis, and that in its inner
displacement the head has also to detach a part or
the whole of the cotyloid fossa together with a
104 Anterior Wall Fractures

Fig. 7.1. A Scheme of anterior wall fracture. B Horizontal sec-

tions of an anterior wall fracture associated with a plate of bone \
from the cotyloid fossa and the quadrilateral plate involving the \. ,r · ..... · '
inner border of the posterior wall. 1 Before displacement, 2 after '-(_ ./ \
displacement )
\ \
r \
,

i I
large part of the quadrilateral surface, this plate of i /
bone usually keeping a posterior hinge, but it may
/
I
be free (Fig. 7.1 B). i
a

Fig. 7.2A-C. Anterior wall fracture. A Antero-posterior radio-

7.2 Radiology
The integrity of the ilio-ischialline and the posteri-
or border of the acetabulum seen on the antero-
posterior view, together with the intact posterior
border of the innominate bone as seen on the iliac-
oblique view, confirm that the posterior column is
intact.
The fracture involves part of the anterior col-
umn, but never the whole.
graph, a diagram, B obturator-oblique radiograph, b diagram,
C iliac-oblique radiograph, c diagram. The detached fragment
has rotated about the fracture of the superior pubic ramus and
the internal segment of the roof is impacted into the subjacent
spongy bone
7.2.1 Antero-posterior View
The anterior inferior iliac spine and the angle of
the pubis are spared (Figs.7.2A, 7.3A and 7.4A).
The only radiological landmarks broken are:
Radiology 105

B C

.1-',
\
--._.-("'\ '-
'1

\ \ ')
I \
I

,
J I
i
i \
b i c

Fig.7.2B-C
 106 Anterior Wall Fractures

Fig. 7.3A-C. Anterior wall fracture. A Antero-posterior radio-

graph, a diagram, B obturator-oblique radiograph, C iliac-
oblique radiograph (taken 6 months after injury) C
Atypical Examples
the anterior border of the acetabulum, generally
in its upper third;
the ilio-pectineal line in its middle part, and
always at two points: the posterior one divides
the posterior half of the pelvic brim, while the
anterior one breaks the superior pubic Tamus,
which is generally crossed more or less frontally.
These two points of breakage of the pelvic brim
delineate a segment of the ilio-pectineal line
which forms the larger parallel side of the
trapezoid-shaped fragment so characteristic of
the anterior wall fracture.
The teardrop is sometimes invisible (as in two of
our examples) but in most cases it is displaced in-
wards with respect to the ilio-ischial line and is
tilted together with the fragment of the pubic
ramus to which it nearly always remains related;
sometimes, however, it forms an isolated fragment
or remains attached to the accessory plate-like frac-
ture formation involving the quadrilateral surface,
described above.
The fragment of the anterior wall may be split
transversely (Fig. 7.4A), as in two of our cases, or
longitudinally. A longitudinal split, if it is recognis-
able radiographically, appears as a duplication of
the fragment or of the pelvic brim (Fig. 7.5 A);
sometimes it is only discovered at operation.
The roof is always involved to some extent. Its
thin internal part is detached and, in one case out
of two, a more significant fragment involving the
inner third or half of the roof either remains at-
tached to the fragment and is displaced with it, or
is separated from it, and is in fact a marginal im-
paction, similar to the ones described in the poste-
rior wall group of fractures. In all cases, on the
antero-posterior view, the outer segment of the roof
was not disturbed and remained in place. The
region of the detached anterior wall is often clearly
visible as a radiolucent area situated just inside the
internal limit of the undamaged roof (Fig. 7.3). If
not evident on this view, it may be visible on the
iliac-oblique (Fig. 7.5 C).
The nature of the displacement of the fragment
can be estimated from the following features:
it may be displaced inwards in one piece by the
head, creating a full displacement of a segment
of the ilio-pectineal line (Fig. 7.3 A);
or it may be tilted by the femoral head, an angu-
lation appearing at the level of the superior
pubic ramus. Cooper's ligament acts as a stabi-
lising hinge at this point if it is not ruptured
(Figs. 7.2A and 7.4A).
107
The femoral head is dislocated anteriorly and is ex-
ternally rotated. Inwardly it overlaps the ilio-ischial
line.
7.2.2 Obturator-oblique View
The fracture of the anterior wall is unmistakable.
The view also confirms the integrity of the
posterior border of the acetabulum and shows the
site of fracture of the obturator ring to be at the
level of the ischio-pubic notch and the roof of the
obturator canal (Fig.7.2B). Above all it displays
perfectly the trapezoidal shape of the detached
anterior wall fragment; this is seen to be driven in-
wards by the femoral head. The extent of the ar-
ticular surface which it bears can also be assessed.
The amount of the roof detached with the frag-
ment is delineated and may appear as the inner seg-
ment relating to the anterior wall and continuing to
lie adjacent and symmetrically with the femoral
head. Alternatively the inner segment of the roof
may be detached, tilted, and can be impacted into
the spongy bone of the anterior column
(Fig.7.2B).
The obturator-oblique view is essential for as-
sessing fractures of the anterior wall.
7.2.3 lliac-oblique View
This confirms the integrity of the posterior border
of the innominate bone and displays very clearly
(in 12 cases out of 18) the commonly present elevat-
ed bone plate on the inner aspect of the pelvis
which is seen in section (Figs. 7.4 C and 7.5 C). It
establishes the point of rupture of the anterior wall
of the acetabulum and confirms the integrity of the
anterior inferior iliac spine together with the anteri-
or border of the ilium above the acetabulum
(Figs. 7.4C and 7.5C).
7.3 Atypical Examples
Unusual radiological appearances may be created
by the following features:
fracture lines which run longitudinally due to
splitting of the detached fragment and which
are only apparent near the ilio-pectineal line
and may be confused with it (Fig. 7.5 A);
the elevated bone plate from the quadrilateral
surface, which on the antero-posterior view may
 B

.;-
)
/
/ . .,. l.
j \
! <:" \
I ( -
\ . \
\ \ .
.(
\ / .
. I
i I j
i r
i I
a

Quadrilateral
/
... surface of
I the ischium
I
I
/
/
\
\
\
\
I
I
I
/
I
I
c , I
Fig.7.4A- C C
Atypical Examples 109

B
Fig. 7.5 A - C. Associated fracture of anterior column and ante-
" ./ rior wall. A Antero-posterior radiograph, a diagram, B obtura-
(-\r-'-'
,I I .
tor-oblique radiograph, C iliac-oblique radiograph, c diagram.
I ' The fracture line detaching the anterior fragment runs longitu-
i '. dinally adjacent to the pelvic brim, causing an appearance of
\ reduplication on the radiographs. A fracture lifting a plate of
\ bone from the quadrilateral surface is limited anteriorly by the
\ main fracture line, There is also a fracture of the ischio-pubic

"
a ramus

Fig. 7.4A-C. Anterior wall fracture. A Antero-posterior radio-

graph, a diagram, B obturator-oblique radiograph, C iliac-
oblique radiograph, c diagram. The extension of the fracture
lifting a plate from the quadrilateral surface of the ischium is
shown clearly in C. The anterior fragment is in two parts, both
applied to the head. The teardrop is not visible. There is an un-
displaced fracture of the inferior pubic ramus
110 Anterior Wall Fractures

Elevated cortex
of quadri lateral surface
limited anteriorly
by break in pelvic brim

8
 CT Study of Anterior Wall Fractures 111

c
Fig.7.6A-F. A pure anterior wall fracture. A-C The three
standard views. The main fragment is divided by a secondary
fracture line perpendicular to its long axis. A plate of bone is
elevated from the cotyloid fossa and the quadrilateral surface
and is obvious on the iliac-oblique view. The inner part of the
roof is impacted on the three views. D-F CT sections demon-
strating the separated anterior wall (D, E), which is subdivided
in section E, and the quadrilateral surface bone plate with a
posterior hinge involving the inner border of the posterior wall F

appear as a reduplication of the main fragment;
it becomes precisely delineated on the iliac-ob-
lique view (Fig. 7.5 C);
associated fractures of the ischio-pubic ramus
(six cases) without displacement in three cases
(Fig. 7.4) and displaced in three cases (Fig. 7.5),
so a displaced segment of the anterior column
lies between the fracture of the anterior wall and
the fracture of the ischio-pubic ramus. In fact,
this configuration forms an association between
a fracture of the anterior wall and a fracture of
the anterior column, this relationship being sim-
ilar to that between a posterior wall fracture and
a posterior column fracture described in
Chap. 11. We confine these cases to the group of
anterior wall fractures, partly in order to keep
the classification as simple as possible, and
partly because of the dominant anterior wall
element; also because they need the same surgi-
cal approach - fixation of the ischio-pubic
ramus fracture is never carried out.
7.4 CT Study of Anterior Wall Fractures
The CT scan confirms that the iliac wing, the pos-
terior column, the ischio-pubic ramus and the pub-
ic angle are intact. The only disturbed part is the
middle portion of the anterior column.
In Fig. 7.6 the fracture line detaching the ante-
rior wall and the quadrilateral surface plate is
112 Anterior Wall Fractures

Fig. 7.7A-G. An associated anterior wall and anterior column

fracture. A Antero-posterior view, B obturator-oblique view
demonstrating the trapezoidal anterior wall fragment, the tear-
drop making a separate fragment which appears in G between
the displaced anterior wall and the undisplflced posterior col-
umn. C Iliac-oblique view with the quadrilateral plate seen in
profile. D - G CT sections (see text)
 CT Study of Anterior WaH Fractures
D E
F G
directed obliquely, i.e. forwards and outwards at
about 60° at the level of the acetabular roof. At the
level of the superior pubic ramus there is regularly
considerable comminution. The fracture involves
the medial part of the roof as seen in Fig.7.6D.
Generally, the fragment of anterior wall takes
with it a full segment of the pelvic brim, and is then
limited posteriorly by a fracture line transgressing
the anterior part of the quadrilateral surface to
reach the comminution of the obturator canal area.
It sometimes happens that the fracture line divides
the pelvic brim itself longitudinally, or is situated
just a little below it (Fig.7.6D).
In Fig. 7.6 E a part of the internal portion of the
roof appears isolated as it is impacted into the un-
derlying cancellous bone.
The degree of displacement of the anterior wall
is evident; it generally rotates externally.
Very frequently the anterior wall fragment is as-
sociated with a fractured plate of bone from the
113
quadrilateral surface, which is pushed inwards by
the head. CT scanning is an excellent technique for
studying this. The anterior limit of this fragment
runs with the posterior limit of the anterior wall
fragment. This plate of bone usually keeps a bony
hinge posteriorly, and it can happen that the
posterior hinge involves the innermost part of the
posterior wall (Fig. 7.6 E, F).
Figure 7.7 shows the case of an associated ante-
rior wall and anterior column fracture, shown by
the fracture of the ischio-pubic ramus. The tear-
drop appears on a separate fragment, which is seen
in Fig. 7.7 G between the displaced anterior wall
and the posterior column.
At the level of the roof, the fracture line separat-
ing the two main fragments slants obliquely for-
wards and outwards at an angle of about 55°. The
inner part of the roof is impacted (Fig. 7.7B). The
hinge of the quadrilateral plate does not involve the
posterior wall.
8 Fractures of the Anterior Column
In fractures of the anterior column of the acetabu-
lum (JUDET and LETOURNEL 1960) a segment of
variable size is separated from a sector of the bone
extending from the middle of the ischio-pubic
ramus below, to any point above as far backward as
the mid-point of the iliac crest. This description has
contributed to the basis of our fundamental con-
cept of the architecture of the acetabulum (see
Fig. 1.1).
8.1 Morphology
As depicted in Fig. 8.1 these fractures fall broadly
into four categories, depending upon the level at
which the upper end of the fracture line cuts the
border of the innominate bone: very low, low, inter-
mediate and high (Table 8.1).
8.1.1 Very Low Fractures
The displaced fragment comprises the lowest part
of the anterior wall and its related acetabular ar-
ticular surface (Fig. 8.1 A). Seen from the outer
aspect of the pelvis, the fracture line cuts the anteri-
or wall horizontally, then descends vertically
through the cotyloid fossa and finally divides the
ischio-pubic ramus in the same line, or towards the
pubic angle or even at both points. From the inner
aspect, the superior surface of the pubis is crossed
horizontally and the ilio-pectineal line is cut in its
middle part; from this point the fracture line
descends vertically across the anterior part of the
quadrilateral surface.
Very low fractures leave undisturbed a large part
of the articular surface, which explains why in five
instances out of six the femoral head reduced spon-
taneously and was stable. Only in one case did a
dislocation remain (Fig. 8.2).
Table 8.1.
Fractures of anterior column 42 4.46070
Very low (anterior horn articular surface) 6 0.63070
Low (psoas gutter) 13 1.38070
Middle (anterior interspinous notch) 5 0.53070
High (iliac crest) 11 1.17070
Incomplete 7 0.74070
8.1.2 Low Fractures
The upper limit of the fracture line transgresses the
psoas gutter (Fig. 8.1 B).
From the outer aspect the fracture line begins
just below the anterior inferior iliac spine, crossing
the lip of the acetabulum and the articular surface,
and detaching a fragment of the latter which in-
cludes the anterior wall and some part of the roof.
It then becomes more vertical, traversing the coty-
loid fossa to reach the ischio-pubic notch. Inferior-
ly, the ischio-pubic ramus is fractured.
From the inner side the fracture line begins
below the anterior inferior iliac spine and cuts the
posterior third of the iliopectineal line obliquely
posteriorly and superiorly; it then descends to meet
the superior border of the obturator foramen by
separating the anterior part of the quadrilateral
surface.
Next to the main fragment we have seen in one
case a supplementary detached segment including
part of the iliopectineal line behind the point de-
scribed above. This has been associated with a bone
plate from the quadrilateral surface driven in by the
head and similar to that which has been described
in Chap. 7.
In all cases the femoral head has been sublux-
ated anteriorly and has remained congruent with
the displaced anterior fragment.
In four particular instances we thought for years
that the low anterior column fracture was associat-
116
Very low
A
c Intermediate
Fig. 8.1 A-D. Schemes showing the four types of fracture of the anterior column. The hatched area in D represents the outline
of two of our exceptional cases
ed with an extended plate of bone which was
detached exclusively from the quadrilateral surface.
In fact this plate also comprised a large piece of the
posterior part of the articular surface divided by a
more or less vertical fracture line which finally
crossed the posterior lip of the acetabulum at a very
low level (see Figs. 8.16 and 8.17). This plate was
limited anteriorly by a roughly vertical fracture line
dividing the fossa acetabuli, limiting the anterior
column fragment posteriorly; posteriorly it was
limited by, on the exo-pe/vic aspect, the fracture
line crossing the posterior part of the articular sur-
face, and on the endo-pe/vic aspect, a fracture line,
convex posteriorly, dividing the quadrilateral sur-
face very posteriorly, leaving the posterior edge of
the bone undisturbed (see Fig. 8.17). This plate of
bone, pushed by the femoral head, performed a
considerable internal rotational movement around
its posterior limit, which is perfectly shown on the
CT sections (Fig. 8.17).
Fractures of the Anterior Column
B Low
D High
What is remarkable in these fractures is that the
retroacetabular surface was totally undisturbed, the
obturator foramen being divided in its superior
border and in one or two other places along the
ischio-pubic ramus.
These cases were interpreted up to 1992 as asso-
ciated anterior column and incomplete posterior
hemitransverse fractures. Their exact features were
recognized on CT and 3 D reconstruction together
with better intra-operative analysis of one recent
case not included in the statistics in this book.
Their radiological description will be given among
the atypical examples as they were difficult to ana-
lyse for a very long time.
8.1.3 Intermediate Fractures
The displaced sector is bounded by a fracture line
which passes above from between the anterior
Morphology
superior and inferior iliac spines and below
through the ischio-pubic ramus (Fig. 8.1 C).
From the external aspect, beginning between the
anterior iliac spines, the fracture line curves
downwards so as to include a large part of the supe-
rior segment of the acetabulum. It descends fur-
ther, crossing the cotyloid fossa, and passes
through the ischio-pubic notch; finally, it cuts the
ischio-pubic ramus usually well towards the body
of the pubis.
From the inside, the fracture line, which may be
somewhat concave postero-superiorly or more
sharply angulated, meets the ilio-pectinealline and
cuts it a few centimetres in front of the sacro-iliac
joint. It then descends across the quadrilateral sur-
face diverging from the pelvic brim and reaches the
superior border of the obturator foramen.
Often, in addition to this main fragment being
driven forwards and inwards, a supplementary frag-
ment was detached as a result of damage to the
hind part of the iliopectineal line, or a bone plate
from the quadrilateral surface, and hinged posteri-
orly, was seen in half of our examples.
Atypical examples are seen:
The anterior column may be split through or
below the anterior wall (Fig. 8.6).
The ischio-pubic ramus has been broken in one
instance in its middle part.
The fracture line passing through the iliac wing
may stop a few millimetres short of the anterior
edge of the bone, the fracturing energy having
been expended and insufficient to break the
dense trabeculae which strengthen the region of
the bone between the iliac spines.
The head of the femur always follows the displaced
fragment and remains subluxated anteriorly.
8.1.4 High Fractures
These fractures result in separation of a massive
segment of the anterior column which comprises
the anterior wall of the articular surface, nearly all
of the roof, and the front part of the iliac wing
(Fig. 8.1 D). The important feature which they have
in common is the separation of the anterior part of
the wing, while in the lower part, the fracture line
is variable in position. The photograph of the dried
bone (Fig. 8.9) is a good example.
From the outer side, the fracture leaves the iliac
crest at a variable point in front (four cases) of the
anterior pillar of the iliac wing, or behind it and
117
sometimes very posteriorly disposed. From the
crest, the line pursues a direct, curved or somewhat
jagged course to reach the margin of the acetabu-
lum behind the roof. It traverses the articular sur-
face, detaching nearly all of the roof, and enters the
cotyloid fossa. From this level it can pursue a varie-
ty of courses:
Becoming horizontal, it may cut the anterior
wall and border of the acetabulum (two cases,
Fig. 8.11). In this instance the obturator ring is
not broken and the lowest part of the anterior
wall also remains undisturbed. This fracture
line limits the shaded zone shown in Fig. 8.1 D.
It may reach the ischio-pubic notch and beyond
this break the superior pubic ramus where, in
addition, an intermediate fragment may be de-
tached (Fig. 8.8).
After reaching the ischio-pubic notch it may
cross the obturator foramen to break the ischio-
pubic ramus at the inferior border of the sym-
physis (two cases), or in the middle of the
ischio-pubic ramus (two cases) or even at both
points.
From the inner side of the pelvis the fracture line,
having left the iliac crest by a jagged course,
sometimes corresponding to the one on the outer
side, reaches somewhere towards the back portion
of the brim of the true pelvis and then descends
over the quadrilateral surface more in its anterior
part, diverging from the ilio-pectinealline to end in
the ischio-pubic notch. As is occasionally the case,
if it recrosses the iliopectineal line (Fig. 8.1 D) the
obturator foramen is spared.
The massive anterior column fragment is rarely
detached in one piece and it is usual to find second-
ary splits in the bone. These have been seen at the
level of the interspinous notch (Fig. 8.8), at the level
of the anterior wall, or at the level of the superior
pubic ramus.
In addition to this main fragment or fragments
we have seen also:
on nine occasions out of eleven, a bone plate
hinging posteriorly from the hind part of the
quadrilateral surface (Fig. 8.8 C) and not bear-
ing any articular surface;
impacted fragments, which are found more fre-
quently than the radiographs would indicate.
The amount of articular surface involved varies
and the fragments become impacted in the un-
derlying cancellous bone of the column.
118 Fractures of the Anterior Column

./- ·_·-i
. . j.;,
/

I' \., .
(. ' \.
\ \
\
\
i
\
a

Fig. 8.2. Very low fracture of the anterior column. A Antero- Fig. 8.3. Very low fracture of the anterior column. A Antero-
posterior radiograph, a diagram posterior radiograph, a diagram. The reduplication of the tear-
drop indicates that the fracture line crosses the cotyloid fossa
 Radiology 119

8.1.5 Atypical Examples

In two cases the fracture line in the iliac wing did

not quite reach the iliac crest, being arrested
1 - 2 cm from the edge by the strong bone which oc-
curs at this level.
In one case the iliac wing fracture line was very
posterior and reached the sacro-i1iac joint, taking
with it the superior and vertical segment of the
auricular surface. Continuing inferiorly, the line
skirted the pelvic brim and spared also the margin
of the greater sciatic notch, running downwards to
reach the ischio-pubic notch (Fig. 8.10).
Another special example is described at the end
of this chapter (Sect. 8.2.5).
Remark: Anterior column fractures detach an ante-
rior segment of the ilium which is variable in size
and orientation. Inspection of the outlines in
Fig. 8.1 shows that the more inferior the site of the
fracture through the obturator ring, the lower is the
iliac wing component at its site of rupture of the
anterior border of the bone.

8.2 Radiology

The most significant radiological character in alI

fractures of the anterior column (as was the case
for fractures of the anterior waIl of the acetabu-
lum) is the integrity of the posterior column as in-
dicated in the antero-posterior view by the un-
broken posterior border of the acetabulum and the
\
\.. ..._ .-.'\ ilio-ischial line, and in the iliac-oblique view by the
("
.
.\ perfect outline of the posterior border of the in-
1 . nominate bone.
(
. \.
...... . \. The exception to these generalisations is the ilio-
)
ischial line. In 28 out of 39 cases it was intact but
"

\ .I
I in the others, its outline was disturbed due to the
i
i i associated fracture of the quadrilateral surface ele-
; I vating a posteriorly hinged plaque of bone.
a " " \ I The latter can present as:
a simple irregularity of the ilio-ischial line creat-
ing an interruption in the normal curve (seen in
two cases presenting with elevated bone plates
from the quadrilateral surface);
a forked appearance produced by splitting of
the line in its middle part corresponding to a
smaIl elevated bone plate (one case, Fig. 8.5);
Fig. 8.4. Very low anterior column fracture. A Antero-posterior
radiograph, a diagram. We cannot prove that the horizontal
rupture of the ilio-ischial line in two places with
fracture line crossing the anterior column cuts the articular sur- a medial displacement of the middle segment
face; perhaps it passes just below it with respect to the basic line (four cases).
 oj(_ _-+- Area of detachment
of anterior column
Posterior border
of ilium

....... _( .....\
\ -
/
. \.
/ I
i Fig. 8.5A - C. Low fracture of the anterIor column A Antero-
/ posterior radiograph showing branching of the iIio-ischial line
/ and the site of detachment of the anterior column, a diagram,
/ B obturator-oblique radiograph, b diagram, C iliac-oblique
b i radiograph (post-operative) showing the undisturbed border of
Radiology
The different varieties of anterior column fractures
are diagnosed by a direct radiological study of the
anterior column.
8.2.1 Very Low Fractures
(a) Antero-posterior view. The landmarks of the
posterior column and the whole of the roof are un-
disturbed (Figs. 8.2, 8.3).
The ilio-pectineal line is broken in its middle
part below the level of the roof of the acetabulum
and the detached segment extends as far as the
pubic symphysis. The anterior border of the
acetabulum is fractured in its lower half below its
characteristic notch. The teardrop and the ilio-
ischial line are displaced relative to each other so
that the teardrop lies on the inside of the line (in
four cases out of six). Sometimes one can see a
reduplication of the teardrop, the outer image
keeping its relationship with the ilio-ischial line
while the other is displaced within it (Fig. 8.3). The
obturator ring is broken, perhaps in the middle of
the ischio-pubic ramus, at the level of the angle of
the pubis, or at both of these points. The femoral
head stayed in place or spontaneously reduced in
five cases out of six. In only one case did an
anterior subluxation persist (Fig. 8.2).
(b) Obturator-oblique view. It shows clearly the
point of fracture of the ilio-pectineal line and of
the ischio-pubic ramus; the size of the detached
fragment of the anterior column is delineated. In
some instances it may be difficult to affirm how
much of the anterior horn of the acetabular ar-
ticular surface has been involved (Fig. 8.4). It will
be remembered that normally this part of the ace-
tabulum terminates inferiorly at approximately the
centre of the middle segment of the acetabulo-
obturator line.
(c) Iliac-oblique view. This confirms the integrity
of the posterior border of the innominate bone but
does not show clearly the level of the fracture line
of the anterior border of the acetabulum.
8.2.2 Low Fractures
The upper point of detachment of the anterior col-
umn is situated a little higher than in the previous
instance and encroaches on the anterior part of the
roof of the acetabulum.
121
(a) Antero-posterior view. The roof of the acetabu-
lum is seen to be injured, an internal portion rep-
resenting a quarter or more being detached
(Fig. 8.5 A); this part of the roof may stay attached
to the anterior column or may constitute a separate
fragment which has been elevated or driven into the
subjacent cancellous bone. The latter is an example
of marginal impaction being produced by pressure
from the head of the femur as it dislocates.
Most significantly, the ilio-pectinealline is frac-
tured in its posterior part at about the level of the
roof of the acetabulum, and the sector of the pelvic
brim which is displaced inwards and forwards ex-
tends from this point to the pubic symphysis.
Also, the teardrop is seen to maintain its rela-
tionship with the ilio-pectinealline and is displaced
inwards with respect to the ilio-ischialline. In addi-
tion to linear displacement its attitude is further
disturbed for it appears to have pivoted. The dis-
placement of the teardrop with respect to the ilio-
ischial line occurs in the reverse direction to that
which we see in fractures of the posterior column,
a most important point.
The anterior border of the acetabulum is broken
in its superior segment. The bony outline of the ob-
turator foramen is broken at two points, one at the
level of the anterior half of the ischio-pubic ramus
and the other in the region of the ischio-pubic
notch but the latter point of rupture is not always
visible.
The femoral head lies in external rotation and is
dislocated anteriorly, being clearly displaced in-
wards of the ilio-ischial line.
The upper area of damage of the column is
sometimes visible as a triangular clear zone at the
level of the roof, between it and the pelvic brim
(Fig. 8.5 A), exactly as was observed at the zone of
detachment of the posterior column fracture
(Figs. 6.3 and 6.8 A).
(b) Obturator-oblique view. This is the most impor-
tant view, showing the displacement of the anterior
column by the femoral head (Fig. 8.5 B). The head
is seen to rest and be framed congruently by the ar-
ticular surface borne by the displaced segment of
the anterior column. The points of rupture of the
obturator ring can be determined and the posterior
border of the acetabulum is seen to be intact.
(c) Iliac-oblique view. The following features are
visible:
the fracture line through the anterior border of
the acetabulum, although its level may be dif-
ficult to define precisely;
122 Fractures of the Anterior Column

A B

Anterior limit 01 part

of innominate bone which
Fragment 01 rool remains in placp.
'\ remaining in place
"' -I~
( -....'"
\
Ii ' /"'\
\
)--" \ ,

,/
/ \
i \
I \
/ i
I i
i i
a b /

Fig. 8.6A - D. Middle fracture of the anterior column. A Ante- cation of the roof outline and the supra-acetabular fracture
ro-posterior radiograph, showing teardrop on a separated frag- line, C iliac-oblique radiograph, c diagram, D scheme of the
ment, a diagram, B obturator-oblique radiograph with redupli- spur sign
 Radiology 123

Segme nt of fract ure

seen in profile as spur sign

the uninterrupted outline of the posterior bor-

der of the innominate bone;
the frequent occurrence (in two-thirds of cases)
of a bone plate, seen in section, lifted from the
quadrilateral surface and driven in by the fe-
moral head;
the integrity of the iliac wing and of the iliac
crest.
The typical features of these low fractures of the
anterior column reflect point by point certain char-
acteristics of fractures of the posterior column with
which they have a definite symmetry.
Atypical examples occur:
C On the antero-posterior view, the ilio-ischialline
although visible may appear bent, forked, or
Fracture line crossing bifurcated (Fig. 8.5 A) in its middle part. This is
internal iliac fossa
the result of elevation of a bone plate from the
quadrilateral surface which is seen most clearly
on the iliac-oblique view.
An additional fragment may be present, asso-
ciated with the ilio-pectineal line posterior to
the main detached segment of bone; this is easi-
ly recognisable because one surface is identifi-
able as the typical curve of the pelvic brim.

8.2.3 Intermediate Fractures

(a) Antero-posterior view. Radiologically these are

very similar to the low fractures of the anterior col-
umn but the detached segment is somewhat dif-
c ferent in shape and the fracture line cuts the ob-
turator ring inferiorly at the level of the body of the
Fig.8.6C,D. pubis or the highest part of the ischio-pubic ramus;
124
above, the fracture line crosses the iliac fossa to
reach the notch between the anterior iliac spines -
the interspinous notch. The ilio-pectinealline is cut
much more posteriorly. Sometimes, the large ante-
rior detached fragment is itself split across the ante-
rior wall of the acetabulum (Figs. 8.1 C and 8.6A).
A significant part of the roof, or sometimes the
whole roof, with the detached fragment of the
anterior column to which it remains attached, tilts
with it and remains approximately congruent to the
displaced femoral head (Fig.8.6A).
(b) Obturator-oblique view. A fracture line above
the acetabulum (Fig. 8.6D) may be visible. It resem-
bles the spur sign which will be explained later
(Chap. 14), and is of the same significance. This
fracture above the acetabulum is never as displaced
as it is in the both-column type in which the projec-
tion of the fragment above the fracture line pro-
duces the typical spur appearance (Figs. 14.12 and
14.29). Reduplication of the image of the roof com-
monly occurs (Fig.8.6B).
(c) Iliac-oblique view. As before, these fractures can
be accompanied by elevation of a bone plate from
the inner aspect of the pelvis visible in the iliac-
oblique view (Fig. 8.7 C).
In two cases out of the five, the fracture line in
the iliac wing did not quite reach the notch between
the anterior iliac spines. The true trajectory was
only discovered at operation.
Figure 8.7 A shows the antero-posterior radio-
graph of an intermediate anterior column fracture,
apparently very comminuted. The corresponding
iliac-oblique view reveals that the posterior column
remains intact (Fig. 8.7 C).
8.2.4 High Fractures
(a) Antero-posterior view. Again, the principal
landmarks of the posterior column (the posterior
border and the ilio-ischial line) are not damaged
(Fig. 8.10A). The greater part or the whole of the
roof is tilted and the landmarks of the anterior col-
umn are broken or displaced; above these there is
a fracture line in the iliac wing which reaches the
crest. The iliac wing fracture segment can appear
doubled because the individual fracture lines
through the adjacent cortices may not be in the
same plane. If the line is fairly straight on the exter-
nal face, the appearance may be one of an inverted
"V" on the internal iliac fossa surface. The break
in the iliac crest may be situated a little behind the
Fractures of the Anterior Column
anterior superior iliac spine, towards the summit or
even considerably behind this; we have mentioned
already the single case of sacro-iliac joint involve-
ment with detachment of the fragment (Fig. 8.10).
The roof is detached with the anterior column,
usually in its entirety or leaving behind only its
outer part, representing hardly one-quarter of the
normal anatomical roof; the displaced fragment of
roof remains approximately concentric with the
anteriorly dislocated femoral head.
The site of fracture of the pelvic brim is variable.
Thus the entire related bar of bone may be de-
tached from its extreme posterior end to the pubic
angle; a secondary fracture line may occur at the
level of the anterior wall of the acetabulum. Alter-
natively, if the pelvic brim is fractured very posteri-
orly, the anterior fracture line may be at the root of
the superior pubic ramus (Fig. 8.8) or at the level of
the anterior wall of the acetabulum (Fig. 8.11), and
then the obturator foramen is not involved.
The teardrop is tilted and displaced inwards with
respect to the ilio-ischial line, but it can constitute
an isolated fragment supplementary to the main
fragment (Fig.8.8A).
The obturator ring can be broken at the level of
the ischio-pubic notch (not always visible in this
view), along the superior pubic ramus, at the angle
of the pubis, in the ischio-pubic ramus, or at two
of these sites. In some instances the ring is not
broken (Fig. 8.11).
(b) Obturator-oblique view. In this view (Figs. 8.8B,
8.10 Band 8.11 B) the long section of the anterior
column is seen to be displaced and driven inwards
by the femoral head which remains approximately
congruent to the articular surface borne by the
displaced fragment. The points of rupture of the
bony ring forming the obturator foramen are clear.
Nearly always in these cases a fracture line is seen
above the acetabulum in the iliac wing.
Atypical examples of high fractures (7 cases) in-
cluded in our series are those in which the fracture
line in the iliac wing failed to reach the iliac crest
by a centimetre or two. In another, a split divided
the upper part of the anterior column and passed
through the notch between the anterior iliac spines
(Figs. 8.8 and 8.9).
Fig. 8.7 A-C. Middle fracture of the anterior column. A Ante-
ro-posterior radiograph, a diagram, B obturator-oblique radio-
graph, C iliac-oblique radiograph, c diagram. The antero-poste-
rior view appears complicated because of numerous accessory
fracture lines resulting from osteoporosis. The iliac-oblique
view shows that the posterior column is intact
A
B

)(-
/-,,
\\ \I
,
I
I
I \
\
I
I

,
I
I
I

I
\
a \

Posterior border
of the pelvic bone - - ---\

Elevated
segmental
quadrilateral- --'c---I \
\
surface \
\
of ischium \

,
I
J

\
\
\
\
\
c \
c
A

_._</.. . . .
/ i
i i
/ i
a
\
\
/ .- ......
/
I
i
\
\
\
"' \
b !

rtexof I
/ QUadri lateral .schium
. \ surface 0 I

I . . . . J\_.. _.'~)
,./
/ /7
! i!
!
I .
/
,i (
'\
I
,...,,/
c c ./
Radiology 127

(c) Iliac-oblique view. As before, this (Fig.8.10C)

confirms the intact posterior border of the innomi-
nate bone. It has shown in one instance (Fig. 8.8 C)
a bone plate, visible in section, lifted internally
from the quadrilateral surface.

8.2.5 Atypical Examples

A special example of a fracture of the anterior col-

umn which has been seen only once (Fig. 8.12) was
associated with an anterior dislocation of the hip.
This was really a low fracture of the anterior col-
umn. The fracture cut the anterior border of the
acetabulum near the level of its notch. It then
descended across the cotyloid fossa and seemed to
strike the densely trabeculated bone, limiting the
ischio-pubic notch (which was not fractured) from
where it deviated towards the ischium from above
to below and skirting the bony ring of the ob-
turator foramen. This low fracture of the anterior
column appears therefore to have included in the
detached fragment the whole of the obturator
foramen (the integrity of which was preserved)
separating it from the rest of the posterior column. A
In another four cases (three in this series plus
one very recent; see Figs. 8.16 and 8.17), in addition
to the previously described characteristics of low
anterior column fractures there was a fracture line
transgressing the posterior wall. The fracture began
at a low point on the posterior lip of the aceta-
bulum on the antero-posterior view and climbed up
through the posterior wall in a upward and inward
direction, rarely straight, most often sinusoidal in
form, and appearing to reach the internal border of
the roof of the acetabulum. Initially (in the first
edition of this book) we classified these cases
among the anterior fractures with associated
posterior hemitransverse fractures. However, one
recent case (outside the series reported in this
book), operated on in February 1992, thoroughly
studied using CT and 3-D reconstructions, enabled
us to attain a perfect understanding of the fracture
line dividing the posterior wall on the antero-
posterior and obturator-oblique views. It repre-
sents, in fact, the posterior hinge and limit of the
quadrilateral surface plate of bone commonly asso-

..
Fig. 8.8A - C. High fracture of the anterior column. A Antero-
posterior radiograph, a diagram, B obturator-oblique radio-
graph, b diagram, C iliac-oblique radiograph, c diagram. There B
is an additional fracture line extending forwards to the inter- Fig. 8.9A,B. High fracture of the anterior column. A Lateral
spinous notch view and B medial view of dried bone corresponding to Fig. 8.8
128 Fractures of the Anterior Column

ciated with an anterior column fracture. In these in-

stances, the plate of bone had an unusual posterior
extent and took with it the inner half of the posteri-
or part of the acetabular surface, without apparent-
ly disturbing the ilio-ischial line. In fact, the ilio-
ischial line may appear slightly more curved or
thinned in its mid portion, as a part of it may
belong to the quadrilateral plate of bone. There
may be an associated marginal impaction of the in-
ner part of the roof (Fig. 8.17).
On the iliac-oblique view the large quadrilateral
surface plate of bone is seen in profile due to its
rotation, and a teardrop image may appear on it.
On the obturator-oblique view the fracture through
the posterior wall appears more vertical; the low
anterior fracture column itself is perfectly demon-
strated, as are the ruptured points on the obturator
foramen.
These cases deserve to be classified among the
anterior column fractures as the major part of the
posterior column is undisturbed. Only the most in-
ferior part of the posterior lip of the acetabulum
and the posterior segment of the obturator fora-
men are involved. The rest of the column appears
totally uninjured.

A
8.3 CT Study of Anterior Column Fractures

The successive CT sections demonstrate clearly

that the posterior column is undamaged and in
perfect connection with the intact part or whole of
the iliac wing.
The level of detachment of the anterior column
either from the iliac crest or from the anterior
border is precisely defined. seen in
In the case of a high or intermediate anterior section
column fracture, the fracture line above the roof
and through it is most frequently angled obliquely
forwards and outwards at 45 0 (Fig. 8.13), some- / ."..;
/~ . - .

times a little more. It happens in some instances i .i

that the fracture line is roughly coronal (Fig. 8.14) .I "",\
( /
above and through the roof, but below it slants \ I
slightly forwards and outwards. i i
As for the anterior wall fracture when the head i \.
i I
dislocates anteriorly, displacement of the anterior /.J
a
column is not enough to allow pelvic penetration,
so a plate of bone from the cotyloid fossa and con-
sequently the quadrilateral surface is mobilised to
allow the head to dislocate. This bony plate may be
totally detached, but in most cases it retains a
posterior hinge which generally does not disturb Fig.8.10A
CT Study of Anterior Column Fractures 129

8 C

F-racture surface
of ili ac wing

Greater sciatic
notch in situ

/ '
/ D
i
i
I
\ "

"

\
\
b \
Fig. 8.10A - D. High fracture of the anterior column involving a diagram, 8 on obturator-oblique radiograph only the lower
the sacro-iliac joint. A Antero-posterior radiograph with horse- parts of the sacro-iliac joint lines are visible, b diagram , C iliac-
shoe-shaped outline of the sciatic bullress seen in profile, oblique radiograph , D scheme
130 Fractures of the Anterior Column

A B

,
/1----,
/
i
( I /
i
i 1/ .~ ---_/
j i
. I ./ .
\
! I I i.
I
I i, \ \.
a / / \
'.\
\
\
b
\.

Fig. 8.11A-C. High fracture of the anterior column. A Antero-
posterior radiograph, a diagram, B obturator-oblique radio-
graph, b diagram, C iliac-oblique radiograph (after repair). The
fracture line spares the obturator ring, cutting the anterior wall
transversely and following the hatched area shown in Fig. 8.1 D.
It might be thought from the antero-posterior view that the
pelvic brim is intact; this is disproved on the obturator-oblique
view. The iliac-oblique view proves the continuity of the posteri-
or border of the innominate bone. Contrast this fracture with
the pure superior wall fracture shown in Fig. 5.12
CT Study of Anterior Column Fractures 131

the posterior wall (Fig. 8.13). In fractures which are

accompanied by only a slight displacement of the
head, there is no detached plate (Fig. 8.14) from the
quadrilateral surface. Thanks to CT and 3-D recon-
struction, we are now able to see that in four cases
out of 44 (see Sect. 8.2.5) the quadrilateral surface
plate of bone extended much more posteriorly than
we had for years believed. Figure 8.17 shows that
this bone plate may extend considerably posteriorly
and take with it the inner vertical half of the
posterior wall, thus dividing the posterior lip of the
acetabulum but leaving undisturbed (or nearly so)
the ilio-ischialline and the posterior edge of the in-
nominate bone, i.e. the major part of the posterior
column. The obturator ring is fractured in different
places according to the variant of anterior column
fracture concerned.
At the level of the CT scan roof, a marginal im-
paction of the inner part of the roof is often visible;
it appears as a polyhedral fragment of the roof,
seemingly totally independent (Fig. 8.17).
In the low variant the roof is intact, the upper
part of the anterior column below the is involved
roof, and the anterior part of the cotyloid fossa is
divided (Fig. 8.15). In the high variants the fracture
line through the iliac wing may cut the two tables
c at different levels and the fracture may be com-
minuted on one or the other aspect of the wing.
Fig. S.llC
The displacement of the anterior column and
the quadrilateral surface plate may be seen very
precisely on CT.
132 Fractures of the Anterior Column

Fig. 8.llA-C. Atypical fracture of the anterior column. A An-

tera-posterior radiograph before reduction of the anterior
dislocation. There is a low fracture line cutting the anterior wall
at its notch; in the cotyloid fossa it is deflected by the strang
bone of the ischio-pubic notch but it continues and splits the
ischial body. The obturator ring is spared. B Antero-posterior
radiograph after reduction, b diagram, C iliac-oblique radio-
c graph after reduction
 CT Study of Anterior Column Fractures 133

A B

Fig.8.13A-F. Intermediate anterior column fracture. A Ante-

ro-posterior view, B obturator-oblique view, showing that the
anterior column fragment is divided by a fracture line trans-
gressing the root of the pubis. The fracture line above the roof
corresponds to the track of the fracture line above the acetabu-
lum. C Iliac-oblique view, showing the track of the fracture line
through the iliac wing. D - F CT sections. The overall fracture
line detaching the anterior column and the quadrilateral surface
plate runs forwards and outwards at an angle of about 50°. The
c quadrilateral plate does not involve the posterior column
 134 Fractures of the Anterior Column

E F
Fig.8.13E-F

A B
Fig. 8.14A- F. Intermediate anterior column fracture, slightly line above and through the roof is roughly coronal. F shows
displaced, not involving a quadrilateral surface plate of bone. clearly that there is no detached quadrilateral surface plate of
A - C The three standard views, D - F CT sections. The fracture bone
CT Study of Anterior Column Fractures 135

E F

Fig. S.14C-F
 136 Fractures of the Anterior Column

Fig. 8.1SA-D. Low anterior column fracture. A Obturator-ob-

lique view, B-D CT sections. The roof is intact (B), the upper
part of the anterior wall is involved (C), the cotyloid fossa is
divided anteriorly D
CT Study of Anterior Column Fractures 137

\ / .
,.."\
.-~ . \
\. \
} i
I
i
i
a b

Fig. 8.16A,B. Low anterior column fracture. A Antero-posterior
radiograph, a diagram, B iliac-oblique radiograph, b diagram.
The fracture line traverses the posterior wall very obliquely up-
wards and inwards, stopping before the greater sciatic notch and
apparently disturbing neither the ilio-ischial line nor the
posterior edge of the bone in B. This fracture line is in fact the
posterior hinge-limit of the quadrilateral plate of bone detached
conjointly with the low anterior column fracture and which
takes with it the inner vertical half of the posterior part of the
articular surface. Of the landmarks of the posterior column, on-
ly the posterior lip is ruptured inferiorly
 138
A B
Fig. 8.17 A- M. Atypical low anterior column fracture. A Ante-
ro-posterior radiograph, B obturator-oblique radiograph, C ilio-
oblique radiograph, a, b diagrams of the fracture. A very oblique
fracture line divides the posterior wall of the acetabulum on the
antero-posterior radiograph and appears nearly vertical on the
obturator-oblique view. On the iliac-oblique view a large bone
plate from the quadrilateral surface is seen to have been driven
in by the femoral head; due to the rotation of the plate the part
of the cotyloid fossa that it brings with it produces a teardrop
image on this view. A marginal impaction of the inner part of
the roof is obvious on all three views. In A and B the post-
traumatic wear of the supero-external quadrant of the head can
be seen. D-J CT sections. The head has pushed in front of it
the anterior column fragment, to which it is congruent. Its wear
Fractures of the Anterior Column
can be seen in F. The enormous plate of bone detached from the
quadrilateral plate is in one piece, rotated internally around a
displaced posterior hinge, and has taken with it the inner half of
the posterior part of the acetabular articular surface, the frac-
ture finally dividing the inferior limit of the posterior wall. The
marginal impaction can be seen in D. J - M 3-D reconstructions
demonstrating on J how the head pushes the 2 fragments. On
M the endo-pelvic track of the quadrilateral plate does not in-
volve the posterior edge of the bone. On L the retrocotyloid sur-
face is totally undisturbed and the contour of the exo-pelvic part
of the quadrilateral plate at the level of the ischium is clear, par-
ticularly the division of the most inferior part of the posterior
lip of the acetabulum
 CT Study of Anterior Column Fractures 139

D E

F G

c H

Fig. 8.17C-I
 140 Fractures of the Anterior Column

J K

L M

Fig.8.17J-M
9 Pure Transverse Fractures
The line of a transverse fracture of the acetabulum
divides the innominate bone into two segments.
The upper is essentially ilium and the lower, which
we shall call ischio-pubic, may contain some part of
the embryologic ilium. Other simple fractures of
the acetabulum result in separation of all or part of
a column, whereas transverse fractures cut trans-
versely both columns and divide each into two
parts, but the upper and lower segments of the
divided columns remain intact with respect to each
other.
The reason for including this fracture type as a
member of the simple group pertains to the simpli-
city of the fracture line. It tends to orientate with
respect to the trabecular pattern of the bone in a
radial or horizontal fashion and not, as is com-
moner in the other pure forms, in a more sagittal
direction. The innominate bone possesses a nar-
rowed region or isthmus which has been described
as the elective site of transverse fractures; undoubt-
edly the isthmus exists and in some instances trans-
verse fracture lines have been seen to occur here.
However, since 1958, we have observed that the site
of these transverse fractures is extremely variable
and completely independent of the apparently nar-
rowed zone.
9.1 Morphology
The configuration of a transverse fracture is con-
firmed during surgical operation as being practical-
ly in one plane. The obliquity of the plane can vary
in any direction.
It is useful to regard the lip of the acetabulum
as being a circle, the circumference of which will be
sectioned at two points by a transverse fracture
which resembles a chord (Fig. 9.1 A). The chord
may be orientated horizontally or obliquely and ac-
cordingly the fracture line may be low posteriorly
and high anteriorly or low anteriorly and high pos-
teriorly.
Table 9.1
Transverse fractures 70 7.44070
Trans-tecta! 13 1.38%
Juxta-tecta! 36 3.82%
Infra-tecta! 20 2.12%
Incomp!ete 0.10%
The plane of the transverse fracture of the
acetabulum (Fig. 9.1 B) can be prolonged across the
head of the femur where it appears to be in perfect
continuity with one of the great circles of the
sphere. The direction taken will vary from those
horizontally disposed and therefore situated low in
the acetabulum, to those which are very oblique
and involve the roof. The possible combinations of
obliquity and level in this type of fracture are in-
finite.
The anatomical description of the fracture may
be considered according to the level at which it cuts
the acetabulum, and to the displacement of the
elements concerned.
9.1.1 Orientation of Fracture
According to the level at which fractures break the
acetabulum they may be classified as follows:
(a) Infra-tectal, cutting the inferior part of the
anterior and posterior walls of the acetabulum
or only the horns of the articular surface. The
cotyloid fossa is split horizontally (Fig.9.2A).
(b) Juxta-tectal, cutting the anterior and posterior
walls and passing through the highest point of
the cotyloid fossa in the region of the internal
limit of the roof. It may be impossible to tell
pre-operatively whether the roof is involved
(Fig.9.2B).
(c) Trans-tectal, passing at the level of the roof of
the acetabulum only leaving the outer part of
the roof in situ on the iliac wing (Fig. 9.2C).
142
a quadrilateral surface in order to regain the point of
c This endo-pelvic direction varies with each case.
A
B all fractures of the acetabulum have this character-
Fig.9.1A,B. Diagrams of variations in the plane of transverse
fractures
In effect, a plane section of the innominate bone is
produced which is in line with the fracture where it
cuts the articular aspect of the acetabulum.
Table 9.1 shows the distribution of transverse frac-
tures in our series, according to this description.
Seen from the retro-acetabular surface an infra-
tectal fracture is horizontally disposed and cuts
posteriorly the lesser sciatic notch or may divide
the ischial spine longitudinally. All the other frac-
ture lines have an obliquity which generally
becomes greater as the fracture becomes higher;
they cut the anterior border of the greater sciatic
notch at a variable level. The highest forms involve
the angle of the greater sciatic notch (Fig.9.6A)
and even a segment of the superior border.
From the inside of the pelvis, from the point of
rupture of the anterior border of the acetabulum,
Pure Transverse Fractures
the fracture line, straight in outline, more or less
oblique backwards and inwards, reaches the ilio-
pectineal line and beyond this descends on the
rupture of the posterior border of the iliac bone.
The point of rupture of the pelvic brim is deter-
mined by the obliquity of the fracture surface: it
can approach the sacro-iliac joint in extreme cases.
We have preserved the name, transverse fracture,
in preference to others because it reminds us of the
classical description by MALGAIGNE (1847) "rup-
ture en travers de l'os des iles" (transverse fractures
of the hip bone). We have avoided the term trans-
acetabular (CREYSSEL and SCHNEPP 1961) because
istic.
9.1.2 Displacement in Transverse Fractures
(a) Displacement of the ischia-pubic segment. In 11
cases (4 juxta-tectal and 7 infra-tectal) there was no
displacement. The others exhibited slight or marked
displacement. When present, the displacement may
appear to be maintained by the position of the
femoral head, but in some instances, the head
returns to its proper position under the fragment of
roof left on the upper segment of iliac wing.
The attachments which the fragment maintains
with the pubic symphysis dictate the manner in
which the ischio-pubic segment is able to move.
This appears to occur generally inwards into the
true pelvis but really the displacement of the frag-
ment is more complex than would appear at first
sight. Both theoretically and for practical reasons,
it is important to understand that the displacement
comprises two elementary components:
Fig. 9.2A-C Schemes of transverse
fractures. A Infra-tecta! type,
B juxta-tecta! type.
C trans-tecta! type
Morphology
Firstly, there is a rotation around a vertical axis
passing through the pubic symphysis (Fig. 9.3 A).
Displacement around this axis cannot occur with-
out some distortion of the pubic symphysis itself.
If the movement is pure, the acetabulum, which is
situated posterior to the axis of the symphysis,
appears displaced inwards. The posterior part of
the fracture line is always more displaced than the
anterior, i.e. there is more displacement at the
sciatic notch.
Secondly, a rotation can occur around a fairly
horizontal axis passing from the point of rupture
of the posterior border of the pelvic bone to the
pubic symphysis (Fig. 9.3 B).
At the point of rupture in the greater sciatic
notch there is little displacement, the symphysis
pubis is slightly distorted, and the segment of the
pelvic brim on the ischio-pubic fragment tilts in-
wards, carried by that part of the bone immediately
above the axis of rotation. It is only this part which
appears displaced in an antero-posterior radio-
graph. In fact, the ischial tuberosity also tilts, but
outwards. The axis of rotation which always passes
through some point in the pubic symphysis can
pass, not necessarily through the point of rupture
of the posterior border of the innominate bone
(that is to say, at the level of the fracture), but
through a point on the posterior border situated
below the fracture line. In this instance, the pivot-
ing of the fragment around the axis will be charac-
terised by greater tilting in its upper part than in the
preceding case. The combined displacement of the
posterior border and of the pelvic brim is inwards
while the ischial tuberosity tilts outwards.
143
It is most usual for the displacement of the
ischio-pubic fragment to be produced by an associ-
ation of these two rotations - vertical and hori-
zontal. The pubic symphysis is the site of both the
vertical axis and the horizontal axis, so that the
ischio-pubic fragment displaces inwards more in its
posterior part, and at the same time tilts in such a
manner that its superior part is displaced inwards
while its inferior part is displaced outwards.
To these elements of displacement can be added
some elevation of the fragment which can further
complicate the radiological interpretation. This
special mode of displacement of the ischio-pubic
fragment can only result through a distortion of the
pubic symphysis. In only 11 cases was a more or
less marked disjunction of the pubic symphysis as-
sociated with a transverse fracture and as soon as
the stability of the pubic symphysis is lost, the
mechanism of displacement described no longer
occurs.
(b) Displacement of the iliac fragment. It is the
rule that this part remains in situ, but in a few cases
it had been driven outwards and was associated
with an antero-inferior opening of the sacro-iliac
joint. The femoral head was thrust between the two
fragments, tending to displace the superior out-
wards and the inferior inwards.
We have seen two dislocations and three signifi-
cant anterior openings of the sacro-iliac joint and
it is difficult to say whether they were the direct
consequence of the displacement of the transverse
fracture itself or whether they were associated in-
juries.
(c) Displacement of the femoral head. The fe-
moral head can reduce spontaneously and lie under
~ ~--- _ AxiS of
''-'J
. rotation X Y
Fig.9.3A,B. Diagrams of compo-
Section of pelVIC nents of displacement of ischio-pubic
brim resulting from
maximum displacement fragment in transverse fractures. A
Rotation about vertical axis passing
through pubic symphysis; B rotation
about axis passing from pubic sym-
physis to back of fracture line
144 Pure Transverse Fractures

.,
')
\ '---'\
........ ~.~ '\
-\
., V \
\ I )
\ /
\
i /
\
}..
I
a 1\ / b

Fig. 9.4A, 8. Infra-tectal transverse fracture. A Antero-posterior There is no displacement and the fracture line cuts the greater
radiograph, a diagram, B iliac-oblique radiograph, b diagram. sciatic notch at the junction of the middle and upper thirds
Morphology 145

Posterior .~_-'..,-''--_~
border I
~
\
i
i
i
i
i
a i

Fig. 9.5A,B. Juxta-tectal transverse fracture with pseudarthrosis 120 days after injury. A Antero-posterior radiograph, a diagram,
B iliac-oblique radiograph
146 Pure Transverse Fractures

Fig. 9.6A-C Trans·tectal transverse fracture. A Antero·posteri- displacement with dislocation of the sacro-iliac junction and a
or radiograph, a diagram, B obturator-oblique radiograph, fracture of the femoral neck
b diagram, C iliac-oblique radiograph, c diagram. There is gross
Radiology 147

the intact segment of roof; this was found in 26

cases and at first sight did not appear displaced.
When it remains displaced (44 out of 70 cases) it is
directed towards the inside of the pelvis as a central
dislocation. One could be critical of the term cen-
tral dislocation in consideration of the fact that the
head is not dislocated with respect to the inferior
fragment; however, the same argument could be ap-
plied to a posterior dislocation in that it remains in
contact with a posterior fragment. The term central
dislocation is useful and is commonly employed,
remembering always the inward displacement ob-
served on the antero-posterior radiograph; it also
distinguishes this type of displacement accompany-
ing transverse fractures from posterior dislocation
and from the predominantly anterior displacement
of the head accompanying anterior column frac-
tures that we call anterior dislocation.
Once we observed a transverse incomplete frac-
ture cutting the posterior wall and the posterior
two-thirds of the quadrilateral surface yet not in-
volving the anterior column.

9.2 Radiology

9.2.1 Antero-posterior View

Often, the antero-posterior view is in practice suffi-

C cient for conclusive diagnosis. All of the oblique
and vertically disposed landmarks relevant to the
acetabulum (pelvic brim, ilio-ischial line, anterior
and posterior borders of the acetabulum) are seen
to be interrupted by a transverse fracture. The ob-
turator ring is not broken.
The level of rupture of the vertical landmarks of
the columns varies according to the orientation of
the plane of the fracture. They are divided at ap-
proximately the same horizontal level in infra-tectal
fractures. Juxta-tectal and trans-tectal fracture lines
are progressively more oblique upwards and medi-
ally and the degree of obliquity can be gauged by
comparing the level of rupture of successive vertical
landmarks.
If the plane of the fracture passes through the
cotyloid fossa at a distance from the roof which it
c leaves intact, clearly it is of the infra-tectal type
(Fig. 9.4). If it passes at the level of the limit of the
internal margin of the roof or through its most in-
ternal part it is juxta-tectal (Fig. 9.5), but the exact
position in higher examples is difficult to see
Fig. 9.6Cc precisely, especially if the obliquity in the sagittal
148 Pure Transverse Fractures

Fig. 9.7 A,B. Transverse fracture. Low infra-tectal fracture with fracture line cleavmg the ischial spine: A Antero-posterior radio-
graph, a diagram, B iliac-oblique radiograph, b diagram
Radiology 149

Fig.9.8A-C. Juxta-tectal fracture displaced anteriorly but not

posteriorly: A Antero-posterior radiograph showing impaction
of the supero-external border of the femoral head, a diagram, C
B obturator-oblique radiograph, C iliac-oblique radiograph
150
plane is also marked; the essential feature is that
the dense shadow of the roof of the articular sur-
face is intact. There is separation of a portion of
the roof only in the trans-tectal type of transverse
fracture (Fig. 9.6).
In all types of transverse fractures an unfrac-
tured segment of roof remains attached to the iliac
wing fragment and linked to it by intact trabeculae.
Should part of the roof be detached with the
ischio-pubic segment it is seen to be continuous
with the line of the cotyloid fossa and frames, con-
centrically in most instances, the head of the femur
(Fig. 9.6).
The ischio-pubic fragment, displaced inwards by
variable amounts, is in one piece, and the obturator
ring is not broken. On this fragment the lower seg-
ment of the ilio-ischial line and the teardrop are
clearly visible and preserve their normal interrela-
tionships.
As stated above, the extent of any displacement
is variable; it is more marked in high and oblique
transverse fractures but can be considerable in the
low forms as well (Fig. 9.7). The assessment of the
nature and amount of displacement occurring in a
transverse fracture or, more exactly, of the mutual
displacements of the two main fragments, and the
anterior and posterior ends of the fracture line, is
important in the choice of surgical approach and
can be quite difficult. A correct appreciation is
founded on the following concepts:
(a) On the antero-posterior view of the pelvis taken
strictly symmetrically, if the iliac wing appears
larger than that on the opposite side, this ap-
parent opening of the iliac fossa area indicates
a posterior and outward displacement of the
upper segment of the fracture. This is allowed
by an anterior opening of the sacro-iliac joint.
(b) The driving inwards in one piece of the ischio-
pubic fragment due to pivoting of this fragment
around a vertical axis passing through the pubic
symphysis, can be interpreted on the antero-
posterior radiograph by observing displacement
of the vertical landmarks. The displacements
relative to each other of the pelvic brim, the
ilio-ischial line and the posterior border of the
ilium increase progressively from front to back
during rotation through a vertical axis passing
through the pubic symphysis. Unfortunately,
magnification owing to radiological technique
renders difficult the interpretation of the radio-
logical shift of these landmarks, as this magni-
fication decreases from front to back and its ef-
Pure Transverse Fractures
fects are opposite to the true displacement of
the landmarks.
(c) The rotation around a horizontal axis, passing
also through the symphysis and which allows
the upper part of the fragment to tilt inwards
and the lower part outwards, may be seen more
easily. If the posterior border of the innominate
bone appears without displacement and yet the
pelvic brim and the ilio-ischial line are dis-
placed, the former more than the latter, this is
due to pivoting around a horizontal axis going
from the point of rupture of the posterior
border to the pubic symphysis.
Rarely, the pelvic brim is undisplaced whereas the
posterior border of the bone and the ilio-ischial
line are displaced: this occurs when the ischio-
pubic fragment has rotated around an axis extend-
ing from the point of rupture of the pelvic brim to
the pubic symphysis.
The degree of apparent central dislocation is ex-
tremely variable. In 35 cases out of 70 it was severe,
the external surface of the head of the femur being
in contact with the superior fracture surface and
having lost all contact with the intact portion of
roof on the iliac segment. In 9 cases the head pro-
truded much less into the pelvis, but the upper joint
space had lost its normal congruence. In 26 cases
the head was perfectly framed by the remaining in-
tact segment of the roof on the upper iliac frag-
ment.
It must be said that CT scanning will help a
great deal in the analysis of these displacements.
9.2.2 Obturator-oblique View
This gives the best indication of the orientation of
the plane of the fracture in relation to a particular
great circle of the acetabulum, and it gives a good
idea of the severity of any central dislocation. It
confirms the uninjured state of the obturator ring,
one of the essential characters of the pure trans-
verse fracture. The level of the fracture at the poste-
rior border of the acetabulum is also seen.
As always, it is comparison of the two oblique
views with the antero-posterior view which allows a
full assessment of the relative displacement of the
two fragments produced by the fracture, thereby
providing the decision as to which surgical
approach is necessary.
Radiology 151

Fig. 9.9 A - C Trans-tectal transverse fracture associated with a tached to the inferior segment of the acetabulum. C Representa-
bilateral fracture of the anterior segment of the pelvic ring. tion of the fracture lines on a dry bone, showing the complete
A Antero·posterior radiograph, a diagram, B iliac·oblique independence of the transverse fracture and the vertical fracture
radiograph. The apparently free segment of the pelvic brim is at- through the obturator foramen
152 Pure Transverse Fractures

9.2.3 Iliac-oblique View

This shows the outline of the fracture on the

quadrilateral surface of the ischium and the point
of rupture of the greater sciatic notch.
The fracture line may include the angle and
sometimes even the superior border of the greater
sciatic notch (Fig. 9.5). It may cut the anterior bor-
der of the greater sciatic notch at a variable level:
in the upper third (Fig. 9.4), in the inferior quarter
dividing the ischial spine (Fig. 9.7), or by splitting
the lesser sciatic notch. In contrast, it is often dif-
ficult on this view to locate precisely the point of
rupture of the anterior border of the acetabulum.

9.3 Atypical Cases

(a) Fractures without displacement (11 cases out of

70). These are only of theoretical interest and
their treatment is simple. Accurate diagnosis re-
quires first-class radiography (Fig. 9.4).
(b) Oblique fractures in the sagittal plane. The
plane may be oblique from below posteriorly or A
above posteriorly; the landmarks of the col-
umns seen on the antero-posterior view are not
cut at the same level. The fracture line may
reach the angle of the greater sciatic notch
behind and the pelvic brim in its middle part in
front. The opposite obliquity can be marked so
that the fracture line cuts the inferior margin of
the acetabulum behind and the pelvic brim
/
/ '--',
above and very far back (Fig. 9.7). Displace-
i
ment adds to the problem of diagnosis but the
integrity of the obturator foramen and the stu- -
\ .........
dy of the oblique views should allow the frac-
ture complex to be defined.
a
(c) Trans-tectal transverse fracture associated with
an anterior vertical fracture of the pelvis. Fig-
ure 9.9 shows an anterior fracture line which is
vertical and extra-acetabular. We have included
it here and not with the T-shaped fractures, be-
cause a segment of the pelvic brim remains as-
sociated with the posterior column and the
anterior fracture appears completely indepen- Fig. 9.10A-C Juxta-tectal transverse fracture. A Antero-poste-
dent from the transverse component. Fig- rior radiograph, a diagram, B obturator-oblique radiograph
ure 9.10 gives another example. The ilio-pec- confirming that a segment of the pelvic brim remains with the
tineal line is ruptured twice, together with the ischio-pubic fragment, b diagram, C iliac-oblique radiograph,
c diagram, The ischio-pubic fragment includes the superior
ischio-pubic ramus. border of the greater sciatic notch. The anterior fracture of the
(d) Transverse fracture with anterior dislocation. In superior pubic ramus does not involve the acetabulum, therefore
a single case, a pure transverse fracture was this is not a T-shaped fracture
Atypical Cases 153

/
i
j
i
i
\
\
\
b \ c

Fig. 9.108-C
154 Pure Transverse Fractures

./
I
i - '- '- .
./ 'J
.X. \. / .
"~\ ~.
i.
,
\1
·1 .I
·I .
·I ./
) Fig. 9.11 A, B. Exceptional transverse fracture with anterior dis·
location of the hip. A Anterior·posterior radiograph, a diagram,
a / B iliac-oblique radiograph
CT Scan Study of Transverse Fractures
associated with an anterior dislocation of the
hip (Fig. 9.11).
9.4 CT Scan Study of Transverse Fractures
A transverse fracture will involve the sections from
the superior border of the greater sciatic notch up
to the ischio-pubic notch of the acetabulum, as iliac
wing and obturator ring are intact.
The fundamental point is this: a transverse frac-
ture of the acetabulum appears as a sagittal frac-
ture line; whatever its level, it is oriented in an ante-
ro-posterior direction (Figs.9.12, 9.13). Through
the upper sections, before the fracture line trans-
gresses the posterior border of the bone, it divides
the anterior aspect of the anterior column and the
quadrilateral surface (Fig.9.12E). However, we
have seen one instance of transverse fracture,
typical in all its other features on the three views,
which had a fracture line oriented at a 70° angle
forwards and outwards (Fig. 9.13).
The fracture begins on one section by detaching
a small piece of the pelvic brim. The involved seg-
ment increases outwards in size on successive sec-
tions, always separated by a sagittal fracture line.
On the roof section, the fracture line may not yet
be visible (infra-tectal transverse), or may skirt the
inner border of the compact area Guxta-tectal
transverse), or may involve more or less of the com-
pact roof (trans-tectal transverse). However, it is
essential to note that in pure transverse fractures
CT always also demonstrates at least a lateral piece
of the anatomical roof that is intact and normally
connected to the iliac wing.
Below the roof, on subsequent sections, the frac-
ture line involves the anterior and posterior walls.
These are each divided by an antero-posterior frac-
ture line, in line with one another (Fig. 9.13 E, F).
When the transverse fracture line slants
anteriorly and superiorly, the posterior wall is in-
volved lower than the anterior wall, and the frac-
ture can be seen dividing just the anterior border of
the acetabulum while there is still a full fracture line
dividing the posterior wall (Fig. 9.14). On further
155
sections below this one, the fracture line displaces
outwards, involving only the posterior wall, and
finally cuts the posterior border of the acetabulum.
In these late sections one must be careful not to
take the postero-inferior part of a transverse frac-
ture for an associated posterior wall fracture, in
spite of the fact that in the lowest sections the frac-
ture line may become directed outwards and for-
wards.
The iliac wing and obturator ring are not in-
volved in this injury, and sections through them
show intact bone.
The displacement of the iliac fragment is easy to
analyse: its inferior part may have displaced out-
wards and may sometimes have rotated slightly ex-
ternally; the latter only occurs when permitted by
an anterior opening of the sacro-iliac joint, which
is clearly visible, and the rotation of the iliac wing
can be assessed by comparison with the other side,
if intact.
The displacements of the ischio-pubic fragment
are more complex. An internal rotation around the
vertical axis of the symphysis pubis is perfectly
demonstrated on CT, the ischium displacing much
more than the pelvic brim. On the other hand, rota-
tion around a horizontal axis, which is so frequent
and which displaces the upper part of the fragment
inwards, is more difficult to appreciate, unless the
sections of the pelvis are perfectly symmetrical and
allow analysis of the displacement by comparison
with the other side.
The displacement of the femoral head is also
very easy to assess: the head may have remained
perfectly centred under the intact part of the roof,
with - if the sections are thin - a complete or
nearly complete "roundel" appearance on the sec-
tion just below the roof. Furthermore, when the
head displaces it follows the ischio-pubic fragment,
often conserving perfect congruity with the parts of
the acetabular walls attached to this fragment.
The localised impressions or impactions of the
femoral head at the time of trauma, or, in old cases,
the wear of the displaced head against the fracture
line of the iliac segment, can be seen on standard
CT, but obviously 3-D reconstruction allows a
much better view of them.
 156 Pure Transverse Fractures

A 8

Fig. 9.12A-G. Juxta-tectal fracture without central dislocation of the

head. The fracture divides the middle part of the ilio-pectinealline and
the ischial spine posteriorly. A-C The three standard views, D-G CT
sections. D The fracture line divides the anterior aspect of the anterior
column and the quadrilateral surface. E The fracture is orientated
antero-posteriorly and has spared the roof. In F, the transverse fracture
line can be seen dividing the anterior and posterior walls in line. G The
fracture being oblique inferiorly and posteriorly, the posterior wall is
now just involved, but in a slightly more oblique direction, and this
must not be confused with an additional posterior wall fracture C
 CT Scan Study of Transverse Fractures 157

D E

F G
Fig. 9.12D-G
158
Pure Transverse Fractures

Fig. 9.13 A-F. Pure juxta-tectal transverse fracture with central

subluxation of the head. Four months after injury. A-C The
three standard views. Notice the cartilage wear at the level of the
inner border of the roof. 0 - F CT sections showing the antero-
posterior direction of the fracture line, which can be seen in
E to have spared the roof which obviously belongs to the superi-
or iliac fragment
 CT Scan Study of Transverse Fractures 159

E F

Fig. 9.13E,F
 160 Pure Transverse Fractures

A B

Fig. 9.14A-G. Juxta-tectal transverse fracture running oblique-

ly anteriorly (cutting the ilio-pectinealline superiorly and the in-
ferior part of the greater sciatic notch posteriorly), accompanied
by a central dislocation of the head. A-C The three standard
views, D - K CT sections showing that, unusually, the transverse
fracture is oriented at about 60° forwards and outwards. It cuts
the anterior lip very high up and the posterior lip at a much
C lower level
CT Scan Study of 1fansverse Fractures 161

D E F G

H J K
Fig. 9.14D-G
10 T-sbaped Fractures
It is appropriate to begin the study of associated
and complex varieties with the T-shaped fracture.
Although the outlines are simple, the radiological
interpretation can be difficult. Diagnosis must be
accurate because of the difficulty of open reduc-
tion which sometimes necessitates two surgical
approaches or an extensile approach.
Typical T-shaped fractures associate a transverse
fracture with a vertical split which divi.des the
ischio-pubic component into two parts, and which
passes through the middle part of the obturator
foramen (Fig. 10.1).
In our earlier publications we used to consider
these fractures relatively rare, having had only three
examples amongst our first 75 cases. If we hold to
our definition of the typical lesion, they remain
fairly rare: subsequently we have had only 34 cases
out of the total of 940. With more experience over
the years, we have encountered fractures associat-
ing a transverse configuration with a split in the in-
ferior segment cutting the outline of the obturator
foramen obliquely, sometimes in a forward direc-
tion and sometimes backwards. Now we include
under the heading T-shaped, all fracture forms
which associate a transverse fracture of the
acetabulum, at whatever site, with an oblique or
vertical split traversing the ischio-pubic segment
(Table 10.1, Fig. 10.2).
Previously, associated fractures of the posterior
column with an anterior hemitransverse fracture
had been regarded as a separate group. However,
the outline of this associated fracture complex is so
near to that of a typical T-shaped fracture that it is
difficult to differentiate them on the standard
radiographic views. The upper segment of the pos-
terior column component, together with the
anterior hemitransverse component is approximate
to the configuration of a transverse fracture from
which arises the vertical split of the ischio-pubic
component, i.e. the lower segment of the posterior
column component (Fig. 10.3).
Table 10.1.
T -shaped fractures 66 7.02070
Vertical 41 4.36%
Anterior 12 1.27%
Posterior 13 1.38%
Without changing our decision to include this
association among the T-shaped fractures, CT has
brought proof of its reality. Figure 1O.4A-C shows
what appears to be a T-shaped fracture, but the CT
scans (Fig. 10.4 D - F) demonstrate the posterior
column fracture to be separated by a coronal frac-
ture line and show the typical transverse fracture
line dividing the anterior column in an antero-
posterior direction.
The elimination of this associated fracture sim-
plifies classification and causes no practical in-
convenience, either radiologically or from the point
of view of surgical treatment, as they require the
same surgical approach.
Later we shall describe the symmetrical lesions
to those we have just eliminated from our classifi-
cation, i.e. associated fractures comprising a frac-
ture of the anterior column with a posterior hemi-
transverse component. We do not include these
with the T-shaped fracture group because the
anterior column fracture components and the pos-
terior hemitransverse fractures are easily identified
and cannot be taken for T-shaped fractures.
10.1 Morphology
This can be described according to the transverse
component, the vertical component, and femoral
head displacement.
164 T-shaped Fractures

10.1.1 Transverse Component

In our 66 T-shaped fractures, the transverse compo-

nent was trans-tectal in 18 cases, juxta-tectal in"30
and infra-tectal in 18.
No further description is necessary because the
transverse element is identical to the pure transverse
fracture described in the previous chapter.

10.1.2 Stem Component

(a) Vertical T-shaped fracture. This is the com-

Fig. 10.1. Scheme of a typical T -shaped fracture
monest (41 cases). The split descends through
the middle of the cotyloid fossa to cut the
ischio-pubic notch, and then traverses the ob-
turator foramen to fracture the ischio-pubic
ramus in its middle (34 cases) or at two points
(7 cases) (Fig. 10.1).
(b) Anterior T-shaped fracture. In 12 cases out of
66, the split is obliquely disposed downwards
and forwards. It runs adjacent to the anterior
horn of the articular surface, cuts the ischio-
pubic notch very anteriorly and then, on the
other side of the obturator ring, splits the body
or the angle of the pubis (Fig. 10.5 A).
(c) Posterior T-shaped fracture. In the remaining
~ ,
13 cases, the split of the inferior fragment is
"'A stem 0 fT Inclined
" " " Iy
anterior obliquely disposed posteriorly and downwards
(anterior T-shaped fracture)
(Fig. 10.5 B), but here its direction is variable.
Most frequently it descends to the level of the
ischial body, detaching a sizeable bone frag-
Fig. 10.2. Scheme showmg possible direcllons of the stem of the
T m T -shaped fractures ment from the ischium itself. This fracture line
usually spares the obturator ring (11 cases out
of 13). More precisely, the fracture line
descends in the cotyloid fossa, following the
margin of the posterior horn of the articular
surface and then passing between it and the
edge of the obturator foramen. Descending fur-
ther on the external surface of the ischium it
finally cuts its posterior border. From the inner
side of the pelvis the fracture line is seen to be
slightly concave posteriorly and to cross the
posterior part of the quadrilateral surface so as
to meet the site of rupture of the posterior
border of the ischial tuberosity. In one case, the
fracture line cut the hindmost part of the bony
margin of the obturator foramen to which it
was almost tangential (Fig. 10.11).
At this point a little difficulty arises. We
Fig. 10.3A,B. T-shaped fractures" Schemes showing why it IS
difficult to distinguish between A the classical T configuration have been challenged for including these latter
and B that associating a fracture of the posterior column with types of fractures in the T-shaped group be-
an anterior hemltransverse component cause they leave the obturator foramen intact,
 Morphology 165

A B

c E
Fig. IO.4A - C. The three standard views of an associated posterior column and anterior hemitransverse fracture, D - I typical CT
sections (see text)
 166 T-shaped Fractures

F G

H
Fig. 10.4 F - I

Fig. 10.SA,B. T-shaped fractures. Schemes showing classification of fractures according to the direction of the stem of the T.
A Anterior, B posterior
Radiology
and in Chap. 12 we shall be describing associat-
ed transverse and posterior wall fractures,
among which the inclusion is controversial.
We have demonstrated that every transi-
tional type between all types of acetabular frac-
tures exists, and this is particularly true here.
Between what we call a posterior T-shaped frac-
ture with the stem tangential to the obturator
foramen (A in Fig. 10.6) and a transverse frac-
ture associated with an indisputable posterior
wall fracture (B in Fig. 10.6), every transitional
type is possible, as all obliquities of the stem
have been seen (C in Fig. 10.6). However, I like
to call a "posterior wall fracture" one that has
not disturbed either the quadrilateral surface or
the cotyloid fossa, and to call a "posterior col-
umn fracture" one that involves these. It is in-
disputable that the stem of a "posterior T-
shaped fracture" divides these two surfaces,
which are left undisturbed by a posterior wall
fracture.
10.1.3 Displacement
In our series of 66 T-shaped fractures, the trans-
verse component was severely displaced in 52
instances, moderately displaced in seven and undis-
placed in seven. The vertical component displace-
ment was severe in 32 cases, moderate in 24 and
absent in 10.
Fig. 10.6. Diagram showing the mUltiple possibilities for a frac-
ture line additional to a transverse fracture line to divide the
posterior column. B j and B2 are typical posterior wall fractures
(they do not involve the inner aspect of the bone). A is the ver-
tical stem of aT-shaped fracture tangential to the obturator
167
The head of the femur usually appears dislocat-
ed centrally (50 cases). In 5 of the 13 cases of T-
shaped fractures in which the oblique branch was
disposed backwards, downwards, and trans-ischial,
the head was apparently dislocated posteriorly. In
two cases the head appeared to have dislocated
anteriorly and in six cases it spontaneously reduced
under the intact part of the roof.
Among the associated lesions we have observed
five cases of associated bilateral vertical anterior
fractures of the obturator ring and four single frac-
tures of the superior pubic ramus.
10.2 Radiology
The two components of this associated fracture
should be identified sequentially.
10.2.1 Transverse Component
On the antero-posterior and oblique views all the
vertically disposed landmarks, as well as the more
slanting ones relating to the acetabulum, are
broken. The upper segment of each of these re-
mains attached to the iliac wing, which itself is
intact. The upper segments of the vertical and
oblique landmarks are limited inferiorly by a frac-
ture line which has the regularity of the plane of the
Horizontal diagram
foramen. C (C j , C2 , C3) shows the possible tracks of the stem
of a T -shaped fracture dividing the posterior column but not the
cotyloid fossa (they divide the quadrilateral surface). Infact A
and C have an identical track though the cotybrid fossa, and B
and C cannot be on the same horizontal cut
168 T-shaped Fractures

pure transverse fracture. The level of rupture of the
landmarks varies with the obliquity of the fracture
in its transverse aspect.
The roof is involved or left uninjured according
to whether the transverse element is trans-tectal
(18 cases), juxta-tectal (30 cases) or infra-tectal
(18 cases). Always, there remains an intact segment
of roof attached to the iliac wing, and one to which
at operation it will be possible to bring back and
maintain the head of the femur; this is an impor-
tant difference between T-shaped fractures and
fractures of both columns. Displacement at the
transverse element is almost always significant (65
cases out of 66).
10.2.2 Stem Component
The ischio-pubic component characterises a T-
shaped fracture and is not always easy to recognise.
(a) Vertical T-shaped fractures. While the rupture
of the ischio-pubic ramus in one or two places
is seen readily, the fracture through the cotyloid
fossa and that through the ischio-pubic notch
may be much more difficult to confirm: before
the advent of CT scanning, it remained invisi-
ble in five examples where there was a fracture
of the ischio-pubic ramus accompanied by little
displacement (Fig. 10.7). It is difficult to con-

'./ .......
r-
. "\
I .
'- '".\Ii. .'
I Fig. 10.7 A,B. T-shaped fracture. A Antero-posterior radio-
Y ! graph, a diagram, B iliac-oblique radiograph. This is a classical
a ( I appearance, in which a fracture of the middle of the pubic
ramus is associated with a single fracture of the pelvic brim
Radiology
ceive that a transverse fracture could be asso-
ciated with a fracture of the ischio-pubic ramus
without there being also a fracture in the coty-
loid fossa to complete the vertical segment of
the T. Practically then, whenever a transverse
fracture is seen associated with a rupture of the
ischio-pubic ramus, it should be possible to
confirm that it is a T-shaped fracture. Now-
adays, as will be described later (see Sect. 10.4),
CT scanning will show the fracture through the
cotyloid fossa unmistakably.
It is important to avoid confusing a T-
shaped fracture with a transverse fracture asso-
ciated with an anterior fracture of the obtura-
tor ring, outside the acetabulum (Fig. 9.9). In
both instances, the ischio-pubic ramus will be
broken at one or two points. In a T-shaped frac-
ture there is only one fracture of the pelvic
brim, whereas an anterior fracture of the ob-
turator ring in addition to a pure transverse
fracture will present with two fractures of the
brim and CT scanning will make identification
easy as it will show an intact cotyloid fossa at
the level of the ischio-pubic fragment.
The stem fracture line is easy to confirm
when the two fragments of the column are sep-
arated on the antero-posterior view or on an
oblique view, or when the fracture through the
ischio-pubic notch is visible on the antero-pos-
terior view or more easily on the obturator-
oblique view (Fig.l0.8).
In the few cases where it was not possible to
see the fracture through the ischio-pubic notch,
especially when the obturator-oblique view was
not available (as in earlier cases of our series),
we believe that we can confirm the fracture in
the cotyloid fossa by the following indirect
signs:
superimposition of inferior fragments of
both columns on the antero-posterior view
(Fig. 10.9);
displacement of the teardrop with respect to
the ilio-ischial line on the antero-posterior
view, explainable only by a split in the coty-
loid fossa and a separation of the columns
(Fig. 10.8);
the crossing of the two columns on the iliac-
oblique view which gives an X-shaped con-
figuration.
Thus in order to establish the presence of a ver-
tical T-shaped fracture, the rupture of the
169
ischio-pubic branch seems to us sufficient but
one should search further for absolute proof:
a fracture through the ischio-pubic notch
especially visible on the obturator-oblique
view;
the vertical fracture line passing through the
quadrilateral surface as seen on the iliac-
oblique view, although on many occasions
this is so difficult to see (Fig. 10.8). Now-
adays, CT scanning demonstrates perfectly
the fracture through the cotyloid fossa.
(b) Anterior T-shaped fractures. The fracture
through the obturator ring (Fig. 10.10) is sited at
the most anterior part of the ischio-pubic ramus
or at the angle of the pubis. In a few cases,
moreover, the bony ring has been ruptured in
two places. In 7 cases out of 12 the line dividing
the cotyloid fossa was very anterior so that the
teardrop was detached with the fragment of the
posterior column. The ischio-pubic notch was
apparently broken at the level of the acetabulo-
obturator line, in its most anterior part.
In two cases, although the teardrop was de-
tached with the fragment of the anterior col-
umn, the split between the two column frag-
ments remained visible in the antero-posterior
view. In these cases one could observe on this
view a separation of the U and the ilio-ischial
line.
In one particular transitional case, a trans-
verse fracture was associated with a split in the
cotyloid fossa and the ischio-pubic notch; at
the same time, the ischio-pubic ramus was not
fractured. It was an incomplete T-shaped frac-
ture.
(c) Posterior T-shaped fractures. In these the ver-
tical split is in general clear and does not pose
any particular problem. On the antero-posteri-
or view, it may be apparent that it cuts the most
posterior part of the obturator foramen
margin, detaching a segment comprising the
body of the ischium which forms the posterior
part of the obturator ring (Fig. 10.11). Most
frequently, it splits the body of the ischium and
detaches an enormous fragment of this bone
while leaving intact the obturator foramen
(Figs. 10.12 and 10.13). It is of interest to com-
pare the radiographic appearances shown in
Figs. 8.12 and 10.12, which are at first sight so
similar.
The teardrop and the ilio-ischial line are
always disturbed with respect to each other
and, in 4 cases out of 13, there was an ap-
170 T-shaped Fractures

pearance of duplication of the ilio-ischial line

(Fig. 10.12).
The oblique views only confirm the ex-
istence of the vertical split and the usual in-
tegrity (11 cases out of 13) of the obturator
foramen. It is the obturator-oblique view which
best delineates the posterior fracture line.

10.3 Atypical Examples

10.3.1 Additional Vertical Fracture
of Obturator Ring
Sometimes, associated with the T-shaped fracture,
there is an accessory vertical fracture line through
the obturator ring. Such an additional fracture
presents no therapeutic difficulty (Fig. 10.14). The
obturator-oblique view shows the main vertical
split in the cotyloid fossa, and more anteriorly, the
associated vertically-disposed fracture of the obtu-
rator ring which divides the superior pubic ramus
and which may "share" the foot of the stem of the
main vertical component of the T through the
ischio-pubic ramus. In most instances, the antero-
posterior view of the pelvis reveals a fracture of the
obturator ring of the opposite innominate bone. A
The fracture is recognisable as T-shaped, for ad-
jacent to the transverse component, there is a split
in the cotyloid fossa clearly visible on the ob-
turator-oblique view. The other fracture line, more
anterior, interrupts the outline of the obturator
foramen and cuts the ilio-pectineal line below the
transverse fracture component, isolating therefore a
segment of the pelvic brim. This segment is a frag-
ment from the anterior column relating to the T
complex proper.

10.3.2 Additional Fracture Line

in Cotyloid Fossa

We once operated upon a patient with an old frac-

ture (Fig. 10.15) where, in addition to a juxta-tectal
transverse fracture line, there was a double vertical
split almost totally isolating the cotyloid fossa; the
femoral head was dislocated posteriorly and the
acetabulum was occupied by the psoas and the ob-
turator muscles, rendering the dislocation irreduci-
ble. Moreover, the two main column fragments had Fig.l0.8A-C. Vertical T-shaped fracture. A Antero-posterior
come together and had united at their adjacent ar- radiograph showing the impacted fragment of roof, a diagram,
ticular borders. The intermediate segment carrying B obturator-oblique radiograph on which the fracture line of the
superior part of the obturator ring is visible, b diagram, C iliac-
most of the cotyloid fossa had displaced inwards oblique radiograph showing fracture line separating the two col-
with respect to the two preceding segments and umns, c diagram. The stem component of this fracture has two
complicated the malunion further. fracture lines in the inferior pubic ramus
 AtYPlcal
' Examples

.".- .- ........ '

/ ",-, ~/
\, I '
</'j

Fig. l0.8B-C
172 T-shaped Fractures

/
/ b Anterior column
I
\
\
\
\
a i

Fig.lO.9A,B. Vertical T-shaped fracture. A Antero-posterior position in A and their separation in B furnishes sufficient
radiograph, a diagram, B iliac-oblique radiograph, b diagram. evidence of a stern and therefore this is a T-shaped fracture
The separation of the columns is not visible, but their superim-
Atypical Examples 173

....... ,
'- \ ( '\
\" \. \ i
i
/ i
i i
I i
I
a I

Fig. lO.lOA-C Anterior T-shaped fracture. A Antero-posterior a separated fragment. The fracture line in the cotyloid fossa lies
radiograph, a diagram, B obturator-oblique radiograph on very anteriorly and the obturator ring is cut at the level of the
which the fracture of the ischio-pubic notch is visible, b dia- angle of the pubis
gram, C iliac-oblique radiograph, c diagram. The teardrop is on
174 T-shaped Fractures

Fig. 10.11. Posterior T-shaped fracture. A Antero-posterior ra-

diograph, a diagram. Note the separation of the teardrop from
Fig.IO.loe the ilio-ischial line
Atypical Examples 175

'\
\ .,--
' --!- -\
\ \
) \
/ .i
i
/
/
I i
\ i
a
'1 i
A

~ . ~\

i
'-- j i
!
J
i
/
b /
8
Fig.l0.12A-C. T-shaped fracture with trans-ischial stem. reduplication of the ilio-ischial line proves that there is a frac-
A Antero-posterior radiograph, a diagram, 8 obturator-oblique ture through the posterior part of the quadrilateral surface. The
radiograph with a well-exposed split of the ischio-pubic frag- image of the more internal of the ilio-ischiallines has a normal
ment through the ischium, b diagram, C iliac-oblique relationship with the teardrop. The iliac fragment is cut inferior-
radiograph, c diagram. The obturator ring is intact. The ly as in a transverse fracture
176 T-shaped Fractures

(
\_.-.
,,\. '.\
\ \
.----". ! .)
\ I .
\J.
.
!\
j .
. \
I. .
,.,.,
c
Fig.tO.t2e

Fig.tO.13A,B. T-shaped fracture with posterior stem splitting

the ischium. A Antero-posterior radiograph with pelvic brim
only a little displaced, B obturator-oblique radiograph on which
it is even less displaced. Note the purity of the transverse ele-
ment
Atypical Examples 177

Fig.lO.14A-C. T-shaped fracture with associated fracture of

the superior pubic ramus. A Scheme, B iliac-oblique radiograph,
~ diagram, C obturator-oblique radiograph showing the fracture
lines of the ischio-pubic ramus, the superior ramus and the
Ischio-pubic notch, c diagram

c
178 T-shaped Fractures

I
I
i
\. i
\\ \
\ \
b \ i
B

Fig.l0.15A-D. T-shaped fracture with double stem in the A Scheme, B antero-posterior radiograph, b diagram, C obtura-
cotyloid fossa separating a segment of the quadrilateral surface. tor-oblique radiograph, D iliac-oblique radiograph, d diagram
CT Study of T-Shaped Fractures 179

10.3.3 Association of a Posterior Column

and an Anterior Hemitransverse Fracture

Figures 10.16 and 10.19 show very convincingly a

posterior column fracture associated with an ante-
rior hemitransverse fracture, an example which we
include with the T-shaped fractures. The case of
Fig. 10.16 is also unusual because the anterior frac-
ture line appears passing below the tip of the
anterior horn; starting from the pelvic brim it
crosses the root of the superior ramus of the pubis,
as is seen on both the antero-posterior and ob-
turator-oblique views. This is not an associated ver-
tical anterior fracture of the pelvis for the pelvic
brim is broken only at one single point. We include
it with the T-shaped fractures for there is some dif-
ficulty in distinguishing it radiographically from
the case shown in Fig. 10.10. Note that only the
posterior wall of the articular surface had been
detached in this particular case.
CT scanning has now made the identification of
such cases much easier (see Fig. 10.4), as the sec-
tions through the roof and below it show the com-
bination of a more or less coronal fracture line
separating off the posterior column with a sagittal
fracture line dividing the anterior column (see
Fig. 10.19).
D

10.4 CT Study of T-Shaped Fractures

/ . - ,-.... . ...---~---

/
. / The sections through the iliac wing show intact
/F' ,...._) ___ ..../
bone.
i
Depending on the level of the transverse compo-
!
I nent, the sub-chondral bone plate of the roof is
\. either involved (T-shaped fracture with a trans-
\ tectal transverse component) or spared (juxta-tectal
\ or infra-tectal transverse component).
I The transverse fracture component of the T
I shape is, as always, antero-posteriorly oriented and
d \ extends further and further laterally as the sections
progress downwards.
If the transverse fracture is oblique downwards
and anteriorly or upwards and posteriorly, the
anterior wall or the posterior wall may appear in-
volved solely by the inferior part of the transverse
component in the lower CT sections, and this must
not be taken to be an additional wall fracture. It is
an easy mistake, but it should be noticed that the
fracture line separating the wall is antero-posterior-
ly oriented and not obliquely outwards and for-
Fig. IO.ISD wards as in pure wall fractures.
180 T-shaped Fractures

A B

Anterior wall
of acetabulum
.""'"" """"- 0

I
.I I
.
i \..
\. \\
l.
\
\
\
\
a \

Fig.l0.16A-C. Fracture of the posterior column with a hemi- but does not displace the anterior column. It is probably extra-
transverse anterior fracture. A Antero-posterior radiograph, articular for it cuts the pelvic brim, in Band C it transgresses
a diagram, B obturator-oblique radiograph, b diagram, C iliac- the root of the superior pubic ramus
oblique radiograph, c diagram. The hemitransverse element cuts
CT Study of T-Shaped Fractures 181

The vertical stem divides only the ischio-pubic

fragment and is particularly easy to recognise on
the CT sections. It divides the cotyloid fossa close
to the anterior wall (anterior T-shaped fracture) or
close to the posterior wall (posterior T-shaped frac-
ture; Fig. 10.18) or more or less medially (vertical T-
shaped fracture; Fig. 10.17).
The ischio-pubic notch is divided in vertical or
anterior T-shaped fractures, but is spared in most
posterior T-shaped fractures (Fig. 10.18). In the lat-
ter the fracture line dividing the ischium is perfectly
visualised on CT. The involvement or otherwise of
the ischio-pubic ramus cannot escape detection in
a complete CT examination.
Of course, the displacement of the femoral head
and marginal impaction of the inner part of the
roof, if any, are perfectly displayed on CT sections.

Fig.l0.16C
 182 T-shaped Fractures

A B

c E
 CT Study of T-Shaped Fractures 183

t~V
. '·· i.• .,
R
~ r; L

F G

Fig. 10.17 A-G. T-shaped fracture with a vertical stem and two
points of rupture of the ischio-pubic ramus. A-C The three
radiographic views show a separated segment of the inner aspect
of the sciatic buttress. The transverse component involves the in-
nermost part of the roof. D - G CT sections. Above the roof the
fracture line slants slightly inwards and forwards in a nearly
antero-posterior direction. E A thin segment of the pelvic brim
separates the segment of the two columns (arrow). F The inner
part of the roof appears to be involved and the fracture line
slants slightly inwards and forwards. G The vertical stem skirts
the inner limit of the anterior wall, with some localised com-
minution
 184 T-shaped Fractures

A B

Fig.l0.18A-H. T-shaped fracture. A-C The three standard

views. The stem slanting posteriorly and backwards, transgres-
sing the ischium, is well seen on the antero-posterior and ob-
turator-oblique views. A small additional posterior wall frag-
ment appears on the obturator-oblique view, apparently not
bringing with it a segment of cartilage. D-H CT sections. The
transverse fracture line has an antero-posterior orientation and
the inferior or inner fragment is undoubtedly divided by a cor-
onal fracture line, i.e. the vertical stem. F The inner part of the
roof is involved, and so is its posterior part, which has lost some
articular surface, infact borne away by the small posterior wall
fragment. G The stem of the T does not disturb the ischio-pubic
notch and begins to run through the ischium, as is confirmed in
C
H
 CT Study of T-Shaped Fractures 185

E F

G H

Fig. lO.lSE-H
 186 T-shaped Fractures

A B

Fig.l0.19A-G. Associated posterior column and anterior he-

mitransverse fracture. A-C The three standard views. The asso-
ciated fracture could be made out because the fracture line
separating the posterior column has a quite different orientation
from that of the fracture line transgressing the anterior column;
the anterior fracture looks like a transverse on the antero-
posterior and, especially, on the obturator-oblique view. D-G
CT sections. The posterior column is separated by a fracture line
directed forwards and inwards at about 30 0 - 40 0 (like a true
posterior column fracture), and F, G an additional fracture line
with an antero-posterior orientation (like the line of a transverse
C fracture) divides the anterior column
CT Study of T-Shaped Fractures 187

D E

F G

Fig. lO.19D-G
11 Associated Posterior Column and Posterior Wall Fractures
This association comprises a fracture of the poste-
rior wall of the acetabulum in one or several frag-
ments, perhaps with marginal impaction, together
with a fracture of the posterior column. The latter
is not always complete and frequently little or not
at all displaced.
11.1 Morphology
11.1.1 Posterior Wall Component
This can be represented by any of the types that we
have already described. In our series these were
postero-superior in 10 cases, typical in 21 cases,
and postero-inferior in one case; in 18 examples the
displaced posterior wall fragment was in one piece,
and in 14 cases it was comminuted. There was mar-
ginal impaction in some examples.
The posterior wall fragment or fragments re-
mained congruent with the femoral head when it
was dislocated, but remained displaced and away
from it when it had been reduced.
The posterior dislocation of the femoral head
was almost constant (23 cases out of the 32). In
four cases the head had spontaneously reduced and
was found concentric with the roof and the anterior
wall, which were intact. In five cases the dislocation
was central.
11.1.2 Posterior Column Component
The fracture line begins in the cavity created by the
posterior wall fracture and reaches the greater
sciatic notch (Fig. 11.1), most commonly a little in-
side the angle, exceptionally below this point, from
the upper half of the anterior border of the greater
sciatic notch.
On the quadrilateral surface the fracture line has
the characteristics of the posterior column fracture
Table 11.1
Associated posterior column and posterior 32 3.40070
wall fractures
Posterior column and posterior wall 21 2.230/0
Posterior column and posterior superior wall 10 1.05070
Posterior column and posterior inferior wall 0.12070
and, like it, most commonly reaches the ischio-
pubic notch, but never transgresses the ilio-pec-
tineal line (Fig. 11.1 A). In six cases, the column
separated from the ischial tuberosity and did not
disturb the obturator ring (Fig. 11.1 B).
The ischio-pubic ramus was broken in only one-
quarter of our cases and more usually the posterior
column fracture line split the ischio-pubic notch
but spared the ischio-pubic ramus. It is almost as
if the fracture energy exhausted itself before being
able to achieve a rupture here. This incomplete frac-
ture of the posterior column suggests that the in-
jury is primarily a fracture of the posterior wall and
the other component, the detachment of the poste-
rior column, is secondary.
The displacement of the posterior column is
variable in magnitude: there was no displacement
in one case in three; in only 10 cases of the 32 was
the angle of the greater sciatic notch significantly
displaced into the pelvis.
11.2 Radiology
11.2.1 Antero-posterior View
Two sets of features are striking: firstly, the un-
broken ilio-pectinealline and anterior border of the
acetabulum, which confirm that the anterior col-
umn is intact; and secondly, the posterior disloca-
tion of the femoral head which displaces, in one or
several fragments of significant size, the posterior
190
A and therefore of the anterior column, and delin-
B
Fig. 11.1A,B. Schemes of associated posterior column and pos-
terior wall fractures
wall (Figs. 11.2, 11.5) or the posterior superior sec-
tor of the acetabulum or the postero-inferior seg-
ment (Fig. 11.3). If it is just the central part of the
posterior wall which is displaced, the roof is seen to
be intact but if it is a postero-superior segment,
part of the roof is included (see Fig. 11.7). A
postero-inferior fragment remains concentric with
the femoral head and includes the sub-cotyloid
groove and the upper part of the ischial tuberosity.
In one case out of two, the head of the femur re-
mained related to some fragments due to marginal
impaction (Fig. 11.4).
The posterior column fracture is recognised
either by its obvious displacement inside the outline
of the pelvic brim, by the clear visibility of the
posterior border of the innominate bone with the
ischial spine and the angle of the greater sciatic
notch (Fig. 11.3), or because the ischio-pubic ramus
is broken at a variable point, being sometimes asso-
ciated with a fracture of the postero-superior mar-
gin of the obturator foramen (Fig. 11.2).
Associated Posterior Column and Posterior Wall Fractures
The displacement of the posterior column may
be slight, in which case the ilio-ischialline keeps an
almost normal relationship with the teardrop. In
such cases, it is rarely possible to diagnose this pos-
terior column fracture from the antero-posterior
view alone (Fig. 11.2).
11.2.2 Obturator-oblique View
This confirms the integrity of the ilio-pectinealline
eates the size and the nature of the posterior
fragments. In particular, the direction of the in-
ferior part of the fracture line detaching the poste-
rior column can be seen. This line sometimes leaves
undisturbed the ischio-pubic ramus (Fig. 11.3) or
may break it in its middle part (Fig. 11.2) unless it
splits the ischium without involving the obturator
foramen. There is no fracture of the superior
border of the obturator foramen when the fracture
line splits the ischium.
11.2.3 Iliac-oblique View
This shows the amount of displacement of the
posterior column, the direction of the upper part of
the fracture line on the inner aspect of the pelvis,
and the level of rupture of the greater sciatic notch.
It confirms that the anterior border of the acetabu-
lum is intact (Figs. 11.3, 11.4 and 11.5).
11.3 Atypical Examples
The posterior dislocation does not always persist.
In four cases the head was in place, centred about
the undisplaced roof and the anterior wall; since
the posterior fragments remained displaced, it is
certain that the dislocation had spontaneously re-
duced.
The fracture of the posterior column, if it ex-
tends very far forwards including the teardrop and
the ilio-ischialline on the displaced fragment, pro-
duces a thinning of the upper part of the pelvic
brim seen in the simple forms described earlier in
extended fractures of the posterior column.
In one unusual case there was no fracture of the
ischio-pubic ramus but the obturator ring was
broken at the level of the superior pubic ramus.
Comment 191

i
/ .. ::::-/- \
I. .I
i !
\. \\
\
i
i
a b

Fig. n.2A,B. Associated posterior column and posterior wall border high up but reaches only the middle part of the greater
fracture. A Antero-posterior radiograph, a diagram, B iliac-ob- sciatic notch
lique radiograph, b diagram. The fracture line cuts the posterior

11.4 Comment column and posterior wall lesions are associated,

This association between a posterior column and a
posterior wall fracture must be differentiated from
the pure posterior column fractures. Certainly, the
latter are accompanied sometimes by detached
fragments in the upper part of the fracture com-
plex, but these fragments are small and do not
carry articular cartilage. Their removal does not
compromise the overall reconstitution of the ar-
ticular surface. The posterior column is always sig-
nificantly displaced. In contrast, when posterior
it is the fracture of the posterior wall which pre-
dominates, and the fracture of the posterior col-
umn is sometimes incomplete or often only slightly
displaced.
The special examples which associate a posterior
wall fracture and an incomplete posterior column
fracture, detaching a fragment from the ischium
and not involving the obturator foramen, form the
transition between certain very extended fractures
of the posterior wall and the more typical forms of
the association which we have just studied.
192 Associated Posterior Column and Posterior Wall Fractures

b /

Fig.11.3A-C. Associated posterior column and postero-inferi-

or wall fracture with posterior dislocation. A Antero-posterior
radiograph, a diagram. The postero-inferior fracture is clearly
visible and the posterior border of the iliac bone is projected on
the inner aspect of the pelvic brim. The ischio-pubic ramus is
not fractured. B Iliac-oblique radiograph showing the isolated
posterior border of the iliac bone, and the posterior column
displaced with the ischio-pubic ramus which acts as a hinge, b
diagram, C obturator-oblique radiograph. Note that the oblique
radiographs were taken with the beam at 45° to the film,
C
resulting in marked distortion
Comment 193

A
8

i
/+--
" '\
( r-\ \
I
.
L,!
\ I
//
I- ', r'
\ \
.
j \.I
b I ./
"

Fig. ll.4A-Co Associated posterior column and posterior wall turator-oblique radiograph, c diagram. The impacted fragments
fracture with marginal impaction. A Antero-posterior radio- are separated from the head by a translucent area in A and B
graph, a diagram, B iliac-oblique radiograph, b diagram, C ob-
194 Associated Posterior Column and Posterior Wall Fractures

11.5 CT Study of Associated Posterior

Column and Posterior Wall Fractures

CT allows recognition of the more or less displaced

posterior wall fragment, usually accompanied by
the femoral head (Figs. 11.6 and 11.7); the size of
the fragment can be seen and an associated margin-
al impaction cannot be missed (Fig. 11.6F). The
direction of the fracture line dividing the posterior
wall is forwards and outwards at an angle of
45°-60°.
In addition to this, a fracture line oriented cor-
onally or obliquely forwards and inwards detaches
the posterior column from the level of the greater
sciatic notch or a little below it. This fracture
(Fig. 11.6E) mayor may not involve the sub-chon-
dral bone of the posterior part of the roof, and
then divides the posterior column to reach the
cotyloid fossa, which is fractured frontally.
Sometimes the ischio-pubic notch is intact and the
fracture line divides the ischium, but usually the
ischio-pubic notch is ruptured and beyond the ob-
turator ring the fracture involves the ischio-pubic
ramus. The medial and external rotational
displacements of the posterior column are of a
variable amount.
c The head may reduce spontaneously and remain
stable under the intact part of the roof (Fig. 11.6).

Fig. 11.4C
CT Study of Associated Posterior Column and Posterior Wall Fractures 195

A B

I
i
...../

)
I
,//-----'
,
b

Fig. 1l.SA-C. Associated posterior column fracture and poste-

rior wall fracture with posterior displacement of column and
femoral head. A Antero-posterior radiograph, B obturator-
c oblique radiograph, b diagram, C iliac-oblique radiograph
 196 Associated PosterIor Column and Posterior Wall Fractures

A B

c D

Fig. 1l.6A-J. Associated posterior column and posterior wall tion. Several posterior fragments (A, B) and an incarcerated
fracture. A Antero-posterior view before reduction of the poste- piece of bone (D) are obvious. E-H CT sections. Above the
rior dislocation of the head, B - D standard views after reduc- roof, the fracture line separating the posterior column is angled
 CT Study of Associated Posterior Column and Posterior Wall Fractures 197

E F

G H

J
forwards and inwards at 60° (E); at the level of the roof (F) it carcerated piece of bone is visible in G, while the posterior wall
is coronal. The fracture line separating the posterior wall is as fragments can be seen in F-H. I,J 3-D reconstruction
usual angled at about 60° forwards and outwards. The in-
 198 Associated Posterior Column and Posterior Wall Fractures

B
A

Fig. 11.7 A-G. Associated posterior column and postero-supe-

rior wall fracture with a posterior dislocation. A - C The three
standard views. There is a big posterior wall fragment which has
detached the lateral part of the roof as seen on the obturator-
oblique view (8). D-G CT sections. Unusually, the posterior
column fracture has started by detaching the inner aspect of the
posterior column, leaving its external aspect in place. This is dif-
ficult to identify from the iliac-oblique view (indicator in C), but
is evident on the CT scan. In E the missing sector of the roof
can be seen attached to the posterior fragment. In F a marginal
c impaction is visible
 CT Study of Associated Posterior Column and Posterior Wall Fractures 199

o E

F G

Fig. 1l.70-G
12 Associated Transverse and Posterior Wall Fractures

The association of a transverse fracture with a frac-
ture of the posterior wall is frequent and represents
the second largest group of our complex fractures,
ranking just behind the both-column group and far
ahead of our other three groups. The association
has been recognised for a long time and is accom-
panied by dislocation of the femoral head which is
more commonly posterior (117 cases, 64% of the
total number of cases of this associated fracture) but
sometimes central (60 cases, 32%). The mechanical
theory, outlined in Chap. 2, explains and relates
these sub-categories to the site and direction of im-
pact. From a purely practical point of view, one
could combine all of these fractures into one group
because they are amenable to operative management
through the same approach and have similar
radiological characteristics, apart from the direc-
tions of dislocation. These are sometimes identical
or so similar that it is impossible to conclude, from
a radiological study of the fracture alone, which
might be the direction of dislocation associated
therewith. This notion is reinforced by the fact that
in several cases a posterior dislocation has been con-
verted after "reduction" into a central dislocation
(see Fig. 12.10). The same fracture can therefore be
associated with either type of dislocation.
We shall preserve a division into two subgroups,
depending on the type of dislocation, because me-
chanically they are produced by different modes of
impact. Above all, the incidence of complications
with associated transverse and posterior wall frac-
tures with posterior dislocation is striking: it is in_
this variety that we find most pre-operative sciatic
paralyses and most secondary osteonecroses of the
femoral head.
The 183 cases which will be reported in this
chapter comprise:
95 cases of the typical fracture patterns asso-
ciated with a posterior dislocation;
35 cases of the same patterns associated with a
central dislocation;
29 cases in which an associated T-shaped and
posterior wall fracture was combined in 19 cases
with a posterior dislocation and in 10 cases with
a central dislocation;

Table 12.1.

Total Posterior Central No

no. dislocation dislocation dislocation

}
Infra-tectal transverse and posterior wall 16 13 2
Infra-tectal transverse and postero-superior wall 31
Infra-tectal transverse and marginal impaction 14 12 2

}
Juxta-tectal transverse and posterior wall 52 43 9
Juxta-tectal transverse and postero-superior wall 5 3 2 77
Juxta-tectal transverse and marginal impaction 20 10 9
Trans-tectal transverse and posterior wall
Trans-tectal transverse and postero-superior wall
23
4
12
2
11
2 } 27

Transverse fracture and posterior wall fracture extending to iliac crest 18 3 15 18

T and posterior wall 30 19 to 30

Total 183 117 60 6 183

202 Associated Transverse and Posterior Wall Fractures

1 case ___"",--
16 cases ./'"
I

"
f"
I
8 casesl
I
1-
10 casesl_

o
Fig. 12.1A-D. Associated transverse and posterior wall frac-
tures. A Scheme showing infra-tecta! transverse component,
B scheme showing juxta-tecta! transverse component, C
diagram of plane of transverse component, 0 distribution of
sites of rupture of the anterior border of the greater sciatic notch
(55 cases)

18 cases of this associated fracture which have
a very special feature and will be discussed in
Sect. 12.4;
6 cases in which, with the same fracture pattern,
there was no dislocation. The head was perfect-
ly centred under the roof and the direction of
the original dislocation is unknown.
12.1 Cases with Posterior Dislocation
This association was seen in 95 cases of the whole
group.
12.1.1 Morphology
(a) Transverse component. This embodies all the
characteristics of the pure transverse fracture.
The line commences posteriorly in the notch
created by the separation of the fragment of the
posterior wall. It is equivalent to a transverse
fracture in which a portion corresponding to
the detached segment of the posterior wall is
lacking (Fig. 12.1 C). The fracture line can vary
both in its site and in its orientation as is the
case with the pure transverse fractures.
The exact site of rupture of the anterior
border of the acetabulum could only be local-
ised radiologically in 59 cases out of 95. It was
situated in 25 cases at the level of the notch of
the anterior border; in 26 examples it was
higher and in the remaining 8, lower.
The level of the rupture of the posterior
border of the pelvic bone was very variable;
Fig. 12.1 D shows the levels in 55 cases.
As is the case for the pure transverse frac-
ture, the level of the transverse component
could be classified into infra-tectal (25 cases),
juxta-tectal (56 cases) or trans-tectal (14 cases).
As in the pure transverse fractures the in-
ferior ischio-pubic fragment is displaced in-
wards. The displacement predominates at the
posterior part of the fracture due to the com-
Cases with Posterior Dislocation
pound rotation of the inferior fragment around
vertical and horizontal axes described in
Chap. 9. In 8 cases the displacement was absent
although the transverse component was com-
plete.
(b) Posterior wall component. In the majority of
cases (68 out of 95) this resulted in separation
of an area of posterior wall which was a single
fragment in 38 cases, in two fragments in 11,
and in several fragments in 19. In 5 cases a
postero-superior wall fracture was present. In
22 cases the posterior wall fracture was accom-
panied by marginal impaction, while the trans-
verse component was infra-tectal in 12 cases
and juxta-tectal in 10 cases.
The extent of this fracture is variable both in
relation to the articular surface and to the pos-
terior acetabular surface. Sometimes the whole
anterior border of the greater sciatic notch may
be detached in one separate fragment. The
posterior fragments, displaced posteriorly by
the femoral head, are always very displaced and
they never reduce even if the head has done so
and is apparently stable after reduction.
(c) Dislocation. Posterior or postero-superior dis-
locations are accompanied by a large posterior
capsular rupture in practically all instances.
Rarely, in the unusual cases with marginal im-
paction, the separated fragment remains at-
tached to the damaged capsular remnant.
12.1.2 Radiology
(a) Antero-posterior view. The posterior disloca-
tion is the most striking feature. The femoral
head projects behind the roof of the acetabu-
lum which is outlined and superimposed there-
on; it is in some degree of internal rotation and
its inner pole may appear slightly inside the line
of the pelvic brim.
The femoral head is surmounted or capped
by one or more fragments of the posterior wall
which are frequently visible through the rela-
tively transparent iliac wing: some supplemen-
tary fragments may appear to lie behind the
neck or even below it.
Besides the fragment of the posterior wall
and the posterior dislocation, it is possible to
confirm the presence of an associated trans-
verse fracture because all of the vertical and the
oblique landmarks of the acetabulum (ilio-pec-
tineal line, ilio-ischialline, anterior border and
cotyloid fossa) are broken while the obturator
203
ring is spared; on the inferior fragment, the
teardrop and the inferior segment of the ilio-
ischial line maintain their normal relationship.
The roof is intact in cases with an infra-tectal
or juxta-tectal transverse component (81 out of
95); it is involved in those with a trans-tectal
transverse component (14 cases), but even then
a part of it remains intact and normally con-
nected to the intact iliac wing. The ischio-pubic
fragment is displaced inwards as a whole.
(b) Obturator-oblique view: The posterior disloca-
tion together with the size of the posterior
fragments and the extent of the cavity left in
the posterior border of the acetabulum are all
delineated. The integrity of the obturator ring
and the degree of obliquity and site of the
transverse fracture are confirmed and the latter
is better appreciated.
(c) Iliac-oblique view. This confirms the integrity
of the iliac wing and of the roof, or at least a
part of it. The point of rupture and the dis-
placement of the posterior border of the in-
nominate bone can be located while, with more
difficulty, that of the anterior border of the
acetabulum may be seen (Figs. 12.2, 12.3 and
12.5).
(d) Overall radiological assessment. There are sev-
eral points in the radiological diagnosis which
can be summarised to advantage:
(i) Localisation of the posterior wall compo-
nent. Most of the time the fracture of the
posterior wall presents as a notch in the
posterior border as seen on the antero-
posterior and obturator-oblique views, the
extremities of the notch delineating the
missing segment.
It can generally be recognised on the
antero-posterior view whether the fracture
is simple or involves marginal impaction
(22 cases out of 95).
The volume of the one or several
posterior fragments is variable: sometimes
they comprise the whole of the posterior
wall and all of the retroacetabular surface
and even include the anterior border of the
greater sciatic notch, or they may be very
small.
(ii) Character of the transverse component.
This is most commonly juxta-tectal or in-
fra-tectal (Fig. 12.3), rarely trans-tecta!.
The obliquity of the antero-posterior
direction of the fracture line is often dif-
204 Associated Transverse and Posterior Wall Fractures

Fig. 12.2A, B. Associated juxta-tectal transverse and posterior .

,.-. "\ .
./
I .\
wall fracture with posterior dislocation. A Antero-posterior ra-
diograph, a diagram, B iliac-oblique radiograph, b diagram. The
i I"~
marked inward displacement of the ischio-pubic fragment is \ i
very prominent in its posterior part (A). The fracture line
\ hRoof
detaches the angle of the greater sciatic notch. The uneven con- \ of . \:
tour of the pubic symphysis reveals the site of rotation about a
j .""bbturator \ \
b "'i / canal V
horizontal axis

Fig. 12.3 A - C. Associated infra-tectal transverse and posterior distorted because the X-ray beam was not perpendicular to the
wall fracture with posterior dislocation. A Antero-posterior film. A fracture line passes through the inferior fragment below
radiograph, a diagram, B iliac-oblique radiograph, b diagram, C the transverse fracture line, through the ischial spine, separating
obturator-oblique radiograph, c diagram. The oblique views are the anterior border of the greater sciatic notch
Cases with Posterior Dislocation 205

I·~

i
I
\
\.
\
i
i
a i

\
\
\
\
c

Fig.12.3A- C. C
206 Associated Transverse and Posterior Wall Fractures

I
i
Separation of I
posterior /
border of i
ischium i
i
i

A a
Fig. 12.4. Associated infra-tectal and postero-inferior wall frac- The detached fragment does not remain adjacent to the femoral
ture with posterior dislocation. A Antero-posterior radiograph, head. There is pure dislocation in the right hip. Such fractures
a diagram. The postero-inferior fracture line is typical and does are transitional with respect to posterior T-shaped fractures
not involve the obturator ring. It cuts the lesser sciatic notch.

Fig. 12.5. Associated low infra-tectal transverse and posterior

wall fracture with posterior dislocation. A Antero-posterior ra-
diograph, a diagram. The transverse component splits the ischial
spine

~
-.
'1·-\ ,
. \
I \ ./".
\ \.. ~ \.

\ .
i /
; i
. I
A I I
a
 Cases with Posterior Dislocation 207

Roof
I
'-' ~.,
l \ -
\
L
\
\ Incarcerated
'.. fragment of roof
. \ \ .
I . "
.,,) J
IJ
~
.
.
......
( .--<.......
I) .......
i '"'" ,\. ,-.-....
I. .\

i\ \. \
\1. '\
~\ j
\. 1 ) '---'
'~
.......
'\ I/ Inferior
fragment ' ", . _ . ~
\ \ of posterior
a a \ \ acetabulum Stem of T
crossing ischium

Fig. 12.6. Associated vertical T-shaped and posterior wall frac- Fig. 12.7. Associated trans-ischial posterior T-shaped and poste-
ture with posterior dislocation. A Antero-posterior radiograph, rior wall fracture with posterior dislocation. A Antero-posterior
a diagram radiograph, a diagram. This view shows the vertical split in the
ischium. Note the incarcerated fragment of roof
208
ficult to judge, for as a rule the point of
rupture of the anterior border of the ace-
tabulum is poorly visible. In the coronal
plane the obliquity of the transverse frac-
ture varies a great deal and it may be very
important, e.g. the fracture line divides the
acetabular lips in their inferior moiety and
the pelvic brim very close to the sacro-iliac
joint, the fracture track appearing nearly
vertical on the obturator-oblique view
(Fig. 12.15). The fracture of the pelvic
brim may also be difficult to locate and it
may be only at the point of rupture of the
posterior border of the iliac bone that the
trajectory can be located accurately
(Figs. 12.1, 12.2 and 12.5).
(iii) Displacement of the ischio-pubic frag-
ment. This is identical with that explained
in Chap. 9 for pure transverse fractures,
and is measured from the displacement of
the same landmarks. Recapitulating brief-
ly, the ischio-pubic fragment pivots
mutually around two axes:
(1) The vertical axis passes through the
pubic symphysis and the displacement
is therefore greater posteriorly than
anteriorly; this mode predominates
most frequently (Fig. 12.2).
(2) The horizontal axis extends from the
pubic symphysis to the point of rup-
ture of the posterior border of the
pelvic bone; the displacement is most-
ly at the pelvic brim which tilts inward-
ly while at the posterior border there is
minimal shift. The body of the
ischium rotates outwards.
Both components occur together in
the majority of cases, and at operation
the inferior fragment appears driven
medially and at the same time tilted in-
wards in its upper part. CT is very
helpful in assessing the displacement
of this fragment.
12.1.3 Atypical Examples
(a) Association of the posterior wall fracture with
a T-shaped fracture (19 cases). In addition to
the fracture of the posterior wall and the trans-
verse component, there is a vertical split of the
ischio-pubic fragment with a vertical T-shaped
fracture configuration (Figs. 12.6 and 12.16) or
Associated Transverse and Posterior Wall Fractures
a T with a posteriorly directed oblique branch
dividing the body of the ischium and sparing
the obturator foramen (Fig. 12.7). The latter as-
sociated fracture is transitional with respect to
transverse fractures associated with a postero-
inferior wall fracture (discussed in Sect. 10.1.2).
(b) Incarceration of a fragment of the posterior
wall (6 cases).
(c) Transverse fracture with fracture of the iliac
wing. In 3 out of a group of 18 cases, a poste-
rior dislocation of the femoral head accom-
panied a transverse fracture associated with a
very large postero-superior fragment extending
up to the iliac crest. These cases will be studied
in detail in Sect. 12.4.
12.2 Cases with Central Dislocation
12.2.1 Morphology
A fracture identical in configuration to those de-
scribed above (Sect. 12.1) can exist with a central
dislocation of the femoral head (35 cases). There
are a few minor differences in the extreme examples
of both of these groups.
(a) Transverse component. This was juxta-tectal in
20 cases, infra-tectal in 2 cases and trans-tectal
in 13 cases. It was remarkable that in some jux-
ta-tectal examples, the transverse fracture line
was oblique, being directed very steeply up-
wards and forwards; it cut the ischial spine or
the sub-cotyloid groove posteriorly, and in
either instance the point of fracture of the pel-
vic brim was far posterior and therefore high.
The transverse fractures in this association
were always displaced, this being inevitable be-
cause the ischio-pubic fragment had been
driven inwards by the femoral head. The ele-
ments of the displacement of the inferior frag-
ment were the same as those described in previ-
ous cases.
In ten instances a vertical split of the in-
ferior fragment was associated with the
transverse fracture line, thus amounting to a T-
Fig. 12.8A-E. Trans-tectal transverse fracture dividing the roof
in its inner part, the ischio-pubic fragment taking with it the
angle of the greater sciatic notch. A - C The three standard
views. The posterior wall fracture is visible in B and is slightly
displaced. D, E CT sections demonstrating the transverse line
running obliquely forwards and outwards at 80°, dividing the
roof, and the posterior wall separated by a fracture line running
obliquely forwards and outwards at about 45 ° in D and about
60° in E
 Cases with Central Dislocation 209

A B

C E
210 Associated Transverse and Posterior Wall Fractures

A B

b
a

Fig. 12.9A-C Associated juxta-tectal transverse and posterior ro-posterior radiograph, a diagram, B obturator-oblique
wall (in two segments) fracture with central dislocation. A Ante- radiograph, b diagram, C iliac-oblique radiograph, c diagram
Cases with Central Dislocation 211

shaped fracture with one or two breaks in the

ischio-pubic ramus.
(b) Posterior component. There were 20 pure frac-
tures of the posterior wall, associated with 9
juxta-tectal and 11 trans-tecta! fractures, the
posterior wall fractures comprising one or more
fragments.
In five cases the postero-superior wall seg-
ment of the acetabulum was involved, in associ-
ation with an infra-tectal fracture in one case,
juxta-tectal fractures in two, and trans-tectal
fractures in two. In nine cases a juxta-tectal
transverse fracture was accompanied by a pos-
terior wall fracture with marginal impaction.
The posterior fragment or fragments tend to
be less displaced than in cases with posterior
dislocation, because they have not been moved
by the head in its displacement; they often give
the impression of having accompanied the fe-
moral head in its inwards displacement. They
are usually attached to a flap of joint capsule.
(c) Dislocation. Most commonly this was obvious
and severe, but in a few cases it was not so and
in seven the head had returned underneath the
roof of the acetabulum, and the fractures
appeared without dislocation on the first
radiographs.
c
12.2.2 Radiology

(a) Antero-posterior view. The transverse fracture

is as always confirmed by the rupture of the
vertical and the oblique landmarks related to
. . ., ", the acetabulum (the pelvic brim, the ilio-ischial
I \
line, and the anterior border of the acetabu-
\ lum). The transverse fracture line may be juxta-
I
i tectal (20 cases), trans-tectal (13 cases) or, much
I more rarely, infra-tectal (2 cases), then divid-
\ ing the ischial spine posteriorly and passing
obliquely upwards and forwards, thereby cut-
c
ting the pelvic brim in its middle part. The head
appears dislocated centrally and has lost all
contact with the roof. The ischio-pubic frag-
ment is displaced around the usual axes. A
postero-superior fracture is clearly visible as it
involves the external part of the roof. The
typical posterior wall fracture is more difficult
to see when it affects a middle part of the
posterior wall, with one or more fragments,
because these are superimposed on the shadows
of the femoral head or the femoral neck; the ir-
Fig. 12.9C regular outline of the posterior border of the
212
a
acetabulum, or of a large notch outlined at its
level, should lead the observer to look carefully
for the posterior fracture.
The transverse fracture line divides the
greater sciatic notch and finishes at the notch
created by the posterior wall fracture. The ob-
turator ring is most commonly intact, but the
presence of a rupture of the ischio-pubic ramus
in one or two places should lead to a search for
Associated Transverse and Posterior Wall Fractures
Fig. 12.10A-C. Associated transverse and posterior wall frac-
ture with posterior dislocation converted to a central dislocation
during attempts at reduction. Note pedunculated incarcerated
fragment. A Antero-posterior radiograph, a diagram, B obtura-
tor-oblique radiograph, b diagram, C iliac-oblique radiograph
a vertical split of the ischio-pubic fragment (see
Sect. 12.1.3 (a) above), characterising a T-
shaped fracture associated with the posterior
wall fracture. We have seen 10 cases of such an
association accompanied by a central disloca-
tion.
(b) Obturator-oblique view. This confirms the cen-
tral dislocation of the head, and it shows one or
more posterior fragments of which the dis-
Very Large Postero-superior Fragment Extending to the Iliac Crest
C
Fig. 12.10C
placement is always less than in the case of a
posterior dislocation of the hip. Also, it shows
clearly the obliquity of the plane of the trans-
verse component. It confirms or proves the ex-
istence of a vertical split of the ischio-pubic
notch above, and the ischio-pubic ramus below,
in one or two places, in the case of aT-shaped
fracture.
(c) Iliac-oblique view. In particular, this shows the
level of the fracture of the posterior border of
the innominate bone and sometimes the point
at which the transverse fracture line cuts the
anterior border of the acetabulum.
12.3 Comment
The association of a transverse fracture and a frac-
ture of the posterior wall comprises two sub-groups
which are determined according to the direction of
the dislocation.
213
When a posterior dislocation prevails, the trans-
verse component may be regarded as an extension
of the posterior lesion. The posterior fracture re-
mains the more important component because it
allows the head to escape posteriorly. With a cen-
tral dislocation, the principal lesion is the trans-
verse component which allows the head to dislocate
directly inwards and the posterior fracture is the ac-
cessory lesion, the result of the fracture energy
dissipating at the level of the posterior wall.
The two sub-groups are related by the fact that
a posterior dislocation after reduction can be
transformed into a central dislocation, increasing
the displacement of the ischio-pubic fragment.
Without knowledge of the preceding events, from
the secondary central dislocation films, it is not
possible to conceive that the displacement was ini-
tially posterior (Fig. 12.10). Fragments of posterior
wall can be carried into the joint space as a result
of the reduction, which further complicates the ap-
pearance (Fig. 12.10).
In a group of six patients with the typical frac-
ture patterns of this group there was no dislocation;
the head was perfectly centred under the intact part
of the roof and the original direction of the previ-
ous dislocation, if any, was impossible to deter-
mine. A greater or lesser displacement of the frac-
ture line was present. There were four infra-tectal
transverse components, one juxta-tectal and one T-
shaped.
12.4 Very Large Postero-superior Fragment
Extending to the Iliac Crest
These 18 cases of our series are among the most
difficult to recognise and classify.
For a long time we discussed the exact fracture
type of the case in Fig. 12.11, in which apparently
a juxta-tectal transverse fracture was associated
with a fracture line extending across the iliac wing
to reach the iliac crest; initially, for us, this was an
inverted T-shaped fracture (what is wrong with this
idea is that there is no segment of the articular sur-
face attached to the posterior part of the wing).
This case was operated upon through the ilio-
inguinal approach, so the fracture lines were not
explored on both aspects of the iliac wing, and
the only thing we learned was that the iliac wing
fracture line reached a typical transverse com-
ponent without transgressing the pelvic brim,
meeting it in the iliac fossa (Fig. 12.11 A). More
214 Associated Transverse and Posterior Wall Fractures

~(\ \
\
j
/,/
I
/ i
/ '--1-..... i
/ / ( I
(\ /' \
\ i; c
\ (
.I .I
/ )
/'
b Fig. 12.11A-C
Very Large Postero-superior Fragment Extending to the Iliac Crest 215

precise radiological study of later cases, intra-

operative findings, and, recently, 3-D CT images
have allowed us to identify this particular type of
associated transverse and posterior wall fracture,
which deserves, we think, to be left in this group.
The morphology of the fracture is as follows
(Fig. 12.11 A). The transverse fracture component
may be of any type. From the fracture line separat-
ing off the transverse fracture between the posterior
border of the bone and the posterior lip of the
acetabulum starts a fracture line which goes up-
wards. One can think of it like this: that when the
fracture line meets the posterior aspect of the pillar
of the iliac wing, instead of transgressing it and
going down to divide the acetabulum again, thus
delineating a postero-superior fragment, it goes up
behind the pillar or follows its trabeculae so as to
reach the iliac crest. On the inner aspect of the
bone the fracture line divides the internal iliac fossa
from the crest to the transverse fracture line
(Figs. 12.11-12.13).
Consequently, these two fracture lines divide the
innominate bone into three pieces:
the posterior part of the wing, which is con-
nected to the sacrum and does not bear any ar-
ticular surface segment;
the ischio-pubic fragment, below the transverse
fracture line, which takes with it the walls of the
D acetabular articular surface;
the anterior part of the iliac wing extending
downwards to the transverse fracture line, which
takes with it the whole roof, and in fact all the
acetabular surface above the transverse fracture
line.
Thus, the whole of the articular surface is detached
from the posterior part of the iliac wing (Fig. 12.13
shows a typical case and its 3-D appearance). There
is only one other group of acetabular fractures
/;-._. which achieve the same thing, and that is both-col-
/ I
i umn fractures. In the group of fractures under
I analysis here, however, the ischio-pubic fragment is
i in one piece and limited superiorly by a typical
/ transverse fracture, while a both-column fracture
never comprises a transverse fracture component
d (see Chap. 14). For these reasons, we do think that
this small and very peculiar group deserves to be
left among the associated transverse and posterior
Fig. 12.11 A-D. Associated transverse and superior fracture. It wall fractures.
takes the roof and extends to the iliac crest, giving a false im- It will be recalled that earlier a pure postero-
pression of an upside-down T; the stem does not split the roof
fragment. A Scheme, B antero-posterior radiograph, b diagram, superior fracture not involving the pelvic brim was
C obturator-oblique radiograph, c diagram, D iliac-oblique studied, with an iliac fracture line extending to the
radiograph, d diagram crest of the ilium (Fig. 5.11), and also a fracture of
 216 Associated Transverse and Posterior Wall Fractures

A B

Fig. 12.12A-E. Associated trans-tectal transverse and postero-
superior wall fracture extending to the iliac crest, operated on
120 days after injury. A-C The three standard views. A spur
sign is visible in B and corresponds to the intact part of the
bone. D, E Intra-operative views. D shows the transverse fracture
line dividing the posterior column, running backwards and in-
feriorly; from it starts the fracture line extending up to the iliac
crest. E A lamina spreader distracts the iliac fracture line, allow-
ing visualisation of the cartilage of the roof and the femoral
head. a, b Diagrams of the fracture lines
 Very Large Postero-superior Fragment Extending to the Iliac Crest 217

c E

Fig. 12.12D-E
 218 Associated Transverse and Posterior Wall Fractures

A B

Fig. 12.13 A - M. Trans-tectal transverse fracture with associated

postero-superior fracture extending to the iliac crest. A - C The
three standard views. There is an ischio-pubic fragment in one
piece that takes with it the inner part of the roof so limited above
by a trans-tectal fracture. In addition to that, there is a fracture
running through the iliac wing and not involving the articular
surface; it is clearly seen that it starts from the transverse line
behind the roof on the iliac-oblique view. Consequently, these
two fracture lines divide the articular surface into only two
pieces, and the whole of the articular surface is separated from
the intact part of the iliac wing. Apart from both-column frac-
tures, these fractures are the only acetabular fractures that
separate off the whole articular surface. D - K CT sections. The
transverse trans-tectal fracture is clearly seen in G - I, with its
antero-posterior direction. Besides that transverse fracture, a
coronal fracture line transgresses the iliac wing from its anterior
part (D), divides the iIio-pectinealline at the level of the greater
sciatic notch (F), passes behind the roof, then obliquely for-
wards and inwards at about 45 0, and rejoins the transverse frac-
ture. In addition, in this particular case, the anterior part of the
wing has taken with it a piece of the cortex of the retro-acetabu-
lar surface (I,J). L,M 3-D reconstruction of such a case, show-
ing particularly how the piece of cortex of the retro-acetabular
surface remains connected with the posterior iliac wing frag-
C ment
 Very Large Postero-superior Fragment Extending to the Iliac Crest 219

D E

F c

Fig. 12.13D-K
 220 Associated Transverse and Posterior Wall Fractures

L M
Fig. 12.13L-M
CT Study of Associated Transverse and Posterior Wall Fractures
the upper part of the anterior column, which in-
volved a slightly larger fragment bearing, together
with the roof, a segment of the pelvic brim, but
sparing the obturator foramen (Figs. 8.1 and 8.11).
The relationship between these cases serves once
again to demonstrate the continuity between the
different varieties of fractures of the acetabulum.
Indeed, as a matter of fact, four cases of this series
do not have exactly the same feature, as the fracture
line running through the iliac wing reaches the
anterior border of the bone below the anterior
superior iliac spine (Fig. 12.14-12.17). The
postero-superior fragment delineated by this frac-
ture line is generally in two or more pieces. The im-
portant point is that the whole articular surface is
detached from the intact part of the wing. CT scan-
ning helps to identify these cases.
In this group as well as in both-column frac-
tures, true and full secondary congruence may be
achieved.
12.5 CT Study of Associated Transverse
and Posterior Wall Fractures
CT sections demonstrate perfectly whether the dis-
location of the femoral head is central or posterior.
The displaced part, which appears first on the sec-
tions, may be the upper tip of the posterior wall
fragment or the tip of the ischio-pubic fragment
221
detaching from the pelvic brim. The transverse
fracture has the usual characteristics: it is directed
antero-posteriorly and progresses laterally on more
inferior sections. On its way it involves (trans-tectal
fractures; Fig. 12.16) or does not involve Uuxta-tec-
tal or infra-tectal fractures) the sub-chondral bone
of the roof (i.e. the CT scan roof; Fig. 12.15F).
In some instances the inferior or ischio-pubic
fragment delineated above by the transverse frac-
ture is divided by a coronal fracture line which
crosses the cotyloid fossa and, further down, the
ischio-pubic ramus, forming an associated T-
shaped with posterior wall fracture (Fig. 12.16). As
for the pure posterior wall fragments, the involved
posterior wall is separated by a fracture line run-
ning obliquely forwards and outwards; both its ver-
tical and its transverse extent may be appreciated by
CT. If the posterior wall fracture is postero-superi-
or, it may involve the CT scan roof. In some cases
there is accompanying marginal impaction.
In cases of transverse fractures with a large
postero-superior fragment extending to the iliac
crest (Figs. 12.12 and 12.13), the CT sections show
the fracture lines starting from the iliac crest in an
approximately coronal direction and extending
downwards and posteriorly on subsequent sections
to reach the transverse fracture line at the level of
the CT scan roof, which is left undisturbed or is in-
volved only in its extreme posterior part. The
ischio-pubic fragment is in one piece without any
fracture through the cotyloid fossa.
 222 AssocIated Transverse and Posterior Wall Fractures

Fig. 12.14A-I. Trans-tectal transverse fracture associated with a

total postero-superior fragment extending to the anterior border
of the bone, just above the anterior inferior iliac spine. Together
these fractures detach the whole acetabular articular surface
from the iliac wing. A-C The three standard views. A The
transverse fracture cuts the anterior lip high up and the posterior
lip very low down; the ischio-pubic fragment takes with it the in-
ner part of the roof. 8 The whole roof and the posterior wall
above the transverse are detached; there is no longer any ar-
ticular surface attached to the iliac wing. In A and C the fracture
can be seen to reach the anterior border of the innominate bone
above the anterior inferior iliac spine, and the thinning of the
upper part of the posterior column is explained by the big
posterior wall fragment detached from it. D-I CT sections.
E-H The transverse fracture is angled obliquely forwards and
outwards at 70 0 and takes with it an inner piece of the roof (F).
The intact part of the wing does not bring with it any part of
the articular surface. All the articular surface above the
transverse fracture has been freed in several pieces. The fragment
which bears the anterior inferior iliac spine (1 in D - G) includes
a small part of the roof (F, G)

8 C
CT Study of Associated Transverse and Posterior WaIl Fractures 223

o E f

G H

Fig. 12.14D-I
224 Associated Transverse and Posterior Wall Fractures

Fig. 12.1SA- H. Associated juxta-tectal trans·

verse and posterior wall fracture; the initial
posterior dislocation turned into a central sub-
luxation after reduction. A-C The three stan-
dard views. The transverse fracture cuts the mid-
dle part of the greater sciatic notch, the pelvic
brim very posteriorly, and the anterior lip above
its notch. The posterior wall fragment is not large.
D - H CT sections. The transverse fracture line
divides the bone in an antero-posterior direction
(E-G) and barely involves the roof. An addi-
tional posterior wall fragment is separated by a
fracture line running obliquely at 45 0 forwards
and outwards. G shows a small impacted frag-
ment (recognisable also in B) of the posterior
part of the roof. H shows the end of the
transverse fracture through the anterior wall and
the fracture of the posterior wall. The greater
part of the transverse fracture can no longer be
seen in this and the more inferior sections

C
CT Study of Associated Transverse and Posterior Wall Fractures 225

D E F G

Fig. 12.1SD-H
 226 Associated Transverse and Posterior Wall Fractures

A B

Fig. 12.16A-G. T-shaped fracture with a trans-tectal compo-

nent and an additional posterior wall fracture in several pieces
associated with posterior dislocation. A - C The three standard
views. The stem of the T is visible in A and B, which also show
that the inner part of the roof belongs to the anterior column
segment. The intact sector of the roof is visible on all three
views. D-G CT sections. The transverse line, antero-posteriorly
oriented in D, cuts the middle part of the roof in E. The inner
fragment is divided by a vertical stem and the anterior column
segment takes with it the larger piece of the detached part of the
roof. There is a piece of articular surface in front of the neck;
c it comes from the posterior wall
 CT Study of Associated Transverse and Posterior Wall Fractures 227

o E

F G

Fig. 12.160-G
 228 Associated Transverse and Posterior Wall Fractures

A 8

Fig. 12.17 A - M. Juxta-tectal transverse fracture associated with

a single postero-superior fragment extending to the anterior in-
terspinous notch. The innominate bone is divided by these two
fracture lines into three pieces and the whole articular surface is
detached in two pieces. This is an exceptional case. A-C The
standard views. A The transverse fracture is typically juxta-tec-
tal; the whole roof above it has shifted. The obturator frame is
in one piece, as confirmed on the obturator-oblique view (8).
B There is a spur sign, which can exist in this group and has the
same meaning as in true both-column fractures. C The trans-
verse fracture cuts the angle of the greater sciatic notch. The
fracture line of the internal iliac fossa appears incompletely
displaced close to the anterior border of the bone. D - I CT sec-
tions. The transverse fracture has a typical appearance in E - G,
passes just medial to the roof and does not bring any articular
surface. F The postero-superior fragment apparently brings the
whole anatomical roof and the antero-inferior iliac spine. Just
behind it the end of the intact part of the wing is visible. In H
and I the cotyloid fossa can be seen to be intact. J - M 3-D re-
constructions. J shows the three pieces very clearly. K delineates
the fracture of the internal iliac fossa; it is the inner aspect of
the postero-superior fragment. In L we see the fracture line
delineating the postero-superior fragment starting from the
transverse fracture line behind the acetabulum. In L the
transverse fracture is a little irregular with a small free piece of
c the pelvic brim
CT Study of Associated Transverse and Posterior Wall Fractures 229

D E F

G H

Fig. 12.17D-I
 230 Associated Transverse and Posterior Wall Fractures

J K

L M

Fig. 12.17 J - M
13 Associated Anterior and Posterior Hemitransverse Fractures

We identified this association in 1960, and in our Table 13.1

series of 940 cases it has occurred on 62 occasions
Associated anterior and posterior 62 6.58070
(6.580,10). It comprises an anterior fracture of the
hemitransverse fractures
acetabulum, which may be a fracture of the anteri-
or wall or a fracture of the anterior column, Anterior wall and posterior 14 1.47%
together with a fracture of the posterior column, hemitransverse
which is exactly the same as the posterior half of a Low anterior column and posterior 6 0.63%
pure transverse fracture. Accordingly, we have hemitransverse
called this component a hemitransverse posterior Intermediate anterior column and 5 0.53%
fracture (Table 13.1). posterior hemitransverse
High anterior column and posterior 19 2.02070
hemitransverse
13.1 Morphology Anterior wall and incomplete posterior 4 0.42%
hemitransverse

13.1.1 Anterior Fractures with Associated Anterior column and incomplete 11 1.17%
posterior hemitransverse
Complete Posterior Hemitransverse Fracture
Anterior column and epiphyseal 2 0.20%
separation of the posterior column
13.1.1.1 Anterior Fracture
Epiphyseal separation of anterior 0.10%
column and posterior hemitransverse
If this is a fracture of the anterior wall of the
acetabulum (Fig. 13.1 A) (14 cases) the detached
fragment has the typical characteristics of the pure
form. It may be displaced in one piece or remain at-
which there was no rupture of the ischio-pubic
tached to the superior pubic ramus on which it
ramus);
rotates. Frequently, however, it is split either longi-
tudinally or transversely. The femoral head which intermediate anterior column fracture (5 cases)
(Fig. 13.1 B);
has followed the fragment is dislocated anteriorly.
The displacement is always marked. An associated high complete anterior column fracture (19
cases); however, in 6 of these the anterior col-
fracture of the ischio-pubic ramus in its middle part
umn fracture line did not reach the iliac crest
may be seen, in which case the lesion comprises a
(Fig. 13.15).
triple fracture: the anterior wall, the anterior col-
umn and the posterior hemitransverse components.
As stated earlier, the displacement of the anterior
We have already discussed this in Chap. 8 and have
column is always severe. In the high anterior col-
stated there why we consider for practical reasons
umn fractures, there were three examples in which
that these should be included among the pure
the fracture line extended from the iliac crest to the
anterior wall fractures.
angle of the pubis; in two cases the detached
More frequently there was a fracture of the ante-
fragments were split at the level of the anterior wall
rior column (30 cases) and this may be classified as
of the acetabulum or through the superior pubic
for the pure lesion (Sect. 8.1):
ramus. In one case, at least a very large fragment,
low anterior column fracture, the fracture line in one piece, rotated around the point of rupture
cutting the psoas groove (six cases, in one of which was situated very far back along the arcuate
232 Associated Anterior and Posterior Hemitransverse Fractures

notch is variable: in some instances it divided the

ischial spine longitudinally while in the remainder
it cut the greater sciatic notch at any level, even
reaching its superior border.
The displacement of the posterior component
was less marked than in the case of the anterior col-
umn lesion. There were 23 significant displace-
ments, 16 very slight displacements, and 5 cases
showed none at all.
From anatomical and radiological points of
view, the fracture lines of the two components of
the association appear as independent entities. The
fracture, separating in one piece the anterior col-
umn, is concave antero-superiorly and is directed
obliquely downwards and forwards; it diverges
gently from the brim of the true pelvis on to the
quadrilateral surface. The fracture line which cuts
the posterior column meets the anterior column
component at a right angle. The transverse hemi-
fracture line is almost straight when seen on the
quadrilateral surface and cuts the greater sciatic
notch at a variable level. The retroacetabular sur-
face is divided more or less obliquely by this frac-
ture line as is the case in transverse fractures
(Fig. 13.2B).

13.1.2 Anterior Fractures with Associated

Fig.l3.IA,B. Schemes of associated fractures. A Associated Incomplete Posterior Hemitransverse Fractures
anterior wall and posterior hemitransverse, B associated middle
anterior column and posterior hemitransverse
In 15 cases (4 anterior wall, 11 anterior column
fractures), the posterior fracture line was in-
complete and stopped short of the dense trabeculae
found in the anterior border of the greater sciatic
line (Fig. 13.9). In another case, the iliac wing frag-
notch. This group forms a transition between a
ment included the angle of the sacro-iliac joint.
pure anterior column fracture and the association
We must emphasise again that these anterior le-
with which this chapter is concerned. The specimen
sions, together with the anterior displacement of
shown in Fig. 13.2 is a typical example of this well
the femoral head which accompany them, are iden-
defined group. The inferior fragment of the poste-
tical to those which we have seen in pure anterior
rior column delineated superiorly by the hemitrans-
lesions and they are apparently quite independent
verse fracture rotates around the unruptured (but
of the posterior column lesion.
probably twisted) anterior border of the greater
sciatic notch.
13.1.1.2 Posterior Column Fracture

The posterior column is crossed by a fracture line

13.1.3 Important Remarks
identical with the pure transverse variety and may
occur at any of the levels described in Chap. 9. We
Whether the hemitransverse fracture is complete or
found its point of rupture at the posterior border
not, three fundamental points should be observed:
of the acetabulum to be in its inferior quarter or
below in 30 cases, in its middle part in 8 cases, and 1. While the posterior fracture line detaches the in-
in its upper part in another 6 cases. The point of ferior part of the posterior wall, and the fracture
rupture of the anterior border of the greater sciatic of the anterior column detaches the anterior wall
Morphology
Fig. 13.2A-C. Associated anterior wall and posterior incom-
plete hemitransverse fracture. Clinical specimen relating to
Fig. 13.1! . In A the roof remains in place under the wing. above
the posterior end of the transverse fracture component. In B the
fracture line stops before the anterior border of the greater
sciatic notch. C shows method of fixation. See X-rays on
Fig. 13.!!
and part of the roof. a sector of the articular sur-
face of the roof and the posterior wall always
remains normally attached to the wing of the
ilium. This differentiates these fractures from
the both-column type to be described later
(Fig. 13.7B).
2. The outline of the fracture as seen from the in-
ner aspect of the pelvis is completely different
from those of the T-shaped fracture which also
leave a sector of the articular surface in place on
the iliac component (Fig. 13.7 C).
3. The track of the posterior hemitransverse frac-
ture has some particularities. As was described
earlier in relation to pure anterior column frac-
tures, as soon as there is significant displace-
ment, the fracture of the anterior column is not
233
C
sufficient to allow antero-medial dislocation of
the femoral head. In this particular pattern of
acetabular fracture, the posterior hemitransverse
fracture component frees the inferior part of the
posterior column, which rotates around a more
or less vertical axis, thus allowing displacement
of the head. When it happens that the hemi-
transverse component is incomplete, the rotation
around the unbroken anterior border of the
greater sciatic notch is sufficient to allow
displacement of the femoral head. It may be
useful to recall that in nearly all pure anterior
column fractures with displacement a plate of
bone was detached from the quadrilateral sur-
face, and it was shown in Chap. 8 that this plate
could in some cases involve the inner part of the
 234
Fig. 13.3A,B. Diagrams of CT sections. A Below the roof, the
cotyloid fossa seems to be separated by itself, but a few sections
below (B) it can be seen to belong to a fragment that takes with
it a segment of the posterior wall and is separated from the in-
tact lateral part by a fracture line in an antero-posterior direc-
tion
posterior acetabular articular surface. The
greater damage to the posterior column ob-
served in this category of fractures is associated
with the fact that the site of fracture impact is a
little more posterior. The hemitransverse frac-
ture divides the posterior column downwards
and outwards. 2-D CT scanning and 3-D recon-
structions allow a thorough study of it: on the
upper sections only the cotyloid fossa appears
detached, but on successive sections the
posterior limit of what may appear initially as a
plate of bone similar to that connected to a pure
anterior column fracture reaches the inner limit
of the posterior wall (Fig. 13.3), then trans-
gresses it, according to direction either antero-
posteriorly (as in a complete transverse fracture)
or, less typically, obliquely frontwards and out-
A B
Fig. 13.4 A, B. CT sections of two different cases, showing that the fracture line through the posterior wall may either be antero-
posterior (B) or run obliquely forwards and outwards (A)
Associated Anterior and Posterior Hemitransverse Fractures
Fig. 13.5A-E. Associated anterior column and incomplete he-
mitransverse fracture. A-C The three standard views, D,E CT
sections showing that the posterior hemitransverse fracture,
which appears at the level of the roof section, progressively takes
with it most of the articular surface and most of the spongiosa
of the posterior wall, leaving the retro-cotyloid surface ap-
parently undisturbed. The posterior lip, however, is broken in
one place on the antero-posterior view
wards, finally reaching the posterior lip of the
acetabulum (Fig. 13.4).
In several instances of incomplete hemitrans-
verse fracture, we have seen the fracture line
through the posterior wall take with it nearly the
whole of the articular surface and the spongiosa of
the posterior column, which appears elevated from
the retro-acetabular bone cortex (Fig. 13.5). In
other instances, the inferior segment of the posteri-
or column is divided into two pieces by a more or
less vertical coronal fracture line, the anterior part
representing only the cotyloid fossa and the cor-
responding area of the quadrilateral surface, the
posterior part a segment of the posterior column
(Fig. 13.6).
The track of the posterior hemitransverse frac-
ture, depending on its obliquity, may be completely
different along the different aspects of the posteri-
or column, as seen in Fig. 13.2, which shows the
fracture line more or less transversely disposed
through the retro-cotyloid surface and dividing the
posterior part of the articular surface and the
quadrilateral surface obliquely upwards and in-
wards.
13.2 Radiology
13.2.1 Anterior Fracture
It is not necessary to describe in detail the anterior
lesion which retains the characteristics of the pure
Radiology 235

A B

C E

Fig. 13.5
 236 Associated Anterior and Posterior Hemitransverse Fractures

A B

E
Fig.13.6A-E. Associated high anterior column and posterior
hemitransverse fracture. A-C The three standard views.
D, E CT sections. The inferior segment of the posterior column
is divided into two pieces, its anterior part representing only the
cotyloid fossa and the corresponding area of the quadrilateral
surface, the posterior part a segment of the posterior column.
The fracture line separating the anterior column has its typical
C coronal orientation at the level of the roof and that dividing the
posterior column is roughly in an antero-posterior direction
through the posterior wall
Radiology 237

At A2 A3

Fig. 13.7 A-C. Schemes of various as-

sociated anterior and hemitransverse
fractures. A 1 Anterior column and
posterior hemitransverse, A 2 anterior
wall and posterior hemitransverse. A 3
endopelvic aspect of Al and A2,B
both-column fracture (to show dif-
ference), CIT-shaped fracture (to show
difference), C2 endopelvic aspect of Cl B Cl C2

anterior column fracture or of an anterior wall le-
sion (Figs. 13.8-13.11).
On the antero-posterior and obturator-oblique
views, the anatomical type is easy to define. The
femoral head follows the anterior lesion and may
be truly anteriorly dislocated, having lost contact
with the rest of the remaining roof, or it may only
be subluxed. Occasionally it is displaced inwards
and forwards and remains related to a large seg-
ment of the articular surface carried by the anterior
column; an apparent congruence of the two parts
may be a diagnostic pitfall, for the head may not
seem to be dislocated at first sight, and it is easy to
miss, observing that there is a segment of the roof
which has remained in its proper place on the iliac
wing on the antero-posterior view (Figs. 13.9 and
13.13); it may appear on the obturator-oblique view
albeit sometimes very small. Again, the head can
return under the remaining part of the roof and ap-
pears quite separate from the anterior column frac-
ture.
What may on the iliac-oblique view appear as a
bone plate from the quadrilateral surface
(Fig. 13.9), similar to what is often seen in pure an-
terior column fractures, is in fact most often the
quadrilateral surface connected or belonging to the
displaced inferior segment of the posterior column.
In a few cases a plate of bone was also elevated
from the quadrilateral surface (see Fig. 13.8).
13.2.2 Posterior Column Fracture
The main problem is to recognise and define the
fracture in the posterior column.
13.2.2.1 Antero-posterior View
The posterior wall of the acetabulum is crossed by
a fracture line which can break the posterior border
of the acetabulum at any level. It is almost straight,
being slightly oblique upwards and inwards thereby
238 Associated Anterior and Posterior Hemitransverse Fractures

A B

'-',
\. - ,.,,. \
IS .
-/ .........

'"i
/( \
I I
/ i
i i
a I
I b

Fig.13.8A,B

reaching the ilio-ischial line, crossing it in most
cases. In the fractures without displacement (3
cases), the line follows this pattern and merely
marks the posterior border and the ilio-ischialline.
Displacement, when it occurs, is inwards and asso-
ciated with an internal rotation of the lower seg-
ment of the posterior column around a vertical
axis; this is recognisable by a step in the posterior
border of the acetabulum and separation at the
ilio-ischial line (Fig. 13.8). In the rare cases when
the fracture line is very low, it cuts the ilio-ischial
line and the teardrop, the fracture line of the
anterior column having avoided the zone of the
teardrop and having passed into the anterior part
of the cotyloid fossa.
In most instances, the ilio-ischial line was frac-
tured; it was sometimes only distorted and occa-
sionally apparently intact (Fig. 13.11). The ilio-
ischial line lost its relationship with the teardrop, a
consequence of the anterior fracture, in nearly all
cases.
13.2.2.2 Obturator-oblique View
The posterior fracture line transgresses the
posterior wall following an obliquity which often
seems more marked than that observed on the
antero-posterior view. The point of rupture on the
posterior border of the acetabulum is well visual-
Radiology
c articular surface on the posterior iliac segment is
c doubt as to the continuity of the undisplaced roof
Fig. 13.8A-C. Associated anterior wall and posterior hemi-
transverse fracture. A Antero-posterior radiograph, a diagram,
B obturator-oblique radiograph, b diagram, C iliac-oblique
radiograph, c diagram. Note in A and C the plate of bone
elevated from the quadrilateral surface
ised. When an extended fracture line of the anterior
column is present, taking with it the greater part of
the roof so that on the external surface of the iliac
wing it reaches the posterior part of the roof,
the outer aspect of the supra-acetabular region
presents with a break marked by a slight change in
contour of the wing but with little or no separation.
This fracture line appearing in section above the
acetabulum is the one that detaches the anterior
column. It is an equivalent to the spur sign (see
Fig. 14.22).
239
13.2.2.3 Iliac-oblique View
This shows best the direction of the posterior part
of the fracture line at the level of the quadrilateral
surface. Its most common straight character is em-
phasised. The fracture line in most instances
reaches the anterior border of the greater sciatic
notch at a variable level; in extreme cases when it
is low, it divides the ischial spine but when high, it
terminates at the angle of the greater sciatic notch.
A possible fracture of the iliac crest and the frac-
ture outline across the iliac wing, when this type of
anterior column component is present, are also
seen on this view.
13.2.3 A Special Feature of this Group
A feature that serves to individualise this fracture
is the persistence at its normal place on the
posterior iliac fragment of a segment of articular
surface isolated below by the transverse fracture
line, and above and anteriorly by the upper part of
the anterior column fracture line (Fig. 13.7 A1,A2).
The recognition of the remaining segment of the
essential for the complete diagnosis of the fracture
complex. Unfortunately this can be difficult be-
cause it is not well situated for good radiological
definition in the standard views. The diagnosis is
less difficult when the anterior lesion is a fracture
of the anterior wall, for even when fairly extensive,
such a fracture always leaves in place the greater
part of the roof which can be recognised easily on
the antero-posterior and oblique views. There is no
segment with the iliac wing.
Greater difficulty occurs in cases of fracture of
the anterior column, for the column fragment may
include a large part of the roof. Further, the poste-
rior hemitransverse fracture may be high, so that
the articular segment remaining in place may be
small.
The short articular segment may be invisible on
the antero-posterior view when the anterior column
fracture is high and extended (Fig. 13.9); further-
more, it is not seen in the iliac-oblique view for here
it is perpendicular to the direction of the X-ray
beam. It may appear only on the obturator-oblique
view as a small lateral section of roof above and in-
ternal to the remaining intact part of the posterior
border of the acetabulum (Fig. 13.9B). In short,
this anatomical variety will be identified in the
240 Associated Anterior and Posterior Hemitransverse Fractures

B
A

Elevated cortex of
quadrilateral
surface
/
/ . _ ........ /
/ "r1 DnA. / - :..."\.. .....
(
\
/ .(
'" .\ /
{
/

\ \
\ \
i \
a j
b i

Fig. 13.9A-C. Associated anterior column and posterior hemi-
transverse fracture. A Antero-posterior radiograph, a diagram,
B obturator-oblique radiograph, b diagram, C iliac-oblique
radiograph, c diagram. The anterior column is entirely detached
from the anterior superior iliac spine to the pubic spine and has
rotated around the point of rupture of the pelvic brim. It is asso-
ciated with the hemitransverse component seen in A as a rupture
of the posterior border of the acetabulum and of the ilio-ischial
line. In B it cuts the middle third of the posterior wall of the
acetabulum and the upper third of the greater sciatic notch.
Note the plate of bone lifted from the quadrilateral surface of
the ischium
Atypical Examples 241

antero-posterior view or the obturator-oblique

view, or perhaps both, only on recognising the sec-
tion of roof in situ, even if it is very small and over-
shadowed by the surrounding bone of the iliac
wing, as long as its normal connection to the iliac
wing by intact trabeculae can also be confirmed.
With CT scanning the diagnosis is easier, as it
demonstrates the intact part of the sub-chondral
bone of the roof and the posterior wall in their nor-
mal places. Getting a 3-D reconstruction makes the
diagnosis easy.

13.3 Atypical Examples

(a) Incomplete iliac wing component. A high ante-

rior column fracture can present with an in-
complete break in the iliac wing (6 cases); the
fracture does not reach the crest, stopping just
below it, but the jagged outline in the wing ex-
tending towards the crest is unmistakable on the
antero-posterior and iliac-oblique views
(Fig. 13.14).
(b) Incomplete hemitransverse component. The
hemitransverse component may be incomplete
C (15 cases, 11 associated with an anterior col-
umn and 4 with an anterior wall fracture). The
fracture line is obvious at the level of the
posterior wall of the acetabulum on the antero-
posterior and obturator-oblique views. The ilio-
ischial line can appear broken, bent or unin-
volved, but of most significance is the interrup-
tion of the posterior border of the acetabulum.
The iliac-oblique view shows that it is a transi-
tional fracture because the posterior fracture
avoids the anterior border of the greater sciatic
notch, the fracture energy having been ex-
hausted before having ruptured the dense
trabeculae of the anterior border of the greater
sciatic notch. In these cases, the posterior frac-
ture line can be displaced so that the inferior
fragment of the column rotates around a point
at the anterior border of the greater sciatic
notch (Figs. 13.2 and 13.11).
(c) Associated anterior wall, anterior column and
posterior hemitransverse fractures. In a few
Fig. 13.9C cases we have seen a rupture of the ischio-pubic
ramus at one or two places together with dis-
placement of the anterior wall clearly visible on
the antero-posterior and obturator-oblique
views; the anterior wall segment was detached
and driven in one piece. This amounts to the
 242 Associated Anterior and Posterior Hemitransverse Fractures

Fig. 13.11A-C. Associated anterior wall and posterior hemi-

transverse fracture. A Antero-posterior radiograph, a diagram,
B obturator-oblique radiograph, b diagram, C iliac-oblique
radiograph, c diagram. The ischio-pubic ramus is fractured. The
posterior border of the bone is seen in C to be intact: the
hemitransverse fracture line dies out in the region of dense bone
around the angle of the greater sciatic notch. See also Fig. 13.2
•
Fig. 13.10A-C. Associated low comminuted anterior wall and
posterior hemitransverse fracture. A Antero-posterior
radiograph, showing five fragments between the antero-inferior
iliac spine and the middle of the ischio-pubic ramus, a diagram,
B obturator-oblique radiograph, C iliac-oblique radiograph

" -) ..
.- '.....

,
"

/. ~
I \
. '-\
.I .)
I i
I i
~
i
a
i
"\ \

B C
Atypical Examples

Fig. 13.11A-C c
244 Associated Anterior and Posterior Hemitransverse Fractures

.:;;..---- ........
''\
\
)
i
i
\ B
\
a

." ......
'- -\( '\
I) \
7 I
/
.I
I
I
b I

Fig. 13.12A-C. Associated anterior wall and anterior column
and hemitransverse fracture. A Antero-posterior radiograph, a
diagram, B obturator-oblique radiograph, b diagram, C iliac-
oblique radiograph, c diagram. Interpretation is fairly difficult.
At first sight, A suggests a T-shape; in fact an area of detach-
ment of the anterior column can be seen. This is clearer in B
where a typical anterior wall fracture is associated with a frac-
ture of the ischio-pubic ramus. In C the posterior
hemitransverse component is seen to be incomplete
Radiological Differential Diagnosis 245

separation of the posterior column along the

posterior branch of the Y-shaped acetabular
cartilage (Fig. 13.13).
In the third case, a 17-year-old boy sustained
an epiphyseal separation of the anterior column
associated with a posterior hemitransverse frac-
ture, as the posterior part of the Y cartilage had
already healed. Treated conservatively, he devel-
oped osteoarthritis 15 years later.
(e) A transitional and atypical case (Fig. 13.15). A
case operated on in 1992 and outside the series
reported on in this book showed a low anterior
column fracture (dividing the psoas groove on
the obturator-oblique view), associated with a
fracture line dividing the posterior wall very
obliquely upwards and inwards on the antero-
posterior view and nearly vertical on the ob-
turator-oblique view; the ischio-pubic ramus
was unfractured. This fracture line looks like
the one described in Sects. 8.1.2 and 8.2.5, to
which it can usefully be compared. CT Scann-
ing in this case demonstrated that in fact a
quasi-hemitransverse fracture divides the
posterior column at an unusual forwards and
outwards angle of obliquity. The line separating
the anterior column follows the axis of the
cotyloid fossa in a coronal direction. The 3-D
reconstructions show an irregular horizontal
fracture line dividing the ischial spine posterior-
c ly; through the quadrilateral surface this frac-
ture line runs upwards and forwards to reach the
Incomplete hemitransversf' fracture line separating the anterior column
fracture line
below the ilio-pectineal line.

Break in posterior border

of acetabulum 13.4 Radiological Differential Diagnosis

In this group of associated fractures, the differen-

c tiation from T-shaped and both-column fractures is
of particular importance.
Fig. 13.12C
(a) T-shaped fracture. In such a fracture
(Fig. 13.7 Cl, C2), the anterior column is broken
association of a fracture of the anterior wall at a variable level but the plane of the upper
with a fracture of the anterior column, plus a fracture is identical to that of a pure transverse
hemitransverse posterior fracture. As another fracture accompanied by its regularity and is in
transitional form, it completes the overall unity continuity with the fracture dividing the poste-
of these fractures of the acetabulum (Fig. 13.12). rior column. Even T-shaped fractures with the
(d) Associated epiphyseal separation of the posteri- stem directed obliquely forwards have a dif-
or column. There were three cases, two of them ferent design in that they section the anterior
children with a high anterior column fracture column transversely; they do not result in sepa-
detaching the anterior part of the iliac wing ration of the articular surface of the anterior
with the pubis, associated with an epiphyseal wall with its trapezoidal-shaped fragment of the
246 Associated Anterior and Posterior Hemitransverse Fractures

column (as in the anterior wall variety) or with

a segment much more extensive than this (as in
the fractures of the anterior column).
In associated anterior column and posterior
hemitransverse fractures, which we have just
studied, it is only the posterior component
which is identical to the pure transverse type
(see Fig. 13.7 and compare Figs. 13.8A and
10.8A).
(b) Both-columnjracture. Such a fracture in which
all the articular surface is detached embodies a
characteristic spur formation (seen on the ob-
turator-oblique view) resulting from marked
displacement. This is very different from the
slightly displaced fracture line above the
acetabulum which is seen in the associated frac-
ture complex which we have just studied. No
sector of the roof remains in place in a both-
column fracture (Fig. 13.7Al,A2,B).

13.5 CT Study of Associated Anterior

and Posterior Hemitransverse Fractures

The anterior fracture appears with its typical char-

acteristics as follows:
it is separated by a coronal fracture line (column
fracture) or a fracture line that runs obliquely
forwards and outwards (anterior wall fracture);
the fragment driven in by the femoral head
displaces in an internal and anterior direction;
the fragment rotates clockwise around a vertical
or oblique axis so that its articular surface faces
posteriorly;
the anterior fracture segment is in one or several Fig. 13.13A
pieces, and is frequently comminuted at the
level of the obturator canal or close to it;
the posterior hemitransverse component divides
the posterior column obliquely downwards and until it crosses the posterior lip of the acetabu-
inwards, and on the CT sections most frequently lum. This inferior part of the posterior column
has the typical antero-posterior direction of the rotates counter-clockwise around its posterior
transverse fracture; sometimes it divides the hinge under the pressure of the head (see
posterior wall obliquely outwards and forwards. Sect. 13.1.2). The posterior column segment
Just below the roof this inferior segment of the may be separated into two pieces (Fig. 13.6).
posterior column seems to involve only the Exceptionally, the plate, involving as usual the
cotyloid fossa; on further sections it increases in posterior column, takes with it the inner margin
size, reaching the posterior border of the of the anterior wall (see Fig. 13.2). Three-dimen-
posterior wall and further on transgressing it sional reconstructions greatly facilitate the
obliquely downwards and outwards, involving recognition of this fracture pattern (Figs.
more and more of the posterior articular surface 13.13-13.15).
CT Study of Associated Anterior and Posterior Hemitransverse Fractures 247

B C

Epiphyseal
separation of
posterior column

Fig. 13.13A-C. Associated extended anterior column fracture anterior fracture extends from the anterior superior iliac spine
and epiphyseal separation of the posterior column. A Antero- to the pubic spine. Note the buckling of the anterior part of the
posterior radiograph, a diagram, B obturator-oblique radio- iliac wing which was difficult to reduce at operation
graph, b diagram, C iliac-oblique radiograph, c diagram. The
248 Associated Anterior and Posterior Hemitransverse Fractures

Fig. 13.14A-I. Incomplete high anterior column fracture sub-

divided into several pieces and isolating a trapezoidal fragment
connected to the anterior wall, with an associated hemitrans-
verse fracture posteriorly dividing the ischial spine and the mid-
dle part of the posterior lip. A - C The three standard views. In
C (the iliac-oblique view) the inferior segment of the posterior
column has rotated so much that it shows an ilio-ischialline and
its associated teardrop, indicating that the cotyloid fossa is
divided in its extreme anterior part. The ischio-pubic ramus is
broken twice. The fracture through the wing does not reach the
iliac crest. D - 1 CT sections. In D, E the fracture line separating
the anterior column through the roof and the wing is roughly
coronal. In F, G a plate of bone from the cotyloid fossa and the
quadrilateral surface seems isolated; it is in fact (H, I) in con-
tinuity with the inferior part of the posterior column. The
hemitransverse fracture transgresses the posterior wall in an
antero-posterior direction and obliquely downwards and out-
wards, finally (I) leaving the lip of the posterior wall in place
 CT Study of Associated Anterior and Posterior Hemitransverse Fractures 249

D E

F G

H
Fig. 13.14D-I
250 Associated Anterior and Posterior Hemitransverse Fractures

Fig. 13.1SA-J. Associated low anterior column and posterior

hemitransverse fracture. A - C The three standard views. The
fracture line transgressing the posterior wall is strongly oblique
on the antero-posterior view (A). nearly vertical on the ob-
turator-oblique view (8); it starts very low along the posterior
acetabular lip. D-G CT sections. H-J 3-D reconstructions
CT Study of Associated Anterior and Posterior Hemitransverse Fractures 251

D E F G

H J
Fig. 13.15D-J
14 Associated Both-Column Fractures

In the author's thesis (LETOURNEL 1961), the asso- Table 14.1

ciated both-column fracture was described, but it
Both-column fractures 213 22 .65010
was not until 1963, following the description of
eleven operated cases, that a further account was Iliac fracture line to the crest 176 18.72010
submitted. RIGAULT (1962) collected the first cases Iliac fracture line to the antero-superior 12 1.27010
of D'AuBIGNE under the name "central displace- iliac spine
ments of the acetabulum". This description, which Iliac fracture line to the interspinous notch 19 2.02010
Iliac fracture line to the psoas groove 6 0.63010
alludes to the method of production of the lesion,
has unfortunately the inconvenience of using yet
again a term we find imprecise: it does not give an
adequate account of the injury to the bone frame-
work which limits the acetabulum and which is
split in a special and typical manner.
The present study concerns 213 cases, of which
196 have been treated operatively. The conclusions
enable us to advance with certainty the anatomical
and radiological features which follow.
We reserve the term associated both-column
fracturefor those which isolate mutually from each
other the two columns together with the related
segments of articular surface. The only part to re-
main attached to the sacrum is a piece of the iliac
wing, varying amounts of which may remain. From
A
this latter fragment, all segments of the articular
surface have been separated (Fig. 14.1).

14.1 Morphology

These fractures appear very complicated and they

are usually described as "comminuted", this term
masking an insufficient study of the radiographs.
Their complexity is in fact very variable but never-
theless with experience and operation through an
appropriate approach, we have achieved 61 % ex-
cellent reductions.
Exceptionally, each of the columns separates as
one entity, the roof remaining with the anterior col- B
umn. In the large majority of cases, however, the Fig. 14.1A,B. Both-column fractures. Schemes showing iliac
fractures become much more complex because the component A extending to iliac crest, B extending to the anterior
principal fragments are split by secondary fractures border of the ilium
254
which cut the anterior column in two or three
fragments or which isolate a posterior segment or
a postero-superior segment of the acetabulum as a
separate fragment. One fragment may not bear any
articular surface, coming from the iliac fossa or
from the brim of the pelvis. Sometimes, further
local fragmentation occurs along anyone of the
multiple fracture lines crossing the pelvic bone, fur-
ther complicating the basic structure of the injury.
It is convenient to study separately the fragment
of each column.
14.1.1 Posterior Column Components
The posterior column is detached above by a frac-
ture which begins at a variable level on the posteri-
or border of the pelvic bone.
(a) Fracture line at angle of greater sciatic notch. In
about one case in two (Fig. 14.11) the fracture
line begins at about the angle of the greater
sciatic notch and descends on the retroacetabu-
lar surface obliquely downwards and forwards
to reach the acetabular lip. The line then cuts
the articular surface and reaches the posterior
limit of the upper margin of the cotyloid fossa.
C C the posterior fragment, D a special ex-
Associated Both-Column Fractures
From here it skirts the upper margin of the
cotyloid fossa adjacent to the articular surface
of the roof for a short distance and then
descends in the anterior moiety, following the
edge of the anterior articular surface. Lower
down it reaches the ischio-pubic notch in its
anterior part. On the opposite side of the ob-
turator foramen the fracture line cuts the ischio-
pubic ramus at a variable position. In 60 cases
there was a fracture through the body of the
pubis, which may be straight in outline but fre-
quently presenting a special appearance, con-
cave from above, and characteristic because we
have seen it only in both-column fractures (44
cases out of 213) (Fig. 14.11). In 92 instances,
the ischio-pubic ramus was fractured at one
point, in 22 instances it was fractured at two
sites, and in 33, fractures were present at both
the ischio-pubic ramus and the body of the
pubis. In only five cases was the inferior margin
of the obturator foramen spared.
On the inner surface of the pelvis (Fig. 14.1)
the fracture line, from inside the angle of the
greater sciatic notch, is directed forwards ob-
liquely and downwards or horizontally, follow-
ing a straight or jagged course. Before reaching
the brim of the true pelvis it changes direction,
Fig. 14.2 A-D. Fracture lines detaching the
posterior column. Diagrams of A distribu-
tion on retroacetabular surface (66 cases),
B possible sites of secondary fracture lines,
ample. See Fig. 14.27 for radiographs
Morphology 255

becoming more vertical before it reaches and

fractures the ischio-pubic notch. Very rarely it
reaches the upper margin of the obturator fora-
men more posteriorly than this. In a few cases
a part of the ischio-pubic notch and the roof of
the obturator canal are detached and appear as
a separate fragment.
(b) Fracture line at anterior border of greater
sciatic notch. In a little more than one case in
two (Fig. 14.2B) the fracture line detaches the
posterior column from the anterior border of
the greater sciatic notch below the angle, usual-
ly high, but sometimes at a lower point. From
there, descending obliquely downwards and for-
wards, it reaches the posterior lip of the acetab-
ulum at a variable site. It traverses the posterior
part of the articular surface dividing it horizon-
tally, reaches the anterior part of the cotyloid
fossa and then divides the ischio-pubic notch.
From there the ischio-pubic fracture configura-
tion is as before.
On the inner aspect of the pelvis, the frac-
ture line, apart from its point of departure from
the sciatic notch which is lower, has a trajectory
similar to that described in (a) above.
(c) Exceptional cases. There have been several ex-
ceptions to the descriptions in (a) and (b)
above:
Fig. 14.3. Both-column fracture involving sacro-iliac joint.
In five cases of both-column fractures the
Schemes of two examples, in both of which the fracture line
posterior column took with it the sciatic but- detaches the superior border of the greater sciatic notch and the
tress, i.e. involved the whole superior border lower part of the sacro-iliac joint surface in one isolated frag-
of the greater sciatic notch, the posterior in- ment. See Figs. 14.23 and 14.24 for radiographs
ferior iliac spine and a part of the articular
iliac surface of the sacro-iliac joint, as the
fracture line detaching the posterior column 14.1.2 Additional Posterior Components
abutted at the posterior interspinous notch
(diagram of case 7, Chap. 30). It remains to describe the additional fracture lines
In five other cases, the sciatic buttress was which may confuse the basic pattern:
detached as above but was separated from
the posterior column fragment by a second- (a) Secondary fracture of the posterior column. In
ary fracture line dividing the anterior edge 44 cases (20070 of the both-column fractures)
of the greater sciatic notch (Fig. 14.3). Thus, there was a secondary fracture line which split
among 213 cases we had 10 with an associat- the posterior column (Fig. 14.2B). This fracture
ed fracture dislocation of the sacro-iliac line was commonly little or not at all displaced
joint. and cut the posterior border of the bone, being
In four cases, there was a particular mode of adjacent to the superior border of the ischial
detachment of the posterior column in spine, or splitting it longitudinally, or extending
which this fracture line was spiral in con- from a variable point on the anterior border of
figuration. It began low on the anterior bor- the greater sciatic notch. These secondary frac-
der of the greater sciatic notch and joined ture lines of the posterior column were isolated
the acetabulum, the spiral effect existing be- (i.e. the only addition to the simplest outline of
tween the outer and inner surfaces of the a both-column fracture) in 7 cases, associated
pelvis (Fig. 14.2A). with a split of the anterior column in 13 cases,
256 Associated Both-Column Fractures

and associated with a split of the anterior col-

umn and a segmental fracture of the pelvic brim
in the remaining 24 cases.
These fracture lines traverse the retro-acetab-
ular surface rejoining the posterior border of
the acetabulum and dividing the articular sur-
face a little above the posterior horn.
In a single case, the anterior border of the
greater sciatic notch was detached in an isolated
fragment (Fig. 14.2B).
In one case (Fig. 14.2D), there was a hori-
zontal split without displacement which divided
the ischial spine longitudinally and cut the
posterior wall. The segment of the posterior
column supra-adjacent to this fracture line was
divided by another split, coronal in plane,
which followed the profile of the anterior bor-
der of the greater sciatic notch and therefore
separated one fragment which carried with it
part of the cotyloid fossa, the other comprising
a fragment of the posterior wall.
(b) Fracture of posterior wall. In 33 cases (Le.
15.5%) a secondary fracture line detached a
fragment of the posterior wall (Fig. 14.2C)
which carried some articular cartilage. Leaving
above the main fracture of the posterior col-
umn, this secondary fracture line reached lower
down to a variable point on the posterior bor-
der of the articular surface, freeing a posterior Fig. 14.4. Scheme of an atypical both-column fracture. The
fragment and a portion of the articular surface. posterior column includes a segment of the anterior wall. See
In one case, this posterior fragment was very ex- Fig. 14.29 for radiographs
tensive, extending inferiorly on to the ischium.
These additional posterior wall fractures
were the only supplementary fractures to the umn. This situation explains how in these cases,
typical both-column fracture lines in 18 cases, which had in other ways all the radiological ap-
and in 15 other cases they existed together with pearances of both-column fractures, a perfect
a secondary fracture of the posterior column as reduction was obtained by operating through
described in (a) above. the posterior approach (Fig. 22.63).
(c) Involvement of pelvic brim. In five cases, the In the two other cases, the ilio-pectinealline
posterior column took with it a segment of the was followed by the fracture line which liberat-
pelvic brim (Fig. 14.4). In three of these we are ed the posterior column, and the anterior wall
certain that a segment of the anterior wall was detached as a separate fragment.
remained attached to the fragment of the poste- These fractures differ from the T-shaped
rior column. The fracture line above which fractures associated with a vertical anterior
separated the posterior column, instead of fracture of the same innominate bone, because
avoiding the ilio-pectineal line, cut it and then no section of roof remains attached to the iliac
divided the anterior wall of the acetabulum; the wing segment (see Fig. 10.14).
anterior column was also split a little lower at
the level of the obturator canal by a fracture
roughly parallel to the main fracture, or having 14.1.3 Anterior Column Component
a bayonet-shaped configuration. Thus a seg-
ment of the anterior wall, or the anterior wall In all our cases, the anterior column has been
itself, remained attached to the posterior col- separated by a fracture which, on the external sur-
Morphology 257

face of the pelvis, begins at a variable point on the

retro-acetabular surface, confluent with the frac-
ture component which separates the posterior col-
umn; most frequently this was towards the middle
of the surface but sometimes near the posterior
border. The fracture line travels across the iliac
wing, perhaps reaching the iliac crest or extending
more anteriorly to the anterior border of the iliac
bone (Fig. 14.1).

(a) Fracture line extending to the iliac crest. This is
the commonest occurrence (82070 of the both-
column group). The fracture line can be regular
and curved, cutting both inner and outer cor-
tices approximately in the same place, but much
more frequently has different configurations on
the two cortices, being regularly curved on the
external aspect and having a zig-zag configura-
tion along the internal aspect of the wing. Even-
tually, the two cortical fractures meet each
other again at the iliac crest, splitting it
variably. The most frequent site is in the
anterior quarter, level with or behind the
anterior pillar, but it may be at the summit of
the crest or in its posterior part, rarely in its
most anterior part in front of the pillar, or even
at the antero-superior iliac spine. Sometimes
(Fig. 14.5) this iliac wing component bifurcates
in the middle of the iliac fossa, reaches the crest
at two points and isolates a triangular segment
of the iliac crest. Rarely, the iliac fracture line
just reaches the iliac crest without breaking it
and forms therefore a hinge on which the
anterior column tilts (Fig. 14.16).
On the inner surface of the pelvis (Fig. 14.5)
the fracture line cuts the cortex of the iliac fossa
in a zig-zag fashion, passing forwards from the
posterior nutrient foramen, and reaches the ar-
cuate line 2 - 3 cm in front of the angle of the
sacro-iliac joint. It then crosses the brim of the
pelvis and after a short distance, generally ob-
lique below and forwards, rejoins the fracture
line which separates the posterior column.
(b) Fracture line extending to the anterior border
(28070 of the both-column group). The iliac
fracture line always begins on the retro-
acetabular surface at a variable point on the
fracture line separating the posterior column; it
then crosses the iliac wing without reaching the
crest.
The trajectory is generally curved and regu-
lar, skirting 2-3 cm above the superior border
of the acetabulum, and most often reaches the
Fig. 14.5. Scheme of possible
fracture lines detaching the
anterior column
interspinous notch (Fig. 14.1 B). Sometimes it
passes nearer the acetabulum and reaches the
anterior border below the anterior inferior iliac
spine in the psoas groove. Rarely, the line starts
at the back of the acetabulum passing horizon-
tally to split the roof of the acetabulum and
allowing a small segment of the articular sur-
face to remain on the iliac wing. The latter is a
transitional form with respect to fracture of the
posterior column associated with a hemitrans-
verse anterior fracture which we included in our
study of T-shaped fractures.
On the inner surface, the fracture in the iliac
wing, of which the outline is zig-zag or curved,
always cuts the ilio-pectinealline in front of the
sacro-iliac joint and rejoins the fracture line
cutting off the posterior column.
14.1.4 Result of Both-Column Fracture
Between the fracture line of the iliac wing and the
fracture which liberates the posterior column, the
anterior column finds itself totally detached, carry-
ing with it the roof and the anterior wall of the
acetabulum. This large anterior column fragment
was not detached in a single piece except in four
cases where the iliac wing component reached the
crest, and in four cases where the iliac fracture went
to the anterior border; most commonly, it is split
(Fig. 14.6). This split may occur at the level of the
root of the superior pubic ramus, at the level of the
superior pubic ramus (perhaps at these two points),
at the level of the iliac wing or at the level of the
acetabular roof in the following ways:
when the secondary split is at the level of the
root of the superior pubic ramus (Fig. 14.6) it is
258
} Fracture lines } the roof of the acetabulum which isolated a poste-
cutting
anterior wall Fractures of
the root
} Extra-articular of pubic ramus
fracture lines
} Fracture lines
through superior pubic ramus
Fig. 14.6. Scheme of possible fracture lines on the anterior col-
umn in involving the superior pubic ramus. (The articular sur-
face is shown hatched)
most commonly perpendicular with respect to
the axis of the column; it is situated above,
below or at the level of the notch of the anterior
border of the acetabulum and crosses the ar-
ticular surface of the anterior wall. Sometimes,
the split is oblique or bayonet-shaped, and can
therefore pass extra-articularly leaving the ante-
rior wall attached to the superior part of the an-
terior column. A particular accessory fracture
line can occur here so that the roof of the obtu-
rator canal becomes a small isolated fragment;
at the level of the superior pubic ramus we find
again a fracture line which may be perpendicu-
lar to the axis of the pubic ramus, but which is
more often oblique; it stays little displaced and
is difficult to discover because the fragments are
maintained by the pectineal part of the inguinal
ligament which always remains unruptured,
although it has in a few cases been stripped off
the pubic ramus;
rarely the anterior column is fractured inferiorly
at two points, one at the root of the superior
pubic ramus and one more medially. The inter-
mediate fragments may be split by further longi-
tudinal fracture lines;
in some instances, at a variable point along the
iliac fracture line, there was a related horizontal
split which reached the anterior border of the
bone at the level of the interspinous notch or at
the psoas groove (Fig. 14.5), isolating therefore
an anterior fragment of iliac wing. Occasionally
this split stopped short of the region of the ante-
rior border which was not ruptured.
Associated Both-Column Fractures
In ten cases in which seven fracture lines reached
the iliac crest and three reached the anterior border
of the bone, there was a split situated at the level of
rior superior fragment carrying a segment of arti-
cular cartilage (Fig. 14.1 B).
In conclusion, two main groups of fractures of
the both-column type are classified according to
the nature of the iliac fracture. They are related (1)
because certain fractures extending towards the il-
iac crest are associated with a split directed towards
the anterior border; or (2) because certain fractures
with an iliac component directed towards the
anterior border also have an element which is in-
complete but ascends towards the iliac crest with-
out reaching it (see Fig. 14.5).
The configuration of the fractures of both col-
umns can be further complicated by the existence
of supplementary fragments. For example:
a detached segment of bone from the posterior
part of the pelvic brim limited by the fracture
line separating the anterior column and the
angle of the sacro-iliac joint;
fracture components in the iliac wing separating
triangular or polygonal fragments, which only
involve one cortex, particularly of the posterior
and inferior part of the fracture of the internal
iliac fossa a little in front of the sacro-iliac joint;
comminution of the fracture line at the level of
the pubis.
14.1.5 Displacement of the Fragments
and the Femoral Head
The force acting on the trochanter which brings
about these fractures accounts for the nature of the
internal displacement sustained by the columns.
There is also an accompanying rotation around
their main vertical axes which results in an angular
displacement of their articular surfaces relative to
the femoral head which is displaced centrally
(Fig. 14.7). This rolling effect maintains in a signifi-
cant number of cases congruence between the head
and the various segments of the fractured articular
surface. This secondary congruence is accompa-
nied of necessity by separation of the fragments
composing different segments of the socket and by
a narrowing of the mouth of the acetabulum, the
segments at the lip margin being obliged to twist
one with respect to the other in order to maintain
contact with the head.
Morphology
Fig. 14.7. Scheme showing concept of displacement of frag-
ments leading to secondary congruence
14.1.6 Atypical Examples
In two cases where the fracture line reached the
anterior border of the ilium, an associated break in
the anterior column at the level of the psoas groove
isolated a segment of the anterior column which
itself was split by a vertically disposed fracture line
separating the two cortices, the outer carrying the
roof and the inner a segment of the pelvic brim
(Fig. 14.8). They are extremely difficult to reduce
intraoperatively.
Table 14.2
1. Additional posterior wall fracture
2 Additional fracture of the anterior column
3. Secondary fracture of the posterior column
4. Supplementary fragment of the pelvic brim
5. Supplementary fragment of the iliac fossa
6.4+2
7.5+2
8.4+2+3
9. Sacro-iliac fracture dislocation (fract. of the
sciatic buttress)
to. 2+3
II. Additional posterior wall fracture and sec-
ondary fracture of the anterior column
12. Spiral fracture detaching the posterior column
13. Posterior column brings with it a segment of
the pelvic brim
259
In a few cases there existed other lesions of the
pelvis, namely, separation of the pubic symphysis
or a fracture of the opposite obturator ring.
In one injury (Figs. 14.9 and 14.27) we found an
incarceration of the posterior column within the
pelvis which even retrospective study of the radio-
graphs did not define. We established through the
ilio-inguinal operative approach that inside the
pelvic brim the posterior column had been dis-
placed inwards and was rotated internally by about
60° on its vertical axis. Its retro-acetabular surface
rested against what remained of the quadrilateral
surface below the brim of the pelvis. The integrity
of the capsular attachment and upward displace-
ment of the posterior fragment made reduction,
and indeed understanding, of the lesion very dif-
ficult. CT scanning would have been of great help
in such a case.
To conclude this detailed anatomical study:
besides the main fracture lines characterising the
both-column fractures, additional fractures lines
are frequent, giving particular features to many
cases. Table 14.2 classifies these features according
to the point of rupture of the iliac crest or the
anterior edge of the innominate bone.
14.1.7 The Key to Reconstruction
Whatever the complexity of the fracture or the
number of fragments, each segment of articular
Iliac wing fracture line reaching:
Iliac crest Anterior superior Interspinous Psoas groove Total 0/0
iliac spine notch
14 2 1 18 8.49
49 5 2 57 26.89
6 7 3.30
3 3 1.42
6 7 3.30
19 5 25 11.79
14 1 16 7.55
21 2 24 11.32
10 10 4.71
II I 13 6.13
11 2 15 7.08
4 4 1.89
3 2 5 2.36
260 Associated Both-Column Fractures

A B

C o E F

Fig. 14.8A-F. Atypical both-column fracture. A,B Clinical specimen, C-F schemes showing fracture configuration. See Fig. 14.28
for radiographs
Radiology 261

14.2.1 Antero-posterior View

(a) Central dislocation of femoral head. The head

(b) Inward displacement of posterior column. The
Upper segment of
posterior column in situ
Posterior column
entrapped and rotated
60 0 about Hs axis
Fig. 14.9. Diagram of displacement and incarceration of posteri-
or column. See Fig. 14.27 for radiograph
surface belongs to one or the other column. There
remains attached to the sacrum a part of the iliac
wing which never bears articular surface but is the
key to reconstruction.
14.2 Radiology
Although both-column fractures appear at first
sight very complex on radiography, the broad diag-
nosis is in fact relatively simple, even on the antero-
posterior view alone; the precise diagnosis of the le-
sion necessitates study of the oblique views in order
to appreciate the displacement, to recognise the
secondary fractures, and eventually to choose the
best surgical approach. CT scanning and 3-D re-
constructions are now of great help.
related to the displaced inferior posterior col-
umn fragment:
(i) The ilio-ischial line is less clear than nor-
Fig. 14.10. The fracture lines through the two cortices often do
not follow the same track. Their displacement may give rise to
the appearance of four fracture lines dividing the iliac wing
lies clearly displaced inwards, often appearing
within the pelvic brim in external rotation.
femoral head drives inwards and tilts a large
bony fragment which comprises at least the in-
ferior part of the ischium.
This fragment constitutes a major part of
the posterior column and it is bounded below
by a fracture line through the ischio-pubic
ramus or by a fracture line concave from above
through the body of the pubis (Figs. 14.11 Band
14.16B) which we have met only in fractures of
both columns.
Above, it is bounded by a fracture line which
breaks through the posterior wall of the acetab-
ulum at a high but variable level. One does not
always see the point of rupture of the posterior
border because of tilting of the inferior seg-
ment. Nevertheless, if the posterior border of
the inferior fragment is traced upwards, this
leads significantly inwards of the lateral ex-
tremity of the radiological roof, proving that
there is a high fracture of the posterior border
and its displacement. The line of the fracture of
the posterior wall is sometimes easily visible.
On the medial aspect, the posterior border
of the iliac bone is recognised in its upper part
by the characteristic outline of the greater
sciatic notch which is detached, perhaps at the
level of its angle or at the level of a point along
its anterior border. Lower down, the ischial
spine can be defined, often within the area of
the pelvic brim. It is important to identify the
spine and to avoid confusing it with other frag-
ments of bone which when outlined radiologi-
cally can have a remarkably similar appearance.
(Figure 14.29 shows such an example in which
the middle segment of the ilio-pectineal line
simulates the ischial spine. The ischial spine is
visible below and externally.)
Certain other features of importance can be
mally but easily recognisable; it is often in-
complete below (because of the rotation of
the fragment), and in most cases it has lost
its normal relationship with the teardrop
which seems to be displaced inwards from
it.
262
(ii) The teardrop and the cotyloid fossa belong
to this fragment. They may remain aligned
with the head of the femur (Fig. 14.14) but
in all cases they have lost their normal rela-
tionship with the superior pubic ramus and
the ilio-pectineal line, which indicates a
separation and the passage of the fracture
line in the cotyloid fossa near the anterior
wall of the acetabulum. Sometimes the
teardrop and the ilio-ischialline have main-
tained their relationship fairly normally
(Fig. 14.11).
(iii) The posterior horn of the acetabuluar arti-
cular surface is sometimes visible, more or
less concentric with the head of the femur.
(iv) The ilio-ischial fragment has lost its nor-
mal relationship with the segment of the
brim of the pelvis belonging to the superior
pubic ramus and which appears much more
horizontal than normal. The crossing of
the superior pubic ramus and the ilio-
ischial line confirm their relative displace-
ments.
(v) Sometimes two points of rupture of the
posterior border of the acetabulum can be
seen on the antero-posterior view, one
situated high and the other lower, isolating
a fragment of the posterior wall (which is
confirmed on the obturator-oblique view)
(Fig. 14.13). In other instances, a fracture
line across the ilio-ischial fragment meets
the greater sciatic notch at a variable posi-
tion; it is generally little or not at all
displaced (Figs. 14.26 and 14.27; see also
Sect. 14.1.2(a».
(c) Ilia-pectineal line fracture. The pelvic brim is
broken posteriorly in a constant pattern at a
point difficult to locate with precision on the
antero-posterior view. This proves the rupture
of the anterior column (but not necessarily its
separation from the posterior column). In addi-
tion, it is common for additional fractures to
occur at the level of the root of the superior
pubic ramus (the fracture line then traversing
the anterior wall of the acetabulum), or at the
level of the superior pubic ramus, when it is ex-
tra-articular. These secondary fractures can be
comminuted.
(d) Roof of the acetabulum. This is tilted and
displaced as a whole, so as to look more or less
downwards and inwards. To a varying degree it
continues to frame the femoral head. This vari-
able degree of tilt is evident by comparison with
Associated Both-Column Fractures
the opposite side on the routine radiograph of
the whole pelvis.
(e) Iliac wing fracture. According to the case, this
extends to the iliac crest or to the anterior
border of the ilium. The iliac crest is fractured
in a variable position. The fracture line through
the iliac wing is most commonly straight or
regularly curved along the outer aspect of the
wing, whereas along the internal aspect of the
wing it has a zig-zag configuration. When these
two differently shaped fracture lines are dis-
placed, it appears at first sight that there are
four fracture lines through the iliac wing
(Fig. 14.10).
Fairly frequently, in the middle part of the il-
iac fossa, the iliac wing fracture line bifurcates
so as to reach the iliac crest at two points and
isolates therefore a triangular fragment of the
iliac wing with its base superiorly disposed
(Figs. 14.12 and 14.15).
In other cases the fracture line reaches the
anterior border at very variable positions:
at the level of the antero-superior iliac spine
(Fig. 14.18);
at the level of the interspinous notch
(Fig. 14.20);
at the level of the psoas groove just below
the antero-inferior iliac spine;
exceptionally, the fracture line crosses the
roof in order to reach the psoas groove and
allows therefore a short segment of roof to
remain on the iliac wing, constituting a tran-
sitional feature (Fig. 14.21).
The iliac fracture may stop before quite reach-
ing the crest (Fig. 14.16).
(f) Obturator ring. This is broken in its lower part,
at the level of the ischio-pubic ramus in one or
two places or at the body of the pubis or in both
places. Exceptionally (five cases) the ring was
not broken.
(g) "Curved image" (Fig. 14.19). In the great ma-
jority of both-column fractures, the antero-
posterior radiograph includes a curved image
which starts from the lower pole of the sacro-
iliac joint, at first describing in an outward
direction a curved trajectory, dense, regular,
and like the initial part of the pelvic brim (one
can superimpose it on the first part of the op-
posite pelvic brim). It may then appear to
straighten and stop, but if examined carefully it
can be seen as a variable outline, thin and
sinuous, which continues outwards, perhaps
curving upwards and obliquely, or in describing
Radiology
a spur pomtmg downwards (Figs. 14.25 and
14.27) becoming increasingly indistinct. The
latter corresponds to the profile of the fracture
which separates the posterior column and
forms the lower limit of the iliac fragment
which remains attached to the sacro-iliac joint.
The curved image has two possible sources:
(i) It may be produced only by the initial seg-
ment of the radiological pelvic brim. It then
has the same density and the same cur-
vature possessed by the ilio-pectinealline of
the opposite side (Figs. 14.18 and 14.28) and
corresponds to the internal face of the
sciatic buttress. This zone is independent of
the anatomical pelvic brim which may be
detached without altering the curved image
(Fig. 14.19, type I). It can appear more open
than on the opposite side by virtue of
separation of the sacro-iliac joint which
may allow outward rotation of the iliac
wing; even so, its curvature and its density
are perfectly regular, similar to the opposite
side.
(ii) In other cases (Fig. 14.19, type II), careful
examination shows that the curved image
loses thickness from its internal aspect quite
sharply and becomes more vertical. The in-
ferior segment represents the highest part of
the ilio-ischial line (Figs. 14.16 and 14.17)
and corresponds to the incidence of the x-
ray beam tangential to the upper part of the
ilio-ischial zone continuous with the sciatic
buttress. Its lower limit is formed by a V-
shaped fracture, open above, which is the
optical section of the fracture which
separates the posterior column from the un-
disturbed iliac fragment. This second ver-
tical thin segment corresponds therefore to
the outline of the upper part of the ilio-
ischial line after separation of the pelvic
brim with the anterior column on which it
is normally superimposed. Following an an-
atomical reduction, it can be seen to coin-
cide perfectly with the ilio-ischial line.
When the second segment of the curved im-
age, dense but thin, does not exist, it is
because the fracture line which detaches the
posterior column at the level of the
quadrilateral surface is horizontal and
situated above the ilio-ischial zone.
In summary, the centrally displaced femoral head
driving inwards the ilio-ischial fragment, together
263
with the tilt of the whole roof and an iliac fracture
line, establish the injury as a both-column fracture
of which the characteristics will be more accurately
delineated on the oblique views.
In practice on this antero-posterior view, all the
six typical and fundamental radiological land-
marks of the acetabulum are ruptured at one or
more sites and displaced. Not one of them remains
undisturbed. Accessory features which may be ob-
served on the antero-posterior view are:
accompanying lesions of the pubic symphysis or
the sacro-iliac joint (Figs. 14.23 and 14.24);
the spur sign (to be described later), seen much
more clearly on the obturator-oblique view
(Fig. 14.12);
another injury of the pelvis, in particular a frac-
ture of the opposite pubic rami;
a reduplication of the pelvic brim image seen
when the fracture line detaching the posterior
column splits the ilio-pectineal line longitudi-
nally.
14.2.2 Obturator-oblique View
This contributes the following information:
(a) Central dislocation of the femoral head.
(b) Separation of the roof. Its outline is continuous
with that of the anterior wall. These two often
remain congruent with the femoral head.
(c) Fracture of the pelvic brim. The site is often ex-
tremely posterior, and this is the best view to
locate it. It is commonly broken again, perhaps
at the level of the root of the superior pubic
ramus or more medially, the fracture line in-
volving or sparing the anterior wall of the
acetabulum. These two areas of damage can be
comminuted. Whereas on the antero-posterior
view, as a result of the tilting, one sometimes
loses the outline of the middle segment of the
pelvic brim; here it is always clearly visible.
(d) Rupture of the obturator ring locates at the
ischio-pubic notch where the break is often
clear and corresponds to the separation of the
two columns, at the ischio-pubic ramus or in
the body of the pubis.
(e) The spur sign (JUDET-LETOURNEL; Fig. 14.22).
On the obturator-oblique view the area of the il-
iac wing, just above the roof, which is tangen-
tial to the X-ray beam constitutes here the outer
limit of the image of the iliac wing. In all both-
column fractures the fracture line in the iliac
264
wing passes through this external cortex of the
supra-acetabular region and medial displace-
ment of the lower segment is manifest by the in-
terruption of the outer cortex. The iliac wing
fracture line slopes upwards and medially or de-
scribes a V-shape, open upwards; in either case
the configuration forms a characteristic bony
spur. The variable outline of the spur depends
on the orientation, shape and direction of the
iliac wing fracture line. The lower component,
adjacent to the roof of the acetabulum, belongs
to the antero-inferior iliac segment and is
displaced inwards by the head. We have seen
this spur sign in 95070 of cases of both-column
fractures. It will not be visible if the obliquity
of the pelvis is insufficient, for it can be masked
by the anteriorly displaced segment of the iliac
wing, the posterior part of the iliac wing being
hidden. It can be difficult to see when the frac-
ture is situated low and obscured by the femoral
neck.
The spur sign was lacking in two cases when
the fracture extended to the psoas groove
(Fig. 14.21). It was represented in a few cases by
local kinking in the supra-cotyloid region with
slight lateral opening if the central displace-
ment was slight. It can be difficult to see in
cases where a displaced fragment of the poste-
rior wall becomes radiologically superimposed;
careful scrutiny will, however, reveal the spur
(Fig. 14.14).
The spur sign can reduplicate, this being due
to the bifurcation in a Y-shaped iliac wing frac-
ture with the isolation of a large triangular
fragment (Fig. 14.15).
In cases of iliac wing fractures which extend
to the iliac crest, it is quite often possible to see
two clear dense lines forming the outer limits of
the two segments of the iliac wing separated by
the iliac fracture line. They converge towards
the iliac crest, the more internal of the two
reaching the outer aspect of the roof of the ace-
tabulum. The outer border of the more external
of the two reaches the spur (Fig. 14.16).
Where an iliac wing fracture extends to the
anterior border, a short segment of variable
height of the iliac wing appears medial to the
spur and surmounting the roof of the acetabu-
lum (Fig. 14.20).
A similar appearance to the spur sign has
been seen in certain pure fractures of the
anterior columll, namely those in which the il-
iac wing fracture line reached from the iliac
Associated Both-Column Fractures
crest to an area behind the roof of the
acetabulum. It is easy to understand that in the
obturator-oblique view, the supra-acetabular
segment of such a fracture line may be visible
but in general there is very little displacement.
The fact that all the outlines of the posterior
column are intact makes it unlikely that this
may be misinterpreted as a true spur sign.
(f) Posterior or postero-superior fragment. If
present, it is seen usually as a triangular seg-
ment limited on the outer aspect by part of the
posterior border of the acetabulum and tilted
with the femoral head (Figs. 14.14 and 14.18).
(g) Sacro-i1iac joint anterior opening. This injury
occurs occasionally and is clear in the obtura-
tor-oblique radiograph.
(h) Fracture through the root of the superior pubic
ramus. This extends into the anterior wall of
the acetabulum. Comminution of this compo-
nent can isolate the roof of the obturator
canal.
On this view a strict study of the landmarks dem-
onstrates that:
In regard to the posterior column:
the posterior lip of the acetabulum is broken;
- the ischio-pubic ramus is severed in one or two
places.
In regard to the anterior column:
the pelvic brim is interrupted in one or more
places;
the anterior lip of the acetabulum is broken;
the roof is tilted;
the characteristic spur sign is practically always
present.
14.2.3 lliac-oblique View
(a) The displacement of the posterior column frac-
ture is usually clearly manifested. In a few cases
the greater sciatic notch at first sight appears in-
tact but close examination reveals an abnormal
angulation of its anterior border. Also appreci-
ated is the degree of congruence between the
femoral head, the cotyloid fossa, and some-
times the posterior horn of the acetabular arti-
cular surface, as well as the point of rupture of
the greater sciatic notch. A supplementary frac-
ture of the superior or anterior border of the
greater sciatic notch is sometimes identifiable
(Fig. 14.25).
Summary
A secondary fracture line may exist which
splits the posterior column (Fig. 14.26). In rare
cases, a fracture line of spiral configuration,
situated low on the greater sciatic notch,
separates the posterior column fragment
(Fig. 14.13).
(b) The iliac fracture line is accurately delineated.
Its trajectory (curvilinear or zig-zag on both
cortices), the point of rupture of the fracture
line on the iliac crest or the anterior border of
the bone, and the existence of a possible split
towards another point of the iliac crest isolating
a triangular fragment or towards the anterior
border at the level of the interspinous notch or
at the psoas groove are all seen.
The anterior iliac fragment is seen to bear a
fragment of the articular surface comprising
the whole roof of the acetabulum which is tilted
with it and frames the femoral head with vary-
ing accuracy.
In cases where the fracture line in the iliac
wing reaches the anterior border and is asso-
ciated with a split of the anterior column which
cuts superiorly the anterior wall of the acetabu-
lum (see Fig. 14.8), the roof finds itself isolated
with a segment of the wing, polygonal in shape,
and which appears totally separated from the
rest of the anterior column (Fig. 14.20); this im-
portant fragment can, moreover, be split in the
coronal plane as we shall see (Fig. 14.28).
Note that the fracture line of the quadrilateral sur-
face which separates the two columns is only rarely
visible, and that the point of rupture of the anterior
border of the acetabulum is equally as difficult to
delineate.
On this view a strict study of the landmarks of
the two columns demonstrates:
In regard to the posterior column:
Table 14.3
Both-column fracture to: Post-traumatic congruency in both-column fractures
No congruency Congruency only
with the roof
Iliac crest 74 67
Anterior superior iliac spine 9 2
Anterior interspinous notch 8 7
Psoas groove 3 2
Total 94 77
265
the break of the posterior edge of the innomi-
nate bone,
the fracture line separating the two columns
through the quadrilateral surface.
In regard to the anterior column:
the break of the anterior lip of the acetabulum;
the fracture lines through the iliac wing, which
are here perfectly delineated;
14.3 Summary
For didactic purposes, we have described the stan-
dard radiographic views separately. With experi-
ence, the views are most rapidly and more usefully
read together, the antero-posterior and the obtura-
tor-oblique views to study the anterior column, and
the antero-posterior and the iliac-oblique views for
the posterior column.
The basis of both-column fractures rests on
features already described but may be summarised
thus:
central dislocation of the head which drives
medially a large fragment or the whole of the
posterior column, confirmed on the antero-pos-
terior and iliac-oblique views;
the iliac wing fracture configuration, seen on
the antero-posterior and the iliac-oblique views;
the spur sign, seen on the obturator-oblique
view.
The standard views indicate to a certain extent the
degree of congruence between the femoral head
and the various fragments of the articular surface
of the acetabulum, but for a full assessment of this
post-traumatic congruence recourse must be had to
tomography and CT scanning, as we shall see later.
Table 14.3 shows the assessment of articular con-
gruence on the basis of the three standard views in
the pre-operative radiographs.
Congruency only with anterior Total secondary
and posterior wall congruency
8 26
1
1 3
12 29
266
14.4 Atypical Examples
(i) In ten cases, the fracture of both columns was
associated with a fracture-dislocation of the
sacro-iliac joint on the same side (Fig. 14.3).
From the inner aspect of the pelvis in five of
these cases, the fracture line traced from the il-
iac crest began in its posterior part and reached
the anterior border of the sacro-iliac joint a lit-
tle above the pelvic brim. It crossed the sacro-
iliac articulation and detached its inferior por-
tion, continuing posteriorly to the posterior
border of the bone above the postero-inferior
iliac spine. A little before reaching the sacro-
iliac joint the fracture line bifurcated and the
inferior branch, running towards the pelvic
brim, cut it 2 - 3 cm in front of the angle of the
sacro-iliac joint, and then descended on the
quadrilateral surface to rejoin the fracture line
which detached the posterior column and orig-
inated with it at the angle or the upper third of
the anterior border of the greater sciatic notch.
Thus a bony segment became isolated which
carried the inferior half of the iliac articular
surface of the sacro-iliac joint, the postero-
inferior iliac spine, the superior border of the
greater sciatic notch, and a variable portion of
its anterior border. This free segment displaced
and rotated considerably.
The fragment and its related articular sur-
face is recognised in the antero-posterior view
when it is pivoted through about 90°
(Fig. 14.24), and on the antero-posterior view
and on the iliac-oblique view when it is less
rotated (Fig. 14.23). This fragment, which
brings with it a part of the iliac articular sur-
face of the sacro-iliac joint is seen in profile on
one or two of the standard radiographs. How-
ever, this piece of bone is in fact at least a part
of the sciatic buttress. Seen in profile it ap-
pears as a huge U-shaped image of dense con-
sistency, sometimes a little square at its base,
and broadly open from above. On the one or
two standard views which do not demonstrate
this U-shaped image, a separate fragment may
be seen which is limited laterally and below by
a line which has a characteristic profile. This is
produced by the postero-inferior spine pro-
longed by the superior border, the angle and a
part of the anterior border of the greater
sciatic notch (Figs. 14.23 and 14.24).
In five other instances the above-described
fragment remained connected to the posterior
Associated Both-Column Fractures
column (see case 7, Chap. 30). The U-shaped
image was not seen in these cases, probably
because the fragment had not rotated so far.
The diagnosis was easy; a posterior column
fragment taking with it the sciatic buttress is
clearly delineated and essentially recognisable
on the oblique views.
(ii) Comminution of one or more segments of the
multiple fracture lines which traverse the iliac
bone can be extensive. It is possible to be per-
suaded (e.g. the case in Fig. 14.25) that there
would be no chance of a successful reconstruc-
tion but this is not necessarily so.
(iii) In a certain number of fractures of both col-
umns where the iliac wing fracture reaches the
interspinous region between the iliac spines on
the anterior border, there exists at the same
time a split in the anterior column crossing the
upper anterior wall of the acetabulum, isolat-
ing therefore a superior segment of the
anterior column comprising the roof and a seg-
ment of the pelvic brim. In two cases this frag-
ment was split sagittally in such a way that the
external part carried the roof of the acetabu-
lum, and the internal part with a part of the
iliac fossa carried a segment of the pelvic brim.
The radiological search for this fracture that
can be the cause of considerable surgical prob-
lems is difficult. The outline can be detected
best on the antero-posterior and iliac-oblique
views (Figs. 14.8 and 14.28) and CT will dem-
onstrate it clearly.
(iv) In five cases a special feature was that the
detached posterior column included a segment
of the pelvic brim; in all of these cases the iliac
wing fracture line reached the anterior border
of the innominate bone. In two of these the
pelvic brim was split by the fracture line which
detached the posterior column.
In the other three cases, the posterior col-
umn took with it a segment of the pelvic brim
and a corresponding segment of the anterior
wall of the acetabulum (Fig. 14.4). One recog-
nises this peculiarity on the antero-posterior
view for on the ilio-ischial fragment the tear-
drop and the ilio-ischial line have kept their
normal relationships and at the upper part of
the fragment is seen a short segment of the
pelvic brim which keeps its usual relationships
with the preceding elements. In this case, the
roof and the polygonal segment above it in the
iliac wing form an isolated fragment very
clearly separated from the rest of the anterior
CT Study of Associated Both-Column Fractures
column (Figs. 14.4 and 14.29). These three
cases approach the limit of the classification of
the both-column fracture. One could regard
them as comprising a pure transverse fracture
line plus an anterior vertical fracture of the
pelvis, dividing extra-articularly the obturator
ring and a postero-superior fracture, isolating
a fragment carrying all the articular surface
situated above the transverse fracture line.
Thus they form the threshold transitional
group between fractures of both columns and
transverse fractures associated with a postero-
superior fracture.
14.5 Differential Radiological Diagnosis
Both-column fractures should be distinguished
from the following:
(a) T-shaped fractures. Although the bony ring of
the obturator foramen is cut above and below,
only an inferior segment of each of the two col-
umns is isolated. The upper limit of section of
the columns forms a plane transverse fracture
surface. Above all, the fracture line (be it infra-,
juxta-, or trans-tecta!) leaves in place, attached
to the iliac wing, an articular roof segment of
greater or lesser size. The iliac wing is not in-
volved by any of the fracture lines. There is no
spur sign in the obturator-oblique view
(Figs. 10.8 - 10.10).
(b) Anterior column and posterior hemitransverse
fractures. These associated fractures are also
very different from both-column fractures, for
although an anterior part of the roof is de-
tached with the anterior column, the posterior
hemitransverse fracture component cuts the
posterior wall and the upper part of the latter
remains in place, attached to the posterior part
of the iliac wing with at least the upper part of
the posterior wall (Fig. 14.30).
At operation, using the posterior approach,
one finds only the transverse component cut-
ting the posterior column; the fracture of the il-
iac wing only joins it at the level of the cotyloid
fossa. The configuration of the fracture is
displayed perfectly through the extended ilio-
femoral approach.
Finally, when the iliac wing fracture line
reaches the acetabulum a little behind the roof,
it appears on the obturator-oblique view as an
inflexion above the acetabulum but not as a
267
true spur sign as is the case in the both-column
group.
This association is distinguishable because
on the antero-posterior or obturator-oblique
view or on both, an external fragment of the
roof, sometimes small, is maintained in place
and appears on the external limit of the bone
between the slightly displaced wing fracture
above the acetabulum and the point of rupture
of the posterior border of the acetabulum. One
should not confuse it with an isolated posterior
fragment which always tilts with the articular
surface it carried (see Chap. 13).
(c) Associated transverse fractures with fractures
of the iliac wing. A vertical split of the iliac
wing leaves from the transverse fracture line in
the posterior acetabular region and ascends
towards the iliac crest. The upper fragment
comprises all the anterior part of the iliac wing,
and the acetabular articular segment situated
above the transverse fracture line (Figs. 12.11
and 12.12). These form the link and transition
between both-column fractures and transverse
fractures associated with posterior fractures.
We have met 18 cases of these particular asso-
ciated transverse and postero-superior frac-
tures; these and the both-column types are the
only two kinds of acetabular fracture to detach
the whole of the acetabular articular surface.
14.6 CT Study of Associated
Both-Column Fractures
When the iliac wing fracture line reaches the iliac
crest (Figs. 14.31, 14.32), the fracture appears in the
upper sections at a variable point which can be
precisely located. Generally the fracture line pro-
gresses downwards and inwards, dividing the cor-
tices of the iliac wing coronally. The thus frontally
disposed fracture line cuts the inner cortex of the
wing more medially that the outer cortex.
Localised comminutions separating plates of
bone from either cortex of the iliac wing are fre-
quent (Fig. 14.32).
When the iliac fracture reaches the anterior bor-
der of the bone (Fig. 14.33), the sections through
the iliac crest show intact bone, while those passing
through the interspinous notch, the anterior-inferi-
or iliac spine or just below it demonstrate the start-
ing point of the iliac wing fracture which is also
coronally disposed and may be associated with
localised cortical comminution.
268
Normally connected (in most cases) to an intact
sacrum and an intact sacro-iliac joint, the intact
posterior part of the iliac wing is limited anteriorly
by the iliac wing fracture line and inferiorly and
posteriorly by the roof of the greater sciatic notch
(of which a part may be detached with the posterior
column). Below the level of this roof, the posterior
intact part of the wing, which is limited anteriorly
(see Fig. 14.1) by the fracture line separating the
anterior column, and from which originates the
spur sign on the obturator-oblique view, appears on
successive sections as an island of bone of decreas-
ing size (Figs. 14.31-14.33, 14.35), limited inwardly
and outwardly by cortical bone, anteriorly and pos-
teriorly by the fracture lines separating the two col-
umns, and, finally, by spongy bone only. It is visi-
ble as far as the inferior tip of the curved image on
the scout view. Identifying this intact part with cer-
tainty needs some attention, and to be sure that it
is in continuity with the intact part of the wing is
not always easy.
The upper tip of the posterior column fragment
generally appears on the section above the roof of
the greater sciatic notch, because of its upward
displacement; it is medially situated with respect to
the intact part of the wing just described above. In-
creasing rapidly in size, the fragment is limited
posteriorly by the thick and angulated anterior
edge of the greater sciatic notch. The fracture line
limiting it at the front is coronally disposed, but
may not appear so at first sight, due to the internal
rotational displacement of the fragment.
At the level of the scanner roof (the compact
area of sub-chondral bone representing the ana-
tomical dome; see Figs. 14.31-14.34), in the typical
case, the fracture line separating the two columns is
pure, nearly straight and coronally disposed. Of
course, depending on the rotational displacement
of the two columns, this fracture line may appear
open medially. Generally this fracture is situated in
the posterior part of the scanner roof, as the great-
er part of the latter remains linked to the anterior
column fragment.
The anterior column segment, which was identi-
fied at the iliac wing level, is displaced inwards and
forwards and rotates externally around its vertical
axis.
Below the scanner roof the two columns are
separated by a coronal fracture line dividing the
cotyloid fossa in its middle or anterior part, then
skirting the anterior wall, and correspondingly di-
viding the ischio-pubic notch.
Associated Both-Column Fractures
The sections through the obturator ring show its
fracture sites: the root of the superior pubic ramus,
where there is very frequently comminution which
mayor may not involve the anterior horn of the
acetabular articular surface; the superior pubic
ramus; and the pubic angle or the ischio-pubic
ramus. The roof of the obturator canal often makes
a separate fragment.
The main column fragments may present sec-
ondary fracture lines. A more or less extended
posterior wall fragment may be detached from the
posterior column, taking with it a part of the
posterior acetabular articular surface; as in a pure
fracture of the posterior wall, the fracture line
dividing it is sometimes oriented outwards and for-
wards. Quite often, however, the fracture line de-
taching the fragment runs in an approximately
antero-posterior direction and transgresses the pos-
terior wall obliquely downwards and outwards on
successive cuts. The posterior wall fragment is
usually only slightly displaced and apparently
detached, as is the posterior hemitransverse frac-
ture in the case of associated anterior column and
posterior hemitransverse fractures (see Fig. 13.15).
In other cases the posterior wall is very small,
detaching little of the articular surface and much
more of the cortex of the retro-cotyloid surface.
Many other possible splits of the posterior col-
umn were described before the advent of CT
(Fig. 14.1 and Sect. 14.1.2). CT scanning will now
greatly help in identifying them.
Other possible fracture lines dividing the anteri-
or column will also be clearly analysed on the CT
sections. The presence of the frequently seen Y
configuration (Fig. 14.31) of the iliac wing fracture
line can be confirmed or demonstrated.
CT allows analysis of the displacement of the
columns and the femoral head. It also helps in the
assessment of residual post-traumatic congruency;
indeed, the congruence between the acetabular walls
and the head is excellently revealed on CT sections.
The congruence between the scanner roof and the
head is more difficult to appreciate accurately, and
tomography appears to be essential for this.
As usual, marginal impaction and intra-articu-
lar fragments may be discovered, as well as asso-
ciated fractures of the sacrum and/or sacro-iliac
joint injuries.
It is in addition true to say that CT has in cases
demonstrated slight displacement or absence of
displacement of fracture lines which were impossible
to distinguish even a posteriori on the standard views.
CT Study of Associated Both-Column Fractures 269

A B

"- ', ".

\
I
) "

'- "-
,,.'"",
\ \
,
I ,
.- -...... i
")
( \ j
\
\
"
....... . \ I
/

\ \ j
\
\ b
I
a

Fig. 14.11 A-D. Both-column fracture. A Antero-posterior ra-
diograph, a diagram, B obturator-oblique radiograph, b dia-
gram, C iliac-oblique radiograph, c diagram, D scheme of frac-
ture configuration. The fracture line in the iliac wing reaches the
crest behind the anterior pillar. This is a rare example in which
the columns are detached in whole pieces. It was approached
from the posterior aspect and the anterior column was not seen;
nevertheless, none of numerous radiographs have shown any
fracture through the anterior border of the acetabulum. The
spur sign is clearly visible in B
270 Associated Both-Column Fractures

,
c
'"\.

Fig. 14.nc,D
CT Study of Associated Both-Column Fractures 271

Break in c rest

/r::.,j
'
/ I
";·{'-tJ,I~~1- ilia-ischial line
\.\ \
\
,
\
I
a I

Fig. 14.12A-C. Both-column fracture in which the iliac compo- fracture of the superior pubic ramus is probably extra-articular.
nent bifurcates and isolates a triangular fragment. A Antero- The anterior border of the greater sciatic notch is separated as
posterior radiograph, a diagram, B obturator-oblique radio- a discrete fragment C and there is a split in the pelvic brim A
graph, b diagram, C iliac-oblique radiograph, c diagram. The
272 Associated Both-Column Fractures

Tri angular segment

of iliac wing

Fragments of cortex
of ni ternal iliac fossa

Fig. 14.12C
CT Study of Associated Both-Column Fractures 273

A B

Redup lication of
iliac wing outline

.y ...........
. ....... . \ '\
I . \
i \ i
i ! i
! I
I
i I
I
a
,'\ i
i b j

Fig. 14.13A-C Both-column fracture with iliac component ex-
tending to the crest. A Antero-posterior radiograph, a diagram,
B obturator-oblique radiograph, b diagram, C iliac-oblique
radiograph, c diagram. The fracture line which detaches the
posterior column starts at the lower part of the greater sciatic
notch and has a spiral configuration. In B, there is no spur sign
but the fracture line above the acetabulum is clear and the lines
tangential to the two parts of the iliac wing are separated and
angulated. The split in the anterior column is extra-articular. It
will be recalled that the anterior horn of the articular surface
reaches one centimeter below the notch in the anterior border of
the acetabulum and therefore any fracture line cutting the col-
umn below this point is extra-articular
274 Associated Both-Column Fractures

Fig. 14.13C
CT Study of Associated Both-Column Fractures 275

Postero-5uperior
fragments
I
{--'._.Y
-, - I r '..... __ .
- I
I
(
\-, i
\ i
\ \
\ \
\
\
a \ b .,
"'

Fig. 14.14A-C. Both-column fracture with iliac component ex- fragment of the acetabulum. The fracture line at the level of the
tending to the crest. A Antero-posterior radiograph, a diagram, root of the superior pubic ramus is extra-articular. The greater
B obturator-oblique radiograph, b diagram, C iliac-oblique sciatic notch is cut in its middle part
radiograph, c diagram. There is an isolated postero-superior
276 Associated Both-Column Fractures

Fig. 14.14C
CT Study of Associated Both-Column Fractures 277

Tria ngular segment of iliac wing

/
.
/ "-......."'"yI . Fracture line
I . in anterior column
. I
l .
\ \
.,
;
i ,---
I
a i i

Fig. 14.15A-C. Both-column fracture in which the iliac compo- extra-articular fracture line at the root of the superior pubic
nent bifurcates and isolates a large triangular wing fragment. ramus. The bifurcated iliac wing fracture complex forms a dou-
A Antero-posterior radiograph, a diagram, B obturator-oblique ble spur sign in B
radiograph, b diagram, C iliac-oblique radiograph. There is an
278 Associated Both-Column Fractures

Fig. 14.1SC
CT Study of Associated Both-Column Fractures 279

A B

Reduplication and
separation at iliac
wing fracture
Outline of
segment of iliac
wing left in place

b
(~
I

Fig. 14.16A- D. Both-column fracture with iliac component ar- tion. There is kinking of the iliac wing fracture outline in A. In
rested 2 cm from the iliac crest. A Antero-posterior radiograph, B the splitting of the lines tangential to the surfaces of the frac-
a diagram, B obturator-oblique radiograph, b diagram, C iliac- tured iliac wing is clear. The fracture of the anterior column is
oblique radiograph, c diagram, D scheme of fracture configura- extra-articular
280 Associated Both-Column Fractures

Crest apparent ly intact

Fig. 14.16C,D
CT Study of Associated Both-Column Fractures 281

.,.-._./
I
/-.,,,),
.
i '\
\ \ I
\ I
\ i
a b

Fig. 14.17 A-C. Both-column fracture with iliac component ex- radiograph. The posterior column is detached at the bottom of
tending to the crest. A Antero-posterior radiograph, a diagram, the greater sciatic notch. The curved image is long. There are
B iliac-oblique radiograph, b diagram, C obturator-oblique several fracture lines across the anterior wall
282 Associated Both-Column Fractures

Fig. 14.17C
CT Study of Associated Both-Column Fractures 283

Curved image

Fragment of
Pos terio r wall /"--\.L-,iIAl<l"__-.J
I"
i
/.r.~..\_ i
I
. t·.~ \
\ . \
\
\ \
\ \
I \
a j b

Fig. 14.18A-D. Both-column fracture with iliac component ex-
tending to the anterior superior iliac spine. A Antero-posterior
radiograph, a diagram, B obturator-oblique radiograph,
b diagram, C iliac-oblique view, D scheme of fracture configura-
tion. An intra-articular fracture of the anterior column extends
to the anterior border at the notch. There is a characteristic up-
wardly concave fracture line of the body of the pubis. A segment
of the pelvic brim is detached with the posterior column; the
posterior wall fracture is not shown in the scheme
284 AssocIated Both-Column Fractures

Fig. 14.18C,D
CT Study of Associated Both-Column Fractures 285

A-E. Different types of curved image

RADIOLOGICAL
DRY BONE APPEARANCE
Bone surface producing posterior
quarter of radiological pelvic brim

;P.
' Radiological

--
pelvic brim

Projection of
fracture line
separating the posterior
column

Type I

Anatomical pelvic brim

Anatomical
pelvic brim

Fig. 14.19. Scheme explaining sources of curved

images (see text)
286 Associated Both-Column Fractures

Spur sign

""
/ - ',"",
- \
\ '1
- -, j i
nr-,*",-,.", 'i i
i i
..... /
\,. ,- '
-, /
a b i

Fig. 14.20A-C. Both-column fracture with iliac component ex- the anterior column is intra-articular and reaches the notch on
tending to the interspinous notch. A Antero-posterior radio- the anterior border of the acetabulum. The posterior column is
graph, a diagram, B obturator-oblique radiograph, b diagram, detached at the level of the angle of the greater sciatic notch
C iliac-oblique radiograph, c diagram. One fracture line across
 CT Study of Associated Both-Column Fractures 287

\
\
\
i
i
i
c
i

Fig. 14.20C
288 Associated Both-Column Fractures

Small segment of
roof inplace

. /,
:--...---. \
\
/
/
'-- . I
a "- b I
\

Fig. 14.21A-D. Both-column fracture in which the iliac compo-
nent extends to the psoas groove. A Antero-posterior radio-
graph, a diagram, B obturator-oblique radiograph, b diagram,
C iliac-oblique radiograph, c diagram, D scheme of fracture
configuration. The fracture line detaching the posterior column
starts near the angle of the sciatic notch. From here a line
crosses the roof horizontally leaving one small segment on the
iliac wing fragment, seen in A and B. This fracture is transitional
between T-shaped and both-column fracture groups. We have in-
cluded it in the latter because the fracture component which
separates the anterior column stems from the retro-acetabular
surface
CT Study of Associated Both-Column Fractures 289

A-.--..
i \ \
. ~. i \
\
V
. \
.
.I I
( I
' .\ .i
c \ I

Fig. 14.21C,D
290 Associated Both-Column Fractures

i
Plane of section of
bone seen in
obturator-ob lique view

Posterior inferior
iliac spine bearing
a part of the iliac
articular su rface
of the sacro-iliac joint
Greater
sciatic notch

Fig. 14.23A
 CT Study of Associated Both-Column Fractures 291

Anterior border
of greater sciatic notch
/~'r-'-"
I! \. Break in anterior wall -
of acetabulum
i
i
\
b i c

Fig. 14.23A-C. Both-column fracture with iliac component ex-
tending to crest and with involvement of the sacro-iliac joint.
A Antero-posterior radiograph, a diagram, 8 obturator-oblique
radiograph, b diagram, C iliac-oblique radiograph, c diagram.
The fracture line reaches the intermediate level of the sacro-iliac
joint and separates in one fragment the lower part of the ar-
ticular surface of the sacro-iliac joint, the superior border of the
greater sciatic notch, and the upper one-third of the margin of
the notch. This is suggested in A and confirmed in C. The frag-
ment is seen in 8 to be limited inferiorly by the profile of the
greater sciatic notch. The fracture of the anterior column is in-
tra-articular. See Fig. 14.3 for scheme
292 Associated Both-Column Fractures

A B

Fig. 14.24A-C. Both-column fracture with iliac component ex-
tending to the crest and involving the sacro-iliac joint. A Ante-
ro-posterior radiograph, a diagram, B obturator-oblique radio-
graph, b diagram, C iliac-oblique radiograph, c diagram. The
configuration is similar to that described in Fig. 14.23 and
shown in the scheme in Fig. 14.3. The detached fragment has
rotated through 90° and is less well seen on the oblique views.
In A the displaced sacro-iliac fragment is seen end-on. In C a
characteristic fragment is seen protruding into the pelvic area. In
B the margin of the greater sciatic notch is very prominent exter-
nally due to the rotation of the fragment
 CT Study of Associated Both-Column Fractures 293

Fragm ent bea ring

art icu lar surface of
sa~ro- ili ac joint

i
i
i
i
i
Fig. 14.24C c
294 Associated Both-Column Fractures

I
Fig. 14.25A-C. Both-column fracture with iliac component ex-
."'-1-') tending to the crest. It is apparently comminuted. A Antero-
J .
.
I
\
. posterior radiograph, a diagram, B obturator-oblique radio-
i \ graph, C iliac-oblique radiograph. There are multiple cortical
\ fracture lines and there is an intra-articular fracture of the
\ anterior column. The anterior border of the greater sciatic notch
i forms a separate fragment. Reconstruction is feasible and the
a i prognosis good
CT Study of Associated Both-Column Fractures 295

Curved image ~'---fr~

\ .- .........
·......I·
L/ \ .
y.1 \
I
i
/
i
a i

Fig. 14.2SC Fig. 14.26. Both-column fracture with iliac component extend-
ing to the crest. A Antero-posterior radiograph, a diagram. The
posterior column fracture line splits the ischial spine and there
is an extra-articular fracture near the root of the superior pubic
ramus
296 Associated Both-Column Fractures

i
".-.-\- ...../
......
I ".
; ,'
\ \.
\
\
\
a
\
\

Fig. 14.27 A - C Both-column fracture with iliac component ex-
tending to the crest. A Antero-posterior radiograph, a diagram,
B iliac-oblique radiograph, b diagram, C obturator-oblique ra-
diograph. This fracture is distinguished by multiple lines. Note
(1) the split of the ischial spine and another fracture line (2)
which divides the posterior column in the coronal plane. This
separates a fragment carrying the quadrilateral surface and
another carrying the retro-acetabular surface and the articular
surface. The rotation of the posterior column and its jamming
in this position led to considerable operative difficulties
(Fig. 22.76). See Fig. 14.2D for scheme
CT Study of Associated Both-Column Fractures 297

Fig. 14.27C
298 Associated Both-Column Fractures

Fracture line
\r-- '-"'-\
-~
I .
. \
/ - .'--....../ ._\
'- ___ . / I
\-I - -
/ I
(
A' /
i i
i
"\ ! /
\ /
a /
b

Fig.14.28A-C. Both-column fracture with iliac component
extending to the interspinous notch. A Antero-posterior radio-
graph, a diagram, B obturator-oblique radiograph, b diagram,
C iliac-oblique radiograph, c diagram. In this particular case, an
additional fracture of the anterior column at the level of the
anterior wall isolates a fragment of the column. This is in two
segments separated by a break in the coronal plane. The one ly-
ing more medially carries part of the internal iliac fossa with a
part of the pelvic brim; the outer one carries articular surface
and part of the external iliac fossa. See Fig. 14.8 for clinical
specimen and scheme
 CT Study of Associated Both-Column Fractures 299

,
'\ Fracture line

. " .\.
( . - .- ' \

.i \
Y i
I
i
(
........
\
c \
\
\ \

Fig. 14.28C
300 Associated Both-Column Fractures

Anterior border .
of acetabulum -"--
' _ ...l>IJlr .
/-, /
/ ~(
I 1\
l False appearance of
\ an ischial spine which
\ is real ly here
\
a \ b

Fig. 14.29A-C. Both-column fracture with iliac component ex-
tending to the interspinous notch. A Antero-posterior radio-
graph, a diagram, B obturator-oblique radiograph, b diagram,
C iliac-oblique radiograph, c diagram. The anterior column ap-
pears broken in three places: at the level of the notch of the
anterior wall, at the root of the superior pubic ramus, and at the
angle of the pubis. Everything is somewhat confused by the fact
that the posterior column fragment takes with it a segment of
the pelvic brim isolating a short fragment of the anterior border
of the acetabulum seen in a and c. See Fig. 14.4 for scheme and
Sect. 14.1.2.C for text
CT Study of Associated Both-Column Fractures 301

Separated .
portion of the ).
anterior wall ( \
I
i
i
j
j
c /

Fig. 14.29C
302 Associated Both-Column Fractures

A B

Fig. 14.30A-D. Associated extended and comminuted anterior
column and posterior hemitransverse fractures (not a both-col-
umn fracture). A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, c diagram,
D scheme of fracture configuration. The hemitransverse element
splits the ischial spine. In C a portion of articular surface
remaining normally attached to the upper segment of the
posterior column is clearly visible. An undisplaced split is direct-
ed towards the sacro-iliac joint
CT Study of Associated Both-Column Fractures 303

Fig. 14.30C,D
 304 Associated Both-Column Fractures

A B

Fig. 14.31A- I. Both-column fracture extending to the iliac

crest. A - C The three standard views. In B the spur sign is pres-
ent but its inferior tip is hidden behind the femoral neck. The
anterior column is split by a secondary fracture line through the
root of the pubis and the anterior wall. The posterior column is
in one piece. There is a Y fracture line through the iliac wing.
D-I CT sections. D The two fractures of the iliac wing repre-
senting the two branches of the Y fracture line are visualised.
E The fracture line separating the anterior column is frontally
disposed; the tip of the posterior column appears. F Through
the roof the fracture line in fact runs obliquely backwards and
outwards at about 30°, but the opposite rotation of the column
gives the fracture line an apparent orientation of 45° backwards
and outwards. G The fracture line skirts the anterior wall
through the cotyloid fossa. H There is some comminution at the
level of the fracture of the root of the pubis. I The ischio-pubic
notch is cut through its anterior part. The intact part of the wing
c is indicated by the letter S (for spur sign) on E and F
 CT Study of Associated Both-Column Fractures 305

o E

F G

Fig. 14.31D-I
306 Associated Both-Column Fractures

B
A

Fig. 14.32A-H. Both-column fracture extending to the iliac

crest. A - C The three standard views. In A all the fractures are
visible. In B the spur sign is shadowed by a posterior wall frag-
ment. C The posterior column is detached by a fracture line
starting from the inferior part of the greater sciatic notch; conse-
quently the spur sign is long and the curved image on A is also
long. D - H CT sections. D The fracture line through the iliac
wing is roughly coronal and there is an isolated fragment of the
internal iliac fossa. E The fracture line detaching the anterior
column is coronal. F The posterior wall fragment (P) appears;
the fracture line dividing the roof runs slightly obliquely back-
wards and outwards. The spur sign (8) is still visible. G The
postero-superior wall fragment is large and separated by an only
slightly displaced, nearly antero-posterior fracture line. H The
anterior fracture line is comminuted at the level of the anterior
wall, of which a part is attached to the posterior column frag-
c ment. The posterior wall decreases in size
 CT Study of Associated Both-Column Fractures 307

D E

F G

H Fig. 14.32D-H
 308 Associated Both-Column Fractures

A D

Fig. 14.33A-L. Both-column fracture extending to the anterior

interspinous notch with an associated posterior wall fragment.
A-C The three standard views. The spur sign is very apparent
in D, and the anterior column is split through the anterior wall.
A postero-superior wall fragment is seen, taking with it a
posterior part of the roof; the fragment still accompanies the
head but is separated from the lateral part of the roof and
breaks the middle part of the posterior lip. In C the posterior
column seems to be in one piece, the upper border of the greater
sciatic notch having been spared. D - I CT sections. D, E The
spur sign (S) can be seen. A roughly coronal fracture has
detached the anterior column. In E the tip of the posterior wall
appears inside the spur. In F the postero-superior wall (P) ap-
pears independent and separated from the quadrilateral surface
by a sagittal fracture line. In the following sections G and I the
posterior wall is always separated from the posterior column by
an antero-posterior fracture line and appears slightly displaced
from its original position. The posterior column takes with it
more and more of the posterior wall. J - L 3-D reconstruction
c of the fracture
 CT Study of Associated Both-Column Fractures 309

D E

F G

Fig. 14.33D-I
 310 Associated Both-Column Fractures

J K

L Fig. 14.33J-L
 CT Study of Associated Both-Column Fractures 311

A c
Fig. 14.34. A Obturator-oblique view of a both-column fracture or wall fragment takes with it a narrow strip of articular car-
extending to the crest, with an additional small posterior wall tilage and much more of the retro-cotyloid cortical surface
fragment. B, C 1\vo CT sections demonstrating that the posteri-
 312 Associated Both-Column Fractures

A B

Fig. 14.35A-H. Both-column fracture extending to the anterior

superior iliac spine. A-C The three standard views. In A and
B the anterior column fragment is divided through the root of
the pubis. The spur sign is obvious in B. In C the posterior col-
umn is in one piece and there is a supplementary cortical frag-
ment (F) from the internal iliac fossa. D - H CT sections. D A
coronal fracture line divides the iliac wing. A cortical piece of
bone from the inner aspect of the wing (F) is isolated. The tips
of the two separated columns appear. In E, due to the special
configuration of this fracture, the iliac wing fracture is still visi-
ble (its anterior part) in front of the spur, and the upper parts
of the two columns are inside them. F The roof is divided by a
coronal or slightly backwards and outwards directed fracture
line. The spur is still there. In G and H the two columns have
rotated around the head and the whole cotyloid fossa is attached
to the posterior column. S, spur sign; nv, iliac wing; AC,
c anterior column; PC, posterior column
 CT Study of Associated Both-Column Fractures 313

D E

F G

H Fig. 14.35D - H
15 Transitional and Extra-articular Forms
Our study of the anatomical and radiological
features of fractures of the acetabulum has been
conducted over three decades. Personal observa-
tions have led to the isolation, from a previously
somewhat amorphous group, of the posterior wall
fracture with marginal impaction, isolated frac-
tures of the anterior and posterior columns, the as-
sociated anterior column and posterior hemitrans-
verse fracture, the associated posterior column and
posterior wall fracture and the both-column frac-
ture. As the series grew, we were led progressively to
accepting all the types of injury as members of a
spectrum; the interrelationship between them is
logical and based on observable mechanical fac-
tors, namely, the description and site of application
of the injury force, the anatomy of the bones con-
cerned, and the composite mechanical properties
of these. The exact analysis in anyone fracture
would be difficult to achieve, of course, but the
broad outline is not too complicated to grasp.
15.1 Transitional Forms
Between the different categories exist the transi-
tional forms, that is to say, those presenting the
typical lesion of one group but having features of
another neighbouring group.
The transitional forms are fundamental in link-
ing together the types we have described. They are
shown schematically in the diagrams in Figs. 15.1
and 15.2. Figure 15.1 shows how, in the different
varieties of fractures of the walls or of the columns,
any segment of the articular surface can be de-
tached. Thus the main divisions of this surface
(postero-inferior, posterior, postero-superior, an-
tero-superior, and anterior - impossible to define
exactly anatomically) can be detached either by
fractures of the wall or by fractures of the columns.
Figure 15.2 outlines the principal transitional forms
between the different categories of fracture of the
acetabulum which we have distinguished, and
relates these to the gross acetabular architecture
and not just to the articular surface.
Thus, referring to Fig. 15.2, between the posteri-
or wall (A) and the posterior column (D) we have
observed a posterior fragment taking with it the
anterior border of the greater sciatic notch (B) as
well as a partial upper fracture of the posterior col-
umn (C). Again, between the pure posterior wall
fracture (E) and the associated transverse and
posterior fractures (G), we have several cases in
which a fracture of the posterior wall with
posterior dislocation of the femoral head becomes
associated with the outline of a transverse fracture,
the latter failing to reach the anterior column and
the innominate line (F), presumably on account of
dissipation of the fracture energy. These are asso-
ciated posterior wall and incomplete transverse
fractures. Comparably, between the pure anterior
column fracture (H) and the anterior column frac-
ture associated with a posterior hemitransverse ele-
ment (1), we have 18 observations in which the
posterior hemitransverse element is incomplete and
ceases anterior to the dense trabeculae at the
anterior border of the greater sciatic notch (I).
Bridging the both-column fractures in which the
iliac component is directed towards the crest of the
ilium or those with the iliac fracture line going to
the anterior border, we find two transitional forms.
There are those in which there exists at the begin-
ning of the fracture line, rising towards the crest, a
second split which joins the interspinous notch;
similarly, when the fracture in the iliac wing reaches
the anterior border there may arise from this main
fracture line at the level of the internal iliac fossa
a split which ascends towards the crest, sometimes
stopping just before reaching it.
Between the T-shaped fractures with a poste-
riorly directed trans-ischial stem and transverse
fractures associated with a posterior wall fracture,
we have seen numerous transitional types. In these,
the stem fracture line becomes more and more obli-
316 Transitional and Extra-articular Forms

Table 15.1

Simple fractures

Posterior wall 223 23.72%

Typical one-fragment 64 6.800/0
Typical multifragment 47 5.00%
Typical with marginal impaction 39 4.14%
Postero-superior one-fragment 7 0.74%
Postero-superior multifragment 10 1.06%
Postero-superior and marginal impaction 9 0.95%
Postero-inferior 5 0.53%
Extended posterior wall 13 1.38%
Posterior wall and incomplete transverse 29 3.08%
Posterior column 30 3.19%
Typical 15 1.59%
Detaching teardrop 7 0.74%
Partial superior 2 0.21%
Partial inferior 3 0.31 %
Epiphyseal separation 3 0.31%
Anterior wall 18 1.91 %
Typical 12 1.27%
Anterior wall and anterior column 6 0.62%
Anterior column 42 4.46%
Extending to the iliac crest 11 1.17%
Extending to interspinous notch 5 0.53%
Extending to psoas gutter or lower 19 2.02%
Incomplete 7 0.74%
Transverse 70 7.45%
Trans-tectal 13 1.38%
Juxta -tectal 36 3.82%
Infra-tectal 20 2.12%
Incomplete 0.10%
Total (Simple fractures) 383 40.74%

que with respect to the ischium, ranging from being
tangent to the posterior margin of the obturator
foramen, to detaching only the upper pole of the
ischial tuberosity.
Nevertheless every example will approximate to
one of the basic types described; herein lies the
diagnostic value of orderly understanding and clas-
sification.
As was explained in Chap. 4, in order to simplify
the classification we have subsumed three groups of
fractures into larger groups, on the grounds that
they require the same surgical approach. Never-
theless, their existence cannot be disputed and the
fan of lesions around the acetabulum is thus
perfect. So far there has only been one case that
was impossible to integrate into one of our catego-
ries; it is reported on as case 16 in Chap. 30.
15.2 Extra-articular Forms
It remains to consider the types of fractures which
at first sight appear to involve the acetabulum but
on close scrutiny do not involve the articular sur-
face itself. They cannot be dismissed as ordinary
fractures of the pelvic ring for they are intimately
concerned with the mechanical architecture of the
acetabulum. Good-quality radiography is impor-
tant if articular damage is to be excluded with cer-
tainty (Figs. 15.3 -15.5). CT scanning is also of
great help in assessing the integrity of the articular
surface.
Extra-articular Forms 317

Table 15.1 (continued)

Associated fractures

T-shaped 66 7.02070
Vertical stem 41 4.36%
Anterior-oblique stem 12 1.27%
Posterior-oblique stem 13 1.38%
Transverse and posterior 183 19.46%
Infra-tectal and posterior 16 1.70%
Juxta-tectal and posterior 52 5.53070
Trans-tectal and posterior 23 2.44%
Infra-tectal and postero-superior 1 0.10%
Juxta-tectal and postero-superior 5 0.53%
Transtectal and postero-superior 4 0.42%
Infra-tectal and posterior with marginal impaction 14 1.48%
Juxta-tectal and posterior with marginal impaction 20 2.12%
T-shaped and posterior 30 3.19%
Transverse and postero-superior extended to the crest 18 1.91 %
Posterior column and posterior wall 32 3.40%
Posterior column and posterior wall 21 2.22%
Posterior column and postero-superior wall 10 1.06%
Posterior column and postero-inferior wall 0.10%
Anterior column and posterior hemitransverse 62 6.58%
Anterior wall and posterior hemitransverse 14 1.48%
Anterior wall and incomplete posterior hemitransverse 4 0.42%
Low anterior column and posterior hemitransverse 6 0.63%
Intermediate anterior column and posterior hemitransverse 5 0.53%
High anterior column and posterior hemitransverse 19 2.02%
Anterior column and incomplete posterior hemitransverse 11 1.17%
Anterior column and posterior epiphyseal separation 2 0.20%
Epiphyseal separation of anterior column and posterior hemitransverse 0.10%
Both-column 213 22.65%
Extending to iliac crest 176 18.72%
Extending to anterior superior iliac spine 12 1.27%
Extending to interspinous notch 19 2.02%
Extending to psoas gutter 6 0.63%
Impossible to classify 0.10%
Total (Associated fractures) 557 59.26%
Total 940 100.00%
318 Transitional and Extra -articular Forms

Fig. lS.lA- D. Scheme demonstrating manner in which any seg-

ment of the articular surface of the acetabulum can be isolated
A by pairs of fracture lines (see text)
Extra-articular Forms 319

~ .. -

Fig. IS.lA-J. Scheme of interrelationship between transitional

forms of fracture (see text)
320 Transitional and Extra-articular Forms

A B

Fig. 15.3 A, B. Vertical anterior fracture of the obturator ring. articular. Close inspection of A shows the teardrop to be in nor-
A Antero-posterior radiograph, a diagram, B obturator-oblique mal relation to the ilia-ischial line, and the anterior border de-
radiograph, b diagram. Despite the appearance in A, it was con- scribes its normal inflexions without interruption. In B the
firmed at operation that the fracture complex was entirely extra- anterior wall is intact
 Extra-articular Forms 321

8 C
Fig. 15.4. A Antero-posterior radiograph, a diagram, 8 obtura- acetabulum all the landmarks of the acetabular region are in-
tor-oblique radiograph. C iliac-oblique radiograph. The fracture tact. The relationship of teardrop to the ilio-ischialline appears
complex is entirely extra-articular. Between the vertical anterior distorted because the central fragment bearing the acetabulum
fracture and the unusual horizontal fracture above the is rotated
322 Transitional and Extra-articular Forms

Fig_ 15_5_ A Antero-posterior radiograph, B obturator-oblique

radiograph, C iliac-oblique radiograph. A vertical fracture line
crosses the iliac fossa cutting the pelvic brim in its middle part
and rejoins the greater sciatic notch after transversing the
quadrilateral area. The other component is the disjunction of
the pubic symphysis. The acetabulum is not fractured
c
16 Associated Injuries
16.1 Injury of the Femoral Head
It is not surprising that there is a considerable in-
cidence of damage to the femoral head in view of
the magnitude of the force necessary to produce a
fracture of the acetabulum and the fact that it is the
femoral head which transmits this force. At opera-
tion, we have had the opportunity of inspecting the
articular surface and have noted the occurrence of
femoral head lesions, recognising that most of the
time only part of the head has been accessible to
view, since we hesitate to reproduce a posterior
dislocation, for instance, merely in order to inspect
the head. The use of extensile approaches allows a
complete view of the head much more regularly.
The lesions can be considered under three
headings: macroscopic, vascular, and microscopic
or molecular; these can be associated.
16.1.1 Macroscopic Injury
This may be detected from examination of the pre-
operative plain radiographs. Tomography and CT
scanning are also very useful for identifying these
lesions pre-operatively.
The figures reported here are probably not ab-
solutely accurate as the head has not been com-
pletely examined in all the operated cases.
We have seen:
eighteen (1.9070) partial or segmental fractures
of the femoral head detaching an infero-medial
segment of variable size which remains in the
acetabulum while the head dislocates posterior-
ly. This fragment may remain connected to the
round ligament and to a synovial flap from the
inferior part of the femoral neck, and it may
bleed considerably at the time of operation. If
it is not perfectly reduced, it heals to the antero-
inferior aspect of the neck, leading to later
troubles in hip mobility such as inability to per-
form pure flexion, the flexion being inexorably
accompanied by some abduction;
twenty cases (2.12%) of localised subsidence of
the femoral head in its supero-Iateral quadrant,
appearing as a localised depression of the sub-
chondral bone 0.5 - 1.5 cm in width on the plain
views. This is produced at the time of trauma by
localised contact between the head and an intact
part of the innominate bone. It is exceptional in
both-column fractures. Such localised impac-
tions or subsidences remain stable, i.e. they do
not increase in size after reduction of the
dislocation;
twenty-five cases (2.66%) of femoral head wear
noted pre-operatively among the delayed cases.
This wear involves the supero-medial or supero-
lateral quadrant of the femoral head and its ex-
tent is variable. It is the result of friction be-
tween the head and, usually, a non-reduced
transverse fracture. In moving, the head wears
against the bone. At operation, what distin-
guishes this condition from a fracture of the
head is that no fragment is found in the joint,
the bone dust being included in the fibrous
tissue which fills up the gap between the head
and the fragments of the acetabular articular
surface. If the head and the fracture are not
reduced, even under traction, this area of wear
increases in size. Visualisation of the extent of
this wear on the three standard views, tomo-
grams and CT is essential in order to plan what
to do at the time of the reconstruction, and to
answer the question of whether the femoral
head should be kept or replaced by an immedi-
ate arthroplasty.
Besides these lesions which are possible to identify
radiographically, we note from our operative re-
ports that in 48 cases (5.10%) there were lesions of
the femoral head which were impossible to detect
on plain radiographs or CT. They consisted of nine
superficial abrasions of the articular cartilage and
324
39 cases of more widespread destruction extending
into the sub-chondral bone to varying degrees. We
repeat that these figures are sure to be under-
estimated, for the reasons already mentioned.
In cases of posterior subluxation, the round
ligament is always detached. It may detach a small
cartilaginous or osteo-cartilaginous plaque which
one should not omit to excise.
16.1.2 Vascular Injury
We have not performed enough contrast radiogra-
phy of the vessels of the femoral head to be able to
give significant data. We suspect that injury is a less
frequent occurrence than that seen after pure
dislocations of the hip as there is a smaller in-
cidence of avascular necrosis, and it is possible to
perform a late reduction of a dislocation associated
with a posterior fracture without this likelihood.
16.1.3 Molecular Injury
STEWART and MILFORD (1954) believe that the
direct shock and the recoil suffered by the femoral
head could produce intracellular disorganisation
sufficient to bring about cellular death of the bone.
It is interesting to note that in nine of the cases
cited by this author, and in two of ours which were
followed by avascular necrosis, there was only a
horizontal transverse fracture, slightly displaced
and probably without significant capsular damage.
16.2 Capsular Injury
Capsular tears must occur in all cases of posterior
dislocation of the head and are frequent in central
dislocations. In the case of posterior dislocation,
the capsular rent may present with particular fea-
tures:
(a) In a pure fracture-separation it may happen
that the one or several posterior fragments are
separated and driven by the femoral head into
the substance of the posterior gluteal muscles.
The capsule is torn along the acetabular poste-
rior margin, at least along the length of the
fragment. The posterior fragment can remain
in continuity with a shred of capsule as has oc-
curred in some incarcerated examples.
Associated Injuries
(b) In cases of marginal impaction, the separated
fracture fragments remain attached to the cap-
sule which is torn at the upper and lower limits
of the fragments. The head passes across the
posterior wall without creating further capsular
damage of particular importance.
(c) In central dislocations, the capsular lesions are
variable and difficult to systematise. In one
case of transverse fracture we noted the total
detachment of the capsule at the level of the
ischio-pubic fragment which perhaps explained
the failure of an attempted closed reduction. In
the majority of cases, the capsular attachment
to the columns is preserved.
16.3 Vascular Injury
16.3.1 Acetabular Wall
During surgical approach to the retro-acetabular
area, in cases of transverse or both-column frac-
tures with central dislocation of the hip, we often
encounter vascular bundles of varying size pressed
against the bone to which they adhere; they nourish
the posterior wall. In contrast, during the posterior
approach for fractures of the acetabulum accompa-
nying a posterior dislocation, whether or not these
are reduced, we do not see these vessels, a fact
which seems to prove that their tearing or their
thrombosis occurred earlier at the time of trauma.
Such vascular lesions of the acetabulum, aggravat-
ed subsequently by periosteal injury during the
reduction, may explain a certain number of post-
operative necroses of the posterior wall. Similar
arguments apply to anterior wall injuries. The in-
cidence of proven osteonecrosis is very low but
nevertheless there is every indication that care
should be taken to avoid unnecessary periosteal
damage.
16.3.2 Pelvic Vessels
The only vessel damage we have seen with any sig-
nificant regularity is to the gluteal arteries: they
may be torn or stretched by the displacement of the
posterior column. When the fracture detaches the
angle of the greater sciatic notch, it menaces the
superior gluteal vessels. In ten instances we have
found the gluteal vessels trapped in the upper part
of the fracture of the posterior column at the angle
of the greater sciatic notch. On several occasions
Other Pelvic Injuries
the liberation of the gluteal bundle produced a
severe haemorrhage because after having been
damaged at the time of the injury, their entrapment
had secured a temporary haemostasis. The dif-
ficulties of haemostasis of the gluteal vessels and
the techniques recommended will be described
along with the surgical approaches in Chap. 20.
We have not encountered any other more signifi-
cant vascular injury among the cases we have
treated.
16.3.3 Retro-peritoneal Haematoma
As stated above, we have never found a serious in-
jury of a major intrapelvic vessel. The extensive
muscular tearing together with the fracture lines
through wide areas of spongy bone appear suffi-
cient to explain the large haematomata which are
always found at operation; sometimes these extend
superiorly as far as the perinephric region. They
elevate extensively the parietal peritoneum of the il-
iac fossa and abdominal wall. The volume of hae-
matoma is difficult to measure but it may be con-
siderable, requiring replacement of a large blood
volume. Although a constant finding, the clinical
consequence is highly variable; on a few occasions,
it has led to an exploration of the abdomen because
of suspected visceral damage. CT scanning allows
an appreciation of the volume and extent of
haematomata.
16.4 Other Pelvic Injuries
Apart from the bilateral acetabular fractures, other
lesions of the pelvic ring are common and were
encountered in 151 out of our 940 cases, i.e. in
16.066,10.
The most frequent injury was an associated an-
terior vertical fracture across the obturator fora-
men. These fractures appear to us to be mechani-
cally independent of the fracture of the
acetabulum. Of 98 such fractures, 23 were on the
same side as the acetabular fracture and 75 on the
opposite side. There were four cases of bilateral as-
sociated anterior fracture. The 98 acetabular frac-
tures (10.53%) with which the anterior vertical
fractures were associated were classified as follows:
4 posterior column;
- 5 anterior column;
325
15 transverse;
18 T-shaped;
25 associated transverse and posterior wall;
3 associated posterior column and posterior
wall;
5 associated anterior column and posterior
hemitransverse;
23 both-column.
Thus, anterior vertical fractures have been met in
all types of acetabular fracture except fracture of
the posterior wall.
Disruption of the pubic symphysis was an asso-
ciated lesion in 39 cases (4.14%) of our series of
acetabular fractures, which were classified as fol-
lows:
1 anterior wall;
1 anterior column;
11 transverse;
4 T-shaped;
17 associated transverse and posterior wall;
1 associated anterior column and posterior
hemitransverse;
4 both-column.
The disruption was the only associated pelvic le-
sion in 13 cases and combined with one or more
other extra-acetabular pelvic injuries in 26 cases.
Associated sacro-iliac lesions are of variable
significance, varying from simple radiological
anterior gaping of one or both sacro-iliac joints to
pure complete disruption of the joint or fracture-
dislocation of the sacro-iliac joint, in which a pos-
tero-superior piece of the iliac wing remains in its
normal place and is the key for the reconstruction.
Simple anterior gaping of the sacro-iliac joint
has been recorded in 35 instances and corresponds
to an incomplete dislocation of the joint with intact
posterior ligaments. However, the gaping of the
anterior aspect of the sacro-iliac joint was hardly
visible on the radiograph in some cases, and the
figures we give are undoubtedly underestimated.
Through an ilio-femoral or ilio-inguinal approach
such an incomplete disruption cannot escape atten-
tion and is easily dealt with, but through the other
approaches it is impossible or difficult to evaluate.
Nowadays, the indispensable CT scan allows ex-
cellent study of the sacro-iliac joints and no type of
lesion can be overlooked. In the future, therefore,
the data about the incidence of associated sacro-
iliac lesions will be more accurate.
Major sacro-iliac lesions, pure dislocation or
fracture-dislocation, seen in 24 cases, necessitated
326
A
B sions in relation to the type of acetabular frac-
Fig. 16.1A,B. Both-column fracture with an iliac component
extending to the anterior border and accompanied by a frac-
ture-dislocation of the sacro-iliac joint. A Antero-posterior ra-
diograph, B scheme. The whole auricular surface is detached
and only the postero-superior segment of the wing remains at-
tached to the sacrum. The fracture detaching the auricular area
is quite separate from the both-column complex. It is unique in
our experience in this latter respect, usually the sacro-iliac in-
jury component is part of the both-column fracture if present.
It was a poliomyelitic hip
Associated Injuries
reconstruction performed through a special
approach or through the approach chosen for the
acetabular repair. In 17 instances these were on the
same side as the acetabular fractures and in 7 they
were on the opposite side. The types of acetabular
fracture with which they were associated were:
1 anterior wall;
3 anterior column;
5 transverse;
5 T-shaped;
5 associated transverse and posterior wall;
5 both-column.
In six cases there was an iliac wing fracture totally
independent of the acetabular fracture.
In 24 instances afracture of the sacrum, now so
perfectly visualised on CT, was present, displaced
or not, passing through or beside of the sacral fora-
mina. The sacral fracture was associated with the
following fracture types:
1 posterior column;
2 anterior column;
2 transverse;
3 T-shaped;
4 associated transverse and posterior;
4 associated anterior and posterior hemitrans-
verse;
8 both-column.
In our series of 940 acetabular fractures, there were
associated pelvic injuries in 151 cases (16.06%):
in 76 (8%) cases there was one associated pelvic
injury;
in 56 cases (6.2%) there were two associated
pelvic injuries;
in 19 cases (2.02 0/0) there were three associated
pelvic injuries. Table 16.1 shows the pelvic le-
ture.
16.5 Associated Hip Injuries
In 11 of our cases (1.16%) a fracture of the corre-
sponding upper femur was present; six of these
were true femoral neck fractures and five per-
trochanteric fractures. In five other instances, when
the patients were admitted each had an associated
segmental fracture of the femoral head and a
femoral neck fracture. The latter was produced dur-
ing attempts to reduce the dislocated femoral head
conservatively (see Sect. 19.1.2). We have seen four
cases of acetabular fractures associated with a pure
posterior dislocation of the opposite hip.
Table 16.1 2,>
13
Post. column Ant. wall Ant. column Transverse T-shaped Transverse + Post. col. + Ant. col. + Both-column Total O.
~
(1)
post. post. hemitrans. P-
:r::
Ipsilateral anterior vertical 3 2 5 5 3 3 2 23 06'
fracture 2.
-
~
Contralateral anterior 3 10 13 22 5 21 75 ::J.
vertical fracture &l
Disruption of pubic 11 4 17 4 39
symphysis
Gaping of ipsilateral sacro- 4 3 7 8 23
iliac joint
Gaping of contralateral 2 2 5 10
sacro-iliac joint
Gaping of both sacro-iliac 2 2
joints
Complete sacro-iliac 2 5 3 2 5 17
disruption
Contralateral sacro-iliac 2 3 7
disruption
Iliac wing fracture 2 2 1 6
Sacral fracture 1 2 2 3 4 4 8 24
Total injuries 5 2 15 39 37 65 3 11 49 226
Total patients 3 10 23 24 45 3 8 34 151

w
tv
....,
328
16.6 Other Skeletal Injuries
A fracture of the acetabulum is always associated
with major trauma, and commonly other skeletal
or visceral injuries are associated. In 425 cases out
of 940 (45.2UJo) the patients presented with one or
more other fractures. Of these, 27 involved the ip-
silateral femoral shaft. Such an association is some-
times responsible for a posterior dislocation and its
associated posterior wall fracture being missed.
16.7 Urinary Tract Injuries
These are related to the associated fracture of the
anterior pelvis and not directly to a fracture of the
acetabulum. We have only encountered nine rup-
tures of the bladder, eight ruptures of the urethra,
and two lesions of the urethra and the bladder in
association with lesions of the anterior arch of the
pelvis which were independent of the fracture of
the acetabulum. Every case is individual and there
are instances when the urinary and osseous injuries
should be repaired at the same operation. For ex-
ample, if there is a rupture of the bladder, this must
be treated urgently and if the condition of the pa-
tient is good enough, the acetabulum can be dealt
Associated Injuries
with at the same time. Repair of the membranous
urethra is often delayed for about 5 days; this is
also a suitable time for operation on the acetabu-
lum.
16.8 Other Visceral Injuries
In 51 cases in our series (5.4UJo) one or more visceral
injuries other than of the urinary tract were present
and a result of the major trauma. These injuries
can involve any constituent element of the thorax
or abdomen.
16.9 Associated Skull Trauma
In 193 cases (20.53UJo) there was a history of skull
trauma. It was accompanied by a short period of
coma in 162 cases (17.23 UJo) and a long period of
coma in 31 cases (3.2 UJo).
16.10 Sciatic Nerve Injuries
These will be studied later in Sect. 18.2.3.
17 Distribution of the Clinical Series

17.1 Distribution According to Age 17.2 Distribution According to Sex

The ages of our 940 patients ranged from 12 to 85 Males preponderate greatly. Of the 940 cases, 617
years, two-thirds of the patients being within 20 to (65.6%) were males and 323 (34.6%) were females
60 years old. Four patients were more than 80 years (Table 17.2).
old and each of these had at least an anterior col-
umn lesion.
Table 17.1 shows the distribution, according to
age, of the various types of acetabular fracture. 17.3 Distribution According
Concerning the existence of any particular age-type to Time After Injury
relationship there is little to say except to mention
the frequency of anterior column type fractures
Patients with fractures of the acetabulum did not
and both-column fractures in patients more than 60
all reach us early (i.e. during the first 3 weeks) and
years old, both fracture types resulting from the
therefore the following will be considered in discus-
same mechanism of injury: a trochanteric blow
sing the indications for surgical management:
(60-69 years, 39 cases out of 95: 40070; 70-79
years, 12 cases out of 21: 57070; over 80 years: 4 the time from injury to the patients' admission
cases out of 4). to our centre, all 940 cases (Table 17.3);

Table 17.1

Type of fracture Age of patient (years) Total

10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89

Posterior wall 3 45 58 38 46 29 4 223

Posterior column 6 5 4 8 3 3 30
Anterior wall 2 5 4 5 2 18
Anterior column 6 5 6 6 8 5 2 39
Transverse 9 20 21 11 6 3 70
T-shaped 9 21 14 7 12 2 66
Transverse and posterior wall 11 42 50 38 25 16 183
Posterior column and 3 7 9 7 5 32
posterior wall
Anterior fracture and poste- 3 14 9 16 8 10 4 65
rior hemitransverse fracture
Both-column 14 68 45 32 29 19 5 213
Not classified
Total 62 223 216 171 148 95 21 4 940
330 Distribution of the Clinical Series

Table 17.2 the time from injury to operation in the 849

cases in which we operated (Table 17.4);
Type of fracture Sex of patient Total the conditions when first seen of the fractures
Male Female
operated on from 21 days to 4 months after in-
jury, evaluated from plain radiographs and op-
Posterior wall 171 52 223 erative charts (Table 17.5);
Posterior column 23 7 30 the conditions of the fractures operated on
Anterior wall 13 5 18
more than 4 months after injury (Table 17.6).
Anterior column 18 21 39
Transverse 40 30 70
T-shaped 38 28 66
Transverse and posterior wall 129 54 183
Posterior column and 27 5 32
posterior wall
Anterior fracture and poste- 49 16 65
rior hemitransverse fracture
Both-column 109 104 213
Not classified
Total 617 323 940

Table 17.3

Type of fracture Time post-injury (all patients)

0-21 days 21 days to >4 months Total

4 months

Posterior wall 141 36 46 223

Posterior column 14 8 8 30
Anterior wall 11 6 1 18
Anterior column 30 4 5 39
Transverse 36 18 16 70
T-shaped 37 17 12 66
Transverse and posterior wall 121 44 18 183
Posterior column and posterior wall 22 5 5 32
Anterior column and posterior hemitransverse 56 7 2 65
Both-column 170 27 16 213
Not classified
Total 638 172 130 940

Table 17.4

Type of fracture Time post-injury (surgical patients)

0-21 days 21 days to >4 months Total

4 months

Posterior wall 129 35 44 208

Posterior column 13 8 7 28
Anterior wall 9 6 1 16
Anterior column 22 3 5 30
Transverse 21 14 17 52
T-shaped 30 17 12 59
Transverse and posterior wall 117 42 16 175
Posterior column and posterior wall 20 5 5 30
Anterior column and posterior hemitransverse 50 4 55
Both-column 158 23 15 196
Total 569 157 123 849
Table 17.5
~
S.
cr
Type of fracture Patients operated on 21 days to 4 months post-injury
S.
0'
Fracture lines Malunion Posterior wall and Posterior wall and Intra-articular Malunion and Head necrosis ::>
Total
still recognisable persistent posterior posterior dislocation incarcerated nonunion and malunion ~
(')

on X-ray dislocation + segmental fracture fragment ..,0

of the head e:
::>
(JQ

Posterior wall 10 2 20 2 35 0
o-j
Posterior column 4 3 8 S'
Anterior wall 4 6 '">-
Anterior column 2 1 3 :::>
Transverse 10 2 2 14 '"..,
T-shaped 14 3 17 2-
Transverse and posterior wall 29 6 5 42
'<
..,'"
Posterior column and posterior wall 1 1 1 5
Anterior column and posterior 2 2 4
hemitransverse
Both-column 22 23
Total 98 22 26 2 3 5 157

w
w
Table 17.6 w
w
N

Type of fracture Patients operated on beyond 4 months post-injury

Malunion Nonunion Malunion Intra-articular Osteonecrosis Malunion and Posterior Stiffness Wear of Intra- Total
and incarcerated of the head necrosis wall and and the head articular
nonunion fragment neglected hetero- and mal- hardware
posterior topic union
disloca- ossifi-
tion cation

Posterior wall 4 14 23 44
Posterior column 3 3 7
Anterior wall
Anterior column 3 1 5
Transverse 10 4 1 2 17
T-shaped 9 3 12
Transverse and posterior 6 3 3 3 16
wall
Posterior column and 2 2 5
posterior wall
Anterior column and
posterior hemitransverse
Both-column 5 6 2 15
Total 40 12 17 17 28 3 3 123

~
....
5'
~
0'
::>
o....,
5-
(1)

Q
s'
(S'
~
til
!:I
~.
18 Clinical Presentation
Clinical findings relating to the fracture of the
acetabulum itself offer little further of note other
than what is evident on the radiographs. A few
points may be dealt with summarily.
In order not to miss a fracture of the acetabulum
which may be the single injury or a component of
multiple trauma, it is necessary always to have
radiographs of all hips involved in trauma or sus-
pected of having been injured. Ensure that
radiography of the pelvis and of the hips is per-
formed in all cases of multiple trauma, especially if
there is a fracture of the femur on the same side.
Visceral injury of any type deflects attention from
the hips.
18.1 Clinical Findings
These are dominated by the direction of the dis-
placement of the femoral head.
18.1.1 Posterior Dislocation
The classical signs are often lacking, for the dislo-
cation may be atypical and the limb does not adopt
the most frequent typical attitude of flexion-adduc-
tion; conversely, on occasions we have thought to
have found, in view of flexion-adduction and slight
internal rotation, a pure posterior dislocation
whereas the radiograph displayed an associated
transverse and posterior wall fracture with
posterior dislocation of the hip (see Fig. 12.2).
18.1.2 Central Dislocation
The deformity of the lower limb is highly variable
and not particularly helpful. Loss of active move-
ment may be total and the shortening is not
marked.
We agree with others that a significant sign may
be the sinking-in of the greater trochanter, which
external rotation alone cannot explain. We have
found no value in palpation and internal examina-
tion of the pelvis, an observation made by COT-
TALORDA (1922). It may be useful to assess the
relative positions of the antero-superior iliac
spines, for central dislocation will often be accom-
panied, especially in both-column fractures, with
eversion of the iliac wing; the anterior superior iliac
spine on the traumatised side is therefore displaced
laterally in comparison with the opposite side and
lies a little inferior.
18.2 Early Complications
Immediate clinical examination is necessary for
baselines against which to monitor the patient for
hypovolaemic shock, retro-peritoneal haematoma,
intra-abdominal visceral complication and sciatic
nerve paresis. Careful examination should be made
for bruising which may indicate the site of action
of the fracturing force and can form a characteris-
tic collection of clear fluid over the trochanter, as
described originally by MOREL-LAVALLE (1891).
18.2.1 lfaumatic Shock
Intensive treatment may be necessary. The cause of
shock relates to the severity of the trauma and to
haemorrhage from the fracture site and often from
other visceral lesions. The respective part played by
the various elements producing the shock may be
difficult to apportion.
18.2.2 Retro-peritoneal Haematoma
This always occurs to some degree due to the loose
tissue in which it can extend easily. It results in a
334
haemorrhagic syndrome of variable systemic effect
and, on local examination in a puffiness or even a
spasm of the parietal muscles of the lumbar region
and sometimes of the anterior abdominal wall,
which should be assessed at frequent intervals dur-
ing the first few hours. On occasion a laparotomy
has been necessary on account of a possible intra-
abdominal lesion requiring haemostasis.
During the following days the retro-peritoneal
haematoma may be responsible for paralytic ileus
or sub-acute obstruction, which has in some cases
necessitated naso-gastric aspiration. It may be the
cause of a fever of otherwise unknown origin which
delays operative intervention for the fracture of the
acetabulum.
Retro-peritoneal haematoma, present in all frac-
tures of the pelvis or the acetabulum, but of which
the extent is variable, is responsible for numerous
laparotomies without positive findings or in which
only a little blood is found. Knowledge of this clini-
cal feature and the frequency of the haematoma
justifies strict and repeated observation, but in
general one should avoid unnecessary abdominal
exploration.
Nowadays, diagnostic peritoneal lavage, ultraso-
nography and CT scanning help a great deal in
identifying an intra-abdominal haemorrhage and
the extent of a retro-peritoneal haematoma.
18.2.3 Pre-operative Sciatic Nerve Injury
These are of considerable frequency and presented
in 115 cases out of 940, that is 12.2070. Their
medicolegal importance must not be forgotten and
their discovery before any operative intervention or
treatment should be recorded. Regrettably, it must
be confessed that in 20 of our cases of post-opera-
tive paralysis of the sciatic nerve (13 recent and 7
late cases) the nerve had not been examined proper-
Table IS.1
Direction of dislocation Extent of sciatic nerve lesion
Complete sciatic Lateral popliteal
palsy palsy
Posterior 6 20
Central 4 5
Anterior
None
Total 11 26
Clinical Presentation
ly pre-operatively; undoubtedly a proportion of
these were due to the original injury.
Sciatic palsy is not always obvious. Naturally,
the injured person tries to move his leg as little as
possible and often has not noted his weakness. It
therefore is necessary to assess in turn all the
muscles of the leg together with the cutaneous sen-
sitivity, sometimes with firmness, in the case of a
reluctant patient with head injury or local pain, by
stressing to him the value of this examination with
which he must cooperate in an effort to show
voluntary movement. Pain can almost completely
obliterate voluntary movement but with patience
and perseverance one can always arrive at a valid
appraisal.
The extent of the sciatic damage is variable and
comprised in our series:
total sciatic paralysis (11 cases);
external popliteal component paralysis, com-
plete or partial (26 cases);
internal popliteal component alone (2 cases);
patchy and involving both popliteal territories
(72 cases);
hypersensitivity or hyposensitivity in sciatic skin
distribution (4 cases).
Sciatic nerve involvement is most likely with a
posterior dislocation of the head (66 cases of the
115 pre-operative palsies and of 360 posterior
dislocations, i.e. 18.3 %), occurs sometimes with a
central dislocation (42 cases of the 115 pre-opera-
tive palsies and of 403 central dislocations, i.e.
10.4%), more rarely with an anterior dislocation (5
cases out of 115 pre-operative palsies and of 93
anterior dislocations, i.e. 5.4%), and even in two
cases in which there was no dislocation at the time
of the initial radiograph. Table 18.1 shows the type
of sciatic pre-operative paralysis with respect to the
type of dislocation. Table 18.2 indicates the paral-
yses associated with the various types of injury; it
Total % of total number
of dislocations
Medial popliteal Patchy Sensory
palsy lesion only
37 3 66 18
31 42 10
3 5 5
2 3
2 72 4 115
Early Complications
Table 18.2
Extent of sciatic Type of fracture
nerve lesion
Posterior Posterior Transverse T-shaped
wall column
Complete 2 4
sciatic palsy
Lateral 10 3 10
popliteal palsy
Medial popliteal - 2
palsy
Patchy lesion 14 3 3 7 24
Sensory only
Total 27 7 7 9 39
is clear that lesions of the posterior column are the
predominant cause.
Out of 940 cases of fractures of the acetabulum
there were 468 with lesions involving predominant-
ly the posterior column, and among the latter there
were 79 sciatic nerve injuries (16.8 0/0). It is striking
that markedly displaced transverse, T-shaped and
both-column fractures are not accompanied by
such a high incidence, no doubt due to the predom-
inantly medial direction of the femoral head dis-
placement. Note, however, that in the 65 examples
of anterior fractures and posterior hemitransverse
fractures there were six sciatic nerve injuries (9.2%)
despite the fact that the dislocation was markedly
anterior. The 213 both-column fractures were ac-
companied by 14 pre-operative sciatic palsies, an
incidence of only 6.6% for such complex fractures,
the large majority of them (11 out of 14) being
patchy lesions.
At operation, various macroscopic lesions of the
sciatic nerve have been demonstrated in 20 in-
stances (of 115, i.e. 17%) namely:
partial division of the nerve in its outer aspect
in one case;
penetration of the nerve by a sharp bone frag-
ment (4 cases) which lacerated it irregularly and
led to a patchy distribution paralysis in two
cases and a lateral popliteal palsy in two others;
stretching of the nerve across a bony bridge
formed by a posterior fragment: the nerve ex-
hibited a flattened, thinned appearance over
1-3 cm (6 cases) and this caused a total sciatic
335
Total
Transverse Posterior Anterior fractures Both-
and poste- column and and posterior column
rior wall posterior hemitransverse
wall fractures
3 11
26
6 4 11 72
2 4
6 6 14 115
paralysis in one case, and external popliteal
paralysis in two cases, and an irregularly distrib-
uted paralysis in three cases;
the inclusion of almost the whole nerve in the
fracture line of a transverse lesion. The displace-
ment was not apparent on the radiographs
(Fig. 18.1). Evidently, the nerve trunk had been
trapped in the fracture line which had displaced
momentarily at the time of the trauma. The sci-
atic paralysis was total (1 case);
a large haematoma extending into the sheath of
the nerve in four cases and associated with one
external popliteal paralysis, one total and two
patchy distributions.
In three cases the nerve was entrapped within the
fracture line detaching the posterior column and
this was accompanied by one total palsy and two
patchy lesions. However, in several operations, we
have seen the sciatic nerve included in the fracture
line which detached the posterior column and with
it the angle of the greater sciatic notch, but without
the nerve suffering any apparent damage from this
interposition.
In four instances, the nerve was already included
in the callus formation, corresponding with one
total palsy and three patchy lesions.
Apart from these 20 cases of macroscopic le-
sions of the sciatic nerve found at operation, in all
the other cases of sciatic paralysis explored at
operations, the nerve had throughout its visible
length a completely normal appearance. We do not
know the cause of paralysis in such cases. Different
336 Clinical Presentation

authors have debated the problem and all hypothe-

Fig.18.1. Undisplaced transverse fracture complicated by a
complete sciatic palsy. Antero·posterior radiograph. The sciatic
nerve was found to be trapped in the fracture
sise a stretching of the nerve by the femoral head or
by fragments which were pushed backwards in
front of this at the time of the injury. However, the
neurological lesion does not locate necessarily to
such a level of damage and DECOULX (1961) in
particular has drawn attention to the possibility of
stretching of the lumbar plexus. The roots at the
uppermost level, and therefore the longest and the
most vertical, are the most vulnerable in this
respect. Lumbar roots four and five contribute to
the external popliteal component which is most
often affected.
We reported (JUDET and LETOURNEL 1966)
confirmation of this lumbar plexus lesion. Evi-
dence based on certain of the electromyographic
studies performed more than three months after
the trauma have revealed indisputable injury to the
nerves innervating quadriceps muscles. This injury
associating both the sciatic and the femoral nerve
can result only from appropriate root damage.
Electromyographic examination of 18 of the 115
cases of pre-operative paralysis showed that seven
exhibited undoubted abnormality of quadriceps in-
nervation while 11 quadriceps were elec-
tromyographically normal. The seven cases of
quadriceps electromyographic abnormality were
related to one lateral popliteal palsy and six patchy
lesions.
Out of the 115 cases of pre-operative sciatic
nerve paralysis, 97 were available to follow-up;

Table 18.3

Extent of sciatic Neurological outcome Functional outcome

nerve lesion
Complete Significant Partial No recovery Normal Residual Significant Orthopaedic Purely
recovery recovery recovery but normal deficit appliance sensory
activity deficit

Complete 2 7 5 3
sciatic palsy
Lateral 9 7 4 2 10 9 2
popliteal palsy
Medial popliteal 2 2
palsy
Patchy lesion 6 21 29 8 27 9 7 5
Sensory only 2 2

Total 19 30 41 2 26 36 14 8 8
Early Complications
however, in five of these recovery was still in pro-
gress, so 92 cases were available for review more
than 2 years after surgery. Their results are reported
in Table 18.3.
Concerning neurological outcome, recovery was:
complete in 19 cases;
significant in 30 cases;
partial in 41 cases;
nil in 2 cases.
As to functional recovery, the final outcome has
been as follows:
26 patients have normal function;
36 patients have some residual deficit not affect-
ing normal life;
14 have a significant disability;
8 patients wear an orthosis;
8 patients have only sensory deficits.
Overall, 49 patients out of 92 (53.2070) had com-
plete or significant recovery, and the function was
normal or good enough to allow a normal life in 52
cases out of 92 (67.3%).
Sciatic nerve recovery is usually complete or
stable after 2 years, but in at least one case the
recovery continued up to 3 years.
In three cases we performed neurolysis of the
nerve trunk to try to facilitate its recovery.
To try to compensate for residual disability we
performed:
5 transfers of the posterior tibial tendon;
2 lengthenings of the Achilles' tendon;
2 combined arthrodesis of the subtalar and me-
diotarsal joints.
18.2.4 Morel-Lavalle Lesion
The trochanteric region must always be examined
since on many occasions we have seen evidence of
a blow resulting in local loss of sensation, abrasion,
bruises, and haematoma formation. It is often later
during the next days that a soft fluctuant area
develops and which carries a risk of becoming in-
fected. If this is allowed to occur, the safety of early
operation is reduced because it is in the operation
zone. Aspiration or decompression should be per-
formed urgently as soon as the condition is ap-
parent or at least at the time of surgery. Typically,
the fluid responsible is clear and pale yellowish. It
was present and significant in 23 cases. Out of the
275 cases with trochanteric blows, it occurred in
8.3%. This reported incidence seems to us to be
337
lower than in reality, and in fact some localised
Morel-Lavalle lesions of small volume have failed
to be recorded in the operative charts.
18.2.5 Intra-articular Incarceration
of Bone Fragments
We have included this as an early complication
although strictly it is a possible feature of any frac-
ture of the acetabulum. The index of suspicion
should always be high, for the late complications
from having failed to recognise the lesion are seri-
ous. The diagnosis is essentially radiological but
not always easy on standard views; fortunately it
cannot be missed on a complete CT examination.
Incarceration of bone fragments has been
recognised and described as an early complication
for a long time. CAUCHOIX and TRUCHET (1951)
and URIST (1948) and D'AUBIGNE (1968) have each
reported a number of cases. Like these authors we
have always insisted that it constitutes an absolute
indication for surgical intervention.
Old, missed examples of incarceration have not
to our knowledge been described. They form part
of our consideration of lesions seen late (Chaps.
8, 9), and we have reported these elsewhere (JUDET
and LElDURNEL 1968). VACHER (1966) collected
our first nine cases which occurred among 305 re-
cent fractures.
Out of 940 cases of fractured acetabula we have
recorded 26 cases of incarceration recognised early
(2.8%). In fact, the cases taken into account here
are those in which there was a fragment of reason-
able size, possible to identify on plain radiographs.
The incidence of tiny pieces removed during clean-
ing of the joint at operation has not been recorded
faithfully. Now the routine use of CT scanning
makes it possible to recognise all incarcerated
fragments whatever their size, which should dem-
onstrate that the real incidence is higher. We con-
sider here the cases recognised early.
(a) Source. According to the type of fracture, we
have seen incarceration of fragments after:
12 posterior wall fractures;
1 posterior column fracture;
1 anterior column fracture;
4 transverse fractures;
4 associated transverse and posterior wall
fractures;
2 associated posterior wall and posterior col-
umn fractures;
338 Clinical Presentation

1 associated anterior column and posterior

hemitransverse fractures.
In 19 instances there was a single incarcerated
fragment, but in six cases there were two pieces
and in one case there were three fragments.
CAUCHOIX and TRUCHET (1951) considered
that fragments came from the cotyloid fossa, but
in most of our examples they were from the
posterior wall of the acetabulum as demonstrat-
ed by radiography and at surgery. Radiographs
showed a corresponding defect in the posterior
border of the acetabulum or from the outer part
of the roof and these were, of course, proved at
the time of exploration of the joint. We have also
removed tiny pieces of bone attached to the
proximal end of the round ligament, which were
undoubtedly detached from the central border
of the articular acetabular surface.
(b) Mechanism. There are two possibilities:
(i) Primary incarceration. One or several frag-
ments, perhaps pedunculated to the hip
capsule, enter the articular space at the mo-
ment of injury, probably at the same time as
Fig. 18.2. Primary incarceration. Antero-posterior radiograph.
dislocation. The head meets and traps them
Two attempts at closed reduction failed. The fragment always
during attempts at reduction. If the frag- lay in the outline of acetabulum
ment is small and free it may lodge in the
cotyloid fossa, being pumped into the ante-
rior part thereof. If the fragment is very
large it will be impossible to reduce the head the posterior wall together with a large segment
(Fig. 18.2). Smaller fragments may allow of the posterior acetabular surface. As a rule,
reduction but some displacement of the the fragments are of small volume; they may be
femoral head will persist, showing as in- free and correspond to a small sector of the ar-
congruence. ticular surface, 1- 2 cm2 in area, with irregular
(ii) Secondary incarceration. This is a conse- contours, and associated with a bed of cancel-
quence of the reduction itself. The frag- lous bone of varying thickness. They may com-
ment, maintaining a capsular flap pedicle, is prise a segment of the posterior border of the
drawn into the joint during the process of acetabulum 1 - 2 cm long and about 0.5 cm
reduction and lodges between the head and wide, bearing a narrow brim of articular car-
the roof. The proof that this incarceration tilage 3 - 5 mm wide.
occurs at the same time as the reduction is The capsular connections of the fragment
seen on radiography: on the views taken determine to some extent what will happen to
with the head dislocated, the little fragment, them. It is essential to distinguish free frag-
which will soon be incarcerated, is perfectly ments, deprived of all connection, which can, if
visible in a different position and lies extra- they are of small volume, be chased into the
articularly (Fig. 18.4). The volume of the joint, and lodge in the cotyloid fossa generally
fragments, being small, does not impede re- in its antero-inferior part. One finds these a
duction but an incongruence persists, with long way from their origin. Pedunculated frag-
downward displacement of the head. The ments which remain attached to a fragment of
reduction is stable. the posterior capsule maintain a blood supply.
(c) Size and attachment. Both of these features are For this reason, neglected, they can actually
variable. In one case of primary incarceration, grow in size. The attachment limits the distance
the fragment was enormous (Fig. 18.2). It em- that these fragments can penetrate the joint.
bodied the outer half of the articular surface of This rarely exceeds 1 - 2 cm from the articular
 Early Complications 339

Fig. 18.3. Secondary incarceration after reduction of a posterior

dislocation with a posterior wall fracture. Antero-posterior
radiograph

margin, and we have always retrieved them from

between the head and the roof.
In a few instances, the pedunculated frag-
ment is associated with one or two free frag-
ments; it is always necessary to check this pos-
sibility when removing the one which appears
obvious.
(d) Diagnosis. An incarcerated fragment is recog-
nised early in two different circumstances.
Firstly, a dislocation may be irreducible due to
a large radiographically obvious fragment
(Fig. 18.2). Secondly, and more commonly, the
incarceration is only recognised on the control
films taken after reduction of a posterior dislo-
cation, even if the hip appears stable and has an
apparently good range of movement. The ante-
ro-posterior radiograph is usually adequate to
display the fragment but it is essential to exam-
ine the oblique views. On the antero-posterior
view, the head, reduced into its cavity, is related
to a small fragment interposed between it and
the roof, which often carries a sector of articu-

8
Fig.18.4A,B. Incarceration following reduction of posterior
dislocation of the hip associated with a T-shaped fracture and
posterior wall fracture. A Antero-posterior radiograph before
reduction showing a little fragment outside the acetabulum.
A
B Antero-posterior radiograph after reduction
 340 Clinical Presentation

A B

Fig. 18.5 A-F. Associated infra-tectal transverse and posterior

wall fracture. A-C The three standard views. The incarceration
may be suspected in the obturator-oblique view (8). D-F CT
sections. D and E demonstrate the comminution of the trans-
verse fracture line. F shows the incarceration of a fragment not
c bearing articular surface
 Early Complications
D E
F Fig. 18.5D-F
lar surface lying adjacent to that of the roof.
The superior joint space is significantly widened
and the head is eccentric inferiorly. The frag-
ment may be seen extruded either towards the
external border of the roof or partly in the
cotyloid fossa (Figs. 18.3 and 18.4). The oblique
views confirm the site of the fragments.
The best way to recognise and analyse an in-
carcerated fragment, however, is the CT scan
(Figs. 18.5 and 18.6). Sections just below the
roof or through the cotyloid fossa will demon-
strate the incarceration.
A theoretical differential diagnosis in regard
to these incarcerated fragments is the recogni-
tion in the cotyloid fossa, on plain radiographs
or CT, of a little fragment which on operation
proves in fact to be attached to the round liga-
ment. This fragment comes from either the area
of insertion of the ligament around the cotyloid
fossa or from the fovea of the femoral head. At-
tached to the ligament, it may keep its blood
supply and increase in size with time, and like
the true incarcerated fragment, it needs to be
removed.
The same also happens in pure dislocation
of the femoral head, and then, if the fragment
341
is very tiny and the head is perfectly reduced, it
can be left in situ, on condition that the patient
is followed regularly for 2 or 3 years so as to be
certain that the fragment does not increase in
size, bringing the risk that it will push the head
outwards. If this happens, surgical removal is
indicated.
Figure 5.20 shows an exceptional case dem-
onstrated by CT. A free fragment of the ar-
ticular surface is incarcerated in the fracture
line of a segmental fracture of the femoral head
after reduction of the posterior dislocation.
18.2.6 Other Types of Palsies
Pre-operatively we were able to identify, in addition
to the sciatic nerve palsies:
1 case of palsy of the femoral nerve in a patient
presenting with a trans-tectal transverse fracture
associated with an anterior dislocation. He died
2 years later from other causes without having
been followed up;
1 case of palsy of the cauda equina in a patient
with a both-column fracture associated with a
sacral fracture; this is a recent case.
 342 Clinical Presentation

A B

Fig. 18.6A-F. Associated transverse and posterior wall fracture

after reduction of the posterior dislocation. The intra-articular
penetration of bone fragments has caused an enlargement of the
joint space and an inferior and lateral displacement of the
femoral head. D- F CT sections. D Section just below the roof,
already showing the fragments incarcerated in the joint. E The
fragments are more clearly seen; the articular surface of the in-
carcerated posterior fragment is orientated forwards. The tip of
the head is visible. F The head is displaced posteriorly and con-
c gruent with an impacted part of the posterior wall
 Special Cases
D E
F
18.3 Special Cases
18.3.1 Children
We have had the opportunity to study five cases
with an associated epiphyseal displacement in the
region of the Y-shaped cartilage. In three cases
there was a separation of the posterior column
(Fig. 6.10), in another a separation of the posterior
column associated with a fracture of the anterior
column (Fig. 13.14), and the fifth case comprised a
complete separation of the elements of the Y car-
tilage (Fig. 18.7). This last case was in fact operated
on elsewhere and is not included in our series.
The poor prognosis is a strong indication for
reduction and fixation surgically. In order to avoid
an epiphysiodesis, any plate or screw crossing or
spanning the epiphyseal cartilage should be re-
moved later.
The radiographic diagnosis of the fracture is
helped by comparative views of the opposite hip.
18.3.2 Elderly Patients
We have operated on 103 fractures of the acetabu-
lum in patients more than 60 years old (83 were
60 - 69 years old, 18 were 70 - 79 years old, and 2
were aged 80 or more). We have found all types of
fracture but undoubtedly there is a higher incidence
of anterior column and both-column fractures. Prob-
ably, it relates to the most frequent cause, namely,
pedestrians being knocked down by a vehicle and
receiving a direct blow in the trochanteric region.
343
Fig. 18.6D-F
Treatment was almost always surgical and it was
only in relatively rare cases where the radiograph
showed a severe degree of osteoporosis that we
found the fracture so comminuted and so difficult
to fix solidly (the screws could not gain a solid
hold) that we were not able to get an exact reduc-
tion. As will be seen later, the late results of these
freshly operated fractures were pleasing.
18.3.3 Pathological Fractnres
We had one case of interest which was a pathologi-
cal pure low transverse fracture. A chondrosarcoma
was proved histologically during the course of try-
ing to treat a painful pseudarthrosis by a total hip
arthroplasty (Fig. 18.8).
In 17 cases significant osteopenia, not always
identified pre-operatively, led to some technical dif-
ficulties at the time of operation. The fractures in
these cases were:
1 posterior column;
3 anterior wall;
2 anterior column;
1 T-shaped;
3 transverse and posterior;
1 posterior wall and posterior column;
4 anterior column and posterior hemitransverse;
2 both-column. The long-term results in these
cases will be analysed later.
Operations performed in three pagetic bones did
not meet with any problems; even a painful non-
union was operated on and healed (see Fig. 12.6).
344 Clinical Presentation

·A

Fig. 18.7 A-Co Epiphyseal separation of Y-shaped cartilage.

A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph
r
Special Cases 345

Fig. 18.8A,B. Pathological juxta-tectal transverse fracture

(chondrosarcoma). A Antero-posterior radiograph, B iliac-
oblique radiograph
A
19 General Principles of Management of Acetabular Fractures
COTTAWRDA (1922) stated that "while modern in-
dustrial life gives us the means to inflict serious
fractures of the acetabulum, despite sporadic at-
tempts, the results of treatment have stayed as they
were at the time of Ambroise Pare'
By 1955, methods of treatment had without
doubt changed, but really, the results remained
roughly those which prevailed 30 years previously.
On the one hand, there was, we felt, a tendency to
overestimate the quality of results by conservative
management, and on the other, Robert JUDET was
considerably disappointed by his own results using
conservative methods; consequently we decided to
manage all fractures of the acetabulum with
displacement by open reduction. Thus from 1955 to
1980 our attitude did not change and our indica-
tions remained simple and absolute. The improve-
ment in functional result which we observed as a
consequence of this active approach, and the
steady increase in surgical management around the
world seemed to confirm that we had chosen the
best approach.
Since 1980, in accordance with the findings of a
review of our patients, we no longer advise strict
adherence to this rule of systematic surgical treat-
ment. Some fractures resulting in a small post-
traumatic incongruency may be left unoperated
upon and have an excellent long-term result. The
problem is to define what constitutes a "small" in-
congruency. That is the difficulty! Undoubtedly,
further studies are needed reporting the long-term
results related to residual post-traumatic incongru-
ency measured by the computer from the standard
views or, perhaps better, from the 3-D reconstruc-
tion. However, this work, relying as it does on the
proof of time, will take another decade or so to
complete. Other fractures, in which the acetabular
articular surface is totally separated in several
pieces, are sometimes by chance followed by a
restoration of the congruency between the dis-
placed femoral head and the various fragments of
the acetabulum. These happy cases may also be left
unoperated upon and have proved to have very
good long-term results.
To achieve perfect surgical reconstruction it ap-
peared early that it was necessary to study new
methods of surgical access and to adapt and im-
prove those which had been tried earlier. It became
obvious that poor surgical reductions were in most
instances due to failure to recognise the nature of
the lesion pre-operatively and therefore to choose a
correct surgical approach. The results of inade-
quate open reduction are comparable to those of
conservative management of displaced, unreduced
fractures, with the added risk of infection and all
other forms of post-operative complication; conse-
quently, unless operative management can be
shown to achieve a better result, the risks make it
unacceptable. It must, of course, be recognised that
many complications such as embolic phenomena
occur also with conservative management.
So, are there any indications for conservative
management?
19.1 Conservative Theatment
19.1.1 Indications
We treat acetabular fractures conservatively in the
following conditions:
(a) Medical contra-indications. It must be said that
there have been a number of patients for whom
we have regretted not having taken the risk of
an early reduction, given the difficulties im-
posed by the later problems of total hip replace-
ment for the post-traumatic osteoarthrosis
which developed. Nevertheless, it is obvious
that for reasons of age or infirmity, there will
be patients for whom the risks of operative
treatment would be too great.
(b) Pre-existing osteoarthrosis. A pre-existing os-
teoarthrosis precludes operation, although even
348
here, care should be taken to ensure that a rea-
sonable acetabulum is available for future total
hip replacement. Early-stage osteoarthritis is
not a contra-indication (Fig. 19.1), as even if
the osteoarthritis continues its evolution, a
perfectly restored acetabulum will facilitate the
total hip replacement when this becomes need-
ed and will improve its prognosis.
(c) Local infections. In a few cases we have been
forced to avoid operation in patients seen early,
but who had already been treated conservative-
ly elsewhere and developed complications, no-
tably from pin-track infections consequent to
skeletal traction to the lower end of femur, neck
of the femur, or the pelvis. While waiting for an
infection to heal, conservative methods of
treatment may be needed for the fractured ace-
tabulum.
(d) Osteopenia of the innominate bone. In our
opinion, osteopenia is the most important con-
tra-indication to operative treatment. Unfor-
tunately it is difficult to detect or measure and
it is not directly linked to age. Nevertheless,
each time we have found an osteopenic bone
the operation was extremely troublesome, as
the forceps could not hold the bone without ad-
ding more fragments and the screws never got
a good purchase. We have never achieved either
a satisfactory reduction or a reliable fixation in
these cases, and they are among our failures.
(e) Other contra-indications. From our experience
we also think that three kinds of acetabular
fractures must be treated conservatively:
naturally, all fractures without displace-
ment, which heal rapidly without sequelae;
fractures with only a small amount of
residual post-traumatic incongruency. We
know now that the very low transverse frac-
tures which involve only the horns of the
acetabular articular surface, even if they are
displaced, create little incongruency and
their long-term prognosis is good. The same
applies to the low anterior column fracture
which displaces 1 cm or 1.5 cm of the
anterior horn of the articular crescent. In
these fractures, if a careful check shows that
the femoral head is perfectly reduced and
stable, operation is not necessary. For a long
time we have operated upon all types of
posterior wall fractures, but we think now
that if we are sure from the CT examination
that there are no incarcerated fragments,
General Principles of Management of Acetabular Fractures
and if the posterior fragment is so small
that it would accept one or at the most two
screws, surgery is probably not necessary;
both-column fractures achieving a second-
ary congruence (see Sect. 14.1.5). Once this
secondary congruence has been demon-
strated on the standard views, CT scanning
and tomography, the fracture may be
treated conservatively. Healing will be ob-
tained in 5 - 6 weeks and the long-term
results, as already proved by D'AuBIGNE
and RIGAULT and in our own series, will be
very good. However, due to the narrowing
of the entry of the socket, a consequence of
shifting of the fragments, patients will
generally lose a large part of their external
rotation.
19.1.2 Methods
If we do not operate, it is because the hip is stable,
the post-traumatic incongruency is minimal, or sec-
ondary congruence is achieved (and we have never
observed secondary displacement in the cases
treated as described here). The aim is to keep the
things as they are, without attempting to reduce the
fracture or improve its reduction. Therefore:
patients are left in bed for 5 weeks;
they are allowed to mobilize their hip actively as
much as they want and can;
continuous passive motion is begun on the 3rd
or 4th day post-trauma, for several hours each
day, and is continued until they get up;
prophylactic anticoagulant therapy is given;
walking on crutches with progressive partial
weight-bearing is allowed from the 35th day.
We would like to stress the fact we never use any
kind of traction, and when a patient arrives with a
pin allowing traction our first attempt at treatment
is to remove the traction pin. Why do we take this
attitude? The reason is, we feel traction to be ab-
solutely unnecessary: it cannot keep a dislocated
femoral head securely reduced, it cannot achieve a
perfect reduction or at least maintain it until heal-
ing occurs. Traction is useless if the femoral head
or the fracture is stable, and it may be followed by
local problems, and, in the middle and long-term
above all by pain in the knee. I feel very sorry when
I think of all those wounded patients attached to
their bed by 5 - 6 kg of useless traction which gains
them nothing but discomfort. Nevertheless, al-
Conservative Treatment 349

though we have been waging war against it for 25 Table 19.2

years, traction still remains a very common feature
Time from injury to admission
of acetabular fracture treatment today. in conservatively treated cases
Manipulative reduction of a posterior disloca-
tion of the head recognised on radiography is, of 0-21 days 63 69.20/0
course, urgent whether or not it is associated with 21 days to 4 months 19 20.8%
a fracture of the acetabulum. The only contra- >4 months 9 9%
indication to manipulative reduction is the ex- Total 91
istence of a segmental fracture of the head of the
femur; the serious possibility exists of adding to
this a complete sub capital fracture of the femoral any problem. Fifteen cases were reviewed at an av-
neck. Commonly, the dislocation once reduced is erage follow-up of 7.2 years. All results were
stable and the patient can just be left in bed with clinically excellent; no patient developed signs of
a gentle external rotation of the hip. When the osteoarthritis. One transverse fracture, in spite of a
dislocation is impossible to reduce, or if it is very good clinical result, failed to unite in the
unstable, rather than attempt traction we would posterior half of the fracture line.
always prefer emergency operation. Delay in the Sixty-nine patients with displaced fractures were
reduction of the posterior dislocation prejudices also treated conservatively.
the circulation of the head of the femur and The reasons for choosing this treatment were
perhaps the sciatic nerve. not in all cases those outlined in Sect. 19.1.1. At the
Table 19.1 shows the results in our series after beginning of our experience, unidentified type of
manipulative reduction of posterior dislocation of fracture, late presentation of the case and patient
the femoral head. age were the most common factors which led us to
decide not to operate.
We chose the conservative method of treatment
19.1.3 Results for:
26 patients out of 76 from 1957 to 1960, i.e.
Conservative management was employed in 91
34070;
cases. Table 19.2 shows the time lapse from injury
27 patients out of 376 from 1960 to 1970, i.e.
to when the patient was admitted to our care.
7.2%;
Twenty-two patients had an undisplaced frac-
15 patients out of 186 from 1970 to 1980, i.e.
ture. These were of all types (except both-column,
8%;
of which we have never seen an undisplaced in-
23 patients out of 246 from 1980 to 1990, i.e.
stance), namely: two posterior wall, one anterior
9.3%.
wall, four anterior column, nine transverse, one T-
shaped, two associated transverse and posterior, Table 19.3 reports the post-traumatic congruency
and three associated anterior and posterior of 69 fractures treated conservatively according to
hemitransverse. All these fractures healed without anatomical fracture type. As 17 patients were lost

Table 19.1

Quality of closed reduction Type of acetabular fracture

Posterior Posterior T-shaped Transverse and Posterior column and Total

wall column posterior wall posterior wall

Stable 67 2 36 4 110
Unstable 2 2
Re-dislocation neglected 11 8 4 23
Additional femoral neck fracture 4 4
Additional incarcerated fragment 15 3 18
Posterior dislocation converted 11 6 17
to central dislocation
Total 99 2 58 14 174
 350 General Principles of Management of Acetabular Fractures

A B

c D
 Conservative Treatment 351

E F
Fig. 19.1A-F. Both-column fracture with pre-existing osteoar- appearance 18 months after surgery. The osteoarthritis has not
throsis operated on through the ilio-inguinal approach. A-C deteriorated further. The joint space seems to be larger than pre-
The three standard views pre-operatively. D - F Radiographic operatively. The joint is rated 5.6.6

Table 19.3

Type of acetabular fracture Post-traumatic head-roof congruency in Total

conservatively treated cases

Normal Nil Loss of

parallelism

Posterior wall 12 13
Posterior column 1 2
Anterior wall 1
Anterior column 5 1 6
Transverse 4 3 2 9
T-shaped 3 1 3 7
Transverse and posterior wall 5 1 6
Posterior column and posterior wall 1 1 2
Anterior fracture and posterior hemitransverse fracture 5 6
Total 26 19 7 52
352
to follow-up, only the results of 52 cases could be
assessed.
The results of the cases in which initially the
whole of the articular surface was displaced - the
both-column fractures - will be taken separately
here and discussed first. The 17 both-column frac-
tures achieved complete secondary congruence in
14 instances and partial congruence (with the walls
or with the roof) in the other three. Fifteen cases
were followed up over an average of 4.3 years. Of
the 13 cases with immediate complete secondary
congruence, 11 have a very good result (Figs. 19.2
and 19.3) and two a good result. Of the two cases
with partial congruence one has a very good and
the other a good result. These results and those of
various authors referred to above led us to recom-
mend conservative treatment of both-column frac-
tures achieving secondary congruence.
All the other cases treated conservatively had a
part of the acetabular articular surface left intact:
13 posterior wall, two posterior column, one
anterior wall, six anterior column, nine transverse,
seven T-shaped, six associated transverse and poste-
rior wall, two associated posterior wall and posteri-
or column, and six associated anterior and posteri-
or hemitransverse fractures. The post-traumatic
incongruency in these 52 fractures of nine different
types is reported in Table 19.3. In 26 cases on the
antero-posterior radiograph the head-roof con-
gruency was normal, in 19 cases it was nil, and in
7 we noticed a loss of parallelism of the upper joint
space.
Table 19.4 reports the long-term results of 35 pa-
tients (excluding cases of posterior wall and both-
column fractures) followed up for more than 2
years, with an average follow-up of 10.1 years. Of
the 26 cases achieving normal head-roof congruen-
cy post-trauma, 18 were followed up, and the find-
ings were 13 very good, two good, one fair and two
bad results.
Table 19.4
Long-term results in Post-traumatic head-roof
conservatively congruency
treated cases ---------
Normal Nil Loss of
parallelism
Very good 13 4 2
Good 2 4
Fair 1 2
Bad 2 3
Total 18 13 4 35
General Principles of Management of Acetabular Fractures
We have known for a long time that partial post-
traumatic incongruency can go along with a very
good result. The problem is that so far we are still
unable to measure precisely the amount of the in-
congruency, and that no statistics relating to the
long-term results in cases of partial incongruency
have yet been published.
Of the 19 cases in which the trauma resulted in
total incongruency between the femoral head and
the intact part of the acetabulum, follow-up clinical
results were available for 13 cases, of which four
were very good, four good, two fair and three bad.
The single, exceptional instance of a persistently
posteriorly dislocated hip, due to an initial extend-
ed posterior wall fracture and which achieved a
very good clinical result apart from the shortening,
is shown in Fig. 19.4. The patient still works as a
helicopter pilot. We agree now that a posterior wall
fracture of small volume with a perfectly stable hip
and no incarcerated fragment does not require
surgery.
How are these apparently paradoxical results to
be explained? Apart from the case just reported
above, in these cases the femoral head, in fracturing
the acetabulum, displaces with it a large part of the
articular surface and remains congruent to it. It
then loses its congruence with the intact part of the
acetabular articular surface, but after healing and
the achievement of malunion, enough of the dis-
placed part of the articular surface remains in
perfect contact with the head to guarantee a good
joint and good function, provided that the dis-
placement of the head and the major part of the
socket is sufficient to prevent the head from wear-
ing against the inner limit of the intact part of the
articular surface.
Figure 19.5 shows a case of a transverse fracture
which, despite normal head-roof congruency,
resulted in complete non-union of the fracture line,
and 6 years later osteoarthritis had set in.
Of the remaining seven cases with a loss of
parallelism of the upper joint space, four were
followed up over 2 years, showing two very good,
Total one good and one fair result.
To conclude the study of these results, it can al-
ready be said that we hope one day to know what
degree of post-traumatic incongruency is compati-
ble with an excellent long-term result, so that other
19 cases may be passed on for surgical treatment.
7
Meanwhile, it will be seen later on that many, if
4
5 not most, of these results of conservative treatment
are comparable to those achieved after imperfect or
incomplete surgical treatment, which seems logical.
Conservative Treatment 353

Fig. 19.2A-C. Both-column fracture treated conservatively on

account of local infection. One year later the radiographs il-
lustrate the concept of secondary congruence. A Antero- C
posterior radiograph, B obturator-oblique radiograph, C iliac-
oblique radiograph
 354 General Principles of Management of Acetabular Fractures

A 8

Fig. 19.3A- L. Both-column fracture extending to the iliac crest,

achieving secondary congruence and treated conservatively by
bed rest only. A-C The three standard views. D-G CT sections
demonstrating acceptable congruence between the head and the
walls of the acetabulum (G), slightly less satisfactory congru-
ence at the head-roof level. HI, 1\vo tomograms demonstrating
the slight persisting incongruence between head and roof. J - L
Radiographic appearance at 2 years. The case, seen at 3 weeks
after injury, was not operated upon as we considered satisfacto-
c ry secondary congruence had been achieved
 Conservative Treatment 355

D E

F G

G
H Fig. 19.3D - I
356 General Principles of Management of Acetabular Fractures

J K

L Fig. 19.3J - L
 Conservative Treatment 357

A
B

Fig. 19.4A-C. Interesting case of a persisting posterior disloca-

tion associated with a posterior wall fracture, accompanied by
a very good functional result which allows the patient to pilot
c a helicopter
 358
A B
Fig.. 19.5A,B. Pure juxta-tectal transverse fracture treated conservatively. A Radiograph on admission. B The painful non-union was
obvIOus 5 years later
19.2 Justification for Operative Treatment
It is far from ideal to leave an important load-bear-
ing joint in a state of subluxation or articular in-
congruence of which the long-term consequence
cannot be other than arthrosis. The relatively few
examples of long-lasting comfort and good func-
tion associated with hips radiologically dislocated
or subluxed do not compensate for the immense
number of disastrous results which have to be
salvaged by arthrodesis or by total arthroplasty, the
consequences of which remain uncertain in young
patients even today.
It is often difficult to compare the results of
conservative and surgical treatment, for the meth-
ods of assessment utilised by different authors are
so variable. Even if the grading of D' AUBIGNE is
used, there is always a subjective element which
cannot be judged by the reader. Further, the overall
grading of the range of motion of the hip is fallible
because, for example, total loss of external rotation
can exist in a hip graded 6 or perfect, whereas it is
not by any means a normal hip.
By our definition, a normal hip is one with a full
range of movement in all directions.
General Principles of Management of Acetabular Fractures
Most surgeons would agree that the best guaran-
tee of restoration to normal function of any injured
joint must relate to accurate reconstruction of the
surfaces. The hip is no exception. The reason for
the relatively infrequent use of open reduction of
acetabular fractures is simply that it is difficult to
gain access to the part; this has apparently justified
the rejection of the rule for all other joints, viz. that
anatomical restoration of the surfaces is essential.
The difficulty of access is due, in our view, to
failure to recognise the exact lesion, understanding
of which can be achieved only following study of
properly orientated radiographs, CT scans and a
knowledge of the possibilities. Certainly, our early
attempts were not so successful, and as in all
things, experience improves quality of performance
and results. The difficulty of access and the com-
plexity of the morphology of the pelvic bone are
poor excuses for abstaining from surgery. We
believe that all examples can be diagnosed
radiologically even without 3-D reconstruction and
the best approach chosen.
The aims of surgical management are, in sum-
mary:
exact restoration of the articular surface of the
acetabulum with respect to the femoral head,
Timing of Surgery
this restoration being a corollary of perfect ana-
tomical restoration of the bone as a whole;
solid internal fixation, thereby dispensing with
the need for post-operative splinting and allow-
ing early post-operative movements and the use
of continuous passive motion.
The surgery is often difficult, long and only
achieved with patience. It can be performed only if
the correct route of access is chosen and it cannot
be fully justified unless the desired perfection of
the osteosynthesis is achieved. Faults of surgical
reduction are errors to be put at the surgeon's door
rather than to be blamed on the method.
19.3 Indications for Operative Treatment
For recent fractures, the indications for operative
treatment are clear: All fractures of the acetabulum
with displacement, seen within the first 3 weeks
after the injury, should be operated upon, with the
exception of:
patients in whom there exists a strong medical
contra-indication;
patients with undoubtedly osteopenic iliac
bones;
patients presenting with three particular types
of acetabular fracture:
fractures without displacement;
fractures resulting in only slight incongru-
ence: small posterior wall, low infra-tectal
transverse, low anterior column;
both-column fractures achieving a second-
ary congruence (see Sect. 19.1.3).
In these cases surgery might achieve perfect reduc-
tion of both the innominate bone and the
acetabulum, but with higher risks, and if this ana-
tomical result is not achieved it would have been
better not to operate at all.
The age of the patient does not constitute a contra-
indication for surgery: we have operated on two pa-
359
tients over 80, the case of one of whom is shown in
Fig. 19.6.
19.4 Timing of Surgery
Apart from manipulative reduction of a posteriorly
dislocated femoral head, which must be performed
as an emergency procedure whatever the associated
acetabular fracture, only three conditions justify
emergency surgery:
irreducible dislocation of the femoral head;
unstable reduction of a dislocated femoral
head;
posterior dislocation of the femoral head asso-
ciated with a segmental fracture of the femoral
head (see Sect. 19.1.2).
So long as any posterior dislocation of the femoral
head has been reduced and remains stable, without
traction (the limb being just left in slight external
rotation), no acetabular fracture is ever an
emergency case, and the difficult surgery involved
in its treatment need never be performed in the
middle of the night by a tired or inexperienced
surgical team.
The ideal is to operate between the 2nd and 6th
day after injury, when the pelvic bleeding has
stopped spontaneously. In the interim traction is
unnecessary: the patient simply lies in bed, the leg
stabilised with cushions if required. Prophylactic
anticoagulants and antibiotics are administered to
prevent the major complications.
Nevertheless, the time at which the injured per-
son arrives is critical. Arrival is sometimes delayed
for any number of reasons, and the patient is seen
later than the 6th day. No hurry is justified; a full
check up to eliminate any evolutive pathology (i.e.
urinary infection, D.Y.T., etc.) and a full study of
the case have to be performed. Even if waiting a few
days more is necessary, it does not matter too
much, provided that prophylactic antibiotics and
anticoagulants are administered.
 360 General Principles of Management of Acetabular Fractures

A B

Fig.19.6A-G. Both-column fracture extending to the iliac crest

in an 81-year-old patient. operated on through the ilio-inguinal
approach in spite of slight radiograph signs of osteoarthritis.
A-C The three standard views. E-G Radiographic appearance
6 years post-operatively. The patient had undergone total re-
placement of the contralateral hip (D) 1 year after the operation
c on the acetabulum
 Timing of Surgery 361

D E

F G

Fig. 19.6D-G
20 Surgical Approaches to the Acetabulum
During so many years of operative management of
acetabular fractures, we have pursued a dream of
finding a method of approach which would permit
us to cope with all problems of reduction, whatever
type of fracture was involved. This approach would
have to enable us to reach at the same time, and
with the same degree of ease, both columns of the
acetabulum and also to allow access to the inside of
the pelvis. The latter is indispensable in the control
of some fracture fragments, and sometimes
necessary in order to explore nerves and vessels in-
cluded in the fracture line or lacerated thereby.
Despite attempts and anatomical research we are
still some way from this goal. Until 1975 we re-
mained forced to choose an approach leading elec-
tively to one or the other column. The column op-
posite to the approach we had chosen was reached
only by means of the finger-tips or an instrument.
On occasions, we have had to accept dealing with
each column by two separate approaches either
during the same operation or separated by an inter-
val of 8-15 days. During the years 1974 and 1975
we developed an extensile approach that makes it
possible to reach most of the innominate bone, if
necessary along both its external and its internal
aspects. This is called the extended ilio-femoral
approach. The only reason why we do not use this
extensile approach routinely is because of the asso-
ciated risks of post-operative ossifications, which
as yet cannot be prevented with certainty. Never-
theless, now as before, an approach giving com-
plete and simultaneous access to both columns of
the bone has not yet been described.
Dana MEARS developed another extensile
approach, the "triradiate", which for reasons which
will be explained later we never use straight away,
but on occasion we transform a Kocher-Langen-
beck into a triradiate in order to avoid a subsequent
anterior approach.
20.1 Classical Approaches
We believe that we have used, at one time or an-
other, the majority of approaches to the hip de-
scribed in the literature with varying success. As
originally described, they have been abandoned or
modified:
(a) Posterior approaches. Langenbeck approach
(1874): With the subject in the lateral position,
and the thigh flexed at 45 0 so that the longitu-
dinal axis of the femur is directed towards the
posterior superior iliac spine, the approach is
basically through the gluteus maximus in line
with its fibres. The incision extends from the
posterior superior iliac spine to the greater
trochanter; it is deepened through the aponeu-
rosis of the gluteus maximus and the muscle is
split throughout its length, starting over the
trochanter where the sub-gluteal space is more
easily found. This approach is adequate for the
posterior part of the joint but gives only a nar-
row field and is limited distally. Even with the
subject prone, the overall access is poor. It also
has the inconvenience of compulsory transec-
tion of the nerve supply of the part of the
gluteus maximus above the incision.
Kocher approach (1907): With the subject in
the lateral position, KOCHER recommended a
two-limbed incision of which the summit is at
the superior angle of the greater trochanter, the
inferior or vertical branch descends along the
outer aspect of the thigh, and the superior
branch is directed towards a point three finger-
breadths in front of the posterior superior iliac
spine. Below the angle, the deep fascia is divid-
ed in the same line, and above, along the ante-
ro-superior border of the fascia covering the
gluteus maxim us. The incision is through apo-
neurotic tissue throughout.
364
GIBSON (1950) with much honesty de-
scribed the Kocher approach also using the
lateral position. The description was widely ac-
cepted and in France at least, the name of
GIBSON remains attached to this approach,
KOCHER being almost totally forgotten.
The KOCHER approach yields good access
inferiorly, but above, one is impeded by the
gluteus maxim us, especially in the region of the
neurovascular bundle of the muscle where it
emerges from the greater sciatic notch, or if it
is desired to gain access to the whole of the
greater sciatic notch. It can be detached from
the bone, but even this does not give the access
which might be expected.
JUDET and LAGRANGE (1958) have com-
bined these two classical posterior incisions,
thereby gaining the avantages offered by each.
Since 1960, we have called this combination the
Kocher-Langenbeck approach.
(b) Anterior approaches. Smith-Petersen ap-
proach: We commenced our attempts to
operate upon the anterior aspect of the acetab-
ulum using the second modification of this
author's classical description. Unfortunately,
even when the gluteal muscles have been
stripped from the iliac crest and from the exter-
nal surface of the iliac bone, the sheet muscles
of the abdomen detached from the crest, and
the iliacus separated as far as the pelvic brim,
the only parts accessible are the anterior seg-
ment of the iliac wing and crest, the roof of the
acetabulum, the anterior border of the upper
part of the ilium, the internal iliac fossa as far
as the sacro-iliac joint and the posterior one-
third of the innominate line. The external iliac
vessels preclude an exploration of the whole
pelvic brim, and access to the quadrilateral sur-
face of the ischium is impossible. These restric-
tions confine the approach to treating a few
simple transverse fractures or a high pure
anterior column fracture.
LEVINE (1943) described a method using
the same cutaneous incision as the Smith-
Petersen approach but including only the
separation of the internal iliac fossa muscles.
This was even more inconvenient than the
classical Smith-Petersen approach and did not
even allow access to the roof or the outer aspect
of the wing of the ilium.
Modification of the vertical component of
the Smith-Petersen approach has offered more
adequate access to the anterior column. This is
Surgical Approaches to the Acetabulum
the ilio-femoral approach which we shall study
later.
(c) Lateral approach. Ollier approach (1892): Of
all approaches to the hip, this is the most
deceptive with regard to the acetabulum and is
full of unrealised promises. It is performed
with the subject in the lateral position. A curv-
ed incision is made, of which the summit is
5 cm below the greater trochanter and the ends
are directed respectively towards the anterior
and posterior superior iliac spines. The
posterior branch traverses the gluteus maximus
as in the Langenbeck incision whereas the
anterior follows the interval between the tensor
and the gluteus medius, care being taken with
respect to the nerve to the tensor fascia lata.
Deeper, the greater trochanter can be detached
and lifted together with the gluteal muscles, or
the tendons of these can be cut from the bone,
their origins being subsequently stripped from
the iliac bone.
This approach gives access only to the outer
face of the iliac bone and to the posterior wall
of the acetabulum but not to the ischium, to
the roof if one strips the muscles, and perhaps
to the upper part of the anterior column. It
provides no access to the inner aspect of the
pelvis, a serious disadvantage. At first sight it
offered the possibility of operating on both col-
umns simultaneously but in fact the access to
both remains very limited.
As a result of practical experience we use the
Kocher-Langenbeck incision as our only poste-
rior approach.
Of the anterior operations the ilio-femoral
approach developed from the second Smith-
Petersen modification is now rarely indicated;
it has been replaced by our ilio-inguinal
approach.
Lateral exposure with access to the whole of
the posterior column, the upper two-thirds of
the anterior column and the whole of the iliac
wing has been achieved by our extended ilio-
femoral approach.
20.2 Kocher-Langenbeck Approach
The operation is performed with the subject lying
prone on an orthopaedic table. The advantages of
skeletal traction as originally contrived were
marred by the effect of tension of the sciatic nerve,
Kocher-Langenbeck Approach
which could not be safely retracted. Post-operative
paralysis (see Sect. 24.3.1) was alarmingly frequent:
23 cases out of our first 125 operations (180/0). All
of these could not, however, be attributed to trac-
tion since for authors using the lateral decubitus
position without traction, the incidence was
similar. Even without taking into account this
serious complication the presence of the nerve
under tension is a source of anxiety to the surgeon
and constitutes an obstacle to reduction and fixa-
tion.
20.2.1 Technique
Since 1965 we have always used the Kocher-Langen-
beck approach with the subject prone on the or-
thopaedic table; traction is exerted by means of a
skeletal traction pin through the condyles of the
femur (Fig. 20.1). The Steinmann pin with a stirrup
allows traction with the knee flexed at about 45°,
the leg being maintained in this appropriate posi-
tion by means of a suitable prop. Flexion of the
knee relaxes the sciatic nerve significantly. It is
essential to identify the nerve as soon as the gluteus
maximus is split for it is so mobile as a result of this
posture that it risks being missed in the fat and can
be injured by a retractor; it is never like the violin
string which it resembles in operations with the
knee extended. Traction on the flexed knee has
reduced the 18% incidence of sciatic palsy to 3.3%.
Nevertheless, there is no place for complacency
because there were seven sciatic palsies out of 211
operations, despite transcondylar traction with the
knee flexed.
Sciatic nerve damage sometimes results from
obvious causes but in other instances, it has oc-
curred despite every care; we believe that the ma-
jority are due to injury from levers and retractors,
even from those specially designed for the purpose.
The part most at risk is the outer aspect of the
nerve (corresponding to the external popliteal com-
ponent), pressure being applied unconsciously
from the bone levers adapted for exposing the pos-
terior column. The surgeon must be constantly
vigilant with respect to both his own activities and
those of his assistants, particularly the one on the
opposite side of the table who can unwillingly exert
traction or pressure on the nerve intermittently
throughout the procedure.
The angle of the incision is located over the
superior border of the greater trochanter
(Fig. 20.2). The superior branch (the Langenbeck
component) is directed towards the postero-superi-
365
or iliac spine, its length varying with the amount of
access to the posterior column required and the
adiposity of the patient, but we end the incision at
a minimum of 6- 8 cm from the posterior superior
iliac spine; for reasons to be explained shortly, we
no longer extend the incision up to the postero-
superior spine.
The inferior branch descends 15 - 20 cm verti-
cally on the outer face of the thigh passing equidis-
tant between the anterior and posterior borders of
the greater trochanter.
The superficial fascia is divided in line with the
skin incision and the gluteal fascia opened in the
region of the greater trochanter. The fascia lata is
divided vertically and the aponeurotic incision con-
tinued upwards using scissors or by blunt dissec-
tion, splitting the fibres of the gluteus maximus;
this process separates the upper third which
receives its blood supply from the superior gluteal
artery from the lower two-thirds supplied by the in-
ferior gluteal vessels.
The innervation of the gluteus maximus is from
only the inferior gluteal nerve; when splitting the
muscle, the nerve trunks destined to innervate the
superior third of the muscle are met about half-way
between the greater trochanter and the posterior
superior iliac spine.
The splitting of the muscle fibres stops as soon
as the first nerve trunk supplying the upper part of
the gluteus maximus is met. This prevents the post-
operative palsy of the part of the gluteus maximus
above the incision that we have noted post-opera-
tively in many cases when we routinely used an inci-
sion extending up to the posterior superior iliac
spine. A gluteal muscular palsy is so inconvenient
that, as we do not know at operation what will hap-
pen or, sometimes, already has happened to the
other gluteal muscles, it is best to preserve the
whole of the gluteus maximus nerve supply if pos-
sible. If action along the posterior iliac crest ap-
pears essential, a separate incision may be used.
The sub-gluteal bursa is opened and divided at
the level of the trochanter and the margins of the
gluteus maximus are retracted; this exposes the
plane of the deep layer of muscles, and inferiorly,
the femoral insertion of the gluteus maximus
(Fig. 20.3). First the plane must be developed from
the level of the lower part of the inter-trochanteric
crest where it gives rise to the quadratus femoris.
Following the posterior face of this muscle medial-
ly, the external aspect of the sciatic nerve can be
located safely. The nerve is followed towards the
greater sciatic notch and should be explored
 366

Surgical Approaches to the Acetabulum

B
 Kocher-Langenbeck Approach
,,,
,,
,,,
I
Fig. 20.2. Kocher-Langenbeck approach. See text
throughout its course, especially if a pre-operative
sciatic palsy has been detected; it should be freed
from haematoma and bone fragments, but we never
put any kind of sling around it. Frequently, we
divide the femoral insertion of the gluteus maximus
at the level of the femur, 1 cm away from the bone,
ligating at the same time a branch of the posterior
circumflex femoral artery which is divided where it
lies just behind the tendon. This has the advantage
of liberating more of the gluteus maximus and
allowing it to be held back more easily, reducing the
chance of damage to the sciatic nerve from retrac-
tors, and exposing more of the ischial tuberosity. If
divided, the tendon will be repaired with separated
stitches at the time of closure.
The amount of damage inflicted on the external
rotators of the hip varies from injury to injury: they
..
Fig. 20.1A,B. Kocher-Langenbeck approach. A Patient in prone
position with transcondylar femoral traction and knee flexed on
the new ludet-Tasserit radiolucent table, 8 detail
367
Fig. 20.3. Kocher-Langenbeck approach
may be contused or lacerated severely in a posterior
dislocation, or they may be intact. In cases in which
the piriformis and the obturator internus are intact,
they should be divided through their tendons of in-
sertion, the proximal ends of which are attached to
long sutures (Fig. 20.4 B - D). The suture attaching
the piriformis is passed through the upper edge of
the cut of the gluteus maximus for the duration of
the operation; elevating the piriformis in this way
allows the surgeon to see the sciatic nerve emerging
from the greater sciatic notch.
The obturator internus tendon is elevated to-
gether with the gemelli (Fig. 20.4 B - E) by blunt
dissection from the capsule and the retro-cotyloid
surface. Progressing inwards, we finally open the
outer pouch of the constant underlying synovial
bursae (Fig. 20.4 C), thus gaining direct and easy
access to the lesser sciatic notch covered with
recognisable white fibro-cartilaginous tissue (Fig.
20.40), and then we enter the true pelvis. Inspec-
tion of the deep aspect of the obturator internus
 368 Surgical Approaches to the Acetabulum

c F
Fig. 20.4 A - F
Kocher-Langenbeck Approach
Fig.20.4A-G. Kocher-Langenbeck approach. E Horizontal
section of the hip through the lesser sciatic notch, showing the
track of the obturator internus and its underlying synovial bur-
sae. F The Kocher-Langenbeck incision at its completion. A
reveals the discrete tendons of origin of the muscle
converging to form the main tendon (Fig. 20.40).
The beak of the sciatic nerve retractor can be in-
serted easily into the notch (Fig. 20.5 B): it will then
retract the nerve steadily, but the obturator internus
tendon will always be between the retractor and the
nerve, acting as a protector for the latter, provided
the assistant in charge of the retractor pulls con-
stantly on the suture attached to the tendon.
If the piriformis and obturator internus have
been lacerated at the time of trauma, their medial
ends should be sought and marked with sutures in
the same way.
Retracting the obturator internus protects the in-
ternal pudendal neurovascular bundle, which is
never seen.
Dividing and elevating the external rotators ex-
poses the capsule and further inwards the retro-
acetabular surface of the posterior column is pro-
gressively exposed and cleaned of soft tissues. In
cases of posterior wall fracture the stripping is pur-
sued from the margin of the fracture site. Eventual-
ly the anterior border of the greater sciatic notch
and the ischial spine come into view.
It must be remembered that whereas, below,
muscles cushion the sciatic nerve, above, the nerve
369
capsulotomy allows visualisation of the head. (Reproduced by
permission of Joel MATTA). G Access to the pelvic bone provid-
ed by the Kocher-Langenbeck approach: left access to the exter-
nal face, right access to the internal face
rests against the bony edge of the greater sciatic
notch. In cases where it is necessary to gain access
to the ischial tuberosity, it is preferable to partially
strip the origin of the quadratus femoris from the
pelvic bone rather than to divide the muscle, in
order to avoid unnecessary damage to the medial
circumflex artery. A Steinmann pin driven into the
upper part of the ischial tuberosity outside the
sciatic nerve acts as an effective retractor of the
gluteus muscles.
20.2.2 Application
Access is available to the whole of the posterior col-
umn, i.e. to the greater and lesser sciatic notches, to
the ischial spine, to all the retro-acetabular surface,
to the sub-cotyloid groove, to the ischial tuberosity
and, with some difficulty, to the posterior part of
the ischio-pubic ramus which can be osteotomised
by this route in the treatment of a malunion of an
acetabular fracture.
The sciatic nerve is conveniently retracted in two
ways: traction on the divided obturator internus by
means of the stay-suture, and using a special retrac-
tor (Fig. 20.5 A) which we have designed by modify-
370
A This approach can also be extended so as to
B the tendinous insertion of the gluteal muscles into
Fig.20.5A,B. Kocher-Langenbeck approach. A Special sciatic
nerve retractor, B retractor in operative position. See also
Fig. 21.6
ing a Leriche retractor; this has a convex surface
which rests against the nerve and at its end bears a
spike which can be inserted into one of the sciatic
notches. We cannot over-stress that throughout the
operation the sciatic nerve should be observed; it
can at any moment insinuate itself under the retrac-
tor during manipulations especially by the assis-
tant. The flat parts of the blade of the instrument
must be parallel to the direction of the nerve and
must remain at all times in contact with bone.
Through the greater sciatic notch access is
gained to the true pelvis. For usefulness of access,
the soft tissue is incised along the medial aspect of
the anterior border of the greater sciatic notch, i.e.
the obturator internus aponeurosis, and with the
elevator this muscle is elevated from the
quadrilateral surface forwards and downwards as
far as necessary allowing the obturator canal and
the posterior half of the pelvic brim to be reached.
Endopelvic access can be improved by dividing the
sacro-spinous ligament or, as proposed by VIREN-
QUE (1956), by dividing the ischial spine itself at its
Surgical Approaches to the Acetabulum
base. One or other of these procedures frees further
the sciatic nerve, which can be held back into the
pelvis to advantage.
In summary, the surgeon achieves perfect access
to the posterior column and some control of the
quadrilateral surface and pelvic brim by instrument
or the finger.
If grasping the posterior column proves to be
difficult, it may be helpful to insert a Schanz screw
or a femoral head extractor into the ischial
tuberosity. They then act as a handle for manipula-
tion of the column as a whole.
reach the anterior column. The external surface of
the iliac wing can be stripped of periosteum above
the hip capsule as far forwards as the anterior
border. Then a Lambotte hook can be slid in
anteriorly to gain purchase on the anterior column.
We prefer to avoid this periosteal stripping of the
muscle origins as we consider it may be responsible
for periarticular calcification and for significant
damage to the gluteal muscles themselves and to
their neurovascular supplies. If it is necessary to
reach the anterior column in the superior
acetabular region (but not sufficiently to warrant
an anterior incision) we would now prefer to divide
the greater trochanter, and to strip only the lower
part of their origins. In fact, if the choice of ap-
proach has been correct this is rarely necessary and
the only stripping required is in the posterior part
of the wing in order to apply a plate. In the vast
majority of cases the gluteus medius is left com-
pletely undisturbed.
Another possibility is to extend the Kocher-
Langenbeck incision in Dana MEARS' triradiate
approach, as will be described later (see Sect. 20.8).
In a few cases where the prone position was con-
tra-indicated by reason of a flail chest injury, or
because there was a double anterior fracture of the
pelvis associated with a fracture of the acetabulum,
the prone position on the orthopaedic table was not
possible; we used the Kocher-Langenbeck incision
but with the patient lying on the opposite side.
In this position it was also easy to keep the knee
flexed about 45°, in order to relax the sciatic nerve,
but we felt less at ease (perhaps from habit).
20.2.3 Closure
The closure of the Kocher-Langenbeck incision is
easy and simple. It is important to reconstitute a
Kocher-Langenbeck Approach
muscular bed under the sciatic nerve in order to
protect it from a plate or screws. The stay-sutures
placed on the obturator internus and the piriformis
facilitate identification and suturing of these to the
stumps of their tendons of origin remaining at-
tached to the posterior border of the trochanter.
Having repaired these tendons, the neighbouring
muscular bellies should also be approximated so as
to provide a complete curtain protecting the nerve.
At least two suction drains are inserted (one drain-
ing the gluteal region and the other the pelvis by
one of the sciatic notches).
20.2.4 Dangers
(a) Sciatic nerve. It is not without justification that
we labour the dangers of damage to the sciatic
nerve. The main trunk in the operative field can
be damaged by a retractor or by some other in-
strument. In one case, when we had just begun
to use skeletal traction with the knee flexed, it
was lacerated with scissors during the incision
because it was so loose and difficult to identify.
It is from this experience that we recommend
identifying the nerve trunk straight away after
splitting of the gluteus maximus. However, we
find it unnecessary to put a sling around the
nerve.
The sciatic nerve may also be damaged by
stretching at the level of its plexus of origin as
a result of strong traction. In this case the post-
operative electromyography of the femoral
nerve and of the sciatic nerve muscle distribu-
tions may show the plexus as being the source
(see Sect. 24.3.1).
(b) Superior gluteal nerve and gluteal vessels.
These are greatly at risk during freeing of the
superior border of the greater sciatic notch,
especially if the posterior column detaches with
it the angle of the greater sciatic notch, as this
segment may be very sharp. Intra-operative
bleeding from the vessels may occur; it may be
due to wounding of the vessels by any of the
surgical tools or to the displacement of a very
sharp fragment, or may be because the freeing
of the vessels can reactivate a post-traumatic
haemorrhage by mobilising a precarious clot.
An arterial wound can sometimes be easily
sutured, or ligation of the vessel may be
necessary and done properly after isolation.
Haemostasis of the veins is much more dif-
ficult: often multiple ligatures are needed and
371
then there is a higher risk of including the nerve
in ligatures. Isolation of the nerve during
bleeding is practically impossible. The use of
clips increases the risk of injuring the nerve.
We believe that the superior gluteal nerve
is so at risk and the inconvenience of gluteal
muscle palsy is so great that when a vascular
haemorrhage occurs, we immediately pack the
area with wet swabs and leave the packing there
for as long as possible. After half an hour or
more, when the swabs are removed, the bleed-
ing has usually stopped, and from then on we
avoid disturbing anything in that region. At the
time of closure we leave some resorbable swabs
in place to improve the haemostasis.
If bleeding persists at the time of closure, it
is generally not very serious and one or two
stitches on the artery or around the veins are
enough to stop it; this can be done without in-
cluding the nerve as the bleeding is so discrete.
In our experience we have never seen any
isolated lesion of the superior gluteal nerve.
The ligation of the gluteal artery which we have
performed several times was without sequelae.
(c) Summary of the intra-operative complications.
Our operative reports of 461 Kocher-Langen-
beck procedures mention:
8 injuries of the superior gluteal artery;
- 6 injuries of the superior gluteal vein;
- 3 injuries of the sciatic nerve.
At a variable time after operation the presence
of an intra-articular screw was demonstrated in
five cases.
20.2.5 Complications
(a) Sciatic palsies. Out of 461 Kocher-Langenbeck
procedures, 46 (9.90/0) were complicated by
sciatic palsies of all varieties. The benefit of in-
tra-operative transcondylar traction will be
demonstrated in the section on early post-
operative complications. Among these 46 cases,
12 patients had not had sciatic nerve function
evaluated pre-operatively, and given our rate of
pre-operative involvement of the nerve
(12.2%), we surely blamed ourselves in some
cases for causing what were in fact unrecog-
nised pre-operative palsies.
Table 20.1 details the sciatic palsies accord-
ing to time of operation.
372 Surgical Approaches to the Acetabulum

Table 20.1
Type of sciatic palsy after Interval from injury to surgery Total
Kocher-Langenbeck approach
0-21 days 21 days to 4 months >4 months
Total 1
Total popliteal 7 2 9
Partial popliteal 9 3 13
Extended patchy 6 3 3 12
Restricted patchy 1 4 5
Sensory only 2 2
Post-operative extensIOn 1 1
Secondary post-operative 2 3
Total 29 13 4 46

(b) Ectopic bone formation. Ectopic bone forma-
tion is classified according to BROOKER into
types I, II, III and IV. Type I is defined as little
spots of ectopic bone, type II as spots up to
2 cm in width, type III as extended ossifications
leaving less than 1 cm free between the ectopic
bone and either the trochanter or the pelvis,
and type IV as apparently ankylosing ossifica-
tions uniting the two bones.
In practice there is no direct correlation be-
tween the radiological appearance of the ec-
topic bone formation and the restriction of hip
motion. In order to introduce this factor in the
evaluation of ectopic bone formations and
their consequences we classify them as
BROOKER does, but we add to the Brooker
figure the Merle d'Aubigne hip mobility score,
grades 1 to 6. Thus, we can have a IV-6 forma-
tion, which is an apparently ankylosing ectopic
bone formation but which is accompanied by a
normal mobility, while a IV-1 formation is a
radiologically ankylosing ectopic formation
resulting in clinical ankylosis of the hip joint in
a malaligned position.
Overall, following 461 Kocher-Langenbeck
approaches, we have observed ectopic ossifica-
tions in 105 cases (22.7070). Out of 314 cases in
which surgery was performed within 21 days
after injury, there were 82 (26.1 %) with ectopic
ossification, which in 15 of these cases signifi-
cantly decreased hip mobility (two 111-2; two
I1I-3; three I1I-3; one IV-2; three IV-3; one IV-4;
two 11-4; one 1-4). In five cases ectopic bone
was removed, and one further excision is
planned.
Out of 84 cases in which surgery was per-
formed between 21 days and 4 months after in-
jury, 21 (25%) showed ectopic ossification,
which in 11 cases decreased hip mobility (one
11-3; three 11-4; one I1I-3; one I1I-4; two IV-1;
one IV-2; one IV-3; one IV-4). One ectopic bone
was excised, one excision is planned.
Out of 63 cases in which surgery was per-
formed beyond 4 months after injury, only two
(3.1 %) developed ectopic bone, and they were
rated 11-4 and I1I-4, respectively). Table 20.2
details the Brooker ossification types and cor-
responding mobility grades.
(c) Haematoma formation. We now drain all
posterior operation wounds although earlier
this was not the case. As a result, post-operative
haematoma necessitating evacuation is now
rare. Out of the 461 cases in which the Kocher-
Langenbeck approach was used, there were 20
with post-operative haematomas that needed
treatment: eight were aspirated and 12 were
evacuated surgically.
(d) Injection. Out of the 461 Kocher-Langenbeck
operations, there were infections after 15
(3.2%). Of these, 12 (2.6%) followed an early
operation. Seven infections were deep and im-
mediate, four superficial and immediate, two
appeared after a few weeks and were associated
with arthritis, and two occurred late.
(e) Post-operative gluteal weakness. The true in-
cidence of this complication is unknown, as we
must confess that we have not checked it com-
prehensively and in detail in our reviews of pa-
tients. We are only able to report 14 cases of
gluteal palsy: six partial and one total palsy of
the gluteal muscles, and seven cases of deficien-
cy of the gluteus maximus above the incision.
In the case of gluteus maximus, the cause is sec-
tion of nerve fibres coming from the inferior
gluteal nerve and crossing the medial part of
the muscle, perpendicular to the incision, while
Ilio-femoral Approach 373

Table 20.2

Interval to surgery Brooker ossification type Mobility (d'Aubigne grading) Total

after Kocher-Langenbeck
operations 2 3 4 5 6

0-21 days 11 12
II 2 5 31 38
III 2 2 3 6 12 25
IV 1 3 1 2 7
Subtotal 3 5 7 11 56 82
21 days to 1 1
4 months II 3 7 11
III 1 1 4
IV 2 1 1 5
Subtotal 2 3 5 9 21
>4 months II
III 1 1
Subtotal 2 2
Total 2 4 8 14 12 65 105

for the gluteus medius and gluteus minim us, it
may be damage of the superior gluteal nerve as
a result of efforts at haemostasis, or of stret-
ching or crushing of the superior gluteal neuro-
vascular bundle by retractors. The nerve may
also be damaged at the time of trauma by the
displacement of fragments. Appreciation of
gluteal muscle function is sometimes either dif-
ficult or impossible before surgery.
Involvement of the gluteus maximus results
in an asymmetrical appearance due to atrophy
of the buttock above the incision but without
any appreciable functional loss.
(f) Skin necrosis. Only one case of skin necrosis
has been reported with this incision.
20.3 llio-femoral Approach
20.3.1 Technique
As stated earlier, this approach has something in
common with that of SMITH-PETERSEN (1917) but
differs sufficiently to warrant further description.
It is performed with the subject supine on either
an orthopaedic or an ordinary operating table. The
incision (Fig. 20.6) follows the anterior half or two-
thirds of the iliac crest as far forward as the anterior
superior iliac spine and then descends along the
outer line of the sartorius muscle for about 15 cm.
It is distinctly more oblique than the Smith-Petersen
incision. The external face of the iliac wing is not
stripped but the iliac crest is freed from the anterior
abdominal muscles and the internal iliac fossa ex-
posed by elevating the iliacus muscle (Fig. 20.7). At
the level of the anterior superior iliac spine, the in-
guinal ligament is detached together with the sar-
torius which is freed along its lateral border taking
care to preserve, as much an possible, the nerve sup-
ply. One is obliged to divide the external branch of
the lateral cutaneous nerve of the thigh. Flexion of
the thigh facilitates the freeing of the lateral border
and of the deep face of the ilio-psoas muscle, the
tendon of which can be easily divided at the level of
the inguinal ligament, so allowing easier retraction
of the muscular portion of the muscle and greater
access to the anterior column - though not so far
as could be hoped! Remember that section of the
psoas tendon exposes the femoral nerve to the risk
of damage and the surgeon must therefore be
vigilant and protect the nerve from tension during
retraction medially.
The origin of the tendon of the straight head of
the rectus femoris at the level of the antero-inferior
iliac spine is cleared but it is not necessary to divide
it. In the region of the interspinous notch at the
anterior border of the iliac wing, the external face
of the iliac wing is stripped over a little area,
thereby enabling the jaws of a Farabeuf bone-
holding forcep to be applied, which will allow a
solid purchase on the upper part of the anterior
column. An artery is always found at this level and
haemostasis is necessary.
374 Surgical Approaches to the Acetabulum

Fig. 20.6. Ilio-femoral approach. See text Fig. 20.7. Ilio-femoral approach

20.3.2 Application

This method of approach (Fig. 20.8) offers good

access to the iliac crest, the upper part of the
anterior column, and the roof of the superior pubic
ramus. It is practically impossible to reach below
the ilio-pectineal eminence, even after dividing the
psoas tendon; certainly no access is possible
beyond the eminence to the superior pubic ramus.
Across the iliac fossa one can reach the sacro-iliac
joint, the brim of the pelvis in its posterior part,
and gain limited access to the internal wall of the
true pelvis with the finger or an instrument as far
as the superior aspect of the quadrilateral surface
of the ischium. There is easy access to the most
medial part of the capsule of the hip and section of
the ilio-femoral ligament along the anterior
acetabular lip may be necessary to obtain perfect
reduction of the anterior column .

..
Fig. 20.S. Ilio-femoral approach
Ilio-inguinal Approach
20.3.3 Closure
This is routine and anatomical. The psoas tendon
can be repaired if the ends are not too retracted.
20.3.4 Dangers
(a) Femoral nerve. Even with the hip flexed, the
femoral nerve is always in danger but was never
actually damaged in our series of operations.
The femoral vessels are easily retracted and
have never been injured.
(b) Lateral cutaneous nerve of the thigh. This is
always an obstacle to the anterior incisions and
in the ilio-femoral approach some of the lateral
branches always have to be sacrificed; the
femoral branches which follow the lateral
border of the sartorius muscle can be spared
and retracted medially with the sartorius and
the ilio-psoas.
20.3.5 Complications
After 15 operations using the ilio-femoral ap-
proach, the following complications occurred:
two complete paralyses or marked weakness of
the ilio-psoas, but no example of quadriceps
paralysis;
two haematoma formations in the iliac fossa re-
quiring aspiration;
one necrosis of the skin at the angle of the inci-
sion; it was less frequent than in the case of the
Smith-Petersen incision in which the angle is
more acute;
no infections, even in complicated cases.
20.4 Dio-inguinal Approach
LEWURNEL (1960) described an original anterior
approach developed in cadaveric studies. The ilio-
inguinal approach, which we have used largely
since 1965, developed from this. It embodies the
particular feature of completely opening the in-
guinal canal, of which a sound anatomical repair is
much more effective than would perhaps be ex-
pected.
375
20.4.1 Technique
Prior to the operation a Foley catheter is inserted
into the patient's bladder. The patient is positioned
supine on an orthopaedic table. A contra-indica-
tion for the use of the orthopaedic table is the
presence of a vertical anterior fracture through the
pubic rami on the opposite side, as this would allow
deformity of the anterior part of the pelvis due to
the pressure from the perineal post (Fig. 20.9). The
orthopaedic table can nevertheless be used in order
to benefit from the traction, but then it is necessary
to stabilise the opposite limb to its table arm and
to dispense with the pelvic post.
In complex fractures, supplementary lateral
traction (Fig. 20.10 A) of the femur may appear
necessary in order to extract the femoral head from
its centrally displaced position and to maintain its
reduction during the acetabular reconstruction. In
this case a femoral head extractor or a Schanz screw
is introduced from the vastus lateralis crest along
the axis of the femoral neck via a small vertical in-
cision astride the vastus lateralis crest. This device
is pulled on by an assistant or, better, connected to
a lateral traction attachment to the orthopaedic
table. Ribbon gauze is packed into the wound. In
the process of this, an undetected Morel-Lavalle le-
sion may be discovered, which must then be excised
as usual and be properly drained when the wound
is closed.
The incision (Fig.20.10B) extends along the
anterior two-thirds of the iliac crest and from the
anterior superior iliac spine up to the mid-line,
which is reached two finger-breadths above the
pubic symphysis, this latter part of the incision be-
ing slightly concave from above and medially. It is
essential to extend the incision along the crest
beyond its most lateral convexity to permit ade-
quate retraction of the ilio-psoas and the ab-
dominal muscles.
By sharp dissection the incision progresses to
the iliac crest without injuring the abdominal
muscles which sometimes tend to overhang the
middle portion of the crest. The insertions of the
abdominal muscles and the origin of the iliacus are
sharply elevated from the crest; then by a sub-
periosteal dissection, the iliacus is elevated from the
internal iliac fossa as far medially as the anterior
aspect of the sacro-iliac joint and distally to the
pelvic brim (Fig.20.10B). It is almost always
necessary to achieve haemostasis, using bone wax,
in the nutrient foramen of the posterior and in-
ferior part of the internal iliac fossa located
376 Surgical Approaches to the Acetabulum

1 - 2 cm from the sacro-iliac joint and the pelvic

brim. The internal iliac fossa is temporarily packed
with wet swabs.
Over the lower abdomen the incision progresses
through the superficial fascia to the aponeurosis of
the external oblique and the anterior aspect of the
rectus abdominis sheath, which are incised in con-
tinuity and in line with the cutaneous incision, to
pass at least 1 cm above the external inguinal ring
(Fig.20.10B).
The inferior edge of the incised aponeurosis be-
ing held with surgical clamps, the inguinal canal is
opened and the inguinal ligament visualised. A
sponge helps in cleaning the inguinal ligament of
areolar tissue. The spermatic cord in the male, or
round ligament in the female, is then visualised, a
finger is passed posterior to it to elevate it and the
adjacent ilio-inguinal nerve, and a rubber sling is
placed around them to allow their retraction.
In this stage the common origins of the internal
oblique and transversus abdominis and the trans-
versalis fascia have to be detached from the in-
guinalligament. In order to facilitate the further re-
construction of the posterior wall of the inguinal
canal, it is recommended that the inguinal ligament
A
itself be sharply incised with a scalpel so that about
1 - 2 mm of the ligament remains attached to the Fig. 20.9A-C. Contra-indication to traction on the orthopaedic
aforementioned structures. Opening the posterior table. A Antero-posterior radiograph. Because of an anterior
fracture on the opposite side, pressure from the pelvic prop or
wall of the inguinal canal in this way allows im- post deforms the ischio-pubic fragment position; correct reduc-
mediate penetration into the psoas sheath, which tion of this fracture could not be obtained. B, C Patient on the
adheres at this level to the inguinal ligament. Great Judet-Thsserit radiolucent table. B General view, C detail show-
care must be exercised during this step not to injure ing the special bracket for the lateral traction
the structures lying directly beneath the inguinal
ligament.
Laterally the ilio-psoas sheath is entered directly These structures lie within one of two compart-
(Fig.20.l0C). The lateral part of this incision ments or lacunae. The laterally situated lacuna
should be performed cautiously in order to musculorum contains the ilio-psoas, the femoral
safeguard the lateral cutaneous nerve of the thigh, nerve and the lateral cutaneous nerve of the thigh,
which lies immediately beneath the inguinal liga- while the medial lacuna vasorum contains the ex-
ment but is often variously situated: it may be adja- ternal iliac vessels and lymphatics. Usually, there is
cent to the anterior superior iliac spine or up to a thickening of the psoas sheath or ilio-pectineal
3 cm medial to it. fascia which separates the two lacunae.
The mid-part of the incision overlies the To allow exposure of the quadrilateral surface
anterior aspect of the external iliac vessels. and gain access to the true pelvis, it is essential to
Medial to the vessels, at the level of the pubic divide the ilio-psoas fascia completely, which
spine, the division of the conjoint tendon of the in- begins by the section of the thickened portions of
ternal oblique and transversus allows penetration the ilio-psoas fascia. To expose this, the ilio-psoas
of the retro-pubic space. If necessary the tendon of and femoral nerve are retracted laterally. Using the
the rectus abdominis is divided 1 cm above its inser- finger as a guide, the medial surface of the ilio-pec-
tion. The haematoma of the retro-pubic space is tineal fascia is freed, by carefully elevating the ex-
evacuated and the space packed with wet swabs. ternal iliac vessels and their satellite lymphatics
The anterior aspect of the structures passing from this fascial septum using blunt-tipped scissors
under the inguinal ligament are thus exposed. or a haemostat. Upon lateral retraction of the ilio-
Ilio-inguinal Approach 377

c
Fig. 20.9 B, C
378 Surgical Approaches to the Acetabulum

Fig. 20.10A-E

c E
Ilio-inguinal Approach 379

B Abnormlll (Accessory) Hb
h

Fig. 20.10 F -J. Ilio-inguinal approach

 380 Surgical Approaches to the Acetabulum

Ka

Kb

Fig.20.10K-O. See text (Kb,L reproduced by permission of

o
Joel MATTA)
Ilio-inguinal Approach
psoas and medial retraction of the iliac vessels, the
ilio-psoas fascia is sharply incised (Fig. 20.10D) up
to the ilio-pectineal eminence using scissors. Subse-
quently, with the same exposure, the scissors are
used to elevate the psoas sheath sharply from the
pelvic brim; in some individuals this manoeuvre
can be completed by finger dissection
(Fig. 20.10 E).
A second rubber sling is placed around the ilio-
psoas, femoral nerve and lateral cutaneous nerve of
the thigh for subsequent retraction (Fig. 20.10F). It
is sometimes necessary to section the aponeurotic
fibres which tether the lateral cutaneous nerve
medially, in order to mobilise the latter easily with
the other elements. In some cases we leave the nerve
out of the ilio-psoas sling. The insertion of a finger
beneath the iliac vessels (Fig. 20.10G) from lateral
to medial permits the application of a third rubber
sling around the external iliac vessels and the adja-
cent lymphatics. It is essential not to dissect the
vessels closely and to leave as much areolar tissue
as possible around them in order to avoid damage
to the lymphatics and post-operative impairment
of the lymphatic drainage.
Prior to retraction of the external iliac vessels, a
search is carried out posterior to the vessels for
either an anomalous origin of the obturator artery
or the presence of an anastomosis between the ob-
turator vessels and the external iliac vessels which
lies close against the medial aspect of the superior
pubic ramus and is recognised easily by finger
palpation. If such an anastomosis is present it is
clamped, ligated and divided. Only in one case
have we found a large vessel, called by anatomists
the "corona mortis" (Fig.20.10R).
All the structures passing under the inguinal
ligament having been thus elevated in their respec-
tive slings, the ilio-inguinal incision is now com-
plete (Fig. 20.10 I).
20.4.2 Application
By medial or lateral retraction of the elements
placed in three corrugated rubber slings, access is
gained to the internal face of the innominate bone.
(a) Medial retraction of the ilio-psoas (Fig. 20.lOJ)
provides access to the entire iliac fossa, the
posterior half of the pelvic brim and distally
along the anterior column as far as the most
superior aspect of the ilio-pectineal eminence.
Two Steinmann pins driven into the most
381
remote part of the iliac fossa in front of
the sacro-iliac joint act as retractors of the
abdominal muscles and the ilio-psoas
(Fig. 20.10 J). A malleable retractor applied to
these two Steinmann pins forms a stable and
highly efficient retractor system.
(b) Lateral retraction of the iUo-psoas and femoral
nerve combined with medial retraction of the
external iliac vessels opens the middle and
probably most important window of the ilio-in-
guinal incision (Fig. 20.10 K). This window pro-
vides access to a large part of the pelvic brim
from the sacro-iliac joint up to, distally, the
origin of the superior pubic ramus. By sub-
periosteal elevation of the obturator internus
beginning from the pelvic brim, we gain access
to the whole quadrilateral surface up to the
sciatic notches, the ischial spine and the ob-
turator foramen.
Medial retraction is facilitated by the inser-
tion of a Ganz retractor or a malleable ribbon
retractor with the tip on the quadrilateral sur-
face or into the greater sciatic notch.
The anterior edge of the innominate bone is
also accessible between the anterior inferior il-
iac spine and as far down as the ilio-pectineal
eminence. During retraction of the vessels the
pulse of the iliac artery should be checked peri-
odically.
(c) Lateral retraction of the vessels with medial
retraction of the spermatic cord opens the
medial window of the ilio-inguinal incision
(Fig. 20.10L), providing access to the superior
pubic ramus, and sub-periosteal dissection of
the pectineus muscle from the superior aspect
of the ramus may help to reduce the fracture
lines involving this segment. Beyond the
superior pubic ramus, access is gained to the
obturator canal and the obturator neuro-vascu-
lar bundle.
(d) Lateral retraction of the spermatic cord pro-
vides access to the pubic angle and the pubic
symphysis (Fig. 20.10 J). If it is necessary to
reach as far as the symphysis pubis or even fur-
ther, the rectus abdominis is divided 1 - 2 cm
from its insertion (Fig. 20.101). This may be
necessary only in its outer part, but the whole
of the breadth of the muscle can be mobilised.
One can go beyond the mid-line either by
dividing the tendon of the opposite rectus ab-
dominis, or by stripping the superior border of
the opposite pubis and then retracting the mus-
cle, preserving its continuity (Figs.20.10L).
382
(e) Limited access to the external aspect of the iliac
wing in order to apply clamps may be obtained
either over the iliac crest at the level of the
gluteal tubercle or through the notch between
the two anterior iliac spines. It is also possible
to elevate a greater part of the iliac wing, either
to facilitate or allow the reduction of bone
fragments inaccessible from the inner aspect of
the bone or to insert a screw from the lateral
aspect of the ilium distally along the anterior
column into the superior pubic ramus. How-
ever, to do this involves the risk of inducing sig-
nificant ectopic bone formation.
In summary, the ilio-inguinal incision pro-
vides exposure of almost the entire inner aspect
of the innominate bone from the anterior
aspect of the sacro-iliac joint to the pubic sym-
physis, including exposure of the quadrilateral
surface (Fig.20.10M). If necessary, access can
be gained to the lateral part of the ala of S 1
along a strip 1 - 2 cm in width, which allows
good control of the sacro-iliac joint and its fix-
ation with a plate across the joint if required.
Only the inner aspect of the ischium is not ac-
cessible through this approach.
It has to be remembered that the iliacus is
the only muscle to be stripped off the bone,
and that the entire hip abductor system is left
intact. It is also true to say that this approach
proceeds through anatomical areas which are
not usually familiar to most orthopaedic sur-
geons. We therefore strongly advise surgeons
that prior to undertaking the ilio-inguinal ap-
proach they practice on a cadaver and assist at
operation a surgeon who is familiar with this
exposure.
20.4.3 Closure
At completion of the operation, at least two suc-
tion drainage tubes are placed, one in the retro-
pubic space and one in the internal iliac fossa up to
beyond the pelvic brim along the quadrilateral sur-
face. Sometimes a third one is found useful to drain
the inguinal canal itself.
The repair must be conducted with great care.
The slings are cut and removed one by one and it
is verified that the pulsation of the femoral artery
is normal. Reattachment of the muscles may be dif-
ficult because the muscles retract; complete relaxa-
tion with muscle paralysis is valuable at this stage.
Surgical Approaches to the Acetabulum
Heavy sutures are employed to secure the ab-
dominal fascia to the gluteal fascia along the iliac
crest (Fig.20.lON). If divided, the tendon of the
rectus abdominis is re-attached to its stump. The
fascia transversalis and the common origin of the
internal oblique and transversus abdominis are re-
attached to the inguinal ligament, and the narrow
strip of inguinal ligament that was harvested with
these structures aids in a solid repair (Fig. 20.100).
The spermatic cord or round ligament is reposi-
tioned and the ilio-pectineal fascia is not repaired.
Closure of the external fascia of the rectus ab-
dominis and the aponeurosis of the external
oblique completes the restoration of the inguinal
canal in its integrity; the superficial and deep in-
guinal rings remain displaced relative to each other.
20.4.4 Dangers
(a) Lateral cutaneous nerve of the thigh. This may
remain apparently intact but can have been
stretched during the procedure. In order to
avoid the lesion, it is necessary to have liberated
the nerve sufficiently superiorly and always to
be vigilant with regard to retractors. Never-
theless, stretching, if not rupture, of this nerve
is still frequent, and the question begins to be
raised whether it would not be better to cut it
at the beginning of the operation if it really ap-
pears in danger and then suture it properly at
the end. At follow-up we found significant
anaesthesia or hypoaesthesia in the related ter-
ritory in 40 cases, i.e. 22.5070 of the patients
operated on via this approach.
(b) Retro-pubic anastomosis. The communication
between the external iliac artery and the ob-
turator or deep inferior epigastric arteries can
tear or be cut at the time of freeing the posteri-
or aspect of the vessels. It should be avoided by
specifically looking for this inconstant anasto-
mosis, separating it and coagulating or dividing
it between ligatures. Fairly frequently the
anastomosis is not found or it may not exist; it
may have been lacerated by the fracture and
haemostasis already achieved. If cut, the veins
can retract into the obturator canal and be very
difficult to control; in practice, this incident is
unusual and the anastomosis is easy to locate
and to control.
(c) Internal iliac vein. In three cases this vein has
been injured by a retractor. Suture was possible
and there were no complications.
Ilio-inguinal Approach
(d) Injection. Secondary infections appeared a
Fig. 20.11. Osteosynthesis performed through the ilio-inguinal
approach. Transverse fracture associated with a separation of
the pubic symphysis and a vertical fracture of the left pubic
ramus. Antero-posterior radiograph showing osteosynthesis
20.4.5 Complications
Certainly, complications occurred, especially at the
beginning of our use of this approach.
(a) Lymphatic damage. In three of the early pro-
cedures, we dissected too near the vessels and in
effect divided the lymphatic channels accom-
panying them, and also divided the lymphatics
passing at the inner part of the femoral ring.
We believe that lymphangitis, oedematous
swelling at the upper part of the thigh with
redness and warmth which were limited above
by the incision, were due to these lymphatic
sections. These phenomena subsided in 8 - 10
days but not without causing us considerable
anxiety. They have not been seen again since we
ceased close dissection of the external iliac
vessels.
(b) Cutaneous necrosis. This has been rare, occur-
ring in only two cases in the region of the
antero-superior iliac spine, limited to 1 cm on
both sides of the incision. There were no conse-
quences other than scarring.
(c) Haematoma. This may occur at four sites: in
the retro-pubic space, in the internal iliac fossa,
in the inguinal canal, or subcutaneously. Hae-
matoma in the retro-pubic space has not always
had to be aspirated but sometimes drains have
had to be left in place more than eight days due
to continued loss. Haematomata in the iliac
fossa are the most dangerous especially because
383
of their frequency, and they are difficult to
detect: palpation of the internal iliac fossa is
always painful in the first post-operative days.
Fluctuation is extremely difficult to elicit. The
frequency of these haematomata persuades us
to prolong suction drainage and to have several
drains disposed throughout the area. In obese
patients haematomata of subcutaneous tissue
or of the inguinal canal have been seen. They
can also be difficult to locate in spite of their
superficial position. Nowadays ultrasonogra-
phy is helpful in detecting them. Besides the
cases that resulted in infection, we have
aspirated two and surgically evacuated five
other cases: an incidence of 4.5070.
high risk with this approach in the early days
when we did not use prophylactic antibiotics.
Infections occurred during the later stages of
healing, developing from haematomata of the
iliac fossa or subcutaneous tissue which had
not been detected. In three cases a direct cause
can be suggested. Two cases were associated
with secondary infection of a Morel-Lavalle
lesion neglected or unperceived and located at
the zone of impact over the greater trochanter.
In another instance there was a sloughing ulcer
of the buttock.
It is perhaps relevant that the ilio-inguinal
incision passes through the upper part of the
pubic hair region and the frequency of sub-
cutaneous infections after hernia operations is
thought to be significant for this reason. We try
to reduce the danger by a meticulous prepara-
tion of the skin and whenever possible by mak-
ing the incision above the hairy zone. Infections
of post-operative haematomata after an ilio-in-
guinal incision may also have a lymphatic
origin; inevitably, the ilio-inguinal approach is
accompanied by division of most of the lym-
phatics draining the lower limb, the buttock,
the perineum, and the external genitalia. Any
superficial or deep infection in these areas
could result in contamination of a haematoma
in the area of the operative wound, the micro-
organisms being poured into it by the lym-
phatics cut at operation.
All operations on fractures of the acetabu-
lum by the ilio-inguinal approach should be
covered by prophylactic antibiotics adminis-
tered pre-, intra- and post-operatively for 6 - 8
days. In the first 22 ilio-inguinal operations,
without antibiotic coverage, there were seven in-
384
fections (31.8070), whereas at the time the 1990
review was closed, out of the last 146 ilio-
inguinal operations performed with good anti-
biotic cover there were two cases of early infec-
tions (1.4%). Of this total of nine cases of early
infections, three were deep immediate and six
superficial immediate. We had also a case of
secondary infection (after more than 1 month)
without arthritis of the hip, which healed after
debridement and has an excellent result at seven
years. These figures can be taken as another
significant argument in support of the efficacy
of antibiotic prophylaxis.
Infections were treated by early excision of
all necrotic material. The majority of them
have not left any residual problems, even those
in which the plates used for the osteosynthesis
were not removed at the time of excision. We
shall discuss this topic further in Chap. 23.
(e) Nerve injuries. Besides the sensory deficit in
the territory of the lateral cutaneous nerve of
the thigh already reported, we have seen two
transcient palsies of the femoral nerve. Perma-
nent loss of ilio-psoas function occurred in a
single case. A few patients with transitory
weakness recovered. More strange and unex-
pected are the post-operative sciatic palsies, of
which we have seen five cases (an overall in-
cidence of 2.7%). Two were extended patchy le-
sions, two were localised patchy lesions, and
one was a post-operative extension of a pre-
operative palsy. The pre-operative palsy may be
easily explained by the significant displacement
of the posterior column in this case; unfor-
tunately it was made worse by the operation.
The post-operative cases are more difficult to
understand. Even using this approach, we have
continuous manual control of the toe move-
ments during the operation, especially when
working in the sciatic area. We are sure that at
least in one case the drill bit was responsible for
the palsy, but for the other three cases we have
no valid explanation.
(f) Abdominal wall weakness. We have had two com-
plications with hernia formation requiring opera-
tive treatment, but earlier in one case there had
been an infection of the retro-pubic space which
had necessitated extensive drainage. In the large
majority of other cases the anterior abdominal
wall regained its normal strength. In eight cases
we detected some asymmetry on coughing. So
we had a significant abdominal wall problem in
only two cases, an incidence of 1.1 %.
Surgical Approaches to the Acetabulum
(g) Ectopic ossifications. Out of 164 ilio-inguinal
operations performed elevating the external il-
iac fossa not at all or just a little, to be able to
position clamps astride the anterior border of
the iliac crest, we saw seven cases of ectopic
ossification (an incidence of 4.2%). In detail,
according to the combined Brooker-Letournel
grading system, these were: one 1-5, two 11-5,
two 11-6, and one 111-3. The last-mentioned
case of ectopic ossification, the type 111-3, re-
quired excision at 2 years and the result is ex-
cellent. So, with the regular ilio-inguinal opera-
tion we have had six cases (3.6%) of ectopic
ossification without sequelae or in which the
mobility was graded 5 or 6, and one case
(0.6%) of significant ectopic ossification (III -3)
that was excised.
Out of 14 ilio-inguinal operations that in-
volved more or less exended elevation of the ex-
ternal iliac fossa for imperious reasons of
reduction, we have seen six cases of ectopic
ossification (42 %): one type 1-6, one II -6, two
III-6, one III-5, one 111-3. Excision is scheduled
for one case. Although the number of cases is
small, these figures emphasize once again the
well-known risk associated with elevation of
the external iliac fossa.
This extremely low rate of ectopic ossifica-
tion after an ilio-inguinal operation is one of
the greatest advantages of this approach.
(h) Vascular damage. Apart from damage to the
retro-pubic anastomosis and to the femoral vein,
neither of which had any serious post-operative
effects, we must report one other major vascular
injury. This occurred in an elderly woman who
had a both-column fracture, and during the
operation which necessitated prolonged traction
on the vascular slings, the arterial pulse had not
been checked nor was it checked after removal
of the slings. Within the first 24 h, arterial ob-
struction of both the external iliac and internal
iliac vessels became apparent. Arteriography
confirmed that there was a secondary extensive
thrombosis reaching proximally to the bifurca-
tion of the aorta. Despite arterial thrombectomy
the impaired circulation in the limb, in the but-
tock particularly, did not improve and the pa-
tient died after 3 days. This complication was
the result of careless use of the slings around the
vessels. It is important to check repeatedly the
circulation in the artery and in the event of vas-
cular impairment, immediate thrombectomy is
advised.
Combined Anterior and Posterior Approaches 385

Table 20.3

Complications after ilio-inguinal operations Interval from injury to surgery

0-21 days 21 days to 4 months >4 months Total

Intra-operative:
injury to bladder 1 1
injury to external iliac vein 2 3
intra-articular screws
Early post-operative:
sciatic palsies 4 6
sensory troubles due to injury of lateral
cutaneous nerve of the thigh 36 4 40
femoral nerve palsy 2 2
infections (all kinds) 9 10
haematomata (all kinds) 7 7
skin necrosis 2 2
lymphangitis 3 3
phlebitis 3 3
pulmonary embolism 2 2
Ectopic ossifications 13 13
Sciatic palsies 4 5

(i) Intra-operative bleeding. This has been record-
ed carefully in 62 cases. It was:
less than 0.5 1 in one case;
0.5 -11 in 13 cases;
1 - 2 1 in 11 cases;
2 - 3 1 in 17 cases;
more than 3 1 in 20 cases.
Table 20.3 summarises the complications of the
ilio-inguinal approach.
20.5 Combined Anterior
and Posterior Approaches
When we use either the Kocher-Langenbeck or the
ilio-inguinal approach, we know that they allow us
easily to reach one column of the acetabulum, the
one to which they lead directly, but that access to
the other column will be far from as good.
Transverse fractures and most of the complex types
need to be reduced perfectly, requiring comfortable
access to both columns. When we had at our
disposal only these two types of approaches in
complex cases, we were never sure of being able to
perform the whole reconstruction through a single
approach and were always ready to combine both
approaches. The only rule or the best advice we can
give is to choose first the approach that will give the
best chance of completing the job. So we combined
the Kocher-Langenbeck and ilio-femoral approach
until 1965, since when we have combined the ilio-
inguinal and Kocher-Langenbeck approaches.
Two procedures can be performed at the same
operation or staged within an interval of 8 -10
days. When conducted in one operation, there is in-
evitably considerable loss of blood. Blood replace-
ment must be adequate throughout the operation,
and especially if haemorrhage has been severe dur-
ing the first approach, a pause for adequate resusci-
tation should be made before changing the position
of the patient.
Whenever safety is in doubt, we perform the
double approach as a staged procedure, but this
carries a serious disadvantage: one can never be
sure that the patient will be fit for the second stage
after the lapse of 10 days.
Out of 849 recent lesions of the acetabulum we
have used combined approaches on 26 occasions
(3 0/0). Seventeen of these were conducted during
the same operation and nine were delayed because
some kind of complication occurred after the first
stage. If we wait for longer than 10 days the second
operation may be very difficult due to the callus
formation that has to be completely removed to
allow perfect reduction.
Since we started to use the extended ilio-femoral
approach regularly, double approaches have been
used four times and the extended ilio-femoral 112
times.
386
The access to the pelvic bone provided by the com-
Fig. 20.12.
bination of the ilio-inguinal and the Kocher-Langenbeck
approaches
Some authors recommend performing two
approaches simultaneously. GOSSET and ApOIL
combine the Kocher-Langenbeck and an ilio-
femoral approach. WINDQUIST et al. combine the
ilio-inguinal and the Kocher-Langenbeck. It is
probably true to say that the latter association
theoretically allows the best access to the bone
(Fig. 20.12).
We personally consider that the best position for
the posterior approach is prone, and that the best
position for the anterior approach is supine, so
operating with the patient more or less on his side
can never give the ease of movement offered when
the patient is positioned as normal for the respec-
tive approaches.
The better the decision-making in regard to
choice of approach, the less necessary it should be
to use one of these combined approaches. However,
it is sure to happen on some occasions that the
anterior or the posterior approach does not permit
the carrying out of a perfect reconstruction. In
such a situation, instead of struggling for hours to
try to complete the reconstruction, the best answer
is to reduce and fix the column ideally approached
without regard to the other (except for not pene-
trating it with over-long screws), and then to close
the wound, correctly resuscitate the patient, and
turn him over to perform the second procedure in
the same operation.
20.6 Extended IIio-femoral Approach
Prompted by the need for an approach permitting
the simultaneous exposure of both columns
through a single surgical incision, this exposure was
Surgical Approaches to the Acetabulum
developed over a period of 2 years. We began to use
it regularly in 1975. This is an anatomical approach
that follows a logical neurovascular interval, re-
flecting posteriorly the muscles innervated by the
superior and inferior gluteal nerves as a unit with-
out injury to their posterior neurovascular pedicles.
The approach is divided into three main stages:
the elevation of all the gluteal muscles together
with the tensor fasciae latae by division of both
their origins and insertions;
the division of the external rotators of the hip,
exactly as in the Kocher-Langenbeck approach;
an extended caps ulotomy along the acetabular
lip. When necessary, the internal iliac fossa is
exposed by elevation of the iliacus.
20.6.1 Technique
Prior to operation, a Foley catheter is placed in the
patient's bladder. A standard operating table may
be used, the leg being freely draped and the knee
maintained in a flexed position throughout the en-
tire operation to avoid excessive tension to the
sciatic nerve. However, we much prefer to place the
patient on the Judet fracture table in the lateral
position, and this is our routine practice. A Stein-
mann pin is placed through the distal femur in the
supra-condylar area to allow longitudinal traction,
the knee being flexed at about 45 0 to relax the sci-
atic nerve, exactly as we do when we use the
Kocher-Langenbeck approach (Figs. 20.13 and
20.14).
A pelvic post, which can be moved along a col-
umn, is positioned horizontally between the thighs.
It can be raised or lowered during the operation
from the head of the table. When required, this
post can exert pressure on the inner aspect of the
thigh and be used effectively to disimpact the
femoral head from its centrally dislocated position
and to maintain it in the right place while the re-
construction is performed.
The incision is in the form of an inverted J,
starts at the posterior superior iliac spine, proceeds
along the whole of the iliac crest to the anterior
superior iliac spine, and from there descends
straight toward a point 2 cm lateral to the outer
border of the patella, extending half-way down the
thigh (Fig. 20.15).
The periosteum of the iliac crest is sharply in-
cised along the top of the crest, and the gluteal
fascia is elevated from the lateral aspect of the crest
Extended Ilio-femoral Approach
Fig. 20.13. Extended ilio-femoral approach. Front view of patient in the lateral position on the ludet-Tasserit radiolucent table
(this facilitates the repair). In continuity the origins
of the gluteal muscles are elevated from the external
aspect of the iliac wing subperiosteally. Next to the
anterior superior spine, the tensor fasciae latae are
elevated together with the glutei (Fig. 20.16).
In this way we reach anteriorly the upper aspect
of the hip capsule and posteriorly the superior
border of the greater sciatic notch, the approach of
which is very well felt by the rugine, and elevation
is then stopped for a while. Some very strong
fibrous origins of the gluteus maximus may need
sharp section, particularly along the crista glutei.
The area is temporarily packed with a large swab.
The anterior part of the approach is then em-
barked upon. From the anterior superior iliac
spine, the fibrous sheath of the tensor fasciae latae
is incised along the belly of the muscle; below it the
fascia lata is split down to the lower end of the inci-
sion. Doing this, some of the external branches of
the lateral cutaneous nerve of the thigh are cut, but
the further dissection remains inside the sheath of
the tensor, avoiding as much as possible dividing
other branches of the nerve. The muscle is freed
from its sheath along its anterior border and deep
387
aspect and then retracted upward and posteriorly as
allowed by its complete elevation from the iliac
wing. Small vessels from the superficial circumflex
artery are divided and coagulated close to the bone
between the two anterior iliac spines.
Elevation of the tensor fasciae latae and gluteal
muscles from the iliac wing progresses inferiorly to
the reflected tendon of the rectus femoris and the
superior aspect of the hip capsule. Posteriorly this
dissection continues to complete the freeing of the
superior border of the greater sciatic notch. This
stage of the approach is critical and has to be per-
formed with extreme care. The superior gluteal
neurovascular bundle is very close to the bone, may
be suspended from it by a ligament, and it can even
happen that it is identified within the fracture line
of the posterior column adjacent to the greater
sciatic notch. This pedicle can easily be injured
either by the surgeon's actions or by uncontrolled
displacement of the upper part of a detached
posterior column, which may be as sharp as a
knife. When the pedicle is safely under control, it
is protected with a wet swab. The bleeding nutrient
foramina of the iliac wing are plugged with wax.
 388 Surgical Approaches to the Acetabulum

B
Extended Ilio-femoral Approach 389

,,
I

,,I
I

,,
,,,

Fig. 20.16. Extended ilio-femoral approach

from the lateral femoral circumflex itself, indicates

the position of the artery that will have to be ligated
Fig. 20.15. Extended ilio-femoral approach. See text
later on.
Once the circumflex vessels are divided and
retracted, through a little layer of fat tissue the
Returning to the front part of the incision,
retraction of the tensor fascia lata posteriorly ex-
poses the thin sheath of the rectus femoris, which
is incised longitudinally and horizontally. Retract-
ing the rectus femoris downwards and medially ex-
poses a very strong and usually not transparent
aponeurosis that is carefully incised longitudinally
and horizontally, allowing visualisation of the
lateral femoral circumflex vascular pedicle, which is
isolated, clamped, ligated and divided (Fig. 20.17).
When the rectus femoris is retracted, a constant
small vascular pedicle reaching the lateral border of
the muscle always has to be coagulated; in fact, it
traverses this strong aponeurosis and, as it comes

Fig. 20.14A,B. Extended ilio-femoral approach. A Back view of
patient in the lateral position on the Judet-Tasserit radiolucent
orthopaedic table. B Detail: the special bracket for raising or
lowering of the thigh (here the device is raised - it was lowered
in A)
Fig. 20.17. Extended ilio-femoral approach. Initial elevation and
posterior retraction of tensor fasciae latae and gluteal muscles.
The lateral femoral circumflex vessels are seen through the inci-
sion in the overlying fascia. (Reproduced by permission of Joel
MATTA)
390 Surgical Approaches to the Acetabulum

sheath of the ilio-psoas muscle is reached. This thin

sheath is longitudinally incised and the ilio-psoas
muscle fibres are elevated from the anterior and in-
ferior aspects of the hip capsule with an elevator,
the muscle being retracted downward. The reflected
tendon of the rectus femoris is then totally resected.
The gluteals and tensor fasciae latae together are
retracted posteriorly and upwards, and access to the
wing is total. Now the superior and anterior aspects
of the capsule have to be completely exposed, pro-
gressively dividing numerous aponeurotic struc-
tures well described by the anatomists.
Finally the anterior inter-trochanteric crest is
reached, and above it the attachment of the gluteus
minimus tendon to the anterior aspect of the
greater trochanter. The gluteus minimus tendon is
isolated and held in a big clamp prior to section.
The tendon is divided 3 - 5 mm from its bony inser-
tion after tagging of each of its borders with two
long, easily recognisable sutures to facilitate its
future repair (Fig. 20.18). If there are any dif-
ficulties in identification of the tendon of the
gluteus minimus, the inferior part of the incision
can be returned to and the vastus lateralis sought,
with its so easily identifiable vertical white fibres:
following these upwards leads infallibly to the inter-
trochanteric crest, and above it, nearly in continui-
ty with the vastus fibres, the tendon of origin of the
gluteus minimus can be identified.
Retraction of the gluteus minimus gives access
to the external aspect of the greater trochanter and
to the deeper aspect of the gluteus medius tendon.
Internal rotation of the hip helps to individualise
the tendon. The gluteus medius tendon is totally
transgressed in the same manner as for the gluteus
minimus tendon (see Fig. 20.18).
B

Fig. 20.19A,B. Extended ilio-femoral approach. The exposure

of the external aspect of the innominate bone is now complete

The result of this wide dissection is a massive

Fig. 20.18. Extended ilio-femoral approach. The gluteus
minimus tendon has been completely transected and the medius
tendon partly transected. Posterior dissection along the iliac
wing has reached the greater sciatic notch. (Reproduced by per-
mission of Joel MATTA)
flap comprising the gluteal muscles and the tensor
fasciae latae together with their neurovascular
bundles, which are protected by a large wet swab.
The flap is retracted posteriorly to expose the
posterior aspect of the hip covered with the external
rotators, which will be cut as in the Kocher-
Langenbeck approach, their identification being
facilitated by internal rotation of the hip
(Fig. 20.19). The piriformis tendon is first severed
and its tendon temporarily attached to the lateral
 Extended Ilio-femoral Approach
margin of the gluteal flap. Lifting up the muscle
allows identification of the sciatic nerve and gives
access to the anterior border of the greater sciatic
notch.
Next the tendon of the obturator internus is
sought, tagged with a stitch, and the obturator in-
ternus and the gemelli are divided 2 cm away from
the trochanter. Lifting up the muscle from the hip
capsule and the retro-acetabular surface makes it
possible to reach and open the constant underlying
synovial bursae. That allows visualisation of the
tendinous bands of origin of the obturator internus
and access is gained through the lesser sciatic notch
to the interior of the true pelvis (see Fig.20.4E).
The beak of the sciatic nerve retractor is then in-
serted into the lesser sciatic notch. Tension applied
by the assistant to the stitch tagging the obturator
internus protects the nerve from the retractor.
The quadratus femoris is usually left undis-
turbed; it protects the ascending branch of the
medial femoral circumflex artery. If necessary, the
muscle is divided along its mid-portion or elevated
from its ischial insertion so that the artery is better
protected.
20.6.2 Application
This standard extended ilio-femoral approach
allows access to (Fig. 20.20):
the whole posterior column down to the upper
part of the ischial tuberosity;
the whole external aspect of the ilium;
the anterior column as far medially as the ilio-
pectineal eminence, but no further.
Whenever it is necessary to inspect the interior of
the joint, either to check the quality of reduction,
or to look for debris, a capsulotomy along the
margin of the acetabulum is performed and extend-
ed as needed (it may involve the upper two-thirds
or three-quarters of the capsular insertions)
(Fig.20.19B). Then the combination of longitu-
dinal traction and lateral traction applied with the
Judet table will distract the femoral head from the
acetabulum and allow visualisation of the inside of
the joint.
If access to the internal iliac fossa appears
necessary, especially in the performance of delayed
reconstructions, the internal iliac fossa can be ex-
posed by the elevation of the abdominal muscles
from the iliac crest in continuity with the iliacus
from the internal aspect of the iliac wing
391
(Fig. 20.21 A, B). Prior to this, the origins of the
sartorius and the inguinal ligament are elevated
sub-periosteally from the anterior superior iliac
spine. Further anterior exposure is achieved by
dividing the origin of the direct tendon of the rec-
tus femoris and by distally elevating the iliacus
down to the psoas groove. The elevation of the il-
iacus is carried out inferiorly and medially as far as
necessary, so at the maximum the sacro-iliac joint
and the posterior half of the pelvic brim can be
reached.
We now have sufficient experience in these eleva-
tions from both aspects of the iliac wing, and
without having encountered any problems related
to any supposed devascularisation of the iliac wing,
to promote this technique when it is needed. Never-
theless, as usual, maximum care should be taken to
preserve bone vascularisation as much as possible,
and the following remarks of Joel MATTA should
be kept in mind:
"In case of a T-shaped fracture in which no frac-
ture lines traverse the iliac wing, both sides of
the iliac wing may be exposed without fear of
devascularizing the bone. For a both-column
fracture, however, the anterior column fracture
usually traverses the anterior portion of the
wing to the iliac crest. Here complete exposure
of both sides of the iliac wing may easily devas-
cularize a large segment of the anterior column.
In order to prevent this, soft tissue pedicles must
remain attached to the anterior column for vas-
cularity. At the minimum, the direct head of the
rectus femoris and anterior hip capsule should
be left attached to the anterior column?'
20.6.3 Closure
Closure is simple and anatomical. Several suction
drains (three or four) are essential and must be
placed in all areas where haematoma formation is
possible: the retro-acetabular surface, the external
and internal iliac fossae, the true pelvis, the anteri-
or aspect of the hip.
The hip capsule, if divided and if possible, is
repaired with interrupted sutures.
The tendons of obturator internus and pirifor-
mis are re-attached to the stumps of their tendons
of origin from the greater trochanter. Their repair
is facilitated by positioning the hip in external rota-
tion. We quite often suture both muscles together
to provide a muscular barrier between the sciatic
nerve and the fixation plates applied along the
posterior column.
 392
A B
Fig. 20.20. A Access offered by the standard extended ilio-femoral approach. B The same access allows control of the quadrilateral
surface
With the hip in the same position, the gluteal
tendons are repaired anatomically, guided by the
tag sutures; five or six stitches are advised for each
tendon.
If severed, the direct head of the rectus femoris,
the sartorius and the inguinal ligament are re-
attached by sutures placed through drill holes.
The gluteal fascia is reapproximated to the
abdominal fascia along the iliac crest and closed
antero-laterally over the thigh. The re-attachment
of the gluteal fascia to the abdominal muscles can
be greatly facilitated by abduction of the hip. If the
abdominal muscles have been elevated, they are re-
attached to the iliac crest together with the gluteal
fascia by trans-osseous sutures.
20.6.4 Dangers
(a) Superior gluteal neurovascular bundle. The
care necessary in freeing this from the superior
border of the greater sciatic notch was de-
Surgical Approaches to the Acetabulum
scribed earlier (Sect. 20.6.1). It is of the utmost
importance.
Injuries to the superior gluteal artery oc-
curred in four cases (3.50/0). They may be
caused by the surgeon's tools, by uncontrolled
displacement of a fragment, or perhaps by
compression from retractors or the reactivation
of vascular wounds from the original injury by
mobilisation of the obstructing clots during
freeing of the vessels.
It is certainly to injuries of this vascular pedi-
cle that some cases of post-operative muscular
necrosis described by other authors, some fol-
lowed by infection, are due. So far we have not
seen this complication in a single case. The same
goes for Joel MATTA, Jeff MAST and Claude
MARrIMBEAU, and the four of us can report
more than 400 extended ilio-femoral approaches
between us. In my opinion, the care that has to
be taken to free and preserve these vessels and
nerves can never be emphasised enough.
If a vascular wound occurs, however, what is
to be done?
 Extended Ilio-femoral Approach
A
Fig. 20.21A-C. Extended ilio-femoral approach. B Exposure of
the internal iliac fossa pelvic brim and anterior brim of acetabu-
lum. (Reproduced by permission of Joel MATTA). C This
An arterial haemorrhage does not need
suturing, ligation or clipping immediately. The
best treatment is to pack it tightly with wet
swabs and wait as long as possible. Usually
haemostasis will occur; if not, one or more
stitches performed at the time of closure will
solve the problem, with resorbable gelatin
swabs left in situ.
Venous haemorrhage is more difficult to
control. Again, the first answer is packing, and
further ligation or clipping will gain control
over the bleeding.
The reason we do not advise trying to con-
trol bleeding immediately is because the gluteal
nerve is very difficult to identify or isolate,
especially if there is bleeding. Consequently,
there is a danger that ligating the vessels or at-
tempting to control the bleeding with sutures or
clips may include the superior gluteal nerve
totally or in part, with subsequent gluteal
muscular palsy.
(b) Sciatic nerve injuries. The sciatic nerve is iden-
tified as easily through this approach as
393
approach to both aspects of the iliac wing allows control of both
extremities of a transverse fracture line with proper clamps, and
permits one to go right round the bone
through the Kocher-Langenbeck, and there are
no difficulties in retracting and safeguarding it.
As the lesions treated through this approach in-
volve the posterior column less than do lesions
treated through the Kocher-Langenbeck, the
sciatic nerve is less at risk in this approach.
20.6.5 Complications
In a series of 114 extended ilio-femoral operations,
38 of which included elevation of the internal iliac
fossa, we have had the following complications:
(a) Skin necrosis. There were five cases (4.3070), in-
volving the posterior part of the incision, the
part that follows the posterior part of the iliac
crest, on which the patient rests.
(b) Haematomas. We saw nine haematomas
(7.9%). One instance was aspirated, seven were
evacuated surgically, and one was not treated.
They occurred at the level of the external iliac
fossa.
394
(c) Intra-articular screw. In spite of the intra-
operative testing, we inserted one screw into the
joint in a both-column fracture. It was removed
after 4 months.
(d) Nerve injuries. Among these 114 cases we had
no case of immediate post-operative sciatic
palsy. However, there was one case of sciatic
palsy of secondary onset that we are unable to
explain.
Naturally, sensory troubles in the territory
of the lateral cutaneous nerve of the thigh oc-
cur regularly as some of the branches of this
nerve are compulsorily cut in this approach.
We have also had a case of complete, non-
regressive palsy of the superior gluteal nerve
(0.8070) that remains unexplained as there was
no trouble with the pedicle at the time of
surgery. It was secondarily treated by an ilio-
psoas transfer according to MUSTARD'S tech-
nique. The pre-operative condition of the nerve
was, in fact, not checked.
(e) Thrombo-embolism. We have had one case of
phlebitis and one of pulmonary embolism.
These low figures testify to the efficiency of the
anticoagulant therapy we use.
(f) Infections. So far we have had no case of im-
mediate post-operative infection, but - to
repeat this once again - we give prophylactic
antibiotic treatment as a matter of routine.
We have seen one case of late or delayed in-
fection, but the patient had been previously
Table 20.4
Interval to surgery Brooker ossification type
after extended ilio-femoral
approach
0-21 days I 1
(49 cases) II
III
IV
Subtotal
21 days to 4 months
(42 cases) II
III
IV
Subtotal
>4 months I 2
(23 cases) II
III
IV
Subtotal
Surgical Approaches to the Acetabulum
operated elsewhere through an ilio-inguinal in-
cision, and the intra-operative culture taken
from the track of the previous incision, which
we opened to remove the material, was positive,
so this infection cannot be charged to the ex-
tended ilio-femoral approach.
(g) Ectopic bone formation. Ectopic bone forma-
tion is a frequent occurrence after this ap-
proach, as after all other extensile approaches.
Table 20.4 reports these cases according to time
to operation, Brooker type and d'Aubigne
mobility grading.
Among the 48 patients operated on within
21 days after injury, 28 have developed ec-
topic ossification (57%). Of these, 21 (42%)
have retained grade 5 or 6 mobility and 7
(14%) have reduced mobility. We graded the
latter as follows: one 11-4, two IV-4, two
IV-3, two IV-2. In four cases bone forma-
tions were excised, two are waiting for exci-
sion. Among these 48 patients, 13 received
preventive treatment considered today as ef-
ficient (diphosphonates excluded): out of
four who received indomethacin, two
developed ectopic ossifications (one IV-5
and one 11-6), and of the nine who were
given indomethacin plus radiotherapy, none
developed ectopic ossification. Thus, out of
35 patients who did not receive preventive
treatment, 26 developed ectopic ossifica-
Mobility (d'Aubigne grading) Total
2 3 4 5 6
1
9 10
4 1 5
2 2 2 2 4 12
2 2 3 6 15 28
5
2 4
2 11 15
2
4 5
Addendum: The Kocher-Langenbeck Extended to a Triradiate Approach
tions (74%), which in the seven severe cases
detailed above reduced mobility (20070). Of
these, bone excision was undertaken in four
cases and is planned in another two. The ef-
ficiency of the preventive treatment will be
discussed later in Sect. 25.5.4.
Out of 42 patients operated on between 21
days and 4 months after injury, 15
developed ectopic ossifications (35%). Of
these, 11 have mobility rated at grade 6
(26%) and four have impaired mobility
(9%) rated as follows: one IV-2, two III-4,
one IV-3. Thirteen of these patients had
received preventive treatment and developed
no ectopic ossification (eight had in-
domethacin, five indomethacin and
radiotherapy). Among the 29 patients
without preventive treatment, 15 developed
ectopic ossifications (51 %) and in four of
these mobility was reduced (13%). The rate
of mobility-reducing ectopic ossifications in
patients without preventive treatment was
13%, compared to 20% among the patients
operated on within 21 days after injury.
Among the 23 patients operated on through
this approach more than 4 months after in-
jury, only five developed ectopic ossifica-
tions (21070) and none had reduced mobility.
Of the 23 patients, the seven who received
preventive treatment had no ectopic ossifi-
cation, so out of 16 patients without preven-
tive treatment, five developed ectopic ossifi-
cation without clinical sequelae (31 %).
These figures may be regarded as demonstrating
that the more time has elapsed after the accident,
the smaller and less clinically significant are the ec-
topic ossifications.
20.7 Post-operative Care
Antibiotics are given 1 day before and 8 - 10 days
after surgery. Anti-coagulation therapy is routine.
The leg is kept in slight abduction for a few days.
Active flexion of the hip is encouraged from the 1st
post-operative day. Continuous passive motion is
begun on the 2nd day.
Indomethacin and radiation are now used
routinely to try to prevent ectopic bone formation.
Full weight-bearing is permitted at 75-90 days.
During the partial or non-weight-bearing period
395
and as long as necessary after full weight-bearing is
allowed, physical therapy is directed toward
strengthening the muscles, particularly the abduc-
tion musculature.
20.8 Summary of the Use
of Different Surgical Approaches
The operations on 849 fractures of the acetabulum
have been performed through the various operative
approaches described, as detailed in Table 20.5.
20.9 Addendum: The Kocher-Langenbeck
Extended to a Triradiate Approach
The triradiate approach promoted by Dana MEARS
is another extensile approach that allows an access
to the innominate bone comparable to that offered
by our extended ilio-femoral approach. However,
the dissection of the gluteus maximus along its
fibres, as in a Kocher-Langenbeck approach, leaves
in place a segment of the gluteus maximus. Con-
sequently, unlike our approach, this approach does
not allow mobilisation of the superior gluteal
neurovascular bundle; hence, there is no easy access
to the sacro-iliac joint as there is through an ex-
tended ilio-femoral approach.
Personally we have never used the triradiate as a
first approach. We do think, however - and have
done this on a few occasions - that a Kocher-
Langenbeck approach may conveniently be trans-
formed into a triradiate when further access to the
anterior column appears necessary intra-operative-
ly. As usual, when we perform a Kocher-Langen-
beck approach the patient is lying prone on the
orthopaedic table. At any time during the opera-
tion, if the access offered by the approach appears
insufficient, instead of only dividing the gluteal
tendons or elevating the gluteal muscles as far as
possible from the external iliac fossa (which
damages the muscles), we may decide to extend the
Kocher-Langenbeck into a triradiate approach.
Dana MEARS himself wrote in his book that "the
triradiate concept also provides a way in which an
initial exposure with the posterior and distal limbs
(Le. Kocher-Langenbeck exposure) can be enhanced
rapidly if the initial visualization proves to be in-
adequate. "
396 SurgIcal Approaches to the Acetabulum

Table 20.5

Fracture type Approaches used in 849 cases

..c:: ..c::
u
ro u
ro
...o0- ...o0- o
0-
~
ro 0-
...o ...oro
";:: ";::
E OJ
~ E
ro
...
0-
OJ
...
OJ
..c:: ..c::
(5 (5

Posterior wall 88 95 12 10
Posterior column 14 13 1
Anterior wall 5 9 2
Anterior column 2 5 15 5 2
Transverse 9 16 5 2 7 4 6
T-shaped 16 14 4 2 1 10 7 5
Transverse and 45 75 8 3 21 17 2
posterior wall
Posterior column 11 18
and posterior wall
Anterior column and 5 5 34 2 3 2 2
posterior hemitransverse
Both-column 26 14 71 6 9 5 2 37 8 10 8
Total 216 245 15 147 17 14 29 16 4 76 38 17 9

From the angle of the Kocher-Langenbeck route,
a proximal extension of the incision is made up to
the anterior superior iliac spine. The angle formed
by the superior limbs is about 120°. We have been
permitted to reproduce from Dana MEARS' book
the description of the surgical technique as follows:
"For exposure of the anterior column the fascia
lata and myofascia of the tensor fasciae latae are
incised from the anterior superior spine to the
greater trochanter. The anterior border of the
tensor fasciae latae muscle is sharply incised
from its fascia so that the entire muscle can be
retracted superiorly and posteriorly along with
the cutaneous flap. The origins of the tensor
fasciae latae and the gluteus medius and
minimus are incised from the iliac crest. Sub-
periosteal elevation of the gluteus medius and
mmlmus is undertaken from anterior to
posterior and distally to the capsule of the hip
joint.
On the lateral aspect of the proximal femur
the interval between the gluteus medius and the
vastus lateralis is identified and a transverse inci-
sion is made in the periosteum. With an osteo-
tome or an oscillating saw the greater trochanter
with the attached gluteus medius and minimus is
osteotomized and reflected craniad. With
curved heavy scissors working from proximal to
distal and anterior to posterior, the gluteus
medius and minimus are sharply elevated from
the capsule of the hip joint. The capsule is
carefully preserved. With the gluteus medius
and minimus reflected superiorly, the lateral
pelvis and roof of the acetabulum are ap-
proached by elevation of the periosteum. The
abductor muscle mass is anchored superiorly
and posteriorly by the use of two Steinmann
pins inserted into the illium 2.5 cm and 5.0 cm
above the roof of the greater sciatic notch.
When the capsule is intact it is incised sharp-
ly from its origin around the acetabular rim to
permit access to the hip joint.
As part of the exposure of the anterior col-
umn, often it is necessary to extend the dissec-
tion medially to the ilio-pectineal eminence and
Addendum: The Kocher-Langenbeck Extended to a Triradiate Approach
posteriorly to the sacroiliac joint. The anterior
limb of the cutaneous incision is continued 6 cm
to 8 cm medial to the anterior superior iliac
crest. The insertion of the abdominal muscula-
ture on the anterior iliac crest is sharply incised.
The iliacus muscle is elevated sub-periosteally
from the inner table of the ilium and retracted
medially. The exposure continues posteriorly to
visualize the anterior aspect of the sacro-iliac
joint. In addition to exposure of the sacro-iliac
joint, the inner wall of the ilium and the
superior pubic ramus to within 2 cm of the sym-
physis can be visualized.
To augment the anterior exposure, the origins
of the sartorius and the direct and indirect heads
of rectus femoris are incised from the anterior
superior and anterior inferior spines and the hip
397
capsule, respectively. The aponeurosis of the ex-
ternal oblique and the inguinal ligament are in-
cised as previously described under the ilio-
inguinal approach. The interval between the
psoas muscle and the external iliac vessels is
developed. After adequate identification of the
external iliac vessels, the interval between them
and the spermatic cord or round ligament and
the rectus abdominus is developed. Sub-perios-
teal elevation is undertaken along the superior
pubic ramus by the use of each of these longitu-
dinal intervals. By medial or lateral retraction of
the ilio-psoas, external iliac vessels, and rectus
abdominus virtually the entire length of the
anterior column can be visualized and the
quadrilateral surface can be palpated?'
(MEARS 1988)
21 Operative Treatment of Displaced Fractures
Within Three Weeks of Injury

Experience shows that repair of fractures of the tient. CT scanning should now be regarded as
pelvis is rapid, so that beyond the 21 st day there are indispensable. Only through the discipline of
problems imposed by fairly advanced union, the joint study of the plain radiographs and the CT
fracture lines becoming obscured by abundant sections will an exact anatomical diagnosis be
callus which fixes the fragments in malposition. possible; 3-D reconstructions are very helpful
Perfect cleaning of the fracture surfaces, an ab- here.
solute requirement for anatomical reduction, is (d) Prophylactic antibiotic treatment covering the
much more difficult. In practice, until the 14th widest possible spectrum of micro-organisms is
post-accident day operative intervention is general- begun at least 24 h prior to surgery, earlier if
ly straightforward, but it becomes progressively less there are skin wounds or a Morel-Lavalle le-
so during the 3rd week, at the end of which healing sion, or if a urinary catheter has been placed as
is generally fairly advanced. an emergency procedure, which is very fre-
quent. Anticoagulation therapy is begun pre-
operatively if the operation is delayed, i.e. in
our practice in nearly all cases, as most of our
21.1 Pre-operative Care patients are referred. The type of anticoagulant
chosen depends on the patient's condition
(Sect. 21.5.3).
With regard to the fractured acetabulum itself,
(e) Indomethacin treatment to prevent ectopic
leaving aside aspects of treatment of other injuries
ossification (if a Kocher-Langenbeck or an ex-
and general resuscitation, pre-operative care and
tended ilio-femoral approach is scheduled) is
preparation comprise:
begun the day before surgery.
(a) Local care of the injured limb. If a posterior
dislocation has been reduced and is stable, sim-
ple maintenance of the limb in slight abduction 21.2 Choice of Surgical Approach
and slight external rotation prevents redisloca-
tion. While waiting for operation, in our view, This is a fundamental issue and the errors of
traction is unnecessary whatever the type of approach which we have committed have been
dislocation. repeated reminders of the care that is necessary.
(b) Skin preparation is similar to that for all or- ,The aim is to choose a surgical approach which
thopaedic operative procedures but it must be permits the complete repair of the acetabulum, or
emphasised again that when an ilio-inguinal at least, to choose that which will give the best
approach is anticipated, 24-48 h should be chance of achieving this aim. Sometimes the ideal
allocated for scrupulous preparation by shav- cannot be achieved, and, as stated in the preceding
ing, repeated washing and application of io- chapter, when we choose either the Kocher-
dine, and antibiotics. Langenbeck or the ilio-inguinal approach, which
(c) Radiographic study. During the period of lead electively to one column of the acetabulum, it
preparation of the injured person, the study of is not possible to be certain that the whole
the four standard views must not be rushed. osteosynthesis can be performed through the one
Poor films must be repeated, the surgeon con- incision: a second operation may be necessary,
tributing in the radiology department if dif- preferably immediately but sometimes delayed. To-
ficulty is encountered in positioning the pa- day when a double approach is very likely to be
400 Operative Treatment of Displaced Fractures Within Three Weeks of Injury
necessary we prefer to choose the extended ilio-
femoral approach.
The right choice of approach depends on four fac-
tors:
the anatomical type of the fracture. It is here
that a knowledge of the classification is in-
dispensable;
the age of the fracture. The fast callus forma-
tion increases the operative difficulties with
time: all the callus needs to be removed to allow
a perfect reduction;
the extent of access to the innominate bone of-
fered by each surgical approach;
the rate of ectopic bone formation following the
different approaches, which seems to be in-
dependent of the fracture type.
For five varieties of fractures there is a clear best
approach; the other five entail making a choice.
21.2.1 Kocher-Langenbeck Approach
This is suitable for:
all varieties of fractures of the posterior wall of
the acetabulum;
all fractures of the posterior column;
associated fractures of the posterior column
and the posterior wall.
21.2.2 IIio-femoral Approach
This may be used for high fractures of the anterior
column involving one separated fragment. These
can in fact be treated just as well through the ilio-
inguinal approach, but the ilio-femoral approach is
used if it is certain from the radiological diagnosis
that an easy reduction is possible through this
easier approach.
21.2.3 IIio-inguinal Approach
This is suitable for:
anterior wall fractures;
anterior column fractures of all types;
associated anterior column and anterior wall
fractures, which we have included in the anteri-
or wall fractures group, as they just add to the
latter a fracture of the ischio-pubic ramus that
we do not fix.
21.2.4 Fracture Types for Which There
is a Choice of Approach
21.2.4.1 Transverse, T-Shaped and Associated
Transverse and Posterior Wall Fractures
These three fracture types have in common a
transverse fracture component. The reduction of
this transverse fracture is critical, and the higher it
is situated on the acetabular sphere the more exact
its reduction has to be: a fault in reduction in an in-
fra-tectal transverse fracture is much less important
than the same thing happening in a trans-tectal
transverse fracture involving the weight-bearing
area. To perfectly reduce a transverse fracture we
have to have as easy control as possible of both ex-
tremities of the fracture line, and the choice of
approach depends to a great extent on this. Conse-
quently, for these three varieties of fractures we
currently advise using:
the extended ilio-femoral approach if the trans-
verse fracture component is trans-tectal, as this
is the only approach which will allow complete
and perfect intra-articular control of the frac-
ture line;
the Kocher-Langenbeck approach if the trans-
verse component is juxta-tectal or infra-tectal
and the patient is being operated on no more
than 15 days after injury. After this date we ad-
vise the extended ilio-femoral approach.
In associated transverse and posterior wall frac-
tures in which the transverse and the posterior wall
component must be repositioned, only these two
approaches can be used.
However, in cases of transverse or T-shaped frac-
tures, in addition to these two approaches, which
are the most often used, the ilio-inguinal approach
has some uncommon indications. In these kinds of
fractures it sometimes happens that the displace-
ment of the transverse fracture line is absent or on-
ly slight posteriorly, while the displacement at the
level of the pelvic brim is significant: this is the
situation where rotation of the ischio-pubic frag-
ment or fragments has occurred around a roughly
horizontal axis. In these latter instances it is plainly
logical to choose the ilio-inguinal approach.
In T-shaped fractures, if we choose the Kocher-
Langenbeck or the ilio-inguinal approach and
perfect reduction of the segment of the column op-
posite to the approach cannot be achieved, we then
perform the other approach.
Fracture Types for Which There is a Choice of Approach 401

Table 21.1

Fracture type ~
.~
'" ffi 0;
.... .... ~ 0
:a ....
~
'-
0
'";. '"
·c
.5'"
u
~ :0 0;
~ '" "5 S
:0'"
~ u
.5- '"
'"
Cl
-;.,
os "0 '"00 00
0. "0
2
0
..c::
u
os :a
u
s;., 0; ..c::
u
~ ~
os ~ ~
.,..Su
~
2 .... os
.,.u ~
0
.... '"0.
os ....
os '"
;(
0
....
0;
....
0;
.... '"0
~
0
0. ~
'" .-'"
u 0 0. 0 0
:a '"
0. ~
os -5.... 0. E E
'os"
~ <lJ
.n <l)

eoou~ .n
~

'"
.n
~

'"
....
0
·c
0 5
"0
~
os ~
os
....
.g
<l)
'- ~ '"
<l)
..c::
u '"u
..c::
§,g
00
~
00
~
~ 0; 0; 0; 0;
~
.... ~ ~ os
os os
a'"os
0 0
~ .... ....
5 ·s00<= ·s00 ;;l
~
Cl ·s00
~
....J .-
,
....
....
.0
-;l
....
-;l
.... 0. E '"
0..
"0
<l)
"0
'" '"
"g.s
0.
0.
0.
0.
-5 os os
<l) _
.... "0
u'"
..c:: .... '"u
..c:: ..c:: .... ~ .8 .8 .8 .... ~

'" '" '"

<l)
u ..c:: ·8 ..c:: ~ <l) U 0 0

~ ~ ~ ~
0. ;(
0
~ ~
0
~
0
~ 0 (/J 0 (5 '-ll
- os
~::: E-
~ ~
E-

Posterior wall 58 67 3
Posterior column 8 5
Anterior wall 3 6
Anterior column 1 4 10 5 2
Transverse 6 8 4 2
T-shaped 9 8 3 2 3 1 4
Transverse and 34 54 6 13 3 2
posterior wall
Anterior column and 5 5 34 2
posterior hemitransverse
Posterior column 10 10
and postenor wall
Both-column 25 6 66 6 8 2 26 3 8 7
Total 156 158 3 12 129 16 13 2 5 3 42 7 14 8

The complete statistics relating to our series do
not exactly reflect the indications outlined above,
which are those of today, because up to 1975 we did
not have an extensile approach at our disposal.
Table 21.1 shows how these three categories of
fractures concerned were approached from 1958 to
1989.
21.2.4.2 Associated Anterior and Posterior
Hemitransverse Fractures
These are approached through the ilio-inguinal in-
cision if operated on up to 15 days after injury.
Later than this the ilio-inguinal approach is still ad-
vised if the posterior transverse component is un-
displaced or only slightly displaced, but if the
posterior component is significantly displaced and
needs to be reduced, the extended ilio-femoral
approach has to be used.
21.2.4.3 Both-Column Fractures
For the most complex cases the question of choice
of approach is now fairly clear.
(a) Kocher-Langenbeck Approach. If the fracture
line in the iliac wing goes to the anterior border
of the iliac bone, it is possible, at the price of
some stripping of the lower part of the external
iliac fossa, to control the anterior column frac-
ture from behind through the Kocher-
Langenbeck approach. We did that in 31 cases
up to 1975. In cases of extreme difficulty there
are two possibilities:
close the posterior approach and subse-
quently perform an ilio-inguinal approach;
or
enlarge the Kocher-Langenbeck incision by
dividing the tendons of the gluteal muscles
from their trochanteric insertions or by
transforming the Kocher-Langenbeck into a
triradiate approach.
Nowadays, even for this category of both-col-
umn fracture we prefer to use the ilio-inguinal
or, exceptionally, the extended ilio-femoral
approach.
(b) IIio-inguinal approach. When the fracture line
in the iliac wing extends to the iliac crest and
the posterior column fragment is in one piece,
402 Operative Treatment of Displaced Fractures Within Three Weeks of Injury
the ilio-inguinal approach is chosen. However,
it is true to say that, using this approach, an as-
sociated pure posterior wall fragment has
sometimes been neglected when it did not
appear to justify a subsequent posterior ap-
proach. Complete reconstruction through the
ilio-inguinal approach has proved to be im-
possible in some instances where the posterior
column was incarcerated inside the pelvis and
impossible to mobilise from the front, or when
a bone fragment was interposed and/or im-
pacted into the fracture line dividing the
posterior column and was then impossible to
mobilise from the front. In these cases a subse-
quent Kocher-Langenbeck operation was nec-
essary. These are circumstances where radiolog-
ical assessment is difficult, but it will probably
be possible to analyse and recognise them with
the help of CT scanning.
(c) Extended Ilio-femoral Approach. This approach
is indicated in both-column fractures when:
the posterior column is separated in several
pieces;
there is sacro-iliac involvement;
there is comminution of one or several frac-
ture lines;
a double approach is very likely to be neces-
sary for radiological reasons or features;
the fracture is operated upon more than 15
days after injury.
21.3 Operative Details
21.3.1 Where and How to Insert Screws
Before operating on fractures of the acetabulum, in
addition to the gross architecture of the innominate
bone, its finer structure must be studied with regard
to the fixation of screws to be used alone or with
plates. The bone is so variable in form and structure
that a thick dense zone can change to something
much thinner and less solid within 1 - 2 cm.
The pelvic bone is of spongy texture of varying
thickness limited by two rather thin cortices, except
for the posterior part of the iliac wing, which is fair-
ly uniformly thick to within 4 - 5 cm anterior to the
posterior iliac spines, and whose inferior part offers
a zone even more solid and dense (the sciatic butt-
ress). The most solid areas of dense bone are shown
in Fig. 21.1 A. These are:
the iliac crest and its immediate neighbourhood;
the superior aspect of the sciatic buttress, which
corresponds to the last 2 - 3 cm of the ilio-pec-
tineal line in front of the sacro-iliac joint and the
adjacent bone of the internal iliac fossa in an
area 1.5 - 2.0 cm wide;
the brim of the true pelvis and its immediate
neighbourhood (special care should be taken
when using the section adjoining the acetabu-
lum);
the anterior border of the iliac wing and strips of
the adjacent surfaces about 2 cm wide.
The thickest parts, shown in Fig. 21.1 B, comprise
mostly cancellous bone. Screws hold well but they
must be long, properly sited and accurately orien-
tated in these areas, which are:
the posterior column, at the level of the posterior
wall of the acetabulum and the ischial tuberosity;
above the roof of the acetabulum, in a zone
limited above by a line extending from one
finger-breadth above the superior border of the
greater sciatic notch as seen from the outer
aspect of the bone, or the posterior limit of the
pelvic brim (see Fig. 21.1 A), to the upper pole of
the anterior inferior iliac spine. This zone is easy
to demarcate when operating through the poste-
rior approach; approaching by the ilio-inguinal
route it is more difficult to locate, but one is cer-
tain to be extra-articular if above the line joining
the inferior pole of the anterior inferior iliac
spine to the upper border of the greater sciatic
notch, the drill bit being directed vertically or
upwards and inwards; below this line, near the
pelvic brim, it is safe to place screws but they
must be disposed vertically or obliquely from
above inwards and upwards;
the posterior part of the internal iliac fossa
(Fig. 21.1 A) along a strip about 1 cm broad just
lateral to the sacro-iliac joint. This zone is easy
to demarcate by the ilio-inguinal approach. The
screws will be long and have a good purchase
only if they are inserted parallel to the sacro-iliac
articular surface; if they are inserted obliquely
outwards and backwards, they may be, from the
same point of entry, 3 - 4 times shorter
(Fig. 21.1 C). Through the Kocher-Langenbeck
approach this area is difficult to access due to
the presence of the superior gluteal
neurovascular bundle. One passes underneath
the bundle if possible to position a plate, but it
has to be recalled that the iliac surface faces
laterally when the patient is prone, whereas the
 Operative Details 403

retro-acetabular surface faces postero-Iaterally.

One
finger-breadth
21.3.2 Special Instruments
Sciatic bullres
(last 2-3 cm of pelvic
brim and 1.5-2.0cm
broad)
A
(b) Forceps. We have tried many types of bone-
B The forceps we tend to favour are of the
Fig. 21.1. A, B The most solid areas of the innominate bone.
Vertical or oblique striations represent the thickest zones; cross-
hatched areas correspond to dense areas. C From the same start-
ing point, screws of very different lengths can be inserted
Plates used here must be twisted along their long
axis in order to span the two parts and lie ac-
curately on the bone surface;
the body of the pubis.
(a) Operating tab/e. The advantages of the or-
thopaedic table and traction have already been
emphasised in the descriptions of the various
surgical approaches (Chap.20). Since 1987 we
have had at our disposal a modified Judet or-
thopaedic table (Manufacturer Tasserit, BP 193,
St. Denis-Les-Sens, Sens Cedex 89104, France)
which is completely radiolucent (Fig. 21.2A)
and offers exactly the same possibilities as the
previous one; it can also be used with adapted
brackets for intramedullary nailing of the tibia
and femur. Fig.21.2B is an intra-operative
antero-posterior view of the pelvis taken on the
radiolucent table.
holding forceps for these operations. We have
been working for 25 years to try to improve al-
ready existing forceps or design new ones. We
must confess that the problem is not yet solved
and new improvements are still necessary, as to-
day's armamentarium does not help the sur-
geon to solve easily the very variable problems
of reduction offered by acetabular fractures.
Farabeuf pattern with a ratchet (Fig. 21.3 A).
Large, medium and small models are used. It is
useful to have an asymmetrical pair of forceps
and we have modified the Farabeuf model in
this way (Fig. 21.3 B), the longer jaw being ap-
plied to the quadrilateral surface and the
smaller jaw to the external surface of either col-
umn.
Bone forceps have to be applied at various
angles to the surfaces of the columns and are
likely to skid. A useful trick is to temporarily
insert suitable 3.5- or 4.5-mm screws at an ap-
propriate site, leaving the heads proud. The
jaws of the Farabeuf forceps (or other clamps)
can be modified (Fig. 21.3 C) to grip the screw
heads securely, thereby affording excellent con-
trol of the part desired. Sometimes two screws,
one in each fragment, can be used in this man-
ner to maintain reduction while definitive
screws and plates are inserted (Fig. 21.3 D - F).
 404 Operative Treatment of Displaced Fractures Within Three Weeks of Injury

B
Fig. 21.2. A The new ludet·Thsserit orthopaedic table, fully approach. B Intra-operative plain radiograph taken of a patient
radiolucent; this one is ready for an extended ilio-femoral lying on the radiolucent table
 Operative Details 405

A
~--------------------------____~ B

c D

E F

Fig. 21.3. A The three types of Farabeuf forceps. 8 (top) operative use of a Farabeuf forceps modified in this way. E,F
Modified Verbrugge's forceps, (bottom) asymmetrical Farabeuf Two further examples of application of Farabeuf forceps by
forceps. C The jaws of the forceps need to be modified to grip means of screws temporarily inserted into the bone
closely the heads of screws which have been left proud. D Intra-
 406 Operative Treatment of Displaced Fractures Within Three Weeks of Injury

Nevertheless, the Farabeuf forceps are far from in cross-section (to facilitate light, easy penetra-
sufficient on their own. The curved Verbrugge tion of the bone cortices), with a washer or
clamps (Fig. 21.3. B) in the original version or sphere fixed about 0.5 cm from its extremity, so
slightly modified are sometimes useful. Joel that despite the considerable force often exer-
MATTA designed two pairs of clamps cized on the point, there is little danger of un-
(Fig. 21.4A) which appeared immediately controlled perforation of the fragment.
useful. They are now manufactured in a slightly (d) Retractors. Several types must be available:
modified version, the angle of the tips having
Steinmann pins can be driven into the inter-
been changed and a washer added on each to
nal iliac fossa in the anterior approach to
prevent their penetrating into the bone. We very
maintain retracted muscles and viscera; they
frequently use them with benefit whatever the
are very helpful. Their usefulness is en-
approach chosen (Fig. 21.4 B). MATTA'S two
hanced by using a malleable strip applied
king-tongue clamps (Fig. 21.4A) are also
alongside. A supply of such copper strips of
helpful. These two last types of clamps, due to
various sizes should be at hand (see
their width at the level of their articulation and
Fig. 20.10 J).
to the length of their jaws, make reduction of
Sciatic nerve retractor (Fig. 21.6) has been
the fracture of the column opposite to the ap-
developed specifically for these operations.
proach much easier than the other types of
Its beak is applied in either of the sciatic
clamps, which often appear too short.
notches.
(c) Ball spike. This is a useful instrument
Ganz retractor (Fig. 21.7), especially helpful
(Fig. 21.5) with which to temporarily hold a
for gaining access to the quadrilateral sur-
bone fragment awaiting fixation. It is simply a
face through the middle window of the ilio-
long pointed spike, triangular or quadrangular
inguinal incision.

A B

Fig. 21.4. A Joel MATTA's clamps; those at the bottom of the

picture are the king-tongue clamps. B The last version of Mat-
ta's clamps: all with a washer on each jaw; the jaws are less
c angulated than before. C Prototypes of some EL clamps
Operative Details 407

A
Fig. 21.5. A (top) Curved chisel, (middle) Picador ball spike, (bottom) femoral head extractor. B Detail of the ball spike

(e) Other instruments. the way we use the Farabeuf forceps: its jaws
are applied to screw heads which have been
Lamina spreader has permitted us to disim-
left proud and are then tightened. However,
pact numerous fracture lines and facilitates
we feel it to be too heavy and not very han-
their cleaning (Fig. 21.8).
dy because it is fixed to the bone.
Femoral head extractor (Fig. 21.5A). The
The AO distractor (Fig. 21.10) can be very
instrument used to extract a femoral head
useful in the distraction, disimpaction and,
during arthroplasty operations (such as a
later on, reduction of a fracture line. Its use
"corkscrew") can be driven into the ischial
is of course especially indicated when oper-
tuberosity in order to control and manoeu-
ating without the orthopaedic table, but
vre the posterior column, or into the greater
even for us, who routinely use the fracture
trochanter along the femoral neck, through
table, it has been useful and efficient in
a secondary incision, in order to reduce a
some cases.
central dislocation.
Chisels, straight and curved (Fig. 21.5 A),
can be used to advantage like tyre levers for
dis impaction and then reduction of some 21.3.3 Implants for Osteosynthesis
types of fracture lines.
The AO pelvic reduction clamp (Fig. 21.9) We have tried many possible devices for the fixation
may help the reduction, used somewhat in of a disrupted pelvic bone. We have eliminated

Fig. 21.6. Sciatic nerve retractor Fig. 21.7. Ganz retractor

 408 Operative Treatment of Displaced Fractures Within Three Weeks of Injury

Fig. 21.8 Lamina spreader Fig. 21.9. AO pelvic reduction clamp

staples and cerclages and we do not agree with or

recommend the use of cerclages applied on screw
heads left proud.
For the last 25 years we have remained faithful
to the use of screws and plates (Fig. 21.11). We have
to say that for many years (up to 1982) we exclusive-
ly used chrome cobalt devices (as will be seen from
many of the figures), but following the general
trend and the recent disaffection for chrome cobalt
alloys in trauma surgery, we have been using
stainless steel for the last 7 years and will continue
to do so.
Fig. 21.10. AO distractor

Fig. 21.11 A-D. The Vitallium devices: A 4-mm Vitallium screws, 8 straight
Vitallium plate, C EL curved Vitalhum plates with two radii (88 mm and
D 108 mm), D 7.4-mm Venable screws
 Operative Details
Fig. 21.12A-C. The stainless steel devices: A Different types of stainless steel screws. Left
to right: 3.5-mm screw, 4.5-mm screw, fully threaded 6.5-mm screw, partially threaded
6.5-mm screw (extremes of the range in all cases). B We prefer self-tapping screws. C The
A large 6.5-mm screws may be used in different ways
21.3.3.1 Screws (Fig. 21.12A)
3.5-mm screws are the ones we prefer for fixing
any kind of plates as their heads are nearly flush
with the plate. In isolation they appear to us to
be the ideal in most cases. They should range in
length from 15 mm to 110 mm (with at most
5 mm between sizes). We would prefer them
with a 6-mm head diameter.
4.5-mm screws may be used in isolation, but not
to fix the walls of the acetabulum (they are too
big); they may be convenient to fix column or il-
iac wing fracture lines, being inserted between
the two tables of the iliac wing or along the axis
of either column. They are not appropriate for
fixing of the plates as their 8-mm head is very
prominent above the plate and may conflict
with the elements sliding over the plates.
We always use self-tapping screws
(Fig. 21.12B). We hope that this series and its
long-term results will be a supplementary proof
that self-tapping screws are perfectly suitable
for bones, especially the pelvic bone, and their
use saves so much time!
6.5-mm screws, fully or partially threaded, may
be inserted between the two tables of the iliac
wing or along the axis of the columns
(Fig. 21.12 C), in the same way as the 4.5-mm
screws. The choice depends on the thickness of
the different parts of the pelvic bone to be fixed.
They should range in length from 40 mm to
130mm.
409
Spongiosa screws can be used in place of
6.5-mm screws. However, we find them totally
inadequate for fixing fractures of the walls of
the acetabulum.
21.3.3.2 Plates (Fig. 21.13)
A plate to fix an acetabular fracture should be able
to be contoured so as to fit perfectly against the
reconstructed bone. This means that it must be
capable of being bent and twisted along its long
axis, and of course curved on the flat.
There are in fact three categories of plates avail-
able today:
The Vitallium plates (see Fig. 21.11 B, C);
the AO 3.5-mm reconstruction plate, which can
be bent or twisted in all directions, as it is very
malleable. It is a forgiving plate which does not
need to be perfectly contoured and may, in
some skilful hands, be a help to reduction
(MAST et al. 1989);
the Shermann plates (Fig.21.13) with holes
spaced regularly throughout their length, which
we have always liked. Their advantage is that
when they are bent, the bends occur between the
holes without disturbing them, or hardly. They
were initially manufactured in chrome cobalt
alloy; now they are produced in stainless steel.
We have always asked the manufacturers to
design strong plates which can be bent in all
their axes, but which are not forgiving: the plate
410 Operative Treatment of Displaced Fractures Within Three Weeks of Injury
Fig. 21.13. Plates: (above) stainless steel curved acetabular plates
manufactured with two radii (88 mm and 108 mm) and with
4 - 12 holes; (below) stainless steel straight plates, Shermann
type, with 4-12 holes
never gives towards the bone when the screws
are tightened. We use them regularly as buttress
plates, to reinforce a reduction already fixed
precariously by isolated screws. Occasionally we
also use the plate as an aid to reduction.
We have developed special plates (Fig. 21.13)
precurved on the flat, available in two radii of cur-
vature, 88 mm and 108 mm, which according to our
measurements correspond to the two most frequent
types of curvature of the pelvic brim, and having
4-12 holes.
Recently, a strong AO 4.5-mm reconstruction
plate with a radius of curvature of 98 mm has been
developed. Its mechanical properties are close to
that of the one we have designed.
In fact, perhaps two kinds of plate are necessary,
a malleable one and a strong one, since the latter is
compulsory or at least preferable if there is a loss
A B
Fig. 21.14A,B. How to bend a straight plate "on the flat" with two strong Mole wrenches
of substance or a comminution that the plate has
to span.
21.3.4 Method of Internal Fixation
During the development of our technique, we have
found that internal fixation with screws alone has
become less common, and plates are nearly always
required. In practice, the aim is to apply these
plates to surfaces already perfectly reduced and
maintained, although precariously, by some isolat-
ed screws, and the plates must be contoured ap-
propriately at the time of operation to fit the area.
Obviously, if a strong plate is used, and if its cur-
vature is not perfectly adapted to the bone, a partial
displacement of the fracture site will occur during
screwing. To form the plates into the perfect shape,
strong forceps or Mole-type wrenches are needed
(Fig.21.14), or a specially designed plate bender
which allows bending or twisting of the plates in all
directions necessary (Fig. 21.15 A).
Fixation may also require long or large-diameter
screws. When long 3.5-mm screws are inserted into
the posterior column from the pelvic brim or into
the anterior column from the retro-acetabular sur-
face, they must run parallel to the quadrilateral sur-
face of the ischium. Occasionally they can per-
forate and traverse the cotyloid fossa and yet not
impede rotation of the femoral head, because they
are away from it as long as the articular cartilage is
intact (see Fig. 21.15 B).
Insertion of large-diameter and long screws
along the axis of the columns will be described later
(see Fig. 22.34). Whatever their type - Venable
7.4-mm-diameter, 6.5-mm, or 4.5-mm screws -
they undoubtedly have a solid purchase, but they
 Operative Details
A
Femoral head
B
Fig. 21.15. A The Osteo plate bender and special tool for addi-
tional twisting. B A screw placed parallel to the quadrilateral
surface and perforating the cotyloid fossa need not always be a
danger to the cartilage of the femoral head
also have a tendency to twist and displace the bone
during insertion when the large thread bears
against a nearby cortex.
21.3.5 Reduction of Dislocation
In the case of a persistent posterior dislocation it is
usually easy to obtain reduction of the femoral
head after cleaning the acetabulum, removing
possible incarcerated fragments, and with the aid
of traction afforded by the orthopaedic table com-
bined when necessary with direct manipulation.
The reduction of a central or of an anterior
dislocation is the first thing to achieve if there exists
an undisplaced segment of articular surface which
always comprises a piece of the acetabular roof.
Longitudinal traction can be combined with lateral
traction which is exercised by a Lambotte hook
passed under the lesser trochanter or under the
femoral neck, or by using the femoral head extrac-
tor or a Schanz screw driven into the greater
trochanter. The greatest difficulty is offered in
both-column fractures, due to the absence of any
roof fragment in its correct location.
411
The first action necessary is then to reduce the
anterior fragment of the iliac wing with respect to
the posterior fragments thereof, and it is under
these that the femoral head will be repositioned
correctly during further reconstruction of the col-
umns; while this is being done, the head is main-
tained at approximately its normal place by the
combination of the longitudinal traction made
possible by the table itself and the lateral traction
exerted by an assistant or by the specially designed
bracket on the ludet-Tasserit table.
21.3.6 Reduction of Fracture
The plan adopted is always peculiar to the case con-
cerned and no particular scheme can be offered. It
is not amiss to remember that here, as for all
osteosyntheses, extreme care is necessary to pre-
serve what remains of soft tissue attachments to the
bone fragments, and to reduce periosteal stripping
to a minimum.
All fracture lines must be reduced, and to
achieve this it may be necessary first to mobilise im-
pacted fragments and then to secure a hold on the
main fragments by direct application of forceps or
through the heads of temporary screws. Again, the
orthopaedic table is valuable in aiding manipUla-
tion of inferior fragments of either column by
direct traction.
It is necessary at all stages to be extremely
careful about the quality of reduction and not to
accept approximations. When several fracture lines
are to be reduced, negligence during the reduction
of the first will lead to a progressive deterioration
in the quality of reduction of all the others. Frac-
tures range from those in which one reduction cor-
rects everything in that action to those in which one
is obliged to proceed by steps, reducing one column
first and then the other, or perhaps one fragment
with respect to a column and then proceeding to
complete the ensemble by matching the rest of the
bone. Frequently the various stages of reduction
can be fixed by isolated screws, achieving a suffi-
ciently solid assembly to allow removal of the
clamps and so free the operative field for the ap-
plication of a plate.
It is not exceptional for the repositioning of
marginal impaction of the walls of the acetabulum
to create empty space requiring bone graft from the
iliac crest posteriorly or from the greater trochan-
ter.
412 Operative Treatment of Displaced Fractures Within Three Weeks of Injury
The osteosynthesis achieved, it is necessary to
verify the solidity of the assembly and the total
absence of mobility of the fragments when the hip
is moved in all directions. During these manipula-
tions, in total silence, if grating is heard or felt, this
may indicate that a screw is protruding into the
acetabulum.
21.4 Post-operative Care
21.4.1 Local Care
Suction drainage is maintained for 4 - 7 days, the
tubes being removed one by one when they are not
draining more than 10 ml fluid per 24 h.
Post-operative immobilisation is not required.
Assuming that the osteosynthesis as checked dur-
ing the operation is sound, the patient is simply put
to bed and allowed to move all his joints including
the operated hip. Post-operative traction is ab-
solutely useless. No surgeon treating an acetabular
fracture should ever be satisfied with performing
an internal fixation that appears so unreliable that
post-operative traction seems necessary; except
when osteoporotic conditions are unexpectedly
met, work must continue until a sound fixation is
achieved.
Walking using crutches without weight-bearing
or with partial weight-bearing (one-fifth of body
weight) can start at about 10-14 days after follow-
up radiography. Full weight-bearing is achieved in
10- 12 weeks. We have not regarded a longer
period of protected weight-bearing as of value in
preventing the collapse of an avascular bone
necrosis.
21.4.2 Physiotherapy
Physiotherapy is begun very early. Continuous
passive motion is effective the day after the opera-
tion and is continued for 18-21 days, whatever
approach was used (Fig. 21.16). We use the hip ma-
chine, sometimes night and day, for the first 10
days, progressively increasing the amount of possi-
ble flexion from an initial 40°- 50 0. Flexion of
more than 90 ° is usually attained after 5 - 6 days.
In addition to continuous passive movement,
from the 4th or 5th post-operative day active
mobilisation of the hip in all directions is perform-
ed twice a day under the physiotherapist's
guidance.
21.4.3 Medical Treatment
By JEAN-PIERRE MOULINIE, anaesthesiologist of
the Orthopaedic Department, Centre Medico-
Chirurgical de la Porte de Choisy, Paris
21.4.3.1 Antibiotic Treatment for Prophylaxis
of Infections
Principal Antibiotics Used
First-generation cephalosporins:
- cefadroxil (Oracefal) p.o.
- cefazolin (Cefacidal) i.v.
Macrolides:
- pristinamycin (Pyostacine) p.o.
Quinolones:
pefloxacin (Peflacine)
- ofloxacin (Oflocet)
- ciprofloxacin (Ciflox) p.o. and/or i.v.
Aminoglycosides:
- netilmicin (Netromycin) i.v. or i.m.
Conditions for the use of all of these are that the
patient has no known allergy and that liver and
kidney functions are normal.
Treatment Protocol
The protocol outlined here has been in use in this
hospital for more than 12 years now.
Antibiotic treatment is begun as soon as the pa-
tient enters the department, if possible at least 48 h
before surgery. Because of their anti -staphylococcal
activity, cefadroxil or pristinamycin are given (as
long as there is no known allergy to penicillin or
cephalosporins) at a dosage of 2-3 g per day.
Usually the antibiotics are given orally. If the in-
travenous route is used (as in multiple trauma pa-
tients, those with open fracture wounds, patients
with bladder catheters or ileus syndrome, and cases
in which an ilio-inguinal approach is planned),
cefazolin 1 g is given every 8 h combined with
netilmicin 3 - 6 mg/kg per day in a once-daily in-
jection or pefloxacin 400 mg twice a day.
On the day of operation, cefazolin 1 g is injected
at least 15 min before the incision is made and
every 2 h during surgery. Administration continues
with the same dose every 8 h for 24 h or more, until
oral nutrition is allowed. After this, cefadroxil or
pristinamicin is given at a dosage of 1 g twice or
three times a day until 2 days after the last suction
drain has been removed.
Post-operative Care 413

Fig. 21.16A,B. The Kinetec

hip continuous passive motion
machine
A

In addition to the above, netilmicin (300 mg per

day Lm. in one injection) or a quinolone (peflox-
acin 800 mg per day p.o.) can be given for 10-15
days if there is any risk of infection.
Finally, if analysis of a bone or tissue sample
taken intra-operatively has shown the presence of
bacteria, appropriate specific antibiotic therapy
continues for 30 more days.
In all cases, full blood count, erythrocyte sedi-
mentation rate, serum creatinine and blood urea
are monitored every 10 days.

Conclusion
One of the keys to post-operative infection manage-
ment is prevention. The systematic use of an-
tibiotics is necessary to minimise the sequelae of
post-operative wound infections.
The main bacteria involved in the field of ortho-
paedic surgery and traumatology are Staphylococ-
cus species, Enterococci and Pseudomonas species.
The available agents of choice are cephalosporins
of the first generation, quinolones and amino sides,
all of which have proved their worth in our depart-
ment in many years of routine use. B
414 Operative Treatment of Displaced Fractures Within Three Weeks of Injury

21.4.3.2 Anticoagulant Treatment
for Prophylaxis of Deep Venous Thrombosis
and Pulmonary Embolism
Principal Drugs Used
heparin calcium (Calciparine) 1 ml (= 25000
IV) s.c.
acenocoumarol (Sintrom), a coumarin antico-
agulant, p.o. on transfer from heparin, 1 tablet
=4mg
nadroparine (Fraxiparine), fractionated heparin
or low molecular weight heparin (LMWH),
1 ml = 25000 anticoagulation factor Xa (anti-
Xa) Choay units (1 anti-Xa Choay unit = 0.41
anti-Xa IV)
Treatment Protocol
All patients in our department with acetabular
fractures, whether undergoing surgery or not, re-
ceive prophylactic anticoagulant treatment. Most
of them, some with multiple trauma, are already
under anticoagulant treatment when admitted to
our department; intravenously administered hepa-
rin, Calciparine and LMWH have been used for
this. We usually try to continue the anticoagulant
treatment, but by a route other than intravenous
unless a deep venous thrombosis was detected
before or diagnosed after venography or Doppler
sonography. If a deep venous thrombosis is
detected we immediately choose between
Calciparine or Fraxiparine.
Calciparine: Calciparine is given at a dosage of
0.05 - 0.1 0 ml per 10 kg body weight subcutaneous-
ly three times daily (0.00, 8.00 a.m., 4.00 p.m.) at
least 12 h before surgery.
Activated cephalin time (ACT) or partial pro-
thrombin time (PTT) is measured after 6 h in order
to obtain a reference time. Another injection is
given at HO, the time of surgery, increased or de-
creased by 0.05 m!. ACT is checked again 6 h later
and thereafter daily, and should not exceed 1.2
times the reference time.
Treatment is continued for 8-12 days. A plate-
let count is performed twice weekly to give early
warning of thrombocytopenia. Sintrom is given for
75 days.

Anti Xa activity
5
- - - Plasmatic anti Xa activity
after subcutaneous Injection
4
ofaLWMH

2 3 4 5 6 7 8 9 10 11 12 h

------'v'--------"v_---"'-v_
o h

D
12 24 48

Fig. 21.17. Timing of injection of a

LMWH (Fraxiparine) in order to pre-
surgical act vent major bleeding
Post-operative Care
Fraxiparine: The dosage used is 100 anti-Xa Choay
units per 10 kg body weight, given subcutaneously
once daily. Administration should always start 12 h
before surgery (Fig. 21.17). An immediate coagula-
tion test is not necessary. Another injection is given
24 h later (i.e. about 12 h after surgery). On day 4
a dose of 150 anti-Xa Choay units is given. Treat-
ment lasts about 8-12 days without any testing of
coagulation function except for a twice weekly
platelet count. The patient is transferred to Sintrom
treatment at day 8 for 75 days.
Sintrom: Sintrom is never given before surgery. The
dosage is 0.04 mg/kg per day p.o.
always given as a substitute for heparin (day 15
with Calciparine, day 8 with Fraxiparine).
4-6 days are necessary for a good, efficient
transition and to allow heparin to be discon-
tinued.
Coagulation is monitored every 2 days by measure-
ment of prothrombin time or international
numerised ratio (INR), which must equal 2 if stan-
415
dard prophylactic coverage is aimed at. Treatment
is continued for about 3 months or until walking is
allowed.
Conclusion
Calciparine has been used for prophylaxis for more
than 20 years in our department. With careful and
frequent coagulation tests as we have described,
haematomas have become increasingly rare in the
last 10 years. Kakkar's method was totally aban-
doned 12 years ago. For more than 3 years, we have
been able to substitute a low-molecular-weight
heparin, Fraxiparine (Choay Institute), for Calci-
parine, with advantage: its stronger anti-throm-
botic action gives better protection against deep
vein thrombosis and pulmonary embolism, while
its lower anti-coagulation action reduces intra-
operative bleeding. Furthermore, when used pro-
phylactically it does not require coagulation tests to
be performed daily, as prophylaxis using Calcipa-
rine does.
22 Operative Treatment of Specific Types of Fracture
Table 22.1 shows the types of internal fixation used
in 560 cases (for various reasons, nine cases did not
need fixation) operated on within 21 days after in-
jury.
22.1 Posterior Wall Fractures
(a) Approach. The operation is always through the
Kocher-Langenbeck incision. If there is a per-
sisting posterior dislocation, the reduction is
rarely difficult but it should be delayed until
after the cleaning and inspection of the interior
of the acetabulum. If reduction has been per-
formed before the open operation, it is neces-
sary, by means of traction on the orthopaedic
table, to displace the head inferiorly and exam-
ine the interior of the joint, and to evacuate
bone fragments or muscle and capsular debris
which can remain therein. The capsular disrup-
tion resulting from the posterior wall fracture
and head dislocation is generally large enough
for inspection of the joint; if it is not, the cap-
Table 22.1
Types of internal fixation used in 560 cases Vitallium
operated on within 3 weeks of injury
Isolated screws (Vitallium 4 mm, 85
stainless steel 3.5 mm)
Large screws (Vitallium 7.4 mm,
stainless steel 6.5 mm)
Large and small screws 2 2
Straight plates 85
Curved EL plates 6 2
Straight plates and small screws 158
Curved EL plates and small screws 145
Y plate and screws 15
Plates and large screws 19
Staples
Total 517
sular tear is enlarged along the acetabular lip,
but not more than necessary. A Lambotte hook
positioned on the femur above the lesser
trochanter and pulled on by an assistant may
add a lateral displacement that makes it easier
to see into the joint. Regularly now we com-
pletely excise the round ligament, to see the in-
side of the joint better and not to forget any lit-
tle bone fragment attached to it. There is, of
course, no possible consequence of this excision
since the ligament is always torn at the time of
the dislocation. After reduction of the disloca-
tion an appraisal of the acetabular lesion itself
can be made; the possibility of marginal impac-
tion must be borne in mind.
(b) Reduction. In the case of a fracture with margi-
nal impaction (Fig. 22.1) the impacted frag-
ments should be detached with care using a
small spatula or curved chisel that penetrates
the crushed area of the spongiosa, which can
nearly always be detected with careful examina-
tion. This allows the surgeon to leave as much
cancellous bone as possible beneath the sub-
chondral bone of the impacted fragment. They
Stainless steel Total 0/0
3 88 15.6
85 15.5
8 1.4
2 160 28.5
34 179 31.9
1 16 2.9
20 3.5
43 560
418 Operative Treatment of Specific Types of Fracture

Fig. 22.1A-C. Fracture of

the posterior wall with
marginal impaction. A The
post-traumatic condition on
a horizontal cut of the hip.
B In some cases the reposi-
tioning of the impacted
fragments leaves a big defect
behind them, which may
lead to instability and needs
to be filled up with graft
If defect is too big (C)
graft is necessary

2 or 3 mm of strippi ng o f
can be correctly replaced against the femoral soft tissue are enough
head, and then be held there by repositioning to control the reduction
of the fragment
the surrounding separated fragments. Occa-
sionally we have been left with a cavity in the
posterior wall after repositioning the impacted
fragments, likely to produce instability during
post-operative mobilisation; the defect has
been filled using cancellous bone graft taken
from the posterior part of the iliac crest or
from the greater trochanter, or in some cases
bank bone has been used, and even acrylic ce-
ment in elderly people.
When the fracture separates a large single
fragment, this is as a rule easy to reduce. A
problem is posed sometimes when there are
several pieces, not always easy to reassemble,
especially when one fragment, the key, is dif-
ficult to find. It is necessary to be very
painstaking in the restoration of the retro-
acetabular surface, for on its quality depends
the success of reduction of the posterior part of
the articular surface which it supports.
During all manipulation, care is necessary to
preserve soft tissue attachments to all frag-
ments as much as possible. A little stripping of
2 - 3 mm is all that should be allowed, suffi-
cient to clean the edges of the fragments and
make possible a perfect reduction, while avoid-
ing excessive devascularisation.
(c) Fixation. For a long time, fixation has been
satisfactorily achieved using 3.7-mm Phillips
Vitallium screws. Isolated screws are adequate
for a large fragment when the bone is of good
quality and when there is a large soft tissue
pedicle which needs to be preserved (Fig. 22.2).
By carefully siting sufficient screws (three to c
five) so that they diverge from each other, a Fig. 22.2A-C. Fixation of a single fragment with screws only.
solid fixation can be achieved which permits A Diagram, B horizontal section through the hip demonstrating
how to preserve the soft tissue attachments of the fragment.
post-operative movement immediately. In con- C Fixation of solitary posterior wall fragment with screws.
trast, when there is a posterior dislocation asso- Appearance at 4 months post-operatively (for pre-operative
ciated with a small marginal fracture of the state see Fig. 5.3)
 Posterior Wall Fractures
posterior wall, one or two screws may be suffi-
cient for the fixation.
When the posterior fragment is so small that
it needs no more than one or two screws for fix-
ation, the indication for surgery is ques-
tionable, except if there is an incarcerated bone
fragment.
In fact, for more than 25 years now the rule
has been to use a plate to fix posterior wall
B c
Fig. 22.3A-C. Suitable site for application of a plate and screws
for fixation of a posterior wall fracture. A Scheme, B obturator-
419
fractures. One or more screws maintain the
reduction during contouring and fitting of the
plate. These screws are inserted from the
posterior retro-acetabular surface and find a
variable hold in the cortex of the quadrilateral
surface of the ischium. The plate chosen should
be as long as possible. It is moulded on the flat
and twisted along its axis so as to lie accurately
on the restored surface of the posterior column
bridging the one or more fragments. The mid-
dle screws in the plate traverse the fragments
and should be directed so as to reach the
quadrilateral surface of the ischium. Inferiorly,
the plate is always modelled to lie in the sub-
cotyloid groove and on the upper pole of the
ischium; in this region the screws are placed
obliquely and diverge, being of varying length
(35-45 mm). The most inferior screw finds an
excellent purchase in the tuberosity of the
ischium (Fig. 22.3). Superiorly, the plate must
extend to the supra-cotyloid region or just in
front of the sacro-iliac joint above the greater
sciatic notch, where there is ample bone for the
screws. Well constructed, the assembly should
be very solid.
The plate is chosen so as to have at least two
screws above the superior wall fracture area. We
use either straight plates or the curved ones that
oblique radiograph showing moulding of plate in the region of
the ischial tuberosity, C pre-operative view
420
we designed initially for anterior fractures,
which quite often are very suitable as they lie
along the axis of all the fragments. The screws
we use now are stainless steel screws of 3.5 mm
diameter, either in isolation or to fix the plate.
With a little care there is no risk of joint
penetration when inserting screws into the
posterior wall. There is plenty of room to insert
screws; the problem is to put them in the right
direction, and for this there are several tricks:
one can insert a temporary Kirschner wire
tangential to the articular surface under direct
view while the femoral head is distracted, as ad-
vised by Jeffrey MAST (Fig. 22.4A). Any screws
inserted parallel to the Kirschner wire or at a
less oblique angle cannot penetrate the joint.
Experience will bring familiarity with the shape
and position of the articular surface and how
to insert the screws in the correct direction,
keeping in mind that the closer one is to the
acetabular rim, the more oblique the drill bit
has to be directed (Fig. 22.4 B) when the patient
is positioned prone.
Te:nporary Kirschner wire
I
I
I
A
•
, .
Good directions to CD
give to the drill bit
Fig. 22.4A,B. How to insert screws through the posterior wall
safely. A Under direct view, a Kirschner wire is temporarily in-
serted tangential to the articular surface. The drill bit has to be
parallel or less obliquely inserted than the Kirschner wire. B The
closer one is to the acetabular lip, the more oblique the drill bit
has to be when the patient is positioned prone
Operative Treatment of Specific Types of Fracture
It is advisable to check for intra-articular
screws by:
moving the hip in complete silence in the
operating room;
looking inside the joint with gentle traction
if the capsule is still open.
22.1.1 Postero-superior Fractures
The procedure is similar but in order to bridge the
fragment, the plate must be placed more anteriorly
in the supra-cotyloid region. Inferiorly it must at
least reach the sub-cotyloid groove, but it is also
good if the plate rests on the upper pole of the
ischium (Fig. 22.5), and it is often useful to curve
the plate lightly on the flat or to utilise the specially
curved acetabular plate (Fig. 22.8). For perfect ap-
plication of the plate above the acetabulum and a
total controlling view of the fracture lines, the in-
ferior part of the gluteal muscles needs to be
elevated. If the risk of damaging the muscles is
high, a part of the gluteal tendon insertions along
the upper border of the greater trochanter can be
temporarily cut.
B
Posterior Column Fractures
A B
Fig. 22.5A,B. Osteosynthesis of a postero-superior fracture with
marginal impaction by means of a plate and screws. A Antero-
posterior radiograph and B obturator-oblique radiograph taken
22.1.2 Postero-inferior Fractures
A fragment of the ischial tuberosity is detached;
the fracture surface must be defined clearly. This
necessitates adequate stripping of the quadratus
femoris in order to control the ischial fragment
which is usually in one piece with the detached seg-
ment of the posterior wall. Sometimes sharp inci-
sion of the origins of the hamstring muscles is
necessary for a complete view of the fracture line
through the ischial tUberosity. Along the anterior
border of the greater sciatic notch the aponeurosis
of the obturator internus is elevated and the muscle
itself stripped off the quadrilateral surface so as to
reach the fracture line dividing this surface. A
perfect reduction of the ischial tuberosity is accom-
panied therefore automatically by restoration of
the articular surface. Fixation generally comprises
local screws joining the superior pole of the
tuberosity of the ischium to the rest of the ischium,
and a moulded plate on the tuberosity and the
retro-acetabular surface extending above the frac-
ture line (Fig. 22.6). Comminuted examples are dif-
ficult but are dealt with similarly.
421
14 years after surgery. The hip is rated 6.6.6. See Fig. 5.10 for
pre-operative state
22.1.3 Special Features
When the anterior border of the greater sCIatlc
notch is detached, in order to make the reduction
easier it may be found helpful to reduce and fix this
first.
Large posterior wall fragments including the
anterior border of the greater sciatic notch do not
offer any particular difficulty. They can be reduced
and fixed easily by means of a plate.
A very extensive posterior fragment can be fixed
simply with screws alone (Fig. 22.7) or by screws
and a plate. Figure 22.9 shows the fixation of an as-
sociated posterior wall and posterior hemitrans-
verse fracture.
22.2 Posterior Column Fractures
(a) Approach. Routinely the Kocher-Langenbeck
approach is used. The reduction of the disloca-
tion which accompanied the fracture has never
been difficult, and the interior of the joint is
 422 Operative Treatment of Specific Types of Fracture

A c

Fig. 22.6A - C. Osteosynthesis of a postero-inferior fracture by

means of a plate and screws. A Antero-posterior radiograph,
B obturator-oblique radiograph, C iliac-oblique radiograph, all
taken 26 years after operation. The hip is rated 6.6.6. See
Fig. 5.13 for pre-operative state. Nowadays we would extend the
B plate up to the upper pole of the ischial tuberosity
Posterior Column Fractures 423

Fig. 22.7 A-C. Osteosynthesis of an extended posterior wall

fracture by means of screws. A Antero-posterior radiograph,
B obturator-oblique radiograph, C iliac-oblique radiograph, all
taken 15 years after operation. The hip is rated 6.6.6. See
c Fig. 5.15 for pre-operative state
424 Operative Treatment of Specific Types of Fracture

A B

Fig.22.8A-c' Posterior superior wall fracture operated on

through a Kocher-Langenbeck approach and fixed with a new
stainless steel plate. The three standard views show the appear-
ance 3 months post-operatively. The patient did not undergo
radiotherapy but only received indomethacin. Hip motion is
c rated 6. See Fig. 5.19 for pre-operative state
 Posterior Column Fractures 425

A B

Fig. 22.9A-C. Posterior wall fracture with incomplete posterior

hemitransverse fracture operated upon through a posterior
approach. The three standard views show the appearance at 4
years post-operatively. See Fig. 5.17 for pre-operative state. The
fixation was achieved with the earliest version of stainless steel
plate. The patient received diphosphonates post-operatively
c
426 Operative Treatment of Specific Types of Fracture

easy to inspect using the traction facility of the

fracture table.
(b) Reduction. The column is reduced using bone
forceps of which one jaw is applied to the
anterior border of the greater sciatic notch or
the angle thereof, depending on the site of
detachment. The other jaw grips the area of
bone above the roof of the acetabulum, if
necessary by means of a temporarily inserted
screw (Fig. 21.3 E). A finger passed through the
greater sciatic notch on to the inner aspect of
the quadrilateral surface verifies the reduction
of the fracture line here (Fig. 22.10A). Very fre-
quently, there is a rotation of the column about
its vertical axis that is generally easy to control
in recent fractures; only restoration of the en-
dopelvic fracture line assures the correction of
this rotation. Occasionally, forceps or an
elevator applied to the endopelvic surface
through the sciatic notch help in obtaining the
Fig.22.11. Osteosynthesis of a posterior column fracture by
means of a plate and screws. Antero-posterior radiograph after
operation. See Fig. 6.3 for pre-operative state. This was the first
Vitallium plate of our series

reduction. Restoration of the quadrilateral sur-

face and of the shape of the greater sciatic
notch guarantees anatomical reduction of the
posterior column. If the rotation is particularly
difficult to reduce, the femoral head extractor
or a Schanz screw can be inserted into the
ischial tuberosity, thereby achieving full control
of the detached fragment of the column.
(c) Fixation. Solid fixation necessitates a plate ex-
tending from the posterior part of the iliac
wing above the greater sciatic notch down to
the ischial tuberosity. It has to be passed
beneath the superior gluteal neurovascular
bundle, which is mobilised with great care. It
A may be situated fairly laterally and astride the
fracture line, and the screws must be placed
obliquely in order not to penetrate the articular
surface; the middle screws will gain purchase
only through that part of the quadrilateral sur-
face of the ischium which has been detached
with the posterior column (Figs.22.10B and
B 22.11). Alternatively, the plate may lie near the
greater sciatic notch; long screws can reach
Fig.22.10. A The reduction of the fracture line through the through the fracture line alongside the deep
quadrilateral surface is checked by a finger introduced through aspect of the quadrilateral surface of the
the greater sciatic notch. B Diagram of method of inserting
screws holding plates against the posterior wall so as to gain pur-
ischium and gain excellent fixation in the
chase on the quadrilateral surface of the ischium on both sides anterior column (Fig. 22.10). Plates at both
of a fracture line sites may be required (Figs. 22.12, 22.13).
 Posterior Column Fractures 427

A B

Fig. 22.12A-C. Osteosynthesis of a posterior column fracture

by means of two Vitallium plates. The three views, 4 months
post-operatively. The associated vertical anterior pelvic fracture
was impossible to fix from the back. See Fig.6.13 for pre-
C
operative state
428
Fig. 22.13 A, B. Osteosynthesis of a posterior column fracture by
means of two plates. A Antero-posterior radiograph at time of
22.2.1 Special Features
Fractures of the posterior column in which the en-
dopelvic line detaches the teardrop, i.e. those which
are very extended anteriorly, may be difficult to
reduce. A finger introduced along the inner wall of
the true pelvis can reach the anterior part of the
fracture line and control the reduction. It is par-
ticularly useful in these cases to place long screws
parallel to the quadrilateral surface of the ischium;
they will gain a hold in the anterior or superior sur-
face of the anterior column (Fig. 22.14).
Two epiphyseal separations did not present any
particular problem. The plate which bridged the
epiphyseal cartilage was applied in the usual fash-
ion (Figs. 22.15,22.16), and subsequently removed.
22.3 Anterior Wall Fractures
(a) Approach. This is always by the ilio-inguinal
route.
(b) Reduction. As a rule, reduction of the anterior
dislocation which accompanies these fractures
Operative Treatment of Specific Types of Fracture
operation, B antero-posterior radiograph 10 years after removal
of the plates. See Fig. 6.4 for pre-operative state
is easy and is obtained by traction on the or-
thopaedic table. In the occasional difficult case,
a femoral head extractor inserted along the axis
of the femoral neck through a short separate
vertical incision astride the inter-trochanteric
crest provides additional control. After the
head, the fragments of the acetabulum can be
reduced (the ball spike is particularly helpful).
Only two difficulties sometimes arise. Possibly
the fracture of the superior pubic ramus cannot
be seen or palpated; in this circumstance it is
necessary to elevate the pectineus muscle and,
if necessary, the pectineal part of the inguinal
ligament. The other problem is that the main
outlines of the anterior wall fragments may be
complicated by secondary longitudinal or
transverse fracture lines, making control
awkward.
(c) Fixation. Having obtained the reduction, if
possible one or two isolated screws are inserted
to fix the fragments, albeit precariously - but
that is enough to allow all the time needed to
shape a long curved plate which has to be ap-
plied perfectly congruently to the upper aspect
of the pelvic brim and very close to the brim,
 Anterior Wall Fractures 429

A B

c
 430 Operative Treatment of Specific Types of Fracture

B
A
Fig. 22.15 A, B. Internal fixation of a separation injury of the state. A Post-operative antero-posterior view, B appearance 12
posterior part of the Y cartilage. See Fig. 6.6 for pre-operative years post-operatively. The hip is rated 6.6.6

Fig. 22.17. Scheme of method of fixing an anterior wall fracture

associated with elevation of a plate of bone from the quadrilat-
eral surface of the ischium. The latter is held with two sup-
plementary screws

..
Fig. 22.16. Osteosynthesis of epiphyseal separation of posterior
column. Antero-posterior radiograph 10 years after operation.
See Fig. 6.10 for pre-operative state
Anterior Column Fractures
at least to its inferior two-thirds. The plate
bridges the detached fragment and extends
superiorly to the inner part of the iliac fossa, in
front of the sacro-iliac joint. Inferiorly it ex-
tends to the intact part of the superior pubic
ramus and the body of the pubis (Figs. 22.17,
22.18 and 22.19). Other screws are sometimes
necessary to fix longitudinal fracture lines
dividing the anterior wall fragment, but much
attention must be paid to their direction in
order to avoid their perforating the joint. When
there exists a displaced and elevated plaque of
bone from the quadrilateral surface, it can be
pushed back into position by means of an
elevator and fixed by screws which extend from
the plate or from the surface of the ilio-pec-
tineal region. Although it generally does not
bear articular cartilage, anatomical reduction
helps in the overall stability of the osteosyn-
thesis (Figs. 22.17 and 22.19) and provides more
satisfying post-operative radiographs. CT sec-
tions have shown that in some cases the plaque
bears some articular surface (see Sects. 7.1 and
7.4). In such cases, perfect reduction of the
plaque is compulsory.
In all cases of fractures of the anterior wall
as well as of the anterior column, it is impor-
tant to ensure that the quality of reduction is
good. Even a small step-off at the fracture line
on the anterior surface of the bone will be ac-
companied by incongruence of the joint space
and possibly a loss of parallelism of the
superior joint space, corresponding to a slight
anterior displacement of the head due to the
imperfection in reduction of the anterior wall.
22.4 Anterior Column Fractures
22.4.1 Middle and Low Fractures
(a) Approach. These fractures require the ilio-
inguinal approach.
(b) Reduction of the anterior dislocation is ob-
tained as for anterior wall fractures and is
generally stable. Then the anterior column frag-
ment is pushed back to its place and often
maintained by one or two isolated 3.5-mm
screws.
(c) Fixation. This is achieved (Figs.22.20, 22.21
and 22.22) by means of a moulded pre-curved
431
plate along the superior aspect of the pelvic
brim from the pubic symphysis to the posterior
part of the internal iliac fossa. Very long plates
are required. Accessory screws serve to fix a dis-
placed plate of bone of the quadrilateral sur-
face, and also accessory fractures of the anteri-
or wall of the acetabulum which are sometimes
difficult to see on the radiograph and very dif-
ficult to fix.
A fracture of the superior pubic ramus is
easily bridged by the plate but fractures of the
ischio-pubic ramus are reduced automatically;
they are inaccessible and no attempt is made to
fix them.
For middle fractures, an assembly compris-
ing two plates has been used occasionally. One
is at the level of the iliac fossa to bridge the up-
per fracture line, and the other at the level of
the superior pubic ramus to span the inferior
fracture line.
Middle or low anterior column fractures are
regularly accompanied by a plate of bone
detached from the quadrilateral surface. The
extent of this plate can be perfectly seen on the
CT sections. If it involves the posterior wall
(see Sect. 8.2.5.2), it must be anatomically re-
duced, but if it does not bear any articular sur-
face, perfect reduction is not so compulsory, es-
pecially as fixation may be difficult due to the
thinness of the plate fragment.
22.4.2 High Fractures
(a) Approach. The choice may be debatable. If
there is one large fragment comprising the ante-
rior part of the iliac wing, the ilio-femoral ap-
proach is easy and simple. It must be certain,
however, that only fixation of the upper part of
the column will be necessary (Fig. 22.23), other-
wise this approach will be insufficient. In cases
with multiple fragments or if there is doubt, the
ilio-inguinal approach must be chosen.
(b) Reduction of the dislocation can be difficult
and unstable, for, with the anterior wall, the
fragment takes a large segment of the roof. For
this reason, the head has to be held reduced
against the posterior wall and the intact part of
the roof during the reduction and fixation of
the anterior column. It can be maintained by
simple traction perhaps combined with internal
rotation or, in the most difficult cases, by
lateral traction exerted on a femoral head
 432 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.1SA-C. Osteosynthesis of an anterior wall fracture (ilio-

inguinal approach). A Antero-posterior radiograph, B obtura-
tor-oblique radiograph, C iliac-oblique radiograph, all taken 20
years after operation. See Fig. 7.2 for pre-operative state. The
hip is rated 6.6.6
c
 Anterior Column Fractures 433

A B

Fig. 22.19 A - C. Osteosynthesis of an anterior wall fracture as-

sociated with a fracture of the ischio-pubic ramus (ilio-inguinal
approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac-oblique radiograph, all taken 11 years
after operation. See Fig. 7.5 for pre-operative state. The hip is C
rated 6.6.6
434 Operative Treatment of Specific Types of Fracture

Fig. 22.20. Scheme of three methods

of fixation for anterior column fractures

A B

Fig. 22.21A,B. Osteosynthesis of a low anterior column fracture operation, B antero-posterior radiograph 4 years after opera-
(ilio-inguinal approach). A Antero-posterior radiograph before tion
 Anterior Column Fractures 435

A B

Fig. 22.22A-C. Osteosynthesis of middle anterior column frac-

ture (ilio-inguinal approach). A Antero-posterior radiograph,
B obturator-oblique radiograph, C iliac-oblique radiograph, all
taken 21 years after operation. See Fig. 8.6 for pre-operative
state. The hip is rated 6.6.6 c
436 Operative Treatment of Specific Types of Fracture

extractor inserted along the axis of the femoral 22.4.4 Insertion of Screws Along the Pelvic Brim
neck through a small vertical incision astride
The same problem arises from the anterior as from
the vastus lateralis crest. It is helpful to
the posterior approach: can screws safely be in-
manipulate the fragment of the iliac wing by
serted from the superior aspect of the pelvic brim
gripping it with a small pair of Farabeuf
in the acetabular area? The answer is yes, but it re-
forceps placed across the interspinous notch,
quires great care.
and, if necessary, another astride the iliac crest.
Screws inserted through the holes of a plate or
(c) Fixation may be maintained by two Venable
in isolation that run closely along the deep aspect
screws (Fig. 22.23) inserted from in front back-
of the quadrilateral surface can gain a very good
wards, between the two tables of the iliac wing,
purchase in the retro-acetabular surface of the
one from the anterior inferior iliac spine and
posterior column without entering the joint. These
the other from the crest. They give solid fixa-
screws may be embedded into the floor of the'
tion but may be difficult to place in the wing.
cotyloid fossa if it is thick, they may be partially ex-
To fix the inferior part of the anterior column
posed in the fossa by breaking the outer wall of the
fragment a short plate (Fig. 22.23) or isolated
floor of the fossa, or they may nearly completely
screws can be used. In general, we prefer to use
penetrate the fossa if the floor is very thin. Never-
plates placed according to the fracture con-
th~less, they are kept away from the head by the
figuration. Frequently we begin by inserting a
thIckness of two layers of articular cartilage and do
screw into the medial angle of the fragment in
not touch the head in any position, as long as the
the iliac fossa, while the fragment is maintained
cartilage maintains its normal thickness
reduced by clamps. Then the first plate,
(Fig. 22.27 A). Thus, screws inserted parallel to the
generally with four holes (Fig. 22.25), is con-
quadrilateral surface will not disturb the head, but
toured and fixed on the convexity of the iliac
they have to be inserted very close to the pelvic
crest, spanning the fracture line. The second
brim.
plate is contoured to apply perfectly along the
If for any reason the plate has been placed a lit-
superior aspect of the pelvic brim from the in-
tle more laterally it may be impossible to insert the
ternal iliac fossa, just outside the sacro-iliac
drill bit parallel to the quadrilateral surface from
joint, as far as the superior pubic ramus, with
the holes of the plate. The screws may sometimes be
at least two screws beyond the fracture of this
inserted medial to the plate and apart from it. It
part.
may be recalled that above a horizontal line runn-
In some cases lag screws inserted from the il-
ing at the level of the inferior pole of the anterior
iac crest between the two tables of the wing and
inferior iliac spine screws can be inserted in the
a plate along the pelvic brim have sufficed for
vicinity of the pelvic brim: as soon as they are di-
a solid fixation (Fig. 22.26).
rected antero-posteriorly or obliquely downwards,
Further screws will fix the accessory frag-
backwards and inwards they will be away from the
ments or a fracture of the quadrilateral surface.
joint (Fig. 22.27 B).
Screws inserted below the inferior pole of the
ilio-pectineal eminence, at the level of the superior
22.4.3 Special Features
pubic ramus, will not penetrate the joint as long as
they are not directed posteriorly.
In a few examples of anterior column lesions the
There remains the difficult area which is the seg-
iliac fracture line was incomplete, failing to r~ach
ment of the pelvic brim directly related to the
the iliac crest or the anterior border of the bone,
acetabulum. It extends for about 4-5 cm and com-
and yet reduction was difficult. It was greatly
prises the ilio-pectineal eminence, the top of which
facilitated after deliberately completing the frac-
corresponds to the centre of the femoral head. At
ture.
this level screws may be inserted:
In high fractures of the anterior column which
detach a large part of the iliac fossa and in which strictly parallel to the quadrilateral surface and
the reduction is difficult to maintain during the very close to it;
moulding of the plate, two isolated screws can be obliquely inwards and downwards to come out
inserted to advantage, one into the anterior part of through the quadrilateral surface.
the iliac crest and the other into the medial angle Another alternative is to use only short screws so as
of the fragment in the iliac fossa. to be sure not to penetrate the joint (Fig. 22.27 C).
 Anterior Column Fractures 437

A B

Fig. 22.23A-C. Osteosynthesis of high anterior column fracture

not involving the obturator ring. A Antero-posterior radio-
graph, B obturator-oblique radiograph, C iliac-oblique radio-
graph, all taken 18 years postoperatively. See Fig. 8.11 for pre-
operative state. The hip is rated 6.6.6 c
438 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.24A-C. Osteosynthesis of a high anterior column frac-

ture involving the sacro-iliac joint (ilio-inguinal approach).
A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph. See Fig. 8.10 for pre-
operative state. The anterior column fragment took with it the
upper part of the sacro-iliac joint surface. For this reason,
despite involvement of the sacro-i1iac joint, the operation was
performed through the ilio-inguinal approach. A long plate ap-
c plied to the pelvic brim would have been preferable
 Anterior Column Fractures 439

A B

Fig. 22.2SA-C Osteosynthesis of a high anterior column frac-

ture by means of stainless steel plates. A Antero-posterior
radiograph, B obturator-oblique radiograph, C iliac-oblique
radiograph, all taken 2 years after surgery. See Fig. 8.13 for pre-
operative state. The hip is rated 6.6.6 C
440 Operative Treatment of Specific Types of Fracture

Fig.22.26A-C Internal fixation of a high anterior column

fracture with a Vitallium plate and lag screws. A Antero-
posterior radiograph, B obturator-oblique radiograph, and C
iliac-oblique radiograph, all taken 2 years post-operatively. See
Fig. 8.14 for pre-operative state. The hip is rated 6.6.6
Anterior Column Fractures 441

{'Ii
The dangerous
B segment
(4-5 cm long)
T I
A

~
lIio-pectineal- - - o / -
eminence
, " , Safe area

Short screw ~ r. -

Oblique screw

C Screw parallel
to the quadrilateral
surface

c A Section A-A'

Fig. 22.27. A Diagram demonstrating that a screw strictly paral-
lel to the quadrilateral surface at the level of the cotyloid fossa
may not be harmful. a The screw is totally embedded in the floor
of the fossa. b The screw has broken the floor of the fossa. c The
floor of the fossa is so thin that the screw is nearly totally ex-
posed in the fossa, but it is still away from the head because of
the two layers of cartilage of the socket and the femoral head.
B How to insert the screws along the pelvic brim without
penetrating the joint with patient in supine. There are two safe
areas: (1) above a line joining the inferior pole of the antero-
inferior iliac spine and the superior border of the greater sciatic
notch, provided the screws are inserted vertically, or obliquely
posteriorly and inwards, or posteriorly and distally but parallel
and close to the quadrilateral surface: (2) below another nearly
horizontal line tangential to the inferior part of the ilio-pec-
tineal eminence, provided the screws are inserted obliquely in-
wards, or inwards and downwards but not posteriorly and out-
wards. C Horizontal section through the ilio-pectineal
eminence. At that level screws should either be very short or in-
serted very close to the pelvic brim and obliquely inwards and
distally
442
22.5 Pure Transverse Fractures
The approach to pure transverse fractures is chosen
according to three factors:
the localisation of the fracture, which may be
trans-tectal, juxta-tectal or infra-tectal;
the time elapsed since injury: during the 3rd
week, if callus formation appears significant, it
is best to choose the extended ilio-femoral
approach;
the displacement of the fracture.
22.5.1 Pure Juxta-tectal
or Infra-tectal Transverse Fractures
The fracture line can be reached and controlled
from either extremity. In practice, the decision
depends on the relative displacements of the
posterior and anterior ends as assessed from the
three views and CT. The approach chosen, Kocher-
Langenbeck or ilio-inguinal, is appropriate for the
site of greater displacement. If the displacement is
similar at the greater sciatic notch and at the pelvic
brim or predominates posteriorly, we always
choose the posterior approach, as it is easier and
can offer the strong posterior column for internal
fixation, along which plates are easier to contour
and apply. If there is nil posterior displacement or
the displacement is significantly predominant at
the pelvic brim level, we choose the ilio-inguinal
approach.
22.5.1.1 Kocher-Langenbeck Approach
(Figs. 22.29, 22.30 and 22.31)
(a) Reduction. The central dislocation can usually
be corrected easily, and the head comes to lie
under the intact segment of roof, or what re-
mains thereof, and is stable.
Rotation of the inferior fragment in fresh
fractures does not offer any particular difficul-
ty of reduction, in spite of its possible complex-
ity (see Sect. 9.1.2). The ischio-pubic fragment
is manipulated by appropriate forceps; several
trials are often necessary, as no type of forceps
is appropriate to all cases. Often the reduction
is temporarily maintained by a Farabeuf
forceps applied to the heads of two temporarily
inserted screws with their heads left proud.
Reduction of the pelvic brim is assessed by
a finger introduced into the pelvis through the
Operative Treatment of Specific Types of Fracture
greater sciatic notch; it should be remembered
that if the greater sciatic notch is reduced and
the head lies in place under the roof, and yet
the pelvic brim remains displaced inwards, this
is likely to result from rotation of the fragment
about its horizontal axis and not from simple
inward displacement of the whole fragment. It
is necessary to try to rotate the ischio-pubic seg-
ment by pressure directed medially on the
ischial tuberosity or by inserting a femoral head
extractor or a Schanz screw into the ischial
tuberosity and using it as a handle (Fig. 22.33).
(b) Fixation. Posteriorly, there is no problem in the
placing of plates and screws (Fig. 22.28 A). It is
wise to arrange one or more long screws to run
obliquely across the fracture line, parallel to the
quadrilateral surface of the ischium, to gain
purchase in the quadrilateral surface itself or in
the superior aspect of the pelvic brim. These
screws must be of precisely the appropriate
length (Fig. 22.28 A). They may be inserted
through the plate or apart from it.
22.5.1.2 Ilio-inguinal Approach (Fig. 22.28 B, C)
(a) Reduction. A transverse fracture is reduced by
pushing the inferior part of the pelvic brim out-
wards and downwards. Reduction is maintained
by a clamp across the anterior aspect of the
bone, generally put in place through the medial
window of the ilio-inguinal approach. The
Matta clamps are regularly efficient for this
purpose.
(b) Fixation. A curved acetabular plate fixes the
reduction. As usual it has to be precisely con-
toured to lie perfectly along the superior aspect
of the pelvic brim. The contouring is difficult
at the level of the ilio-pectineal eminence: the
plate has to be twisted so as to allow insertion
of the screws either vertically or obliquely in-
wards but never outwardly (Figs. 22.32 and
22.35). A six- to eight-hole plate generally suf-
fices.
22.5.2 Pure Trans-tectal Transverse Fractures
From experience we have come to the conclusion
that these types of transverse fractures should be
approached through the extended ilio-femoral inci-
sion. The reason for this is that these fractures
transgress the weight-bearing area of the acetabu-
lum in the standing position, so their reduction has
 Pure Transverse Fractures 443

The track of the transverse fracture line

Fig. 22.28A- D. Scheme of three methods of fixation for trans-

verse fractures: A through the Kocher-Langenbeck approach,
B, C through the ilio-inguinal approach. D A screw parallel to
the quadrilateral plate may cross the fracture line and gain pur-
chase at the level of the pelvic brim

A B

Fig. 22.29A,B. Osteosynthesis of a transverse fracture with an- after operation. See Fig. 10.10 for pre-operative state. Operating
terior dislocation (posterior approach). A Antero-posterior ra- through the posterior approach was an error
diograph, B obturator-oblique radiograph, both taken 3 months
444
A B
Fig. 22.30 A - C. Osteosynthesis of a transverse fracture (poste-
rior approach). A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 9 years
to be perfect. If we do not reach this goal, early
wear of the articular surface will ensue and lead to
osteoarthritis.
The best way for perfect access to both ex-
tremities of these fracture lines is to use an extensile
approach such as the extended ilio-femoral. The
standard extended ilio-femoral approach allows
good access to these fractures in cases which have
come for treatment soon after injury. As soon as
some callus is present it may be necessary to elevate
the internal iliac fossa. Then we have access to the
whole of the fracture line, to clean it of clots, callus
and debris. Reduction is achieved with two clamps
positioned astride the anterior and the posterior
borders of the innominate bone, or by clamps, the
jaws of which are applied to temporarily inserted
screws.
Once a perfect reduction is achieved and
checked inside the joint via a routine capsulotomy,
fixation is performed. Most frequently we combine
a lag screw and a plate posteriorly and a long, large
screw along the axis of the anterior column
(Figs. 22.34 and 22.35). Firstly, the posterior part of
the fracture line is fixed with a 3.5-mm lag screw
which allows removal of the posterior clamp.
Operative Treatment of Specific Jypes of Fracture
c
after operation. The anterior fracture of the obturator ring was
not explored. See Fig. 9.9 for pre-operative state. The hip is rated
6.6.6
To fix the anterior extremity of this transverse
fracture line, a 4.5 or 6.5-mm screw (fully or par-
tially threaded) is inserted along the axis of the
anterior column. The point of entry is situated at
least 3 - 4 cm above the roof of the acetabulum, i.e.
more than a thumb-breadth, along the posterior
aspect of the anterior pillar of the iliac wing. In
fact there is an area of possible points of entry,
which is a circle about 2 cm in diameter. Introduc-
tion of the screw is possible, but has to be per-
formed carefully. A guide similar to the one used
for cruciate ligament reconstruction was used for a
while, but it soon proved not to be absolutely
necessary.
First the external table of the iliac wing is trans-
gressed with the drill bit; further progress is not
continuous but by stops and starts, in order to be
constantly checking by feel that the drill is in the
bone, boring into the spongiosa and not the cor-
tices. If drilling becomes harder, the direction of
the drill bit is changed to remain in the spongiosa.
In fact the drilling is guided (Fig. 22.34B, C):
by the surgeon now positioned at the back of
the patient, who controls the anterior column
Pure Transverse Fractures 445

Fig.22.31A-D. Osteosynthesis of a severely displaced trans-

verse fracture, well reduced through the posterior approach.
A Antero-posterior radiograph before operation, B - D the three
standard views taken 17 years after operation. The hip is rated
6.6.5

EL ./(1·':/-

-111- '.'.5'

B c D
 446 Operative Treatment of Specific Types of Fracture

A B

c _'--___ D
Fig. 22.32A- D. Osteosynthesis of a juxta-tectal transverse frac- brim. B-D The three standard views taken 23 years after opera-
ture (ilio-inguinal approach). A Antero-posterior radiograph tion. The hip is rated 6.6.6
before operation. The displacement was greatest at the pelvic
Associated Posterior Column and Posterior Wall Fractures 447

22.5.3 Special Features

In the case of a pure transverse fracture associated
with a vertical anterior fracture of the pelvis, the
posterior approach alone does not allow complete
reduction of the anterior fracture line (Fig. 22.30).

22.6 Associated Posterior Column

and Posterior Wall Fractures
(a) Approach. The posterior approach is always
(b) Reduction. After the posterior dislocation has
Fig. 22.33. To obtain reduction of the pelvic brim we can push
the ischial tuberosity inwards (l) or mobilise the whole ischio-
pubic fragment (2) by inserting into the ischial tuberosity either
a femoral head extractor or a Schanz screw
from above, has the tip of his left index finger
applied along the medial aspect of the ilio-pec-
tineal eminence, and knows that the screw has
(c) Fixation. It is sometimes possible, especially if
to pass a few rnillimetres away from the tip of
his finger;
from anteriorly, by an assistant, whose angle of
view is at 90 0 to that of the surgeon and who
ensures that the direction of the drill bit follows
the axis of the anterior column of which he has
visual control only up to below the ilio-pec-
tineal eminence. With these precautions, screws
of 100-130 mm in length can frequently be in-
serted along the axis of the anterior column.
Sometimes the surgeon's aim is not so good and the
large screw goes out through the outer cortex of the
superior pubis ramus, nevertheless providing an ex-
cellent fixation.
There is a wide choice of screw diameter: I like
to use 6.5-mm, but if the bone is thin I turn to the
4.5-mm screws. Joel MATTA often uses 3.5-mm
screws and that allows him to insert two of them in-
to the column. Figure 22.36 shows an instance of a
trans-tectal fracture approached and fixed in the
way described through the lateral approach.
Then a six- to eight-hole plate is contoured and
applied along the axis of the posterior column as
usual from the sub-cotyloid groove or the upper
pole of the ischial tuberosity up to above the
posterior part of the roof.
used.
been reduced and the acetabulum cleared of
debris, it is first necessary to reduce the posteri-
or column and ensure a perfect alignment both
at the angle of the greater sciatic notch and
along the quadrilateral surface. The posterior
column may be minimally displaced, but never-
theless it bears the lower part of the posterior
wall of the acetabulum and must be reduced ac-
curately.
Marginally impacted fragments, if present,
must be repositioned against the femoral head
and held in place by the overlying posterior
wall and, if necessary, bone graft.
the posterior column is displaced little or not at
all, to achieve fixation using one plate bridging
the posterior fragment and assuring fixation of
the posterior column to the posterior part of
the iliac wing at the same time. The posterior
fragment is sometimes initially fixed with
screws to the posterior column, to further re-
strict the problem to the fixation of the equiva-
lent of a posterior column fracture (Fig. 22.37).
If the column is significantly displaced,
reduction may be more difficult, requiring fixa-
tion in two stages. It is best to fix the posterior
column into a good position using a short plate
positioned just outside the angle of the greater
sciatic notch where screws find a good pur-
chase. The posterior associated fragment may
be fixed with screws but much more frequently
requires another plate (Figs. 22.38 and 22.39).
Soft tissue attachments of the fragments
should be preserved as much as possible.
If the angle of the greater sciatic notch is
detached together with the posterior column, it
is possible to insert a long screw from the angle
of the greater sciatic notch and to drive this for-
wards into the substance of the sciatic buttress.
This isolated screw obtains good fixation and
448 Operative Treatment of Specific Types of Fracture

Anterior pillar
of the iliac wi ng

1-2 cm

c
Screw 4.5 or 6_5 mm

Hip capsule

Fig. 22.34A-C. Diagrams showing the most usual types of fixa-

tion of a trans-tectal transverse fracture treated through the ex-
Surgeon's eye tended ilio-femoral approach (see text)
B looking from above and posteriorly
 Associated Posterior Column and Posterior Wall Fractures 449

A
B

Fig. 22.3SA-C Osteosynthesis of a juxta-tecta! transverse frac-

ture displaced anteriorly, not posteriorly (see Fig. 9.11 for pre-
operative state). The fracture was treated through the ilio-in-
guinal approach, which allowed fixation with the same plate of
the disrupted pubic symphysis. A Antero-posterior radiograph,
B obturator-oblique radiograph, C iliac-oblique radiograph, all
taken 3 years after operation. In spite of a collar of osteophytes
on the femora! head, the hip is rated 6.6.6. The plate has broken
at the hole of the first screw inserted into the right pubis, proba-
bly indicating that the symphysis still moves C
 450 Operative Treatment of Specific Types of Fracture

A B

c o

Fig. 22.36A-D
 Associated Posterior Column and Posterior Wall Fractures 451

E F

Fig. 22.36A- F. Osteosynthesis of a pure trans-tectal fracture.
A Antero-posterior radiograph, B obturator-oblique radiograph,
C iliac-oblique radiograph. The high transverse fracture line in-
volves the inferior pole of the sacro-iliac joint, D - F The three
radiographic views taken 3 months after operation. Through an
extended ilio-femoral approach, the transverse fracture was
fixed with a 3.5-mm lag screw posteriorly and a 7.4-mm screw
along the anterior column. A buttress plate was added along the
posterior column
 452 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.37 A-C Osteosynthesis of an associated posterior col-

umn and posterior wall fracture by means of one Vitallium plate
and one lag screw. A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 22
years after operation. See Fig. 11.4 for pre-operative state. The
hip is rated 6.6.6

Fig. 22.38A-D. Osteosynthesis of an associated posterior col-

umn and posterior wall fracture. A Obturator-oblique radio-
c graph, and B iliac-oblique radiograph, both taken before opera-
tion, C obturator-oblique radiograph, and D iliac-oblique radio-
graph, taken after operation. The small plate holds the posterior
column and the long plate bridges the posterior wall fragment
Associated Posterior Column and Posterior Wall Fractures 453

A
454 Operative Treatment of Specific Types of Fracture

Fig. 22.39 A - C. Osteosynthesis of an associated posterior wall

and posterior column fracture performed with the new type of
stainless steel plate and lag screws. A Antero-posterior radio-
graph, B obturator-oblique radiograph, C iliac-oblique radio-
graph, all taken 3 weeks after operation. See Fig. 11.7 for pre-
operative state
Associated Transverse and Posterior Wall Fractures
maintains the column in place, permitting at
leisure the final osteo-synthesis with a plate.
22.7 Associated Transverse
and Posterior Wall Fractures
The dislocation may be central or posterior. In
either case, and whatever the displacement of the
transverse fracture line, as soon as the posterior
wall fragment requires fixation, an access to the
posterior column is compulsory, and this can be
provided only by the Kocher-Langenbeck or the ex-
tended ilio-femoral approach. An anterior
approach does not allow reduction of the posterior
wall and therefore is an illogical choice. The routine
use of simultaneous anterior and posterior
approaches seems to us disproportionate with
respect to the lesion to be treated.
We have finally come to the conclusion that the
choice of approach to associated transverse and
posterior wall fractures is determined by the same
three elements as the choice of approach to pure
transverse fractures (see Sect. 22.5).
If the transverse component is trans-tectal, the
extended ilio-femoral approach is used.
If the transverse component is juxta- or infra-
tectal, the Kocher-Langenbeck approach is used
up to the first 15 days after injury, and during
the 3rd week the extended ilio-femoral, if there
is significant callus formation.
22.7.1 Kocher-Langenbeck Approach
(a) Reduction. The transverse component does not
pose any particular problem other than those
associated with the pure lesion which have been
detailed earlier. The acetabulum is cleaned and
the central or posterior dislocation reduced by
the usual manoeuvres. As a rule the head re-
mains stable underneath the roof.
The transverse fracture must be reduced
first. By retracting the posterior fragments, care
being taken with respect to soft tissue at-
tachments, if any, through the gap afforded
and by traction on the orthopaedic table which
separates the head of the femur from the roof
of the acetabulum, a view of the inside of the
acetabulum is obtained which allows the
transverse fracture line to be controlled under
455
vision. The reduction of the transverse fracture
is obtained by traction and direct manipulation
of the ischio-pubic fragment using Farabeuf
clamps, perhaps applied by means of tem-
porary screws. Any type of clamp may be used,
if convenient; quite often several must be tried
before the most useful is found. A Schanz
screw or a femoral head extractor inserted into
the ischial tuberosity may appear necessary in
some cases. During this process, the sciatic
nerve should be observed carefully. Intra-pelvic
reduction is verified by palpation of the inner
wall of the pelvis, particularly at the level of the
pelvic brim. When the greater sciatic notch has
been reduced apparently perfectly, should the
pelvic brim remain displaced, it is almost cer-
tainly due to a horizontal rotation, which must
be corrected as described earlier.
(b) Fixation. When the transverse fracture de-
taches the angle of the greater sciatic notch
with the inferior fragment (Fig. 22.40) it is
valuable to insert a 3.5-mm screw directly into
the angle in an anterior direction which will
maintain the reduction of the transverse frac-
ture (Fig. 22.42 B). If this is not possible, the
transverse component must be fixed by means
of a plate applied very near to the greater
sciatic notch; in the superior fragment, the
screws need be only fairly short (20-25 mm)
but their hold in this dense bony area is ex-
tremely firm. Along the margin of the greater
sciatic notch, the screws need to be progressive-
ly longer, some being parallel to the quadri-
lateral plate surface and able to gain attach-
ment to the pelvic brim, while the majority are
Fig. 22.40. Scheme of two methods of fixation through the
posterior approach for associated transverse and posterior wall
fractures
456 Operative Treatment of Specific Types of Fracture

directed obliquely inwards, so as to gain pur-

chase on the quadrilateral surface of the
ischium (Figs. 22.41 and 22.49). The posterior
wall fragment or fragments are then reduced
and fixed posteriorly by another plate or some
simple screws.
We no longer risk the fixation of these frac-
tures with screws alone for fear of loss of stabil-
ity during early passive movements; we have,
however, a number of good results from such
simple fixations, performed many years ago
(Fig. 22.42), although in those we used a plaster
cast post-operatively.

22.7.2 Extended Ilio-femoral Approach

(a) Reduction. The transverse component is ap-

proached and assessed as easily as if it were a
pure transverse fracture (see Sect. 22.5.2). We
always try to expose only the external iliac
fossa, but if we feel the need for an access to
the internal iliac fossa so as to have a perfect
view of the fracture line and remove the already
formed callus, we elevate the internal iliac fossa
as far as necessary.
The elevation of the posterior wall fragment
or fragments, the capsular attachments of
which we try to preserve, and if necessary a
capsulotomy along the acetabular lip combined
with longitudinal and lateral traction, allow
checking of the reduction of the transverse
fracture component. When the reduction is
perfect and maintained with clamps, applied if
necessary to temporarily inserted screws, fixa-
tion is completed.
(b) Fixation. The transverse component is fixed
posteriorly with a 3.5-mm lag screw and anteri-
orly either with a long 4.5- or 6.5-mm screw in-
serted along the long axis of the anterior col-
umn (see Sect. 22.5.2 and Fig. 22.43) or with a
short plate (Figs. 22.44 and 22.48). Then the
posterior wall fragment or fragments are re-
positioned perfectly and often fixed initially
with isolated screws. A curved acetabular plate B
is then applied along the posterior column,
Fig. 22.41 A- E. Osteosynthesis of an associated transverse and
spanning the transverse fracture line and the posterior wall fracture. A Antero-posterior radiograph, B ob-
posterior wall fragments. turator-oblique radiograph, both taken 7 years after operation.
It may be asked why we do not use the long See Fig. 12.3 for pre-operative state. Clinically the result was very
and large screws to fix both extremities of the good at this time, but note the collar of osteophytes around the
femoral head margins. C-E The three radiographic views taken
transverse fracture? The answer is: the anterior 25 years after operation. The hip is now rated 5.6.5 and there
part may always be fixed with a large screw, are type III or IV osteoarthritic changes. However, the other hip
which in our experience has been of variable (inset in C) is also osteoarthritic, although to a lesser degree
 Associated Transverse and Posterior Wall Fractures 457

c D

E Fig. 22.41C-E
458 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.42A-F. Osteosynthesis of an associated transverse and

posterior wall fracture. A Antero-posterior radiograph, B ob-
turator-oblique radiograph, C iliac-oblique radiograph, all taken
9 years after operation. See Fig. 12.2 for pre-operative state. A
screw (arrow in A) has been inserted directly into the angle of
the sciatic notch. The case was operated upon in 1959. D - F The
three radiographic views taken 27 years after operation demon-
strate some arthritic changes, type III; so far these are purely
radiological
 Associated Transverse and Posterior Wall Fractures 459

D E

Fig. 22.42D-F
 460 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.43 A - C. Osteosynthesis of a trans· tecta! transverse asso-

ciated posterior wall fracture through the extended ilio-femora!
approach. A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph, all taken 4 years after
operation. See Fig. 12.16 for pre-operative state. The hip is now
rated 6.6.5; the ectopic bone, graded Brooker type IV, is accom-
c panied by normal hip mobility
T-shaped Fractures
length as in some cases the drill bit was not
perfectly directed, or the anterior column was
so concave that a good shot was unattainable.
However, to insert a screw all along the axis of
the posterior column, although often possible,
is difficult; in my opinion, it is sometimes not
possible at all: this screw insertion depends on
the shape of the iliac wing, and it can happen
that the curvature of the wing may impede the
drilling of the long axis of the posterior col-
umn.
22.7.3 Special Features
Fractures of the posterior wall are sometimes asso-
ciated with an incomplete hemitransverse element.
If this hemitransverse component is not displaced,
it suffices to perform an osteosynthesis of the
posterior wall with a plate which also bridges the
hemitransverse component. If the hemitransverse
component is displaced, even by only a few
millimetres, this must be corrected by using forceps
or levering one fragment on the other, and then
fixed by at least two of the screws which hold the
plate spanning the posterior fragment (Fig. 22.45).
Concerning the exceptional occurrence (18 cases
out of 940) of a transverse fracture associated with
a postero-superior fragment of which the fracture
line extends to the iliac crest, thus detaching the en-
tire part of the articular surface situated above the
transverse fracture (see Sect. 12.4), the reduction
and fixation obey the same rules as in both-column
fractures.
T-shaped fractures associated with a fracture of
the posterior wall are rare (30 cases) and the vertical
branch of the T is rarely displaced significantly.
Their reduction is difficult. The posterior approach
is essential but sometimes a secondary anterior
operation has been necessary. As in the transverse
and posterior wall associated fractures, the
posterior column segment is reduced first and
brought into alignment with the wing. Then the
fragments of the posterior wall, sometimes with
marginal impaction, are repositioned with respect
to the posterior column. Finally the segment of the
anterior column must be reduced.
In one case associating a T-shaped fracture with
a posterior wall fracture, a posterior dislocation
was converted into a central dislocation during
reduction. The anterior approach was wrongly
chosen for the reduction and subsequently it was
necessary to perform a posterior approach during
461
the same operative procedure because the reduction
was not perfect posteriorly (Fig. 22.46). The signifi-
cance of the original posterior dislocation had not
been appreciated. To approach these associated T-
shaped and posterior wall fractures we now use the
same rules as for the associated transverse and
posterior wall fractures, and the type of the
transverse fracture component remains the essen-
tial factor in the choice of approach (Figs. 22.46
and 22.50).
22.8 T-sbaped Fractures
The reconstruction of T-shaped fractures requires
perfect repositioning of the inferior segment of
each column with respect to the other and of each
with respect to the iliac wing. These are among the
most difficult fractures to treat, and if one elective
route is chosen (Kocher-Langenbeck or ilio-ingui-
nal) one can never be sure of being able to complete
the reconstruction without the other approach
becoming necessary. Appropriate use of the extend-
ed ilio-femoral approach often has to be discussed.
As for the other types of fractures comprising a
transverse component, it is the location of this
component which is the essential element directing
the choice of approach.
If the transverse component is trans-tectal, we
now use the extended ilio-femoral approach.
If the transverse component is juxta-tectal or in-
fra-tectal, the route is chosen according to
where there is greatest displacement of the in-
ferior segments of the columns: if the posterior
column is more displaced we choose the
Kocher-Langenbeck; if the anterior column seg-
ment is more displaced, we choose the ilio-
inguinal.
In the 3rd week after injury, if there is signifi-
cant callus formation, the extended ilio-femoral
approach is used for all T-shaped fractures.
(a) Kocher-Langenbeck approach. The position of
the head is restored under the remaining sector
of the roof, by simple traction, but not before
the posterior column fragment has been re-
tracted for examination of the interior of the
acetabulum and removal of debris. This ma-
noeuvre also allows one to check that the head
has been perfectly reduced. If necessary, access
to the joint is enlarged by a capsulotomy along
the acetabular lip; this will also permit
 462 Operative Treatment of Specific Types of Fracture

E..l. '95

A B

c Fig. 22.44 A - D
 T-shaped Fractures 463

E F

Fig. 22.44A-G. Osteosynthesis of an associated trans-tectal

transverse and posterior wall fracture. A Antero-posterior radio-
graph, B obturator-oblique radiograph, C iliac-oblique radio-
graph before operation. Note the sharp edge of the upper part
of the posterior column, which caused an injury to the superior
gluteal artery. D Intra-operative view showing the trans-tectal
transverse and associated posterior wall fracture. The case was
operated on through an extended ilio-femoral approach, E-G
the three radiographic views 7 years after operation. Despite the
ectopic bone (grade 111-6), hip function is perfect G
 464 Operative Treatment of Specific Types of Fracture

Fig. 22.45A-D. Osteosynthesis of an associated incomplete

transverse and posterior wall fracture (posterior approach).
A Antero-posterior radiograph before operation, B antero-pos-
terior radiograph, C obturator-oblique radiograph, and D iliac-
oblique radiograph 8 years after operation. The hip is rated 6.6.6

A_ _ ~~ _ _ _--,,_.......:o_,--

B C D
 T-shaped Fractures 465

A B

Fig. 22.46A-c' Osteosynthesis of an associated T-shaped and

posterior wall fracture (anterior and posterior approaches con-
secutively). A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph, all taken 22 years after
operation. See Fig. 12.6 for pre-operative state. The hip is rated C
6.6.6
 466 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.47 A - C Osteosynthesis of an associated trans-tectal

transverse and postero-superior fracture extending to the iliac
crest (extended ilio-femoral approach). See Fig. 12.13 for pre-
operative condition. A Antero-posterior radiograph, B obtura-
tor-oblique radiograph, C iliac-oblique radiograph, all taken 1
year after operation. The postero-superior fragment, comprising
the anterior part of the wing, was fixed with a plate along the
crest and one 6.5-mm screw between the two tables of the iliac
wing. The transverse fracture needed a 3.5-mm lag screw and a
stainless steel plate along the posterior column. The per-tro-
chanteric fracture was fixed during the same operative session
with a ludet-Letournel screw-plate
c
 T-shaped Fractures 467

A B

Fig. 22.48A-C. Osteosynthesis of a trans-tectal transverse frac-

ture with a total postero-superior fragment extending to the an-
terior border of the iliac wing (extended ilio-femoral approach).
1\\10 plates were needed to fix the transverse fracture and the
postero-superior fragment. One lag screw was inserted to fix the
anterior column. A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 6 years
C
after operation. Hip function is normal
 468 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.49 A - C. Osteosynthesis of an associated juxta-tectal

transverse and posterior wall fracture (Kocher-Langenbeck
approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph, all taken 4 years after
operation. See Fig. 12.15 for pre-operative state. Hip function is
c normal
T-shaped Fractures
checking of the reduction of the posterior col-
umn segment.
The posterior column fragment is then
reduced, its reduction being maintained by a
clamp directly applied to the fragment and the
intact part of the iliac wing, or by a clamp ap-
plied to two temporarily inserted screws (one in
the detached posterior column, the other in the
intact part of the bone). This reduction is
checked posteriorly and at the level of the
posterior part of the transverse fracture line
dividing the quadrilateral surface, by a finger
introduced through the greater sciatic notch.
The reduction must be perfect.
The posterior column fragment is often ini-
tially fixed with one or two lag screws
(3.5-mm), which allow removal of the clamp. A
6- to 8-hole, straight or curved acetabular plate
is moulded on the back of the posterior col-
umn, as in the pure posterior column fracture
(Fig. 22.51 A). The screws inserted are chosen
carefully so as to fix only the posterior column
and to remain confined therein and not impede
the reduction of the anterior column frag-
ment which is performed as a second stage
(Fig. 22.51 B).
The reduction of the anterior column seg-
ment is now attempted. This fragment can be
mobilised by the finger, an elevator or a clamp
introduced through the greater sciatic notch.
Reduction is maintained by special clamps,
especially the Matta clamps, which can grip the
anterior fragment and the "above-the-roof"
area without making contact with the posterior
column (Fig. 22.51 C).
Reduction of the anterior column fragment
is assessed by means of a finger inserted
through the greater sciatic notch, alongside the
quadrilateral surface. If reduction is possible
and can be maintained in a satisfactory man-
ner, it is possible, from the posterior aspect
of the posterior column above or below the
fracture line, to achieve fixation of the anterior
fragment by means of long screws (Fig.
22.51 D, E). These may be inserted into the
retro-acetabular surface, or sometimes through
the plate, some of the holes of which had been
left deliberately empty at the time it was used
to fix the posterior column component. These
various long screws may be directed parallel
with the quadrilateral surface, crossing the
fracture line, thereby reaching the pelvic brim,
the crossing of which with the drill bit can be
469
clearly felt, avoiding injury to the elements in
front of it. Some pierce and hold the quadri-
lateral surface itself (Figs. 22.53 and 22.54).
If reduction of the anterior column segment
is not possible through the posterior approach
and if the state of the patient permits, he can
be turned over and, through an ilio-inguinal
approach, the anterior column reduction can
be completed and held by a moulded plate laid
along the pelvic brim (Figs. 22.55 and 22.56).
Strength of the fixation is enhanced by long
screws inserted parallel to the quadrilateral sur-
face and uniting the segments of the two col-
umns.
Through a posterior route an attempt may
also by made to fix the anterior column with a
long screw inserted along the axis of the col-
umn from the above-the-roof area, but this
generally requires partial or complete section of
the gluteal tendons from the greater trochanter.
If reduction of the anterior column segment
appears difficult or even impossible through
the posterior route, the incision can be enlarged
into a triradiate incision, as described in
Sect. 20.8.
Through the posterior route it has also
appeared in a few cases that combining longitu-
dinal traction and retraction of the posterior
column segment allows a very good view and
control of the anterior column fragment, which
is reduced and fixed with long screws starting
from above the roof and inserted under direct
intra-articular view into the anterior column
segment. When this fixation is complete, trac-
tion is released, the head repositioned, and
reduction and fixation of the posterior column
performed as described above.
(b) Ilia-inguinal approach. Indications for using
the anterior route to a T-shaped fracture are ex-
ceptional.
The anterior column segment is reduced and
fixed first: one or two isolated screws maintain
the reduction and a plate is contoured to lie
along the upper aspect of the pelvic brim, care
being taken that below the transverse fracture
line the screws penetrate only the anterior col-
umn fragment. Then reduction of the posterior
column fragment is attempted, and, if it is suc-
cessful, the segment is fixed with long screws
entering via the holes of the anterior plate or
apart from it, and running parallel to the
quadrilateral surface so as to emerge at the
retro-acetabular surface, thus strongly uniting
470 Operative Treatment of Specific Types of Fracture

the two segments of the column (Figs. 22.52

and 22.57). Another possibility is to insert a
long screw along the axis of the posterior col-
umn as described in Fig. 22.67E.
If the reduction of the posterior column
segment appears impossible, the ilio-inguinal
incision is closed and a posterior approach is
subsequently performed after complete
resuscitation of the patient, in order to fix the
posterior column in the usual way.
(c) Extended ilio-femoral approach. This approach
provides complete access to and control of the
transverse fracture line, if necessary by elevat-
ing the internal fossa. The posterior column
fragment is easily controlled up to the upper
half of the ischial tuberosity and the anterior
column fragment up to the ilio-pectineal
eminence via a capsulotomy along the acetabu-
lar lip; finally, a complete and perfect view of
the intra-articular tracks of all the fracture lines
is gained through the same incision.
The technique of reduction varies from case
to case. One can begin by reducing either of the
columns. They are manipulated directly or by
means of clamps and their reduction is
maintained by clamps applied directly to the A
bone or to the heads of temporarily inserted
screws.
The fixation also varies. The anterior col-
umn segment is usually fixed with a long 4.5- or
6.5-mm screw inserted along its long axis exact-
ly as described in Sect. 22.5.2 (Fig. 22.59). It
has happened that the reduction of this
anterior column fragment needed the elevation
of a part of the internal iliac fossa; in such a
case the upper part of the fragment may be
fixed with two or three 3.5-mm screws inserted
from the inner aspect of the bone, aiming above
the acetabulum.
The posterior column, when I am operating,
is regularly fixed with a long, curved acetabular
plate, but prior to this one or two lag screws are
used to fix the upper part of the column. It is,
however, possible to fix the posterior column as
well with a long, large screw along its axis
(Fig. 22.58), starting behind the anterior pillar
and 2 - 3 cm above the greater sciatic notch.
This technique appears to be more difficult
than that for the anterior column. The reason
for this is the shape of the iliac wing, which
often does not allow the drill bit to go very far
inside the column. For a reliable fixation, the
8
screw needs to have a long track inside the col-
 T-shaped Fractures 471

C D

Fig. 22.50 A - E. Osteosynthesis of an incomplete T-shaped frac-

ture (dividing the cotyloid fossa but not the ischio-pubic ramus)
associated with a posterior wall fracture with posterior disloca-
tion (extended ilio-femoral approach). A,B Pre-operative views,
CAntero-posterior radiograph, D obturator-oblique radio-
graph, E iliac-oblique radiograph, all taken 16 months after
operation. The transverse fracture was fixed with two long
4.5-mm lag screws and a plate to fix the posterior fragment. The
hip is rated 6.6.6 E
472 Operative Treatment of Specific Types of Fracture

Track of the plate

1-

B
Screw remains
confined within
the posterior column

Fig. 22.51 A-E. Fixation of a T-shaped fracture through a posterior approach:

A fixation of the posterior column fragment with a 3.5-mm lag screw and a curved
plate. B How to confine the screws within the posterior column. C How to reduce
the anterior column fragment through the greater sciatic notch. A Matta clamp
maintains the fragment while it is fixed. D How to insert the screws from the ratio-
acetabular surface to fix the anterior column. E Final appearance

~~
//
//
---,#.::::::::
====:''''"t~~:=:...:.~--------
--- -=--
D
-----~--------

Fig. 22.52. Fixation of a T-shaped fracture through an ilio-in-

guinal approach
 T-shaped Fractures 473

A B

Fig. 22.S3A-C. Osteosynthesis of T-shaped fracture (posterior

approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, both taken 1 year after operation. See Fig. 10.16 for
pre-operative state. Note in B the long screws extending from the
plate to the pelvic brim. C Radiograph taken 16 years after
operation: osteoarthritis has set in. The hip is rated 4.4.4; the pa-
tient needs a total hip replacement C
 474 Operative Treatment of Specific Types of Fracture

Fig. 22.S4A-E. Osteosynthesis of a T-shaped fracture (posteri-

or approach). A Antero-posterior radiograph before operation,
B antero-posterior radiograph 3 years after operation, C antero-
posterior radiograph, 0 obturator-oblique radiograph, and E il-
iac-oblique radiograph, all taken 20 years after operation. The
E hip has normal function
Associated Anterior and Hemitransverse Posterior Fractures
umn, but in trying to follow the long axis of the
column we quite often emerge too early with
the bit, through the quadrilateral surface or the
posterior aspect of the ischium, and therefore
the fixation is not as good as could be wished.
22.S.1 Special Features
In T-shaped fractures with a stem posteriorly
disposed and trans-ischial, the posterior approach
always suffices; it has been necessary to expose the
inferior part of the ischium in order to achieve a
good reduction, at the price of total detachment of
the quadratus femoris muscle origin and sharp
dissection of the upper part of the origins of the
hamstring muscles. The anterior fragment, which
includes with the anterior column a large part of
the quadrilateral surface and even part of the poste-
rior column, is fairly accessible through the posteri-
or approach.
If it is not possible to reduce the two main frac-
ture lines at the same time, it may be easier to con-
centrate on joining the two inferior fragments and
then to complete the reduction and fixation of
what now amounts to a pure transverse fracture
(Fig. 22.55).
T-shaped fractures in which the stem is oblique
and directed forwards can also be approached, at
least initially, through the posterior route. The
reduction of the somewhat slender anterior column
fragment has not always been perfect even though
accepted as sufficient. However, this fragment car-
ries the anterior wall of the acetabulum which stays
slightly displaced, and may be followed by the
head, allowing central displacement of the latter,
visible on the post-operative radiographs by a loss
of parallelism of the superior joint space, or even
a slight central protrusion. Today, this slight
residual head displacement justifies for us a subse-
quent ilio-inguinal approach.
22.9 Associated Anterior and
Hemitransverse Posterior Fractures
The majority of lesions involving the anterior col-
umn necessitate an anterior approach. The ilio-
inguinal approach gives access to the whole
anterior column of which the repair can and must
be perfect. Through the medial or internal window
475
of the ilio-inguinal incision, access is available to
the posterior hemitransverse component, which
must not be neglected. It can be assessed by palpa-
tion as far backwards as the anterior border of the
greater sciatic notch. The lesion of the anterior col-
umn or wall is repaired exactly as would be the case
for the simple fractures. These can be reduced and
fixed by a long curved acetabular plate applied
along the brim of the pelvis and may be sup-
plemented by isolated screws used originally to
hold the reduction while the plate was contoured or
to fix secondary fracture lines, sagittal in direction,
which the plate does not span (Fig. 22.61).
The fractures of the upper part of the anterior
column have on occasion been fixed by long 7.4- or
6.5-mm screws inserted from the anterior aspect of
the iliac crest or through the interspinous notch and
driven posteriorly into the thickest part of the iliac
wing. The assembly is solid but during insertion
there may be a tendency for telescoping to occur,
with consequent loss of anatomical reduction. The
lower part of the column or the lower fragments
thereof (Fig. 22.62) were fixed by isolated screws.
The anterior lesion having been reduced and
fixed, there remains the problem of the hemitrans-
verse posterior element. If this is situated low, split-
ting longitudinally the spine or cutting the lesser
sciatic notch, it is as a rule little or not displaced.
The inferior fragment of the posterior column is in
this case difficult to manipulate from the anterior
approach and its small displacement can be neg-
lected; it is not related to a significant amount of
the posterior wall and will consolidate quickly,
perhaps with a slight articular incongruence at the
lower part of the posterior wall. This little posterior
fault does not justify a subsequent posterior opera-
tion.
If the fracture is high, cutting the greater sciatic
notch at a variable level, perhaps at the inferior or
the superior third, it is not to be neglected, even
though the displacement may appear only slight. It
crosses the posterior wall of the acetabulum and it
must be reduced. If the screws of the anterior col-
umn have been placed with care so as not to pro-
trude from its posterior margins, it may be possible
by means of an elevator or with the jaws of the
asymmetrical forceps applied, one on the long plate
screwed to the pelvic brim, and the other on the
posterior column, to obtain reduction of the
hemitransverse posterior fracture line; the Matta
clamps can also be very helpful. The reduction is
maintained by screws which can be conveniently in-
serted somewhat vertically downwards from the
 476 Operative Treatment of Specific Types of Fracture

c D

Fig. 22.SSA- E. Osteosynthesis of a T-shaped fracture (Kocher-

Langenbeck and ilio-inguinal approaches). A Antero-posterior
radiograph before operation, B antero-posterior radiograph
after completion of the first stage. CAntero-posterior radio-
graph, D obturator-oblique radiograph, and E iliac-oblique
radiograph, all taken 16 years after operation. The hip has nor-
mal function
Fig.22.S6A-C. Osteosynthesis of an anterior T-shaped frac-
ture (Kocher-Langenbeck approach followed by ilio-inguinal).
A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph, all taken 3 years after opera-
tion. See Fig. 10.10 for pre-operative state
 Associated Anterior and Hemitransverse Posterior Fractures 477

B
c
 478 Operative Treatment of Specific Types of Fracture

A 8

c o
Fig. 22.57 A-E. Osteosynthesis of a T-shaped fracture with a ro-posterior radiograph, 0 obturator-oblique radiograph, and E
juxta-tecta! transverse component and major displacement in iliac-oblique radiograph 16 years after surgery. Hip function is
front (ilio-inguinal approach). A Antero-posterior radiograph normal: this patient developed endo-pelvic ectopic bone without
and B obturator-oblique radiograph before operation, C Ante- any functional impairment
 Associated Anterior and Hemitransverse Posterior Fractures 479

Fig. 22.58 A- D. Osteosynthesis of an associated posterior col-

umn and anterior hemitransverse fracture (extended ilio-femoral
approach). It was possible to achieve internal fixation with only
three long 4.5-mm screws. A Antero-posterior radiograph, B ob-
turator-oblique radiograph, and C iliac-oblique radiograph, all
taken 1 year and 3 months after operation. See Fig. 10.19 for
pre-operative state. D Diagram of the internal fixation
T

E D
Fig. 22.57E

A 8 C
 480 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.S9A-C. Osteosynthesis of a T-shaped fracture (extended

ilio-femoral approach). A Antero-posterior radiograph, B ob-
turator-oblique radiograph, and C iliac-oblique radiograph, all
taken 6 months after operation. A stainless steel plate and screws
(3.5 mm and 4.5 mm) were used for the fixation. See Fig. 10.17
c for pre-operative state
 Associated Anterior and Hemitransverse Posterior Fractures 481

Fig. 22.60. Osteosynthesis of a posterior T-shaped fracture using

a Y-plate (posterior approach). Antero-posterior radiograph 9
years after operation. See Fig. 10.13 for pre-operative state

Line of associated
anterior column
and posterior hemi -transverse
fractures

Associated anterior wall

and posterior hemi-transverse
fractures

Fig. 22.61. Diagram of method of

fixation for associated anterior and
posterior hemitransverse fractures
through the ilio-inguinal approach
 482 Operative Treatment of Specific Types of Fracture

8
A

Fig. 22.62A-C. Osteosynthesis of an extended anterior column

fracture associated with a posterior hemitransverse fracture.
A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph, all taken 23 years after opera-
tion. See Fig. 13.9 for pre-operative state. Hip function is nor-
c mal
 Associated Anterior and Hemitransverse Posterior Fractures 483

A B

Fig.22.63A-C. Osteosynthesis of an associated anterior wall

and posterior hemitransverse fracture (ilio-inguinal approach).
A Antero-posterior radiograph, B obturator-oblique radio-
graph, C iliac-oblique radiograph after operation. See Fig. 13.13
for pre-operative state C
484
posterior or middle third of the upper aspect of the
pelvic brim while working through the external
window of the ilio-inguinal incision. The screws
may be inserted through the holes of the plate or
apart from it. Access is easy lateral to the psoas.
The long drill bit is easily protected and with a
finger one can check beyond the pelvic brim that
the reduction is maintained while drilling. The
screws gain purchase in the segment of the
quadrilateral surface situated below the hemitrans-
verse fracture, and the finger checking the reduc-
tion also checks for the penetration of the drill bit
and the screw through the quadrilateral surface
(Figs. 22.64 - 22.66). Additionally, long screws
parallel to the quadrilateral surface and crossing
the fracture line can be driven so as to reach the
retro-acetabular surface (Figs. 22.61 and 22.63).
Note: It is in fact CT scanning which makes it
possible to see how much of the articular surface is
taken along with the segment of the posterior col-
umn below the hemitransverse fracture. Depending
on how extensive it is, perfect reduction of the
hemitransverse posterior fracture mayor may not
be necessary.
If the fracture is being operated on during the
3rd week after injury, and if the posterior hemi-
transverse component is significantly displaced, the
extended ilio-femoral approach is advised and will
allow control of all the fracture lines from the ex-
ternal aspect or both aspects of the iliac wing, the
different fracture lines being fixed by means of
plates, lag screws, or screws inserted between the
two tables of the wing.
22.10 Both-Column Fractures
The surgical complexity of these fractures is the
result of the fact that the whole articular surface of
the acetabulum is detached in several pieces which
belong to different fragments and which no longer
have any connection with the undisplaced part of
the iliac wing. Unlike the T-shaped fracture, there is
no sector of roof under which the head can be
brought as a first step in the reduction procedure.
22.10.1 Approach
A both-column fracture can be reconstructed by
any of the three approaches described: Kocher-
Langenbeck, ilio-inguinal, or extended ilio-femo-
ral, or by the first two sequentially. The indications
Operative Treatment of Specific Types of Fracture
for the choice of approach are given in detail in
Sect. 21.3.3.
22.10.2 Reduction and Fixation
Through Posterior Approach
The Kocher-Langenbeck incision has been used in
40 cases. An immediate striking feature is the ex-
treme mobility of the fragments and of the head.
Traction in abduction and in different degrees of
rotation is necessary, and by trial and error, it can
be established what best leads to extraction of the
head from the pelvis, and to realignment of the
profile of the greater sciatic notch.
Next, it is necessary to reduce the posterior col-
umn. There are several ways of checking the quality
of this reduction:
by the reduction of the retro-acetabular fracture
line which separates the posterior column from
the iliac wing fragment above, the latter remain-
ing in its proper place;
by the restoration of the normal profile of the
greater sciatic notch, which one must learn to
recognise in order to avoid an excessive angula-
tion;
by endopelvic palpation of the reduction of the
column with respect to the posterior part of the
pelvic brim (which remains in place) and to the
upper part of the fracture line which cuts the
quadrilateral surface of the ischium.
Rotation of the posterior column around a vertical
axis may exist and must be corrected.
It is necessary to be very careful about the initial
reduction of the posterior column, for if it is not
exact, the subsequent reconstruction of other frac-
ture components will be compromised and become
successively more difficult and inaccurate.
The posterior column is manipulated using for-
ceps applied if necessary to one or two screw heads.
Correction of central dislocation may be difficult
to maintain; a Lambotte hook placed under the
neck can be used to exert traction, but it is often
much more effective to use the head extractor in-
serted into the vastus lateralis crest along the axis
of the neck of the femur.
Once the column has been reduced, it is fixed
with a plate suitably moulded along the mid-part
of the retro-acetabular surface, about 1- 2 cm
parallel to the greater sciatic notch (Fig. 22.67 A)
and extending above the angle to the posterior part
of the iliac wing in front of the sacro-iliac joints.
This plate is screwed with care so that the screws
 Both-Column Fractures 485

A 8

Fig. 22.64A-C Osteosynthesis of an intermediate anterior col-

umn and associated hemitransverse fracture (ilio-inguinal
approach). Fixation was with a stainless steel plate and lag
screws. A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph, all taken 1 year after
operation. See Fig. 13.15 for pre-operative state. The hip has
C normal function
 486 Operative Treatment of Specific Types of Fracture

A B

c D

Fig. 22.6SA-D
 Both-Column Fractures
E F
Fig. 22.6SA-F. Osteosynthesis of a typical anterior wall frac-
ture associated with a posterior hemitransverse fracture. A
Antero-posterior radiograph, B obturator-oblique radiograph,
and C iliac-oblique radiograph before operation, D antero-
posterior radiograph, E obturator-oblique radiograph, and F il-
remain confined in the posterior column, reaching
only that part of the quadrilateral surface which
belongs to the posterior column (Fig. 22.67 A); they
must not impede the reduction of the anterior col-
umn subsequently. Inferiorly, the plate should
reach as far as the superior pole of the ischial
tuberosity where the hold is excellent. In the iliac
wing, at least three screws are required, and accor-
dingly, plates with six to eight holes are required to
span the whole distance adequately.
After the posterior column has been fixed, it re-
mains to reduce the iliac wing fracture as well as a
posterior wall fragment if present. To reduce an il-
iac wing fracture which reaches the anterior border
of the ilium, it is necessary to have access to its
whole length and to verify the perfect reduction;
otherwise, if we verify only the back part of this
fracture, its reduction may appear satisfactory,
while the anterior part remains slightly displaced
and consequently the continuity of the ilio-
487
iac-oblique radiograph 7 years after operation. The fracture was
operated on through the ilio-inguinal approach and fixed with
a vitallium plate and lag screws. Note the two long screws fixing
the posterior hemistransverse fracture line. At 7 years after
operation, hip function was normal
pectineal line is not restored and an incongruency
of the joint will ensue. In order to assess the frac-
ture line it is possible to elevate the lower part of
the gluteal muscle origins as far forwards as the an-
terior interspinous notch. It is perhaps less damag-
ing, however, to cut their tendons of insertion a few
millimetres from the greater trochanter and then to
elevate them from the lower part of the iliac wing
as far as the fracture line.
In order to reduce the iliac fracture line, it may
be possible to disimpact the fragments by levering
them apart, and then reducing them, or it may
prove necessary to use a screw inserted into the
supra-acetabular region of the anterior column,
and by means of forceps the anterior column can
be drawn backwards. A Lambotte hook slid into
the interspinous notch may help to draw the
anterior column posteriorly. Having achieved
reduction, we have usually fixed the anterior and
posterior columns together with a plate bent so as
488 Operative Treatment of Specific Types of Fracture

A B

Fig.22.66A-C. Osteosynthesis of an associated anterior col-

umn and incomplete hemitransverse fracture (ilio-inguinal
approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac-oblique radiograph 3 years after opera-
tion. Fixation was with a stainless steel plate and screws. Note
the two long screws, one starting from the plate and one apart
from it fixing the hemitransverse fracture line. See Fig. 13.5 for
c pre-operative state
Both-Column Fractures 489

B C

H--'--+ ----:f-Posterior border of the

obturator foramen
Fig. 22.67 A -- E. Diagram of methods of fixa-
tion for both-column fractures. See text.
E shows the area of penetration for the screw
aiming to follow the axis of the posterior col-
Second aim : halfway between
umn when operating through the ilio-inguinal
ischial spine and posterior
approach. The innominate bone is seen nearly border of the obturator foramen
as in the iliac-oblique view. The starting point
is in the posterior part of the internal iliac
fossa: if a coronal line is drawn from the
posterior end of the pelvic brim, the starting
point is located 10 mm in front of this line
and 25 mm lateral to the pelvic brim. The
surgeon directs the drill bit so as to have it
parallel to the quadrilateral surface (his index
finger acting as a guide) and to pass halfway
between the ischial spine and the posterior
border of the obturator foramen. In practice
the point of entry is not strictly a point but a
circle about 15 -- 20 mm in diameter around E
this point
490 Operative Treatment of Specific Types of Fracture

Fig. 22.68 A-E. Osteosynthesis of a both-column fracture (pos-

terior approach), the iliac wing fracture line reaching the inter-
spinous notch. A posterior fragment was discovered only at
operation. A Antero-posterior radiograph before operation,
B antero-posterior radiograph after operation, C scheme show-
ing fracture configuration, D antero-posterior radiograph, and
E iliac-oblique radiograph 7 years after operation. One plate was
used to fix the posterior column to the posterior part of the iliac
wing, and another to fix the iliac wing to the anterior column.
Two isolated screws hold the postero-superior fragment which
was not bridged by a plate. Despite ectopic bone formation a
o very good clinical result was obtained
Both-Column Fractures
to run approximately parallel to the postero-
superior margin of the acetabulum (Fig. 22.67 B).
This plate may have no contact with the part of the
iliac wing which has remained in its proper place,
but it is probably better if it does.
The reduction of the anterior column is checked
by a finger introduced into the inside of the pelvis
where it can detect the fracture line separating the
two columns at the level of the quadrilateral sur-
face, as well as at the pelvic brim.
In the majority of cases approached by this
route, there exists an isolated fragment of the poste-
rior wall of the acetabulum. If this comprises a
mid-posterior fragment, it can be reduced straight
away with respect to the posterior column and held
in place with one or two screws. The entire posteri-
or surface of the pelvic bone is therefore complete
again, and subsequently, the posterior fragment
will be bridged by one or the other of two plates
(Fig. 22.68).
If the fragment is postero-superior, we reduce it
after the reduction of the posterior column, usually
at the same time as the anterior column; it is
bridged by the plate and may be maintained by one
or two supplementary screws (Fig. 22.68).
Finally, it may be wise to reinforce further the
whole reconstruction using one or two long screws,
inserted parallel to the quadrilateral surface per-
haps through the plate fixing the posterior column
(by replacing one short screw, inserted earlier) or
near it.
Never forget to verify in silence that during
movements of the hip, which should be free, there
is no grating sensation indicating contact between
the head and an intra-articular screw.
In one rare case, when the posterior column
took with it a segment of the pelvic brim (see
Fig. 14.29) and in which the fracture line reached
the anterior border of the ilium, the posterior route
allowed a very good reduction which could be fixed
easily (Fig. 22.69).
Note to second edition: It is true to say that all of
this is still possible today, which is why we have left
this description. Nowadays, however, we no longer
use the Kocher-Langenbeck approach to treat a
both-column fracture; whatever the type of the iliac
wing fracture line, we choose between the ilio-
inguinal and the extended ilio-femoral approaches.
The ilio-inguinal is chosen as often as possible
since it is never followed by ectopic bone forma-
tion. Its only drawback in relation to the two other
approaches is that one can never fix an associated
491
posterior wall fragment through an anterior route.
Another possibility is to transform a Kocher-
Langenbeck into a triradiate approach, or to add to
the ilio-inguinal incision a Rueter type anterior in-
cision and elevate the external iliac fossa.
22.10.3 Reduction and Fixation
Through IIio-inguinaJ Approach
The incision must always extend beyond the con-
vexity of the crest posteriorly, being taken further
if the fracture line in the iliac wing is very posterior
of if there is a loose triangular fragment. Anterior-
ly, it reaches the median line but it must be pro-
longed beyond this if access to the pubic symphysis
is required, or if fixation to the opposite pubis is
necessary on account of comminution in the region
of the angle of the ipsilateral pubis.
(a) Reduction of the anterior column. After the
iliac lesion has been explored and after the
displacement of the posterior column has been
judged, the latter must be left for the time
being.
The primary objective of reduction is the
perfect restoration of the anterior column, ap-
plied first to the iliac wing. Experience has
demonstrated repeatedly that an approximate
and therefore inadequate reduction is almost
always due to an error of rotation of the iliac
wing. It is necessary to attempt to restore the
normal concavity of the iliac fossa which is
always much greater than one realises. When
the iliac fossa is properly reduced, the anterior
superior iliac spine is practically perpendicular-
ly above the nutrient foramen of the posterior
part of the iliac wing. Unfortunately, when left
improperly aligned so that the concavity is not
restored, it is much easier to fit a plate along
the posterior part of the pelvic brim and if this
situation is accepted, the appearance on the
standard antero-posterior radiograph will be
similar to that seen on the iliac-oblique view. To
re-establish the concavity, firm hold on the
anterior column is required, using Farabeuf
forceps astride the interspinous notch or the
crest, or both.
To verify the reduction of the fracture lines
through the iliac wing it is advisable to elevate
the soft parts on each side of the part of the
fracture transgressing the iliac crest. This frac-
ture line must be perfectly closed, and if this is
combined with perfect reduction of the fracture
492 Operative Treatment of Specific Types of Fracture

A
Fig. 22.69 A, B. Osteosynthesis of a both-column fracture.
A good reduction was obtained through the posterior approach.
A Antero-posterior radiograph, B iliac-oblique radiograph after
operation. A fracture component extends to the interspinous
notch. The posterior column took with it a segment of the pelvic
brim. One screw fixes the posterior component to the iliac wing; 8
a plate spans the two columns. See Fig. 14.29 for pre-operative
state

Bad Good

Fig. 22.70A,B. Through the ilio-inguinal approach reduction

has to be checked not only at the level of the iliac wing but also
at the iliac crest level. This is made possible by elevating soft
tissue for about 1 or 2 mm along each side of the fracture line
transgressing the crest
 Both-Column Fractures 493

A 8

Fig. 22.71A-C. Osteosynthesis of a both-column fracture with

a component extending to the crest (ilio-inguinal approach).
A Antero-posterior radiograph, B iliac-oblique radiograph,
both taken 5 years after operation. See Fig. 14.12 for pre-opera-
tive state. One plate fixes the iliac wing component, a Y-shaped
plate spans another, more posterior iliac wing component, and
two screws fix the posterior column, CAntero-posterior
radiograph taken 22 years after operation. Osteoarthritis has set
in, making total hip replacement necessary. Is the acetabular
fracture responsible for this late onset osteoarthritis? C
494
line through the inner table of the wing, the ex-
ternal aspect of the wing is consequently well
reduced (Fig. 22.70).
The fracture lines are reduced by direct
pressure using the ball spike instrument, per-
haps after disimpaction by means of a chisel
used like a tyre lever. The action of reduction of
the iliac wing is only possible after a central
dislocation of the head has been reduced. For
this reason, the head must be maintained re-
duced, approximately at its normal position,
during attempts at reduction and fixation of
the anterior column. To achieve this, it is conve-
nient to exert traction on the head from outside
by means of a femoral head extractor inserted
into the trochanter along the axis of the neck
through an external counter-incision (see
Sect. 20.5).
The possible difficulties of reduction may be
manifold:
(i) Triangular iliac wing fragment. Fairly of-
ten, the fracture line of the iliac wing
bifurcates and isolates a triangular frag-
ment of variable size. The fragment must
be fixed from the start, perhaps first to the
posterior part of the iliac wing, or some-
times to the anterior part, taking great care
to recreate the normal concavity thereof.
This fixation is achieved either by inserting
a 3.5- or 4.5-mm lag screw from the con-
vexity of the crest between the two tables
of the iliac wing, or by using a short 4-hole
plate contoured to apply to the convexity
of the iliac crest (Fig. 22.67 D). Subse-
quently the remaining fracture can be
reduced as if it were the single fracture line
concerned (Fig. 22.71).
(ii) Posterior fragment of the pelvic brim. In a
few cases the posterior part of the pelvic
brim, between the fracture line which de-
taches the anterior column and the sacro-
iliac joint, becomes detached as a separate
fragment (Fig. 22.71 C). The reduction of
the anterior column may entail rotation
about the postero-inferior angle of the il-
iac wing fracture line at the level of the in-
ternal iliac fossa; it may prove impossible
due to the instability introduced by the ac-
cessory fragment of the pelvic brim con-
sidered here. Once recognised, this frag-
ment should be reduced and fixed with
screws. Thereafter, further repositioning
proceeds as before. If the accessory frag-
Operative Treatment of Specific Types of Fracture
ment is located on the sloping superior
surface of the pelvic brim, the same pro-
cedure applies.
(iii) Fracture line reaches the anterior inter-
spinous notch. A more or less horizontal
split may divide the upper part of the ante-
rior column, reaching the anterior interspi-
nous notch (see Sect. 14.1 e). It is in general
displaced slightly. In all cases, it has been
easy to reduce and to maintain with a
screwed plate applied along the anterior
border of the iliac wing.
(iv) Dislocation of the sacro-iliac joint. This
may be complete or comprise a simple an-
terior gaping due to rupture of the anterior
ligament. It has always been reduced, but
on a few occasions it appeared necessary
to place a temporary screw into the sacrum
and another into the iliac fossa, to which
a clamp was applied, to attain perfect re-
duction of the joint. One of the plates ap-
plied to the iliac fossa was chosen so as to
span the sacro-iliac joint and maintain it
reduced by one screw inserted into the
sacrum.
(b) Osteosynthesis of the anterior column. On a
few occasions Venable or 6.5-mm screws have
been used, inserted in a roughly horizontal
direction between the two tables of the iliac
wing, penetrating the iliac wing through the
anterior inferior iliac spine or nearby, or from
the crest, thereafter crossing the fracture lines
to gain purchase in the posterior part of the il-
iac wing. They give a solid fixation but are not
so easy to place. Sometimes, they spoil an ini-
tially anatomical reduction by telescoping or
deforming the fragments at the moment of
final impaction. Most of the time we much
prefer to use plates and screws.
Very frequently an isolated screw fixes the
medial angle of the main iliac fragment in the
iliac fossa (Fig. 22.72). Other screws will fix
free fragments of the pelvic brim or the internal
fossa when they are essential fulcra for the fur-
ther reduction of the anterior column. Some
isolated screws (lagged or not) may be used to
fix some secondary fracture lines or splits
through the anterior column.
As to the plates, many different techniques
have been used. Some examples will be de-
scribed here.
We mould on to the convexity of the crest a
short four-hole plate (Fig.22.71) with two
 Both-Column Fractures 495

A
B

Fig. 22.72A-C. Osteosynthesis of a both-column fracture with

an iliac component extending to the iliac crest. A Antero-
posterior radiograph, B obturator-oblique radiograph, C iliac-
oblique radiograph, all taken 11 years after operation. See
Fig. 14.26 for pre-operative state. One plate fixes the iliac crest
fracture, another directed towards the anterior inferior iliac
spine re-establishes the concavity of the fossa, two screws cross
the split in the anterior column, and a long Lambotte screw im-
mobilises the posterior column. The hip has normal function c
496 Operative Treatment of Specific Types of Fracture

screws for each component of the fracture. This

plate, curved on the flat, must be perfectly
aligned and it assures the reduction of the con-
vexity. It cannot alone control the reduction of
the lower part of the iliac wing. Furthermore,
by loosening the screws subsequently, it is
possible to rectify mal alignment of the lower
part of the column fragment, for in certain
cases, it will not be possible to reduce perfectly
the upper and lower parts of the column simul-
taneously. The plate loosely applied to the crest
permits correction in two steps, firstly ensuring
that the concavity of the iliac fossa can be
restored, and secondly preventing upward dis-
placement of the anterior column.
In order to fix the other fracture lines in the
internal iliac fossa further plates are applied.
A long curved acetabular plate is contoured to
be applied along the superior aspect of the
pelvic brim so as to extend from the posterior
part of the internal iliac fossa as far anteriorly
as the body of the pubis. It bridges the iliac
wing fracture line and the one through the
anterior column, which traverses the anterior
wall of the acetabulum or may be extra-acetab-
ular (Fig. 22.77). This plate could, if necessary,
span the sacro-iliac joint.
In other cases, it may be necessary to use two
plates, one extending from the posterior part ofthe
iliac fossa, near the sacro-iliac joint, towards the
solid bone near the anterior inferior iliac spine, Fig. 22.73A
while the other is placed on the pelvic brim and
need not be long (Figs. 22.74 and 22.75).
In a few cases we have fixed a short plate in
the posterior part of the internal iliac fossa, in to leave screws which are intra-articular, and
order to bridge the lower part of the iliac wing not to have used screws which are so long as to
fracture line; another plate is applied along the impede later reduction of the posterior column
root of the pubis and the superior pubic ramus (one case).
to stabilise the lower part of the anterior col-
umn (Fig. 22.73). Note to the second edition: Since the develop-
Finally, we have used a mixture of one or ment of curved acetabular plates specially
two plates for the upper part of the anterior designed to be applied along the pelvic brim
column and screws to maintain the lower part through this approach, the technique has be-
(Figs. 22.71 and 22.72). When the iliac wing come a little more standardised. Today, we rou-
fracture line reaches the anterior border of the tinely use the following to fix anterior column
iliac wing (Figs. 22.77 and 22.78), the upper fractures (Fig. 22.86):
part of the column is fixed with 3.5- or 4.5-mm a lag screw into the medial angle of the frag-
screws inserted between the two tables of the ment in the internal iliac fossa;
wing. A curved acetabular plate is as usual ap- a short plate along the convexity of the
plied along the pelvic brim. Long screws from crest, or along its inner aspect (4-6 holes);
the plate or nearby fix the posterior column. a long acetabular plate along the pelvic
Osteosynthesis of the anterior column in brim, from the front of the sacro-iliac joint
these circumstances must be checked so as not up to the body of the pubis.
 Both-Column Fractures
B C
Fig. 22.73A- C Osteosynthesis of a both-column fracture with
an iliac wing component extending to the iliac crest (ilio-in-
guinal approach). A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 20
years after operation. See Fig. 14.25 for pre-operative state. One
Some additional short plates may be necessary
to fix the iliac wing fracture line and restore the
concavity of the wing (see Fig. 22.72).
Sometimes, however, when conditions are
favourable we use long 3.5-, 4.5- or 6.5-mm
screws inserted as described between the two
tables of the wing (Figs. 22.87 and 22.88). We
have gone so far in this way as to fix a both-col-
umn fracture, each column being detached in
one piece, with screws alone but we still feel
uncertain as to whether this technique is reg-
ularly reliable.
(c) Fixation of the posterior column. Once the
anterior column has been reduced and fixed, it
may be surprising to discover with the finger or
an instrument that the posterior column can
easily be drawn outwards to its proper position,
497
plate spans the fracture reaching the iliac crest, another extends
towards the anterior inferior iliac spine, a third spans the split
in the anterior column, and two Lambotte screws fix the
posterior column. Twenty years after operation the hip is rated
6.6.6
and that the profile of the greater sciatic notch
is restored anatomically, or so close to that as
to make a posterior approach unnecessary.
The posterior column reduction can be
maintained with the finger, an elevator, the
asymmetrical forceps, a Matta clamp, or using
a Lambotte hook; it is then fixed by long
screws, some of which may pass through holes
of the pelvic brim plate or may be inserted
elsewhere independently in order to gain fixa-
tion in the quadrilateral surface of the
posterior column or in the retro-acetabular sur-
face after having passed parallel to the
quadrilateral surface. By inserting one screw
medial to the ilio-pectineal eminence, close to
the pelvic brim, towards the ischial spine, and
another from the hindmost part of the internal
 498 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.74A-C Osteosynthesis of a both-column fracture with

an iliac wing component extending to the iliac crest (ilio-in-
guinal approach). A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 20
years after operation. See Fig. 14.14 for pre-operative state. One
plate is applied to the crest, a second in the iliac fossa extends
towards the anterior inferior iliac spine, a third bridges the frac-
ture of the anterior column, and some screws have been inserted
from the iliac fossa to fix the posterior column. A short plate
applied along the crest has been removed. Twenty years after
operation some osteoarthritic changes have set in, but the hip is
c clinically asymptomatic
 Both-Column Fractures 499

A B

Fig. 22.7SA-C. Osteosynthesis of a both-column fracture with

an iliac component extending to the iliac crest. Reduction
through the ilio-inguinal approach was simple but difficult to
maintain. A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph, all taken 19 years after
operation. See Fig. 14.17 for pre-operative state. The method of
fixation is similar to that in Fig. 22.69. A plate applied along the
crest has been removed. Nineteen years after operation, the hip
C is clinically perfect
 500
Fig. 22.76A,B. Osteosynthesis of a both-column fracture with
an iliac wing component extending to the interspinous notch
and with a split in the anterior column. A Antero-posterior
radiograph, B iliac-oblique radiograph, both taken two years
after operation. See Fig. 14.20 for pre-operative state. One plate
was used for the pelvic brim and two Lambotte screws for the
posterior column
iliac fossa converging with the preceding one,
an excellent fixation of the posterior column
can be obtained (Fig. 22.76). Certainly, this is
not as strong as the osteosynthesis of the
anterior column described above, but it has
always been sufficient to resist the stress of im-
mediate post-operative movement and con-
tinuous passive motion.
In some cases we tried to insert a large screw
(4.5- or 6.5-mm) from the internal iliac fossa
along the long axis of the posterior column. As
we have already mentioned, this is a very dif-
ficult screw to insert. I do think that correct in-
sertion may be prevented by the shape of the
iliac wing, which prevents one from slanting
the drill bit enough. Furthermore, the internal
aspect of the posterior column, i.e. the quadri-
lateral surface, is not visible through the lateral
window, and ideally the screw should be paral-
Operative Treatment of Specific Types of Fracture
B
leI to it. The axis of the column itself is difficult
to appreciate from above, and nobody else can
help the insertion by looking at a right angle
with respect to the surgeon as for the anterior
column long screw. Figure 22.67E shows the
area of entry for the drill bit.
Difficulties in reduction of the posterior col-
umn can be associated with the screws in the
anterior column being too long, protruding, or
even being inserted into the displaced posterior
column and preventing its reduction. Some-
times, additional fracture lines of the posterior
column, in particular those which isolate the
upper part of the anterior border of the greater
sciatic notch, render manipulation with the
finger-tip or an instrument difficult, and there-
fore it is impossible to restore the profile of the
greater sciatic notch. Isolated osseous frag-
ments, with or without associated sectors of
 Both-Column Fractures 501

Fig. 22.77. Osteosynthesis of a both-column fracture with an il-

iac wing component extending to the anterior superior iliac
spine. The ilio-inguinal approach was chosen because of a frac-
ture of the anterior column at the level of the anterior wall. The
posterior column took with it a short segment of the pelvic
brim. Antero-posterior radiograph 4 years after operation. See
Fig. 14.18 for pre-operative state. Three Philips screws held the
fracture during lateral traction on the femur; fixation was com-

..
pleted by the long plate on the pelvic brim

articular cartilage, which become interposed

between the two columns and sometimes im-
pacted into the spongiosa of the posterior col-
umn or the posterior part of the roof, can also
vitiate a perfect reduction and necessitate a sec-
ond approach; in one instance in which the
reduction of the anterior column was impossi-
ble from the front, we restored the anterior col-
umn but not in its right place, and through a
subsequent posterior approach we repositioned
the posterior column against the head. Overall,
we attained secondary congruence surgically,
but we did not restore the anatomy of both the
innominate bone and the acetabulum, which is
our constant goal (Fig. 22.79).

A B

Fig. 22.78A,B. Osteosynthesis of a both-column fracture with posterior radiograph before operation, B antero-posterior radio-
an iliac component extending to the anterior superior iliac spine. graph after operation. Two long screws inserted through the
The high part of the anterior column fracture could be fixed plate on the pelvic brim reach the posterior column
with two screws through the ilio-inguinal approach. A Antero-
 502
22.10.4 Reduction Necessitating
Both Approaches
In 15 cases we have had to employ both
approaches, immediately during the same operative
procedure (8 cases) or as two staged operations (7
cases). Of these, the Kocher-Langenbeck approach
was performed first, followed by an ilio-femoral
operation (2 cases) or an ilio-inguinal procedure (5
cases). In eight cases, the ilio-inguinal approach
was performed first followed by the Kocher-
Langenbeck operation. The indications for these
double approaches were as follows:
(a) Failure to achieve an adequate reduction of the
opposite column. If, through the approach
chosen initially, it soon becomes apparent that
reduction of the opposite column is difficult,
rather than prolong the operation fruitlessly
with consequent risks of infection or damage
to soft tissue, it is better to close the incision,
turn the patient over, and expose the opposite
column. The cause of the difficulty in reduc-
tion may be quickly found. Further, the second
approach is in general rapid, for in effect, it en-
tails only the reduction of an isolated fracture
Fig. 22.79A,B. Osteosynthesis of a both·column fracture with
an iliac component extending to the crest, typical but com-
minuted. A Antero-posterior radiograph before operation,
B antero-posterior radiograph after operation. It was impossible
to mobilise the posterior column through an ilio-inguinal inci-
sion. Two weeks later, the cause was found through a Kocher-
Operative Treatment of Specific Types of Fracture
of a column which will be easy and anatomical,
assuming that the other column has been re-
duced perfectly during the first stage. If the
state of the patient does not permit a second
approach at once, it must be delayed 8 -10
days.
(b) Fracture-dislocation of the sacro-iliac joint.
Fractures of both columns together with a frac-
ture-dislocation of the sacro-iliac joint have
necessitated two approaches. Up to 1975, when
the lateral approach was described, the posteri-
or approach was the one of choice with which
to begin (Figs. 22.80 and 22.81). The fragment
comprising the lower part of the articular sur-
face of the sacro-iliac joint carries also the pos-
terior inferior iliac spine, the superior border of
the greater sciatic notch and the sciatic but-
tress. It is reduced by proper alignment with the
remaining part of the iliac wing and held by
means of a short Venable screw or by a plate.
Then the reduction of the posterior column is
achieved by establishing a proper relationship
between this and the posterior segment of the
restored iliac wing. In this extensive posterior
repair, the Kocher-Langenbeck incision has to
Langenbeck approach: fragments of cortical bone from the
cotyloid fossa and the postero-superior articular surface were
wedged between the columns. Finally surgical secondary con-
gruence was achieved: really a malunion of the pelvic bone, but
forming a congruent joint around the centrally displaced head
Both-Column Fractures
be very extended medially with consequent risk
to the gluteal muscles. Perhaps two posterior
approaches would be preferable, one along the
posterior part of the iliac crest to restore the
anatomy of the sacro-iliac joint and the other
a short Kocher-Langenbeck incision through
which the acetabulum could be repaired; we
have used two such incisions in one case.
Our first four cases of both-column fracture
with involvement of the sacro-iliac joint were
treated as follows. One was through the anteri-
or approach first, and it was of course impos-
sible to cope with the sacro-iliac fragment; a
second approach was necessary but an unsatis-
factory reduction was achieved which led to a
post-traumatic osteoarthrosis. The three others
have been treated by posterior approach first;
the reduction of the fragment was obtained
without too much difficulty, the posterior
column reduced, and then an ilio-inguinal
approach was necessary in order to reduce the
anterior column (Figs. 22.80, 22.81 and 22.82).
Nowadays, a sacro-iliac fracture dislocation as-
sociated with a both-column fracture should
unquestionably be operated upon through the
extended ilio-femoral approach.
22.10.5 Reduction and Fixation
Through Extended IIio-femoral Approach
This approach, enlarged by elevation of the internal
iliac fossa, provides access to the whole posterior
column, the whole iliac wing, and the anterior col-
umn up to the ilio-pectineal eminence but no fur-
ther. This means that fractures involving the ilio-
pectineal eminence will be difficult to reach and fix
(particularly as they are often comminuted), and
those involving the superior pubic ramus, the body
of the pubis and the ischio-pubic ramus are im-
possible to reach. The approach could perhaps be
expanded by continuing the segment of the extend-
ed ilio-femoral incision following the iliac crest
from the anterior superior iliac spine towards or up
to the mid-line, as in the ilio-inguinal approach.
Deeply, we could form the three windows, or at
least the lateral and middle windows of the ilio-in-
guinal incision and through this extension of the in-
cision gain complete access to the anterior column.
However, we have never yet felt it necessary to carry
out such an extended approach.
With good control of most of the fracture lines,
some of them on both aspects of the bone if the in-
503
ternal iliac fossa has been elevated, the reduction
and fixation are performed.
There is no strict plan of work for this, and
many possibilities exist. However, whatever the
chosen schedule for reduction, one must realise
that it is nearly impossible to reduce all the fracture
lines at the same time even with many clamps(!)
because only exceptionally are the columns de-
tached in one piece; most of the time they are divid-
ed by secondary fracture lines. Consequently, these
fracture lines have to be reduced one after the
other, and it must be kept in mind that the reduc-
tion of each one has to be absolutely perfect if the
final reduction is to be excellent. It is impossible to
compensate for an initial error later when perform-
ing further reductions. A bad or a mediocre reduc-
tion of the first fracture line to be reduced is the
guarantee of a poor final reduction.
Prior to their reduction, all the fracture lines are
cleaned of callus, if any, but the maximum of care
is taken to keep some soft tissue pedicles to the
various fragments. Muscles, tendon, capsule: all
these elements bring some blood vessels to the
bone.
I generally begin by reducing the fracture lines
through the iliac wing, as the restoration of the
wing repositions the roof of the acetabulum in its
normal position. These fixations are achieved by
3.5-mm screws and plates.
When there is a triangular fragment of the wing,
it can be spanned completely by a plate applied
along the crest if it is small. A large fragment is
fixed as through the ilio-inguinal approach (see
Fig. 22.67 D). According to the case being treated,
we reduce first either the posterior branch of the Y
fracture line or the anterior one, thus joining the
triangular fragment solidly to the intact part of the
iliac wing or to the still mobile anterior column.
This fixation is achieved by means of either a lag
screw entering at the iliac crest and penetrating be-
tween the two tables of the iliac wing, or by a 4-hole
plate applied on the convexity of the iliac crest.
When this has been done, all that remains is to fix
one fracture line through the iliac wing.
In many instances there is only one fracture
transgressing the wing and the crest. The reduction
of such fracture lines is often achieved by tem-
porarily inserting one or two screws on each side of
the fracture line with their heads left proud. One or
two modified Farabeuf forceps applied to them
allow one to obtain and maintain the reduction;
this trick may be used for reducing all the fracture
lines through the wing (Fig. 22.84). The reduction is
 504 Operative Treatment of Specific Types of Fracture

c
Fig. 22.80A-D. Osteosynthesis of a both-column fracture with
an iliac component extending to the crest and with involvement
of the sacro-iliac joint (consecutive posterior and then anterior
ilio-femoral approaches). A Antero-posterior radiograph after
first stage of operation through Kocher-Langenbeck approach,
B antero-posterior, C obturator-oblique and D iliac-oblique ra-
diographs 21 years after completed operation. See Fig. 14.23 for
pre-operative state. The posterior approach was chosen first ilio-femoral approach, but it was adequate for the completion of
because of the sacro-i1iac joint involvement. The most inferior the reduction of the anterior part of the fracture complex. Twen-
fractures of the anterior column were not accessible through the ty-one years after operation the hip is clinically perfect
Both-Column Fractures
also checked at the level of the iliac crest (see
Fig. 22.70), especially if the internal iliac fossa was
not elevated, in order to be sure that the endo-
pelvic reduction is also satisfactory.
Fixation of this single or the remaining fracture
(if a part of the Y has already been fixed) is per-
formed as follows. A 4-hole plate, bent on the flat
according to its future position on the iliac crest
and contoured to apply perfectly to the convexity
of the iliac crest, is fixed with four 3.5-mm screws
25 - 45 mm in length. The fixation is completed by
contouring one or more plates of 4-6 holes which
must apply perfectly to the external aspect of the
wing, across the fracture line. One of them may
span the inferior angle of a triangular fragment.
Then the upper part of the anterior column is
restored and a large segment of the roof is reposi-
tioned, against which the femoral head may rest
(Fig. 22.89 A). It is useful to remember that to allow
this reduction, the femoral head was extracted from
its centrally dislocated position and maintained ap-
proximately in its anatomical position by combined
longitudinal traction and raising of the pelvic post
disposed beneath the thigh.
It now remains to reduce and fix the posterior
column. Once again, we often use temporary screws
giving purchase to a Farabeuf forceps to maintain
the reduction of the posterior column, which is
otherwise mobilised by usual means (manual or
clamps). Once perfectly reduced (it must not be
forgotten to check the reduction at the level of the
quadrilateral surface through the greater sciatic
notch), the posterior column is fixed as usual.
Quite often one or two lag screws (3.5-mm) fix
the upper part of the column, allowing removal of
the clamp, and a long plate (8-10 holes) is applied
along the posterior column extending from the up-
per pole of the ischial tuberosity up to above the
roof. Screws into the posterior column are inserted
with the precautions described earlier (Sect. 22.1).
To complete the fixation, in some cases there
may be one or two problems to solve. A posterior
or postero-superior fragment may be present, usu-
ally having kept its capsular insertions. It is reduced
and fixed with screws and if possible bridged by the
plate.
There may exist a secondary fracture line divid-
ing the anterior column at the level of the anterior
wall through the articular surface: it has to be
perfectly reduced, verified if necessary by elevating
a segment of the internal iliac fossa. When reduc-
tion is achieved and maintained by a clamp, a long,
large screw (4.5- or 6.5-mm) is inserted along the
505
axis of the anterior column as described in
Sect. 22.5.2 and Figs. 22.34 and 22.89B.
There are many other ways to fix a both-column
fracture through the extended ilio-femoral ap-
proach (Figs. 22.90 and 22.91):
(a) Large screws (4.5- or 6.5-mm) may be used to
fix the anterior part of the wing. Some flat iliac
wings transgressed by a fracture line roughly
perpendicular to their tables are good subjects
for the technique. One screw is inserted from
the apex of the anterior inferior iliac spine and
proceeds into the sciatic buttress 1 - 2 cm above
the superior border of the greater sciatic notch.
It may be lagged and is 110 - 130 mm long.
The other screw progresses from the iliac
crest horizontally or slightly convergent with
the previous one, between the two tables of the
wing. It is generally a little shorter
(Fig. 22.89 D, E).
(b) Another large screw, which may also be lagged,
may be used to fix the posterior column. In-
serted from 3 cm above the acetabulum, a little
behind the area of entry of the anterior column
axial screw, it is directed into the posterior col-
umn. Exceptionally following the main axis of
the column, but generally oblique with respect
to it, it gives a good fixation. I used this tech-
nique although I prefer plating the posterior
column.
(c) The plate fixing the posterior column may be
contoured differently, as Jeffrey MAST has
done in some instances (Fig. 22.89C).
Where there is an associated dislocation or
fracture-dislocation of the sacro-iliac joint, this
is reduced at the beginning of the reconstruc-
tion. A dislocation is fixed with two transar-
ticular ilio-sacral screws (4.5- or 6.5-mm). An
associated fracture-dislocation is fixed with two
or three large screws (4.5- or 6.5-mm) inserted
from the posterior part of the iliac crest be-
tween the two tables of the iliac wing, crossing
the fracture line on their way (Fig. 22.91) and
not penetrating the sacro-iliac joint.
22.10.6 A Particular Both-Column Fracture
In one case of a both-column fracture with an iliac
wing fracture line extending to the crest and treated
by the ilio-inguinal approach, the reduction of the
posterior column was extraordinarily difficult. The
cause was incarceration within the pelvis of the
posterior column. The posterior column was frac-
 506 Operative Treatment of Specific Types of Fracture

B D
Special Examples 507

tured in its upper part, a little below the angle of
the greater sciatic notch. It was totally inside the
pelvis and had elevated so that its upper fracture
surface was in contact with the sacro-iliac joint.
Also, it had rotated some 60 0 so that the retro-
acetabular surface rested against the remaining
anterior part of the quadrilateral surface. Once
recognised radiologically, this situation should be
approached through the Kocher-Langenbeck inci-
sion (Fig. 22.83) as a first approach, or through the
extended ilio-femoral incision.
22.11 Special Examples
ments or debris; multiple fragments are common,
particularly in the cotyloid fossa and the anterior
part of the joint. Having ensured that the joint is
clear, the traction is relaxed. A large loose fragment
should be fixed in the usual manner; similarly, a
pedunculated fragment can be reduced and held by
a single screw. If a loose fragment is small and im-
possible to fix, it is preferable to remove it.
Diagnosed and treated early, the incarceration
of fragments does not impair the prognosis of frac-
tures of the acetabulum. It is the neglected in-
carcerations which can produce such bad results
(Figs. 22.92, 22.93 and 22.94).

22.11.1 Incarcerated Intra-articular Fragments 22.11.2 Bilateral Acetabular Fractures

The intra-articular entrapment of fragments recog- We have treated a number of bilateral fractures of
nised after reduction of a dislocation constitutes a the acetabulum. Figure 22.95 shows a case compris-
strong indication for open operation. It is not ing a symmetrical bilateral transverse fracture asso-
urgent provided the posterior dislocation is reduced ciated with a posterior wall fracture and a posterior
and stable. When reduction of a dislocation asso- dislocation. Both sides were operated upon at the
ciated with entrapment of fragments cannot be same time and by two teams. The reduction was
achieved the situation is more urgent, for the good on both sides but one femoral head under-
removal of the incarcerated fragments offers the went some necrosis.
only chance of reducing the femoral head in the
cavity and minimising the risk of avascular necrosis
of the head.
It is very easy to extract a pedunculated frag- 22.11.3 Fractures of Paralysed Hips
ment, after gentle traction on the lower limb on the
orthopaedic table and following exposure through Two cases of fracture of the acetabulum on hips
the posterior approach. It is necessary to check that afflicted with poliomyelitis have been operated
the acetabulum is clear of other incarcerated frag- upon (see Fig. 16.1). They offered an easy approach
because of the atrophied muscles, but there were
..
Fig. 22.81 A-D. Osteosynthesis of a both-column fracture with
difficulties in reduction on account of the poor
quality of the bone and deformity of the pelvis.
an iliac wing component extending to the crest with involvement The fractures united without problems.
of the sacro-iliac joint (consecutive posterior and ilio-inguinal
approaches). A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 2 years
after operation. See Fig. 14.24 for pre-operative state. It was very
difficult to reduce the sacro-iliac fragment through the first
approach: it was fixed with a spongiosa screw inserted into the
crest and a plate on the external iliac surface astride the fracture
line. This complicated but effective fixation was completed
through the second approach with a plate adjacent to but not
on the iliac crest (because it was still cartilaginous in this young
patient), a screw in the angle of the anterior column fragment
at the level of the internal iliac fossa, a plate extending from the
sacro-iliac region to the anterior inferior iliac spine, and finally
a plate across the inferior fracture of the anterior column. The
screws from the latter plate reach the posterior column and gain
a hold on the quadrilateral plate. D Antero-posterior radiograph
taken 20 years after operation; osteoarthritis (type IV) has set in.
The hip is rated functionally 5.5.5
 508 Operative Treatment of Specific Types of Fracture

c D

Fig. 22.S2A- D. Osteosynthesis of a both-column fracture with C obturator-oblique radiograph, and D iliac-oblique radiograph
sacro-iliac involvement (two approaches). A Antero-posterior 16 years after operation . The hip is clinically perfect
radiograph before operation. B Antero-posterior radiograph,
 Special Examples 509

B
Fig. 22.83A,B. Osteosynthesis of a both-column fracture with
an iliac component extending to the iliac crest (ilio-inguinal
approach). A Antero-posterior radiograph, B iliac-oblique ra-
A diograph, both taken 10 years after operation. See Fig. 14.27 for
pre-operative state. Two plates restore the anterior column and
two Venable screws are implanted into its anterior border.
Another plate bridges the split in the lower anterior column.
Long screws reach and fix the posterior column

Fig.22.84. A Farabeuf forceps applied to the head of screws

which have been left proud allows the reduction to be performed
(here it is nearly complete) and maintained while it is fixed
510 Operative Treatment of Specific Types of Fracture

Fig. 22.8SA-C. Osteosynthesis of a both-column fracture (ilio-

inguinal approach). A Antero-posterior radiograph, B obtura-
tor-oblique radiograph, and C iliac-oblique radiograph 1 year
after operation. The four plates were positioned at the most
favourable sites. See Fig. 14.31 for pre-operative state
 Special Examples 511

A B

Fig. 22.86A-C Osteosynthesis of a both-column fracture (ilio-

inguinal approach). Fixation was achieved using three stainless
steel plates. A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac-oblique radiograph, all taken 1 year
after operation. See Fig. 14.24 for pre-operative state. The hip is
C
normal
 512 Operative Treatment of Specific Types of Fracture

A B

Fig. 22.87 A - C. Osteosynthesis of a both-column fracture (ilio-

inguinal approach). A Antero-posterior radiograph, B obtura-
tor-oblique radiograph, and C iliac-oblique radiograph, all
taken 3 years after operation. See Fig. 14.15 for pre-operative
c state. The hip is normal
 Special Examples 513

B c

Fig. 22.88A-C. Osteosynthesis of a left both-column fracture graph, B obturator-oblique radiograph, and C iliac-oblique ra-
associated with a right anterior vertical fracture through the ob- diograph all taken 1 year after operation. See Fig. 14.33 for pre-
turator foramen (ilio-inguinal approach). Fixation was achieved operative state. The hip is rated 6.6.6
using stainless steel plates and screws. A Antero-posterior radio-
 514
Operative Treatment of Specific Types of Fracture

A B

c o E

Fig. 22.89 A-E. A Complete reconstruction and fixation of a serted along the long axis of the column. C Another way to app-
both-column fracture performed on a dry bone through the ex- ly the plate fixing the posterior column. D, E Insertion of a long
tended ilio-femoral approach. B If there is a secondary fracture screw (4.5 or 6.5 mm) from the antero-inferior iliac spine up to
line through the anterior column it is fixed with a long screw in- the sciatic buttress between the two tables of the wing
 515
Special Examples

Fig. 22.90A-C. Osteosynthesis of a both-column fracture (ex-

tended ilio-femoral approach). Fixation was achieved with stain-
less steel devices. A Antero-posterior radiograph, B obturator-
oblique radiograph, C iliac-oblique radiograph, all taken 11
days after operation: these recent views show the fixation of the
different fracture lines better. See Fig. 14.32 for pre-operative
C
state
 516 Operative Treatment of Specific Types of Fracture

A B

c D
Fig. 22.91 A - D
 Special Examples 517

E F
Fig. 22.91 A-F. Osteosynthesis of an atypical both-column frac-
ture associated with a slightly displaced fracture dislocation of
the sacro-iliac joint (extended i1io-femoral approach). A Antero-
posterior radiograph, B obturator-oblique radiograph, and C il-
iac-oblique radiograph before operation, and D antero-posterior
radiograph, E obturator-oblique radiograph, and F iliac-oblique
radiograph after operation. We first fixed the sacro-iliac joint
with two screws inserted between the two tables of the wing and
then the both-column fracture with lag screws and a stainless
steel plate

Fig. 22.92. Primary incarceration of a posterior wall fragment.

Antero-posterior radiograph taken 23 years after simple remov-
al. See Fig. 18.3 for pre-operative state. The hip is rated 6.6.6
518 Operative Treatment of Specific Types of Fracture

Fig. 22.93 A - C. OsteosynthesIs of an associated T-shaped and

posterior wall fracture with posterior dislocation and secondary
incarceration of a posterior wall fragment (Kocher-Langenbeck
approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac oblique radiograph, all taken 19 years
after operation. See Fig. 18.4 for pre-operative state. The in-
B
carcerated fragment was fixed with a single screw
 Special Examples 519

A B

Fig. 22.94A-C Osteosynthesis of an associated transverse and

posterior wall fracture with immediate incarceration and partial
fracture of the femoral head (Kocher-Langenbeck approach).
A Antero-posterior radiograph, 8 obturator-oblique radio-
graph, and C iliac-oblique radiograph, all taken 3 years after
operation. See Fig. 18.6 for pre-operative state. The hip is rated C
6.6.5 because of sequelae of a pre-operative sciatic palsy
520 Operative Treatment of Specific Types of Fracture

Fig. 22.95A,B. Bilateral acetabular fractures (simultaneous operations by two teams). A Antero-
posterior radiograph before operation, B antero-posterior radiograph after operation
23 Anatomical Results of Operation
Within Three Weeks After Injury
The anatomical results of operation have been
assessed in all the cases from the immediate post-
operative views and in some from CT scans per-
formed during the two weeks following the opera-
tion. Since the first edition of this book in 1981, we
have completely changed the way we assess imper-
fect reductions, but perfect reductions are judged
by the same criteria
Perfect reductions are those in which we were able
to perform an anatomical reduction of both the
acetabulum and the corresponding innominate
bone, confirmed by perfect restoration of the ar-
ticular congruence and the re-establishment of all
the radiological landmarks on the three standard
radiographic views.
Imperfect reductions are those in which one or
more or all the acetabular landmarks are not ana-
tomically restored on the three views, indicating im-
proper reduction of one or more elements of the ar-
chitecture of the acetabulum, the consequence of
which is a more or less extensive post-surgical ar-
ticular incongruence, well demonstrated or visible
at the sub-chondral bone level. The problem which
then arises is to evaluate, as precisely as possible,
the amount of post-surgical incongruence. We hope
that 3-D CT scanning will soon help us to do this,
but so far - and this goes for all the cases in our
series - we have no precise method; we can say
easily that a reduction is imperfect, but the degree
of imperfection is very difficult to judge.
For the last two reviews of our cases (1978 -1980
and 1987 - 1990) we finally decided to assess im-
perfect reduction by evaluating the position of the
femoral head with respect to the undisturbed or
restored anatomical roof of the acetabulum. Thus,
we distinguish:
cases in which the head-roof congruency is re-
stored, but with a fault in reduction of the ace-
tabulum;
cases in which there is a loss of parallelism of
the upper joint space, linked to error in reduc-
tion of the acetabulum. We make a distinction
between these cases and those in which the in-
tact side shows the same appearance, as may be
observed in radiographs of normal adult hips
which are free of clinical signs and have not
been subjected to trauma;
cases in which a more or less marked central
femoral head protrusion is obvious;
technical failures, which are the cases in which
we were unable to achieve our goal due to inade-
quate understanding of the pre-operative views,
or because we chose an inappropriate approach,
which happened too frequently at the beginning
of our practice. Poor condition of bone was also
responsible for failures when undetected pre-
operatively. Undoubtedly, inability to interpret
the radiographs led to problems, for good
reduction cannot be obtained unless it is known
in advance where the fracture lines are to be
found. It is extremely hazardous to attempt to
gain an understanding of the fracture at opera-
tion without having understood the radio-
graphs.
We have tried to increase the precision of our as-
sessment of imperfect reductions by supplementing
the previous method with measurement of the re-
sidual central displacement of the femoral head
and of the most displaced landmark in millimetres.
We are obliged to include in these imperfect
reductions some special cases: those in which we
achieved surgical secondary congruence (so called
by analogy with what can occur in some post-trau-
matic situations, discussed in Sects. 1.4 and 1.5). By
this we mean cases which intra-operatively ap-
peared impossible to reconstruct perfectly despite
many efforts. In these cases we decided to group all
the fragments of the shattered acetabular articular
surface as perfectly as possible around the more or
less centrally displaced femoral head, fix them
522 Anatomical Results of Operation Within Three Weeks After Injury
together, and finally fix the reorganized acetabu-
lum to the intact part of the iliac wing (this oc-
curred mainly in both-column fractures). In these
cases we restored the congruence between the head
and the different parts of the acetabulum but the
new reconstructed socket was not in its normal
place, the innominate bone was not anatomically
restored, and its malunion was accepted from the
outset (Fig. 23.1}. In such cases we know that the
different parts of the acetabulum are often separat-
ed by a gap, and that they also twist with respect to
each other.
Other authors have tried to assess these im-
perfect reductions in different ways. Claude MAR-
TIMBEAU evaluates the reduction of each column
and the head-roof congruence on the three views
and gives a score of 8 to a complete reconstruction
(see Appendix at the end of this chapter). Joel
MATTA measures the residual displacement at the
sub-chondral bone level: "No distinction is made
between a step-off or gap at the fracture site with
regard to displacement, and in cases where inter-
pretation is difficult at the sub-chondral bone level,
the reduction is assessed by measuring displace-
ment of the radiographic lines on the three views?'
He considers the reduction as:
anatomic if a maximum of 1 mm displacement
is seen on any of the three views;
satisfactory if a maximum of 2 - 3 mm displace-
ment is seen on any of the three views;
unsatisfactory if more than 3 mm displacement
is seen on any of the three views.
All our perfect reductions are anatomic according
to MATTA'S criteria. Our imperfect reductions cor-
respond to satisfactory or unsatisfactory results in
his classification, but not in an orderly fashion:
thus, a "satisfactory" reduction may correspond to
"head recentred" (17 of 55 cases) or to "loss of
parallelism" (13 of 34 cases).
23.1 Analysis of the Immediate
Radiological Results
Many factors may influence the quality of the
reduction:
the interval from injury to surgery;
the anatomical type of the fracture;
the choice of the most appropriate approach;
the experience of the surgeon in this field.
The immediate results of surgery are reported here,
taking in account these different factors.
Table 23.1 reports the quality of reduction per-
formed in 567 cases 1 operated on within 21 days
after injury. Overall we achieved 418 perfect reduc-
tions (73.72OJo), whereas 149 reductions were im-
perfect, in whatever degree (26.27070). Table 23.1
also relates the quality of reduction of the acetabu-
lum to the type of the fracture and the percentage
of perfect reductions achieved. Among the simple
fractures, the most difficult to reduce appear to be
the transverse; a better choice of approach, and in
particular, use of the extended ilio-femoral ap-
proach for trans-tectal transverse fractures will im-
prove this figure.
The reduction of anterior column fractures has
been greatly improved by the use of the ilio-ingui-
nal approach: the rate of perfect reductions, which
was 69.2% in the first edition of this book, is now
86.4% overall.
Posterior wall fractures, theoretically the sim-
plest type of acetabular fracture, may offer great
difficulties in cases where there is comminution,
and the overall rate of perfect reduction is only
93.7%.
Among the associated fractures, the associated
posterior column and posterior wall are the easiest
to reduce (90% perfect reductions). For all the
other groups the figures are around 70%, except the
both-column fractures, which have risen only from
55% to 60.7% perfect reductions since 1981 (cases
with secondary surgical congruence are not includ-
ed in these figures).
Concerning the associated fracture patterns, the
order of increasing difficulty is: associated posteri-
or column and posterior wall; T-shaped fractures;
associated anterior fractures and posterior
hemitransverse fractures; associated transverse and
posterior wall fractures; and both-column frac-
tures.
Table 23.2 reports the number of perfect reduc-
tions in relation to the total number of cases
operated on during the first 3 weeks after injury,
according to the type of fracture and the precise
date of surgery during these 3 weeks. Table 23.3
gives the same data expressed in percentages. These
two tables demonstrate no very significant dif-
ference in the rate of perfect reconstruction in cases
1 This figure excludes two perfect acetabular reconstructions of
a posterior wall fracture associated with a femoral head ar-
throplasty performed for additional femoral neck and head
fractures.
 Analysis of the Immediate Radiological Results 523

A B

Fig. 23.1A-C. Surgical secondary congruence achieved 18 days

after injury with an excellent long-term result at 11 years. A
Antero-posterior radiograph before operation, B antero-
posterior radiograph 4 months after operation, and C antero-
posterior radiograph 11 years after surgery. The hip is rated
6.6.6. There may be a little remodelling of the lateral part of the
upper joint space, which was slightly open post-operatively C
Table 23.1 v.
~
Type of fracture Quality of immediate radiological result Total Percentage of
perfect reductions
Perfect Head centred, Slight loss Residual Technical Total
reduction acetabulum of superior central failure secondary
imperfect congruence protrusion congruence

Posterior wall 119 5 2 127 93.711Jo

Posterior column 10 2 13 76.911Jo
Anterior wall 7 1 9 77.7I1Jo
Anterior column 19 1 1 22 86.4l1Jo
Transverse 15 2 2 2 21 71.411Jo
T-shaped 21 1 7 1 30 70.011Jo
Transverse and posterior wall 79 23 5 5 3 2 117 67.511Jo
Posterior column and posterior wall 18 1 1 20 90.011Jo
Anterior column and hemitransverse 34 5 4 5 1 1 50 68.011Jo
Both-column 96 16 14 4 4 24 158 60. 7 l1Jo
Total 418 57 34 19 12 27 567 73.711Jo
Percentage of total number of 73.72l1Jo 10.0511Jo 5.9911Jo 3.3511Jo 2.11l1Jo 4.7611Jo
operated fractures
26.2711Jo
~
8
[
w
'"
g.
'"
o
-.
o
Io·
p

~
s-
;l
~
~
<II
~
:>
;::>
~
....
2.
~
Table 23.2 >
::s

Type of Time from injury to operation ~

fracture 0
....,
1 Day 2 Days 3 Days 4 Days 5 Days 6 Days 7 Days 8-14 Days 15 - 21 Days Total :;-
(1)

Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of Perfect No. of S
reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- reduc- opera- S
(1)

tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions tions ~
~
Posterior wall 7 8 15 17 11 12 13 13 6 8 14 14 3 3 33 34 17 18 119 127 i:?0-
Posterior 2 2 5 7 1 2 10 13
column SI
0
Anterior wall 1 2 2 2 2 1 1 7 9 !!!l.
<>
Anterior 3 3 3 3 1 6 8 5 5 19 22 ~

column ~
Transverse 1 3 4 2 2 1 1 2 2 1 4 5 2 3 15 21 <=
T-shaped 2 2 1 2 1 3 1 3 2 3 1 2 4 4 7 7 2 4 21 30
a-
Transverse and 2 4 5 6 3 4 9 11 3 4 5 7 4 8 30 45 18 28 79 117
posterior wall
Posterior 2 2 3 3 3 3 6 8 3 3 18 20
column and
posterior wall
Anterior 5 6 4 5 2 3 5 8 3 3 4 12 15 2 6 34 50
column and
hemitransverse
Both-column 2 4 6 7 10 16 8 9 14 22 7 9 33 49 18 40 96 158
Total 12 17 32 43 33 39 41 53 33 42 40 56 22 29 136 178 69 110 418 567

v.
N
v.
526 Anatomical Results of Operation Within Three Weeks After Injury

Table 23.3

Type of fracture Percentage of perfect reductions achieved according to time of surgery after injury

Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Days 8-14 Days 15-21

Posterior wall 87% 88070 91 % 100% 100% 75% 100070 97% 94%
Posterior column 100% 100% 100% 71 % 50%
Anterior wall 100% 100% 50% 100% 50% 100%
Anterior column 0% 100% 100% 100% 100% 75% 100%
Transverse 100% 75% 100% 100% 50% 0% 100% 80% 66%
T-shaped 100% 50% 33% 33% 66% 50% 100% 100% 50%
Transverse and 50% 83% 75% 81 % 75% 71 % 50% 66% 64%
posterior wall
Posterior column 100% 100% 100% 100% 75% 100%
and posterior wall
Anterior column 83% 80% 66% 62% 33% 75% 80% 33%
and posterior
hemitransverse
Both-column 0% 0% 85% 62% 88% 63% 77% 67% 45%
Total 70% 72% 84% 77% 76% 71 % 75% 76% 62%

operated on between the 1st and the 14th day, but
due to the increasing operative difficulties caused
by callus formation, which is already significant
during the 3rd week after injury, the rate of perfect
reductions then drops from an average of 750/0 in
the first 2 weeks to 62%.
23.2 Analysis of Imperfect
Radiological Reductions
Although taking the position of the femoral head
with respect to the radiological roof as the criterion
by which to assess imperfect reduction may be
criticised, it has appeared to me up to the present
day the most convenient for this series. Once again,
if the use of post-operative 3-D CT scanning or any
other new technology becomes easy, things may
change and the criteria for later studies be other
ones.
It is important to know that the abnormal posi-
tion of the head with respect to the roof, which
defines the most important imperfections, may be
linked to three elements:
the most frequent: mal-reduction of a part or
the whole of the acetabulum;
wear of the femoral head against a fracture line,
which could not be dealt with up to now,
although this remains rare or generally of little
importance during the first 3 weeks. However, it
is easy to see that total wear of the articular car-
tilage over a large surface may be responsible
for a central displacement of the head of
2-3mm;
a missed incarcerated fragment, which can dis-
place the head either downwards or outwards.
Of the 149 imperfect reductions, in 57 cases the
femoral head is perfectly recentred under the ana-
tomical and radiological roof, but a part of the
acetabulum is imperfectly reduced. So far we have
found it difficult if not impossible to assess ac-
curately the extent of the anatomical roof which is
still in place or has been reconstructed. This condi-
tion represents 38.25% of imperfect reductions and
10.0% of all reductions.
Table 23.1 shows the distribution of these
"head-centred" cases according to fracture variety.
In these head-centred cases we have measured the
maximum residual linear displacement of a land-
mark. The maximum residual displacement was
2 mm in one case, 2-5 mm in 37 cases, 6-8 mm in
seven cases, and more than 9 mm in six cases
(Table 23.4).
Fig. 23.2A- D. Reduction of a both-column fracture operated
on through an ilio-inguinal approach in 1966. The reduction is
imperfect and leaves a loss of parallelism of the upper joint
space, clearly visible on the antero-posterior view (A), not ob-
vious on the obturator-oblique view (B), and doubtful on the il-
iac-oblique view (C). Five years later (see Fig. 22.71 B) remodel-
ling of the lateral part of the upper joint space has occurred,
seeming to fill the previously empty space (a double line con-
vergent with the inner part of the roof is visible). D Antero-
posterior radiograph 12 years later: osteoarthritis has set in.
Twenty-five years later a total hip replacement was necessary (see
Fig. 22.71 C)
Analysis of Imperfect Radiological Reductions 527

c
528
Table 23.4
Type of imperfect reduction Residual linear displacement
Central displacement of the head
Omm 1-2mm 3-5mm
Centred head (57 cases)
Loss of parallelism
(34 cases)
57
5 24
Residual head protrusion
(19 cases) 7 10
Technical failure (12 cases)
In 34 cases radiographs demonstrate a "loss of
parallelism of the upper joint space': representing
16070 of the imperfect cases and 6% of all cases
(Fig. 23.2). The central displacement of the head
was measured on the antero-posterior view and was
assessed as 1 - 2 mm in five cases, 3 - 5 mm in 24
cases, and greater than 5 mm in two cases
(Table 23.4). The maximum displacement of a land-
mark was 1-2mm in six cases, 3-5mm in 19
cases, 6 - 8 mm in four cases, and greater than
9 mm in two cases.
Comparing these figures to those of the head-
centred cases confirms that linear displacement of
a landmark is not directly related to the displace-
ment of the head. The latter is only related to the
amount of acetabular articular surface remaining
displaced and thus allowing the head to displace,
centrally or in any other direction. Thus, 56 cases
of maximum linear displacement of a landmark of
3 - 5 mm are made up of 37 cases with a centred
head and 19 with loss of parallelism (see
Table 23.4), and we will see in a moment that the
same amount of displacement may accompany
residual central head protrusion. This means that,
unfortunately, the displacement of a landmark
does not give precise information about the
amount of acetabular articular surface that it
brings with it.
In 19 cases there was a residual head protrusion,
the head having lost all relationship with the
radiological roof. This figure represents 12.7% of
imperfect cases and 3.35% of all cases (Fig. 23.3).
The displacement of the head was assessed at
3 - 5 mm in seven cases and more than 5 mm in ten
cases. The maximum linear displacement of a land-
mark was 3 - 5 mm in three cases, 6- 8 mm in three
others and more than 9 mm in 11 cases.
Anatomical Results of Operation Within Three Weeks After Injury
Cases not
evaluated
Maximum displacement of a landmark for
different
>5mm 1-2mm 3-5mm 6-8mm ;;'9mm reasons
2
I
6 [ill
19
7
4
6
2
6
3
3 3 II 2
6 6 5
It must be said that it is difficult to distinguish
absolutely between mere loss of parallelism and
residual head protrusion, and there are cases which
are undoubtedly open to discussion. Due to their
pure radiological appearance, we have considered
as "loss of parallelism" two cases with central
displacement of the head by more than 5 mm
(Fig. 23.2) and as "central protrusion" seven cases
in which the head was displaced by only 3 - 5 mm
(Fig. 23.3). It is difficult to explain this, and things
could probably have been evaluated differently, but
we were unable to find a mathematical or measur-
able border between those two categories. All the
same, 14 of the 19 cases with residual head protru-
sion had significant residual landmark displace-
ment, compared to six of the 34 cases of loss of
parallelism.
In 12 cases we were totally unable to achieve our
goal: no satisfactory or nearly satisfactory reduc-
tion of any part was achieved. These are technical
failures. We failed because of misunderstanding the
fracture type, making the wrong choice of
approach, because of poor quality of the bone, in-
adequate tools, or other reasons. These 12 cases
represent 8% of the imperfect cases and 2.11 % of
all cases.
In 27 cases we achieved surgical secondary con-
gruence, representing 18.1 % of the imperfect cases
and 4.8% of all cases.
23.3 The Learning Curve
As usual in traumatology, as in other fields of
surgery and indeed any other job, intensive training
and experience improve results. The cases reported
 The Learning Curve 529

Fig. 23.3 A - C. Trans-tectal transverse fracture associated with a

vertical anterior fracture of the obturator ring, operated on
through a posterior approach on the 14th day after injury.
A Obturator-oblique radiograph before operation, 8 antero-
posterior radiograph 2 months after operation, showing what
was called a persistent protrusion of the femoral head.
CAntero-posterior radiograph 8 years after operation: osteo-
arthritis has set in. A MacMurray osteotomy was performed
with a good functional result for 11 years, at which time a total
hip replacement became necessary
530 Anatomical Results of Operation Within Three Weeks After Injury

Table 23.5

Quality of reduction according to time from injury to surgery

Date of operation 0-21 Days 21 Days - 4 Months >4 Months

Perfect Imperfect Perfect Imperfect Perfect Imperfect

1958 - 1962 681170 32% 0 0 0 0

1963 -1967 69% 31% 71% 28% 57% 43%
1968 -1972 79% 20% 76% 24% 75% 25%
1973 - 1977 83% 17% 76% 24% 71% 29%
1978 -1983 84% 16% 80% 20% 50% 50%
1984 -1990 90% 10% 80% 20% 68% 32%

in this book represent experience gathered over
more than 30 years and include all the cases of
acetabular fractures we had to treat (except those
treated immediately by total hip replacement,
without any attempt at reduction). This means
that, whatever the rate of improvement in perfect
reduction achieved by learning, experience and im-
provements in means of internal fixation, all the er-
rors made at the beginning of our experience are in-
cluded as well, when we were pioneering this field,
both in interpretation of radiographs and in
surgical technique and methods of fixation. I am
sure - and pleased - that present and future
generations of surgeons will not have to pass
through the same difficult period, and that their
learning curve will rise much more steeply in the
direction of the ideal: 100070 excellent reductions.
Table 23.5 and Fig. 23.5 report the percentages
of perfect reductions we performed in periods of
approximately 5 years. Except for the late recon-
structions, the curve is slightly ascending and near-
ly regular. The more irregular curve of the delayed
reconstructions is explained by the smaller number
of operated cases: in the period 1978 -1983 we
operated on 14 cases (7 perfect, 7 imperfect) and in
Fig. 23.4A- D. Juxta-tectal transverse fracture operated on
through a posterior approach on the 4th day after injury in
1966. A Antero-posterior radiograph before operation, B
antero-posterior radiograph 1 month after operation. The
reduction left a noticeable femoral head protrusion, C antero-
posterior radiograph 4 years after operation; the remodelling ap-
pears significant but the wear of the cartilage is already con-
siderable, D antero-posterior radiograph 12 years after opera-
tion. A total hip replacement appeared justified

Table 23.6

Date of Percentage of perfect reductions achieved in the three most complex associated fractures
operation
T-shaped Transverse and posterior wall Both-column a

0-21 Days 21 Days- >4 Months 0-21 Days 21 Days- >4 Months 0-21 Days 21 Days->4 Months
4 months 4 months 4 months

1958 -1962100% 43% 29%

212 b 317 217
1963-1967 33% 50% 66% 66% 52% 33%
113 112 8/12 4/6 12123 113
1968 -1972 100% 100% 0% 69% 50% 60% 66% 0%
5/5 111 0/1 9/13 3/6 18/30 213 0/1
1973-1977 50% 100% 100% 66% 75% 100% 90% 50% 0%
112 3/3 1/1 4/6 3/4 111 18120 2/4 0/1
1978 -1983100% 50% 100% 100% 100% 58% 66% 0%
4/4 214 212 11111 5/5 11119 213 0/4
1984-1990 100% 50% 50% 90% 91% 83% 88% 71% 0%
6/6 2/4 214 30/33 11/12 5/6 29/33 517 0/4

a The cases of secondary surgical congruence are not included in this table.
b Under the percentages are the real number of cases operated upon.
 The Learning Curve 531

A B

C D

Fig. 23.4
532 Anatomical Results of Operation Within Three Weeks After Injury

%
100 ---------------------------------------------------------.---------------------------------------------------------------------- ..

" f==!-:~
90

80

70 ------~~~ .

i
60
1
50

40

30

20

10

1958-B2 1963-B7 1968-72 1973-77 1978-83 1984-90 years

0-21 days Fig_ 23.S. Percentages of perfect reduc-

21 days-4 months tions performed, in approximately 5-year
> 4 months periods

%
100

90

80

70

60

50

40

30

20

10

1958-B2 1963-B7 1968-72 1973-77 1978-83 1984 90 years

Fig. 23.6. Percentages of perfect reductions
Both-column achieved in complex fractures operated on within
Transverse and posterior wall 3 weeks of injury, in approximately 5-year
T-shaped periods
Appendix: Claude MARTIMBEAU'S Method of Assessing Displacement in Acetabular Fractures 533

Appendix: Claude MARTIMBEAU'S Method

Radiological Evaluation
Score of Assessing Displacement in Acetabular
1. Congruence AP (1) Fractures
Obturator (1)
Iliac (1)
2. Posterior Column (Wall) Iliac (Obt.) (2) For a quantitative assessment of the post-operative
3. Anterior Column (Wall) Obturator (2) radiographs, specific aspects of the reduction are
4. Central Protrusion AP (1) recorded in each projection. In the antero-posterior
Optimal Score (8)
view, the degree of congruence of the hip joint and
the magnitude of residual central protrusion of the
Fig. 23.7. Radiological evaluation of congruence, central protru- femoral head are documented. In the iliac-oblique
sion, and displacement of the column (or wall) and the maxi-
view, the reduction of the posterior column or
mum point score for each detail. (From MEARS and RUBASH
1988) posterior wall and the congruence of the hip joint
are recorded. In the obturator-oblique view, the
reduction of the anterior column or anterior wall
and the congruence of the hip joint are measured.
Subsequently, the various displacements are ana-
lysed by resort to a grading system.
To evaluate the degree of persistent central pro-
trusion by a study of the antero-posterior view, a
line is extended from the sciatic buttress in a distal
direction to the ischial tuberosity, which is parallel
to the longitudinal axis of the pelvis (Figs. 23.7 and
23.8). A point score system is employed:

• no displacement of the femoral head = 1

• medial displacement of the femoral head across
the line = 0

The assessment of post-operative congruity is eval-

Fig.23.8. Pre-operative evaluation of the congruence of the
femoral head: measure 8 mm medial to the centre of the femoral
head and 1 mm lateral to the centre of the femoral head - the
difference is 7 mm, score O. Evaluation of the protrusion of the
femoral head: medial displacement of 1 mm of the femoral head
across a line drawn from the sciatic buttress and parallel to the
longitudinal axis of the pelvis - score O. (From MEARS and
RUBASH 1988)
uated in a similar way in the examination of the
antero-posterior, iliac, and obturator views. In each
view, a line is drawn from the center of the femoral
head in a superior direction that is parallel to the
longitudinal axis of the pelvis. One centimeter
medial and lateral to where this line crosses the
edge of the femoral head, the width of the joint
space is measured. A point score is assessed:

the period 1984-1990 on 25 cases (16 perfect, 9 im- • a difference of 1 mm or less = 1

perfect). • a difference of more than 1 mm = 0
Table 23.6 and Fig. 23.6 show the rate of perfect
reductions achieved in operations performed within For the evaluation of the anterior column or wall
3 weeks of injury for the most complex varieties of in the obturator-oblique view or the posterior col-
fractures: both-column, T-shaped and associated umn or wall in the iliac-oblique view, the
transverse and posterior wall. The smaller number magnitude of persistent displacement is evaluated
of the T-shaped fractures explains their more ir- by point score:
regular curve in Fig. 23.6. We have taken the oppor-
• displacement of 0 to 1 mm = 2
tunity in Table 23.6 of adding at the same time data
• displacement of 2 to 3 mm = 1
on the rate of perfect reductions achieved between
3 weeks and 4 months and later than 4 months after
• displacement of more than 4 mm =0
injury as well, for purposes of comparison. (From MEARS and RUBACH 1988.)
24 Early Complications of Operative Treatment
Within Three Weeks of Injury
The early complications directly related to the op-
eration will be considered under the headings of
death, infection, nerve damage, secondary dis-
placement at the fracture site, thrombo-embolism,
local complications and miscellaneous. There were
569 cases operated on within 3 weeks of injury.
24.1 Death
During the period of hospitalisation there were 13
deaths (2.28070). These were attributed to the fol-
lowing causes:
1 brain stem damage
4 massive pulmonary embolism
2 inadequate post-operative resuscitation
2 unexplained circulatory collapse
1 myocardial infarction
1 septicaemia due to Candida albicans
1 thrombosis of the external iliac artery
1 unknown
Seven of the operated patients were more than 60
years old and there was a mortality of 5.7% for
these elderly people.
From a more analytical point of view, two of
these deaths were directly attributable to surgery of
the acetabulum, viz. the cases with septicaemia and
arterial thrombosis. Four deaths could probably
have been avoided by more efficient resuscitation.
Six deaths (1.05%) were associated whh a surgical
complication which could affect any procedure.
In relation to operative approach, eight deaths
followed a Kocher-Langenbeck approach (1 pulmo-
nary embolism, 4 inadequate resuscitations, 1 brain
stem damage, 1 myocardial infarct, 1 septicaemia),
two followed an ilio-inguinal approach (1 throm-
bosis of the iliac artery, 1 unknown), one followed
an ilio-femoral (1 pulmonary embolism) and two
followed various other approaches (2 pulmonary
embolisms).
24.2 Infection
There were 24 post-operative infections of all types
(4.2 %). This high incidence relates to 30 years of
surgery of the acetabulum, but there were two
peaks. The first was during the early stages when
we were inexperienced. The pathological anatomy
had not always been thoroughly understood before
the operative treatment, and errors of surgical
approach led to long surgical procedures which
were followed by several bad infections. The second
peak of incidence followed early utilisation of the
ilio-inguinal approach when there were 7 infections
out of the first 22 operations, that is, 31 %. Better
understanding of this approach, precautions to
preserve the lymphatics, and the systematic use of
antibiotics during the peri-operative period have
contributed considerably to reduce the frequency
of this complication.
Out of the 114 subsequent ilio-inguinal opera-
tions, one was followed by infection (0.8%).
24.2.1 Analysis of Post-operative Infections
Infections have occurred after:
13 Kocher-Langenbeck incisions out of 314
(4.10/0);
8 iIio-inguinal approaches out of 158 (5%);
1 Smith-Petersen approach out of 2;
1 double approach (staged) out of 22 (along
with the ilio-inguinal approach);
1 Ollier approach out of 3.
These post-operative infections have been very vari-
able in type and in clinical presentation and seri-
ousness. We have regrouped them schematically ac-
cording to their site with respect to the operated
hip.
536 Early Complications of Operative Treatment Within Three Weeks of Injury
(a) Superjicial injections. Of the nine cases of su-
perficial infection, three followed Kocher-Lan-
genbeck approach incisions and six followed
ilio-inguinal incisions. Five were sited superfi-
cial to the gluteus maximus or external oblique
muscles. A wide excision of the infected area
was performed. Included here are four infected
haematomata of the retro-pubic space after
ilio-inguinal incisions which were recognised
from the 12th to the 30th post-operative day.
One of these was preceded by an infected
Morel-Lavalle lesion in the trochanteric area.
Infected haematomata were debrided and ex-
cised together with their lining membranes and
cures were obtained without hip infections.
The nine cases were followed up and the
long-term results are: three excellent, three
good, one very good but with radiological
osteoarthritis, one fair and one poor due to ad-
ditional osteoarthritis, probably unrelated to
the infection.
(b) Immediate deep injections. There were ten
cases, of which seven followed Kocher-Langen-
beck incisions, one an ilio-inguinal, one an
Ollier and one a Smith-Petersen approach.
The infection manifested between the 8th
and 15th day, but in all cases the hip appears to
have been involved from the start. An excision
of the infected area was performed, but the ac-
counts do not mention the cleaning of the joint
itself. Healing was never obtained straight-away
and further excisions were subsequently neces-
sary. Clearly the excisions were not as radical as
we would now judge necessary.
Following these ten deep infections:
1 patient died from Candida albicans sep-
ticaemia;
9 are now dry but the radiographic ap-
pearance at long-term follow-up is that of
sequelae of septic arthritis; clinically 2 are
good, 3 fair and 4 poor. Two of these pa-
tients, after years of clinically good results,
developed osteoarthritis of the hip which
required total hip replacement.
(c) Delayed injections. Three patients developed
an infection a few months after surgery. These
secondary infections occurred after one
Kocher-Langenbeck approach, one ilio-ingui-
nal, and one double approach (Kocher-Langen-
beck and ilio-femoral).
One case, following an ilio-inguinal
approach, did not involve the hip. It was ex-
cised and healed uneventfully; there was a very
good result at 10 months but it has not been
reviewed since.
Two cases involved the hip. In one, bony
ankylosis of the hip resulted. The other case
followed a Kocher-Langenbeck approach in a
patient with serious sequelae of poliomyelitis
and in whom the early post-operative course
had been accompanied by pyrexia over a long
period. Debridement was performed and the
incision healed. Clinically the result is poor,
with radiological sequelae of infected arthritis,
but after 17 years the patient is still able to have
a nearly normal life (taking her polio sequelae
into account as well).
(d) Late injections. Two late infections developed
after an initial Kocher-Langenbeck approach.
In the first case, the sinus developed after 2
years; it corresponded to infection around deep
stitches. Excision was performed and fixation
devices were removed. At 4 years osteoarthritis
developed, with good mobility of the hip,
which was treated by a MacMurray osteotomy;
the result of this was satifactory 4 years later.
In the second case an abscess developed
along the incision track 6 years after surgery.
The infection was debrided and the devices
removed; the patient healed uneventfully and
had a perfect result at 18 years.
24.2.2 Cause of Infection
As usual, this was not known in the majority of
cases. On one occasion, a sinusitis was probably re-
sponsible, otherwise, the difficulty and length of
the operation, and errors of approach may be in-
voked.
The lymphatics draining the foot, genital or-
gans, the groin, the perineum and the anal canal,
which are damaged during the ilio-inguinal
approach, are probably a significant source of in-
fection, especially in the presence of haematoma
formation.
24.2.3 Prophylaxis
(a) Do not operate on febrile patients or on those
having a leucocytosis. First look for and treat
the focus responsible for the condition.
(b) Recognise and evacuate a Morel-Lavalle lesion
over the greater trochanter.
Nerve Damage
(c) Use multiple suction drains in all recesses of the
operative wound so as to prevent haematoma
formation.
(d) Look for post-operative haematomata, particu-
larly in the internal iliac fossa; evacuate them
surgically.
(e) Administer prophylactic antibiotics 24-48 h
before operation and continue these, especially
after the ilio-inguinal or extended ilio-femoral
approach, for 10- 15 days.
24.2.4 Treatment
Once an infection is recognised, remove a few
stitches to allow the haematoma to evacuate. Vigor-
ous antibiotic therapy starts as soon as material has
been taken for culture.
As soon as the inflammation of the tissue
around the incision has been controlled (usually
necessitating 6 - 8 days of antibiotic therapy), early
excision of the infected area should be performed.
This comprises complete removal of all the necrotic
and infected tissue, beginning at the incision, of
which the two margins should be excised together
with the fistula. It is necessary to follow the fistula
into its depths, excising its wall and the deep re-
cesses of the abscess, but leaving any healthy tissue.
This excision procedure is easy when the infec-
tion is confined to the soft tissue superficial to the
gluteus maximus or the external oblique. Having
established with certainty that the abscess cavity
does not communicate with the hip by manipulat-
ing the joint, and that haemostasis is adequate,
numerous suction drains extending to the depth of
the abscess cavity and its recesses are inserted, and
the wound sutured.
Sometimes deeper exploration has led to the ex-
posure of plates or the heads of screws. Curettage
of the bone on the surface is necessary in order to
remove all aponeurotic devitalised tissue; then the
plate should be partially loosened in order to clean
under it. After checking that the hip joint itself is
not involved, and inserting multiple suction drains,
the wound can again be closed.
Finally, if a communication with the joint is
discovered, the prognosis becomes much less cer-
tain. It is essential to clean the joint. If operating
through the posterior approach, we would advise a
capsulo-synovectomy and division of the transverse
acetabular ligament in order to clean thoroughly
and drain the deep part of the joint. From the ante-
rior aspect, it is probably preferable to supplement
the ilio-inguinal approach by a vertical anterior ap-
537
proach in order to reach the joint and to perform
the capsulo-synovectomy. After copious lavage
with Dakin's fluid or Betadine solution, the wound
is closed in layers with suction drainage in as many
planes as necessary.
24.3 Nerve Damage
24.3.1 Sciatic Nerve Damage
Altogether there were 36 post-operative cases
(6.3%), but their extent and date of onset were very
variable.
Thus, there were 34 immediate sciatic nerve
palsies (5.90/0), as follows:
1 total sciatic (after a Kocher-Langenbeck
approach);
7 total external popliteal component (after 7
Kocher-Langenbeck approaches);
11 partial external popliteal (after 11 Kocher-
Langenbeck approaches, one of which was part
of a double);
4 mild patchy lesions (2 after Kocher-Langen-
beck approaches, 1 of which was double, 2 after
ilio-inguinal approaches);
7 severe patchy lesions (6 after Kocher-Langen-
beck, 1 after ilio-inguinal);
2 pure sensory impairments (after 2 Kocher-
Langenbeck);
2 post-operative extensions of a known pre-
operative sciatic palsy (1 after Kocher-Langen-
beck, 1 after ilio-inguinal).
Besides these, we have seen two cases of secondary
sciatic palsies: that is to say that they appeared
after a post-operative interval during which there
was normal function of the sciatic nerve. These two
cases followed Kocher-Langenbeck approaches.
24.3.1.1 Immediate Post-operative Sciatic Palsy
We deplore the fact that of the 34 patients with
post-operative sciatic palsy, 19 had not been ex-
amined adequately pre-operatively, meaning that
the sciatic nerve functions were not tested. In view
of the frequency of pre-operative paralysis of the
sciatic nerve (12.2%), a certain number of these
paralyses now recorded as post-operative probably
existed before the interventions. The need for pre-
operative clinical examination of sciatic nerve func-
tion cannot be overestimated.
538 Early Complications of Operative Treatment Within Three Weeks of Injury
Post-operative sciatic nerve damage occurred
immediately after 30 Kocher-Langenbeck
approaches out of 337, including 22 double
approaches (8.9070), and after four ilio-inguinal
approaches out of 179, including 19 combined
approaches (2.2%).
The importance, when operating through the
posterior approach, of maintaining the knee flexed,
and of extreme vigilance with the use of retractors
has been stressed during the description of the
Kocher-Langenbeck operation (Chap. 20). The pre-
dominant involvement of the external popliteal
component fits in with the mechanism of stretch-
ing, as in pre-operative paralyses.
Of 21 patients with post-operative sciatic palsies
studied by electromyography of both the femoral
and sciatic nerve territories, seven showed an elec-
trical lesion of the quadriceps muscle, suggesting
involvement of the lumbar plexus.
Before the use of transcondylar femoral trac-
tion, out of 126 Kocher-Langenbeck approaches
we had 23 sciatic palsies (18.4%) and since the
use of traction, out of 211 Kocher-Langenbeck
approaches we have had seven immediate sciatic
palsies (3.3 %). Thus the precautions advocated
have not eliminated the complication. The seven
patients concerned presented as follows:
1 total sciatic palsy;
1 total external popliteal palsy;
2 partial external popliteal nerve palsies;
2 scattered but not severe sciatic palsies;
1 post-operative extension of a known pre-
operative palsy.
We believe that these neurological deficits which
occurred despite transcondylar femoral traction are
due to direct trauma to the sciatic nerve at opera-
tion. The total sciatic paralysis was the result of
complete division of the sciatic nerve with scissors
and this was at the beginning of our use of trans-
condylar femoral traction. The nerve was very re-
laxed and was not recognised. It was repaired and
a partial recovery has occurred - more than ex-
pected, for the triceps surae has regained grade 4
strength.
Other cases are due to damage of the nerve by
retractors in two different circumstances: (1) when
the sciatic nerve lies near an ordinary retractor
under which it readily skids; or (2) when the tip of
our special sciatic nerve retractor is not maintained
against the bone, so that the nerve becomes caught
and pinched or scraped against the retro-acetabular
surface.
One sciatic paralysis after the ilio-inguinal
approach might be explained by direct damage with
the drill introduced from the internal iliac fossa
and reaching the angle of the greater sciatic notch.
Alternatively, in this case there may have been a
sciatic palsy which was not detected pre-operative-
ly; the electromyogram in this instance suggested a
lumbar plexus lesion of a patchy distribution. The
other three cases remain unexplained. However, the
frequency of sciatic damage after the ilio-inguinal
approach has increased since the first edition of
this book from 1.6% to 2.2%. We do not consider
this figure high enough to justify performing the il-
io-inguinal approach with the knee flexed, but we
must pay more attention to this problem.
The types of fracture associated with sciatic
nerve damage were as follows:
7 posterior wall fractures;
1 posterior column fracture;
1 associated posterior wall and posterior col-
umn fracture;
10 associated transverse and posterior wall frac-
tures;
2 T-shaped fractures;
12 both-column fractures;
1 anterior column and posterior hemitrans-
verse fracture.
The frequency of sciatic paralysis after operation
on both-column fractures by the posterior
approach (8 cases out of 40 single approaches, i.e.
20%, and after 2 out of 19 double approaches) sug-
gests that operation in this field is particularly
hazardous. However, through the ilio-inguinal
approach we have had four sciatic palsies out of 86
cases treated (4.6%).
24.3.1.2 Prognosis
The treatment of the post-operative paralyses has
comprised only the maintenance of passive mobili-
ty of the joints of which the nerves were involved.
We have not explored any of the nerves subsequent
to the main operation to which the damage relates.
Table 24.1 shows the outcome of the 34 im-
mediate post-operative sciatic palsies according to
two criteria: the degree of recovery and the func-
tional impairment. With regard to progress:
9 cases have recovered totally;
12 cases have significant recovery; together
these represent 65.6% of the cases;
Secondary Displacement of Fracture Site
Table 24.1
Extent of immediate post -operative sciatic palsy
1 Total
7 Total external
11 Partial external (10 reviewed)
4 Mild patchy
7 Severe patchy (6 followed-up)
2 Pure sensory
2 Post-operative extension of known pre-
operative palsy
8 have recovered partially; 3 have a normal life,
nevertheless, 4 wear an orthopaedic appliance,
and 2 have purely sensory residua.
Except for one patient with pure sensory involve-
ment, all the patients recovered more or less over a
period which can extend up to 3 years. Two-thirds
of the patients have practically no sequelae. So,
while serious, the prognosis of a post-operative
sciatic nerve palsy is not especially gloomy.
24.3.1.3 Delayed Sciatic Palsy
We have seen two cases of delayed sciatic palsy, ap-
pearing on the 14th and the 45th day after a
posterior approach. One patient with significant
sciatic pain without muscular deficit was operated
upon on the 58th day after the first operation and
the nerve was found to be enveloped by bone for-
mation and fibrous tissue. Neurolysis improved the
clinical state and the patient now only has purely
sensory residua. The other case partially involved
the external popliteal component; this patient re-
covered totally and has normal function.
24.3.2 Other Nerve Damage
(a) Damage to the lateral cutaneous nerve of the
thigh is nearly compulsory through the dif-
ferent anterior approaches, the ilio-inguinal
and the extended ilio-femoral. At the time of
follow-up 45 patients were complaining of
more or less intense neuralgia, after 351
approaches in which the nerve was at risk
(about 12070).
539
Recovery Corresponding function
1 Partial 1 Orthopaedic appliance
3 Total 3 Normal function,
4 Considerable 1 Normal, 3 sequelae but normal life
6 Total 6 Normal
3 Considerable 2 Sequelae but normal life, 1 pure sensory deficit
1 Partial 1 Orthopaedic appliance
2 Considerable 2 Sequelae but normal life
2 Partial 1 Sequelae but normal life, 1 orthopaedic appliance
3 Considerable 3 Sequelae but normal life
3 Partial 2 Sequelae but normal life, 1 orthopaedic appliance
2 Partial 2 Pure sensory deficit
1 Nil 1 Considerable disability
1 Recent 1 Improving
(b) Femoral nerve palsy was seen in two cases after
ilio-inguinal approaches chosen for two com-
plex anterior column fractures. The two pa-
tients recovered over a period of 1 year.
(c) Partial gluteal nerve palsy was recorded in four
cases after Kocher-Langenbeck approaches.
One case could be explained by the fact that we
were obliged to ligate the superior gluteal pedi-
cle in order to control a veinous haemorrhage.
Recovery was not perfect at 4 years, but the gait
is reported as normal. The patient died 5 years
after surgery from other causes.
In the other three cases there were no intra-
operative problems at the level of the superior
gluteal neurovascular bundle, and the partial
weakness of the gluteal muscles had disap-
peared completely or to a significant extent
when the muscles were rated 4 at 2 and 3 years
respectively in two cases and 5 in one case at 22
years post-operatively.
24.4 Secondary Displacement
of Fracture Site
Of the cases operated on within 21 days, six have
had secondary displacements of the fracture site,
an incidence of 1.05%. Two were partial: one in a
both-column and one in a T-shaped fracture. One
occurred after a lateral approach, but in that case
we found very poor osteopenic bone. One occurred
after an Ollier approach, which we used at the
beginning of our experience but which gave bad ex-
posure.
540 Early Complications of Operative Treatment Within Three Weeks of Injury
Four displacements were important. Two of
these were in transverse and posterior wall fractures
approached through a Kocher-Langenbeck inci-
sion. One case also developed an immediate deep
infection; this patient's result is bad with radio-
graphic signs of the sequelae of infectious arthritis.
The other case was treated by an arthrodesis.
Of the two other cases, one was a both-column
fracture approached through the ilio-inguinal inci-
sion. The complexity of the case would have
justified using the extended ilio-femoral approach.
In spite of the malunion resulting from the second-
ary displacement, the hip was rated 6.6.6 at 2 years
after surgery.
The other case was of a complex anterior col-
umn fracture approached through the ilio-inguinal
incision. The osteopenic bone, poorly evaluated,
did not allow a perfect reduction; at 1 year the
destruction of the joint was such that a total hip re-
placement was performed.
In addition to these six cases, we have had three
instances of posterior redislocation of the femoral
head. Two were in cases of posterior wall fracture
and one an associated transverse and posterior wall
fracture. In spite of a very good reconstruction of
the joint, with the patient lying free in bed, the
femoral head redislocated in the course of the first
weeks. One case was explained by a missed in-
carcerated fragment; the head redislocated at 6
months, was repositioned and grafted and the hip
stabilised. At 20 years the patient needed a total hip
replacement.
In another case of transverse and posterior wall
fracture approached posteriorly, the head redis-
located on the 16th post-operative day. It was re-
positioned and remained stable, but the case was
lost to follow-up after 2 years.
The last case added a fracture of the femoral
neck to the posterior wall fracture. Repositioning
of the wall and fixation of the neck with screws was
imperfectly performed. Redislocation was followed
by surgery. At 2 years osteoarthritis had set in.
In spite of analysis of the radiographs and oper-
ative reports, in the last two cases the exact cause
of the redislocation is not known.
24.5 Thrombo-embolism
Apart from the massive pulmonary embolisms
responsible for four deaths, we observed three
severe cases of thrombo-phlebitis before the routine
use of anticoagulant prophylaxis. In spite of sys-
tematic carefully controlled anticoagulant prophy-
laxis, we have had 14 cases of patent thrombo-
phlebitis, and in addition eight patients sustained a
pulmonary embolism.
For a long time now we have conducted our
operations under some anticoagulant regimen,
various regimens being started at different times.
As we were apprehensive of the possibility of
haemorrhagic complications, up to 1978 the anti-
coagulant treatment was begun on the 3rd post-
operative day. Since then we have felt more confi-
dent in using a well-monitored anticoagulant pro-
phylaxis, set out by the physician of my depart-
ment, Dr. J. P. Moulinie, at the end of Chap. 21.
24.6 Wound Complications
(a) Ten cases of skin necrosis were treated by local
care or excision and closure, depending on their
size. One case occurred after a Kocher-Langen-
beck incision, five after the extended ilio-
femoral approach, two after ilio-femoral
approaches, one after an Ollier and one after
an ilio-inguinal approach.
(b) Two cases of problems with the lymphatics
after ilio-inguinal approaches were described in
Chap. 20.
(c) Haematomata of the operated area:
12 were aspirated: 6 after a Kocher-Langen-
beck approach, 2 after an ilio-femoral, 2
after an ilio-inguinal, 1 after an extended
ilio-femoral, and 1 after a double approach
in one stage.
25 were evacuated surgically after 12
Kocher-Langenbeck, 8 extended ilio-femo-
ral, and 5 ilio-inguinal approaches.
1 retroperitoneal haematoma of importance
in a case operated on through a Kocher-
Langenbeck incision needed surgical de-
bridement.
24.7 Miscellaneous Complications
For completeness, it must be mentioned that during
the period of hospitalisation we had to treat the
following intercurrent complications:
2 cases of septicaemia (already mentioned);
2 cases of delirium tremens;
2 cases of obstruction or occlusion of the bowel;
2 cases of gastro-intestinal haemorrhage.
25 Late Complications of Operative Treatment
Within Three Weeks of Injury
25.1 Pseudarthrosis
Four of the 569 patients with fractures of the
acetabulum operated upon within 21 days after the
accident developed pseudarthroses (0.7070). These
occurred in two both-column fractures and in two
associated transverse and posterior wall fractures.
All four cases had a slightly imperfect reduction.
The first had been operated upon by the posteri-
or approach, as the ilio-inguinal was not yet in use.
At 6 months a pseudarthrosis of the posterior col-
umn component was recognised; the plate had
broken away with respect to the fracture line.
Another operation was performed through the pos-
terior approach, the pseudarthrosis being excised
and the fracture line maintained by a plate put
under compression. Consolidation was achieved
after 3 months. The patient developed osteoarthri-
tis requiring total hip replacement 20 years later.
The second pseudarthrosis occurred in a both-
column fracture operated upon through the ilio-
inguinal approach; the plate had broken away from
the anterior column fracture, at the level of the
anterior wall of the acetabulum. There remained no
pain and function was excellent. In spite of a slight
secondary displacement and loss of parallelism of
the upper joint space, we judged further interven-
tion unnecessary. Seven years later the patient
developed osteoarthritis and the condition of the
hip justified a total hip replacement.
The third case (Fig. 25.1) was in an associated
transverse and postero-superior fracture detaching
the whole of the acetabular articular surface (see
Sect. 12.4). The postero-superior component was
detached by a fracture line reaching the iliac crest,
and a secondary fracture line through the anterior
column isolated the whole roof of the acetabulum.
Comminution of the fractures lines and already
formed callus on the 17th day after injury made the
operation treacherous and the reduction was not
perfect, but the fixation was apparently sound
(Fig. 25.1 B). Over 4 years the patient had several
periods of pain, some of them intense. The diagno-
sis was possible on plain views (Fig. 25.1 C) and
confirmed on tomograms, which showed non-
union of the transverse fracture, at least in its
posterior part. A second operation was performed
through the Kocher-Langenbeck incision. The non-
union reached the middle part of the greater sciatic
notch; it was cleaned and plated under compres-
sion. Some associated decortication of the hyper-
trophic edges was possible. At 6-month follow-up
the non-union had healed. Nine years later the hip
is rated 6.6.6 and the radiographs do not show any
sign of osteoarthritis (Fig. 25.1 E) except for an im-
portant collarette of osteophytes.
The fourth case occurred in an associated high
anterior column and posterior hemitransverse frac-
ture operated on on the 19th day after injury
through the ilio-inguinal incision. The complexity
of the fracture escaped the CT scan; for instance,
there were five horizontal fracture lines extending
from the main iliac wing fracture line towards the
sacro-iliac joint. The reduction of the anterior col-
umn was obtained with difficulty and was im-
perfect; the posterior hemitransverse fracture was
not fixed. This fracture line did not heal and was
responsible for pain and limping. The non-union
was approached through the posterior route. It was
mobile, and once a stable position of the mobile
posterior column fragment was found it was fixed
with a plate. To achieve this stable position we fixed
the non-union line with a clamp applied to two
temporarily inserted screws, and the position of the
fragment was modified until it did not move while
the hip was mobilised. Healing was obvious at 6
months. At 2 years the hip was rated 5.6.5. Radio-
graphically some changes in the femoral head
structure detected 1 year earlier are stable. The hip
is now rated fair and the long-term result is very
questionable.
No special indications arise from the study of
these four cases, but we found it interesting never-
theless to report them in some detail.
 542 Late Complications of Operative Treatment Within Three Weeks of Injury

A 8

Fig. 2S.1A-E. Juxta-tectal transverse fracture associated with a

postero-superior wall fracture extending to the iliac crest. A An-
tero-posterior radiograph before operation, B antero-posterior
radiograph 2 weeks after operation, Cantero-posterior
radiograph 4 years after operation, showing the non-union,
D antero-posterior radiograph 2.5 years after treatment of the
non-union and 7 years after the original acetabular fracture,
E antero-posterior radiograph 9 years after treatment of the
C non-union. The hip is rated 6.6.6
 Cartilage Necrosis
D E
Fig. 25.1D,E
25.2 Cartilage Necrosis
In six cases of our series of 569 early operated cases
(1070), between the 3rd and the 6th month post-
operatively pain developed in the hip and on
radiographs, we detected progressive and regular
narrowing of the joint space apparently without
alteration in the head or acetabular bone (Fig. 25.2).
Of these six cases, three followed posterior disloca-
tion of the head and a posterior wall fracture, while
the three others followed an anterior dislocation ac-
companying two associated anterior column and
posterior hemitransverse fractures and one pure high
anterior column fracture. Five of them were
recognised within 6 months after surgery and one
between 6 months and 1 year after. For several years
the joint space remained regular but very narrow
and the head was spherical apart from slight
osteophyte formation at the margin; clinically the
hips maintained a good range of movement. After
some years one case remained clinically very good
but had radiological signs of osteoarthritis, while
543
the other five had turned into one fair and four poor
results, because of the marked osteoarthritic
changes which developed. We regard these cases as
having developed a pure cartilage necrosis. It must
be emphasised here that early narrowing of the joint
space makes it necessary to discuss two diagnoses:
cartilage necrosis;
- an intra-articular screw, now easily identifiable
on CT.
Two other cases, initially labelled "cartilage
necrosis" but in which we discovered an intra-artic-
ular screw, were included in previous published
results. In the six cases described above we stick to
this diagnosis because it seems that only the car-
tilage of either the femoral head or the acetabulum
or both has practically disappeared, and the
presence of an intra-articular screw has been
eliminated. Nevertheless they are difficult to
understand, when we know how the cartilage is fed.
Chondrolysis must always suggest the intra-artic-
ular presence of the tip of a screw; we agree on that
with many authors.
544 Late Complications of Operative Treatment Within Three Weeks of Injury

A B c

D E F

Fig. 2S.2A- H. Pure cartilage necrosis. A Multiple-fragment progress at 3 (D) and 13 months (E), F appearance at 5.5 years,
fracture of the posterior wall with posterior dislocation of the G appearance 15 years after operation; signs of osteoarthritis
head, before operation, B appearance 10 days after operation, are also visible, H total hip replacement 25 years later
C - E pure chondrolysis beginning 2 months after surgery and its
Avascular Bone Necrosis
Fig. 25.2G, H
G H
25.3 Avascular Bone Necrosis
Avascular bone necrosis has up to now been far
from easy to diagnose. It seems that magnetic
resonance imaging (MRI) will facilitate the recogni-
tion of this complication at least at the femoral
head level. In our opinion avascular necrosis is a di-
agnosis much too often put forward to explain a
post-operative complication. Since it is known that
there is nothing we can do about it, as the trauma
is considered solely responsible for it, there is much
too great a tendency to blame necrosis for what is
really a wearing of the femoral head against a mal-
reduced fracture line. If wear takes place there is
disappearance of a segment of the head but no se-
questrum formation, and the shape of the loss of
substance is the negative imprint of the shape
responsible for the wear: the step in the acetabular
reconstruction. For instance, wearing against a
transverse fracture line appears on the antero-
posterior view as an orange-slice-shaped missing
part of the femoral head without any sequestrum.
In all instances of necrosis there is a period of
so much pain that the patient has to be radio-
graphed, which finally demonstrates a sequestrum
of the head over a more or less extended area. We
545
have reviewed all cases carefully in order to try to
ascribe to avascular necrosis only what is really
caused by that, and to pin on ourselves what is wear
of the head against a mal-reduced fracture and not
a necrosis; consequently there may be some varia-
tion between the figures in the first and in the sec-
ond edition of this book.
In 1966 Robert JUDET and I reported osteone-
crosis of segments of the acetabulum which can be
isolated or associated with involvement of the
femoral head. This may be the direct consequence
of surgical trauma and appears after difficult
operations accompanied by extensive stripping and
devitalisation of fragments of bone, or it may ap-
pear after the repositioning of fragments detached
completely at the time of the accident. However,
the proportion of completely devitalised and repo-
sitioned fragments which become necrosed is surely
low. It is impossible to overemphasise the necessity
of preserving musculo-aponeurotic pedicles of
bone fragments and muscular origins on the col-
umns themselves as far as possible.
After 569 fractures operated upon during the
first 3 weeks after injury, we have observed 22 cases
of necrosis of various types (3.10/0, in addition to
the 1.0% cartilage necrosis). These comprised:
 546 Late Complications of Operative Treatment Within Three Weeks of Injury

A 8

c D
 Avascular Bone Necrosis
E without having being damaged initially by
Fig. 25.3A-E. Associated necrosis of the femoral head and the
posterior wall. A This fracture, treated elsewhere, redislocated,
B appearance 3 weeks after reoperation, C nine months and
D 1.5 years after operation, function was excellent, E twenty-two
months after operation the associated necrosis of the femoral
head and the posterior wall is obvious. A total hip replacement
was necessary
(c) Necrosis of the anterior column. In the first
7 superior segmental but limited femoral head
necroses;
7 superior segmental but extended femoral head
necroses;
5 massive femoral head necroses;
3 mixed necroses of the femoral head and of the
posterior part of the acetabulum.
(a) Pure avascular necrosis. The clinical and mac-
roscopic aspects of femoral head avascular ne-
crosis which are in no way exceptional are well
known and will not be enlarged on here.
(b) Associated necrosis of the posterior wall and
the femoral head. In three instances after a sat-
isfactory reduction, pain and stiffening devel-
..
Fig. 25.3A-D
547
oped that were associated on the radiographs
with progressive upwards and outwards dis-
placement of the head due to progressive ero-
sion of the posterior wall together with destruc-
tion of the femoral head (Fig. 25.3). At revision
operation, it was found that this damage was
due at least partly to the fact that some of the
screws were by then intra-articular, a conse-
quence of the crumbling of the wall, which was
so extensive in one case that the plate was ex-
posed and was bearing against the femoral
head. There were ample mechanical reasons for
the head erosion, and this makes it difficult to
apportion the roles played by osteonecrosis and
mechanical wear in these examples. These
situations prompt two questions:
Was avascular necrosis of the head, with its
consequent irregular flattening, responsible
for the wear of the postero-superior wall
which finally reached the screws, the fric-
tion of which against the already altered
femoral head increased the damage?
Was there prior involvement of the walls of
the acetabulum which became necrotic and
disappeared, creating an irregular surface
against which the head destroyed itself
avascular necrosis?
The answers to these questions are unknown.
The involvement of both the head and the walls
of the acetabulum may make difficult a subse-
quent arthroplasty, on account of the loss of
bony articular structure.
edition of this book we described three cases of
necrosis of the anterior column. In fact, two of
them can easily be subtracted, as it has been
proven that they had an intra-articular screw.
The first case (Fig. 25.4) was described as fol-
lows.
"After surgical treatment of one extended
fracture of the anterior column by the ilio-
femoral approach, pain developed about 13
weeks after the operation with stiffening of
the hip and an external rotation deformity.
Radiography showed the joint space to be
generally reduced and while the head re-
mained spherical or had only very slight ir-
regularities in its contour, it had risen and
there was progressive protrusio acetabulae in-
dicated by deformity inwards of the ilio-
548 Late Complications of Operative Treatment Within Three Weeks of Injury

A 8 c

Fig. 25.4 A-E. Avascular necrosis of the

anterior wall. Appearance A before and
B after operation, C two years after
operation; the hardware was removed a
few months previously, D, E appearance
at 15 and 25 years respectively after
operation
o E
Avascular Bone Necrosis 549

ischial line. The upward displacement was
not due to a loss of substance of the head and
could be related only to bone destruction due
to post-operative necrosis of a segment of the
anterior column:'
In fact we still hesitate in analysing this case ac-
cording to the clinical evolution. We have classified
the case among the cartilage necroses, as it initially
looked very much like that. It is true that as the
head remained approximately spherical, some wear
of the middle part of the anterior column was de-
tected, but, once again, was it necrosis or wear
because the reduction was imperfect?
25.3.1 Aetiology
(a) TYpe of dislocation. Of the 22 cases of avascu-
lar necrosis, 17 followed a posterior dislocation
of the head, 15 of them involving only the head
and two the head and the posterior wall. Of the
five other cases of avascular necrosis, four
followed central dislocations (three cases in-
volved the head and one the head and the pos-
terior wall), and one, following an anterior dis-
location, was considered a case of isolated head
necrosis.
So, among the 569 cases operated on within
21 days after injury the rate of avascular necro-
sis is:
after a posterior dislocation: 171227
(7.5 0J0);
after an anterior dislocation: 1/63 (1.5070);
after a central dislocation: 41243 (1.60J0).
Table 25.1 shows the different types of necrosis
in relation to the initial dislocation, the
approach used and the type of acetabular frac-
ture.
(b) Surgical approach. It is not surprising that
necrosis was more frequent after the Kocher-
Langenbeck incision, since a large proportion of
the posterior dislocations are approached by this
route. Out of 22 necroses, 18 followed a Kocher-
Langenbeck incision, while two followed an ilio-
inguinal approach and the last two cases an ex-
tended ilio-femoral approach, which was chosen
to reconstruct two cases of associated transverse
and posterior wall fracture.
(c) TYpe of fracture. As would be expected, most
necroses occurred after elementary or associat-
ed fractures accompanied by a posterior dislo-
cation. There were:
5 posterior wall fractures;
4 associated posterior column and posterior
wall fractures;
10 associated transverse and posterior wall
fractures;
1 anterior column and posterior hemitrans-
verse fracture;
2 both-column fractures.

Table 25.1

Extent of necrosis Type of Surgical approach Type of fracture

dislocation

7 Limited superior 7 Posterior 7 Kocher-Langenbeck 3 Posterior wall

head necrosis 3 Transverse and posterior wall
1 Posterior column and
posterior wall
7 Extensive superior 2 Posterior wall
head necrosis 5 Posterior 5 Kocher-Langenbeck 3 Transverse and posterior wall
1 Anterior 1 Ilio-inguinal Anterior column and
posterior hemitransverse
Central Ilio-inguinal Both-column
5 Massive femoral 3 Posterior 3 Kocher-Langenbeck 2 Transverse and posterior wall
head necrosis Posterior column and
posterior wall
2 Central 2 Extended ilio-femoral 2 Transverse and posterior wall
3 Mixed necrosis 2 Posterior 2 Kocher-Langenbeck 2 Posterior column and
(femoral head posterior wall
and posterior Central Kocher-Langenbeck Both-column
acetabular wall)
550 Late Complications of Operative Treatment Within Three Weeks of Injury
(d) Time of reduction of the posterior dislocation.
It has often been stated that avascular necrosis
of the femoral head is more likely if reduction
is delayed, but since 1966 we have discounted
the prognostic value of the time of reduction of
the posterior dislocation. In this series, the rela-
tionships for posterior dislocations reduced
pre-operatively (167 cases) were as follows:
4 necroses (5070) out of 79 posterior disloca-
tions reduced within the first 6 h after in-
jury;
5 necroses (8%) out of 60 posterior disloca-
tions reduced 6 - 24 h after injury;
1 necrosis (16%) out of 7 posterior disloca-
tions reduced on the 2nd and 3rd day after
injury;
1 necrosis (4%) out of 21 posterior disloca-
tions of which the date of reduction is not
recorded.
As for the 58 intra-operative reductions of the
posterior dislocation, we have observed 6 ne-
croses, an overall rate of 10.3%. In detail:
1 occurred among the 7 reduced on the 1st
day (14%);
2 occurred among the 10 cases reduced on
the 2nd day (20%);
1 occurred among the 5 cases reduced on
the 3rd day (20%);
1 occurred among the 4 cases reduced on
the 4th and 5th days (25%).
However,
none occurred among the 8 cases reduced
on the 6th and 7th days;
1 occurred among the 16 cases reduced
surgically during the 2nd week;
none occurred among the 8 cases reposi-
tioned during the 3rd week.
So, in accordance with STEWART and MILFORD
(1954), it appears at first sight that the in-
cidence is lowest following closed reduction
performed within the first 24 h, if we take ac-
count only of close reductions performed with-
in 3 days of injury. This is not axiomatic, how-
ever, since we had one subsequent necrosis
among 33 cases reduced surgically between the
5th and the 21st day after injury.
Thus, necrosis is far from inevitable if the
reduction takes place after 24 h. In fact, the fate
of the femoral head appears to be decided from
the outset, because its vessels are either dam-
aged or not damaged at the moment of the ac-
cident. Intact vessels can nevertheless be jeop-
ardised while the fracture remains unreduced,
and would perhaps recover their patency if the
reduction took place fairly early.
Surgery does not seem to raise the incidence
of femoral head necrosis, whereas there proba-
bly is a relationship with acetabular necrosis.
(e) Quality of the acetabular reduction. Of the 22
operations followed by avascular necrosis, the
acetabular reduction was:
- perfect in 14 cases (3.3%);
- imperfect in 8 cases (5%).
(f) Conclusion. We must accept that femoral head
necrosis occurs and it seems practically im-
possible to prevent or to avoid it. The trauma
of accident practically always determines the
future of the femoral head in destroying or
sparing all or some of the vessels, and whatever
the quality of the surgical reduction, necrosis
may occur. However, during the operation,
avoiding stripping the periosteum should help
to reduce the incidence of necrosis of fragments
of the acetabulum.
25.3.2 Time of Presentation
Of our 22 cases of avascular necrosis, 19 (86%) ap-
peared after an interval of 3 -18 months and three
were diagnosed clinically between 18 months and 2
years (Table 25.2). It is our opinion that if the fem-
oral head has a normal appearance at 18 months,
avascular necrosis will not occur, particularly if ap-
parent demineralisation of the whole femoral head
has been seen on radiographs.
This concept, contrary to classical views, has
been confirmed by the study of necrosis after treat-
ment of old lesions. However, it is logical that
necrosis, a consequence of devascularisation creat-
ed by the injury or by the surgical treatment, does
not wait years to manifest itself.
25.3.3 Clinical and Radiological Course
The evolution and clinical course is variable and
unpredictable. In a few cases, despite marked
radiological changes, there is little pain or further
deterioration. Before the advent of MRI, it was dif-
ficult to assess extent of the necrosis in those in-
stances in which function is tolerable and involves
Post-traumatic Osteoarthritis
Table 25.2
Avascular necrosis Time of presentation
Within 6 Months
6 months to 1 year
Limited superior 1 1
Extensive superior 2 3
Massive 3 5
Associated femoral head
and acetabular
little pain. Medical treatment has not appeared to
influence the evolution of osteonecrosis and we are
forced merely to observe its spontaneous course. If
it reaches the stage of coxarthrosis, surgical man-
agement along conventional lines must be con-
sidered.
25.3.4 Clinical and Radiological Results
It is true that in some cases avascular necrosis of
the femoral head is tolerated for a long time, in
spite of the remodelling of the head and the
acetabulum, or possibly because of that. The hip is
not normal, but little pain, an acceptable range of
motion, and a moderate limp make life liveable and
the need for further surgical treatment is postponed
regularly. Such are the results described as "very
good or good clinically with or without radiologi-
cal osteoarthritis". Some of them do not deserve to
be cited as more than "fair" or "mediocre".
However, in the majority of cases avascular
necrosis of the head or of the head and the
acetabulum is not well tolerated, osteoarthritic
changes becoming superimposed on those due to
the necrosis. After a few years the conditions are
clinically bad and radiologically the osteoarthritis
predominates, so it is time for total hip replace-
ment.
For our 22 cases of avascular necrosis the
clinical evaluation is as follows:
2 very good results with clinically silent radio-
logical osteoarthritis;
2 good results;
4 fair or mediocre results;
14 bad results, with predominant radiological
signs of osteoarthritis.
Whatever the extent of the necrosis, at least 6011,10 of
the cases in each category progressed to a bad result
551
Total
1 -1.5 years 1.5 - 2 years
4 7
7
3
and have now benefited from or are waiting for a
total hip replacement.
25.3.5 Conclusion
The main problem appears to us to be how to tell
the difference between necrosis and wear on a mal-
reduced fracture. For a long time the presence or
absence of a sequestrum was the main criterion by
which to identify avascular necrosis. The role of
MRI is not yet settled, but it appears to us promis-
ing. Now we do not reposition femoral heads which
have been dislocated for weeks or months if their
MR image shows necrosis; in these cases we com-
bine total hip replacement and reconstruction of
the acetabulum. However, in fresh cases we carry
on repositioning the head, as the incidence of os-
teonecrosis is low on the whole and not enough
studies have been carried out to make it certain how
reliable MRI is in the early days following a dislo-
cation.
25.4 Post-traumatic Osteoarthritis
25.4.1 Osteophytes
Early in our practice we noticed that in a noticeable
proportion of cases, rapid development of a collar
of osteophytes occurred around the periphery of
the femoral head. These collars of osteophytes ap-
peared to have three types of evolution:
remaining stable for years accompanied by an
excellent long-term result;
increasing slightly in size without compromising
the excellent clinical result;
disappearing among the other radiological signs
of post-traumatic osteoarthritis.
 552 Late Complications of Operative Treatment Within Three Weeks of Injury

F==1PTO=====i 3 PTo 3 PTO

6 C2 5C2 3C2

~ ~11~ 1_1:_~~1~~::11r----:~~~:~----~
8 PTO
2 C2
___1_5_C_1__ ____
5 C1

F==1C1====!
2 5 PTO
3C2
3 C1

45 Perfect 44 Perfect 39 Perfect 34 Perfect

A
1966 1971 1978 1990

7 C2 4 PTO 5PTO
3C2 2C2

:~~J
3 PTO
13 C1 4C2
8 C1
I I 6 C1

8 C1 3 PTO
6C2

12 C1

F==1 Miscellaneous=

Fig. 2S.SA-C Overview of outcomes

in patients first reviewed in 1966 (A),
58 Perfect 50 Perfect 36 Perfect patients operated on between 1966
and 1971 (B), and patients operated
on from 1971 to 1978 (C). Patients
are classified as having either a
perfect radiograph or a stage 1 (C 1)
or stage 2 (C2) collarette of
B osteophytes. PTO, Post-traumatic
1971 1978 1990 osteoarthritis
Post-traumatic Osteoarthritis
5C2 3PTO
?C?
1====1 PTO=====!
8 C1
?c?
5 C1
4PTO
? ?
9 C1
46 Perfect 31 Perfect
1978 1990
Fig.2S.SC
These collars of osteophytes on the femoral head
are of variable dimensions. The majority measure
2 - 3 mm at the base with a height of 1- 2 mm
(these are described as C 1), but they can have a
base of 4 - 5 mm and a height of 3 - 4 mm (these
are rated C 2).
The significance of factors responsible for the
formation of the collar is not clear, and it must be
admitted that the long-term prognosis must remain
speculative. It is encouraging, however, that
DUPARC and FICAT (1960), in their paper on ar-
ticular fractures of the upper end of the tibia, noted
the frequency of post-operative peripheral osteo-
phytes around the tibia, even after perfect reduc-
tion; these appeared to have no ill effect on func-
tion.
In the first edition of this book and since, we
have not accepted considering the presence of a
femoral head collar of osteophytes as a sign or a
degree of post-traumatic osteoarthritis. The reason
for this is that these collars can be stable for a long
time and co-exist with an otherwise perfect result.
The long-term evolution of the cases with these
osteophytes which we were able to follow up in our
four reviews of cases conducted in 1966, 1971, 1978
and 1990 is reported below.
Fig. 25.5 shows the outcome of the operated pa-
tients who had a perfect radiological result without
a collarette of osteophytes or with a collarette
553
stage 1 (C 1) or stage 2 (C2). It remains true that
some C 1 may be accompanied for a long time by
an excellent functional result. Of the results as
assessed in 1966 (Fig. 25.5A), a third of the C 1
results (33070) were still excellent in 1990; the rest
had progressed, a few (13%) to C2, i.e. with only
an increase in the size of the osteophytes, but most
(53%) to radiologically evident osteoarthritis.
Among the cases operated on between 1966 and
1971 and assessed in 1990 (Fig. 25.5 B), of the 13
C 1 results 1971 six were still C 1 in 1990 (46%), but
four (30%) had progressed to C2 and three cases
(22 %) had developed osteoarthritis.
Among the cases operated on between 1971 and
1978 and assessed in 1978 (Fig. 25.5 C), of the eight
cases rated C 1, five were still C 1 in 1990 (62%),
two cases had become C 2 (25 %) and in one case
(12%) osteoarthritis had developed.
In summary, the early appearance of a collar of
osteophytes around the head has an unpredictable
prognosis. The rates of osteoarthritis among the
C 1 cases assessed as such at the first three follow-
ups (1966, 1971 and 1978) being 53%, 22% and
12% respectively in 1990. Over the same periods,
some of the cases with perfect results also devel-
oped osteoarthritis: the respective rates were 11 %,
8.5% and 8.7%. So, an early appearance of a collar
of osteophytes can be taken as an unfavourable fac-
tor as far as the very long-term prognosis is con-
cerned.
25.4.2 Osteoarthritis
From 1958 to 1989, i.e. over a 30-year period,
among the 569 patients operated upon within 3
weeks of injury and having at least 1 year of follow-
up, 97 developed post-traumatic osteoarthritis
(17%). Excluded from this group are 16 cases of
osteoarthritis which developed after post-operative
infection or after avascular necrosis of the femoral
head or head and acetabulum, which are taken in
account elsewhere. In this section are included the
cases of osteoarthritis which occurred:
after perfect reductions: 43 cases of 418
(10.2%);
after imperfect reduction: 54 cases out of 151
(35.7%).
For a better evaluation of the importance of this
problem, we have considered the possible and
known factors which could be responsible for such
an outcome in operated patients. However, not all
554 Late Complications of Operative Treatment Within Three Weeks of Inju
osteoarthritic cases have the same appearance, and
their courses can be very different. In some cases
osteoarthritis is present radiologically accompa-
nied by an excellent clinical grading and remains
this way for years. In other cases the hip deterio-
rates fast and needs total replacement within a few
years.
We have divided the radiological appearances of
post-traumatic osteoarthritis into five types:
- type I: osteophytes enlarge progressively
around the head and appear around
the acetabulum. All other components
of the joint remain normal. This type
of osteoarthritis is clinically silent and
the overall result is "very good with
radiological osteoarthritic changes" or
"good";
- type II: significant osteophytes and densifica-
tion of the roof. Type II osteoarthritis
is generally clinically silent or causes
only slight symptoms.
type III: adds to the pattern of type II a reduc-
tion of the upper joint space. Type III
osteoarthritis generally causes clinical
symptoms;
- type IV: adds cavities to the pattern of type III;
- type V: fully developed osteoarthritis.
These last two types are practically always accom-
panied by significant clinical manifestations and
rated poor more often than fair.
25.4.2.1 Aetiology
We agree with many authors that trauma may cause
damage to cartilaginous and bony structures of the
hip up to the cellular level or to their nutrient
vessels, but all of these are impossible to assess
accurately at present, so they have to be left out of
account.
The only factors that we believe to be certainly
responsible for a further development of osteoar-
thritis not explained by previous osteonecrosis or a
joint infection are: faults in reduction, unrecog-
nised intra-articular screw, wear of the head after a
secondary displacement (which in fact is equivalent
to a fault in reduction), lesions of the femoral head
well identified at the operation, and perhaps osteo-
arthritic changes that existed before the accident.
Faults in reduction appear to be the most impor-
tant factor in the genesis of post-operative arthro-
sis. The consequences of these faults are various:
they many generate wear of the head (cartilage fil
and then sub-chondral bone) against mal-reduci
fracture lines, or the reduction of the area of co
tact between the head and the acetabular articul
surface may generate a higher intra-articular pn
sure or cause intra-articular instability of t
femoral head, leading to wear.
Of these 97 cases of post-traumatic osteoarthr
sis:
in 44 cases, the cause is unknown;
in 45 cases the surgical reduction was imperfe
5 cases seem to be related to an intra-articu
screw;
in 2 cases important lesions of the femoral he
were discovered at operation;
in 1 case there was a secondary displacemer
25.4.2.2 Osteoarthritis After Perfect Reduction
In the field of acetabular fractures, these cases a
those which cause us most concern. We believe 1
have restored macroscopically normal anatorr
and after various periods of time arthritic chang
develop radiologically or clinically or both. In t
1971 review the rate was 4070, in 1978 it was 5.4~
and in 1988-1990 review it is 10.4%.
It has to be remembered that some patier
developed their osteoarthrosis 20 - 30 years aft
acetabular surgery. We feel obliged to consid
these as cases of post-traumatic osteoarthritis, b
are we right to do so? Are we not blaming surge
for something which would have happened witho
it anyway? We could perhaps rule out the cas
which developed osteoarthrosis on both sides, b
in the unilateral cases it is impossible to deci
after what interval the acetabular fracture could
declared innocent of responsibility.
Table 25.3 shows these 43 cases in relation
fracture type; the three most important tyr;
(posterior wall, associated transverse and posteri
wall, both-column) making the largest contributil
to this complication: 37 cases out of 43.
Thble 25.4 shows the overall long-term results ;:
cording to the degree of osteoarthritis at the h
review. There are 21 cases of types I, II and III ;:
companied by ten very good clinical results and fi
good, with only four fair and two poor resul
Types IV and V, which add up to 22 cases, c:
responsible for four good results (type IV onl:
two fair and 16 poor results.
Post-traumatic Osteoarthritis 555

Table 25.3 two good results, with three fair and 19 poor
(79070).
Type of fracture Number of cases of osteo-
arthritis
Table 25.3 shows that, in the same way as for the
perfect reduction, it is among the three most im-
Perfect Imperfect Total portant fracture types that we find the greatest
reduction reduction number of cases of osteoarthritis: 41 out of 54.
Thble 25.5 tries to analyse in more detail the in-
Posterior wall 19 3 22
Posterior column cidence of osteoarthritis following imperfect reduc-
Anterior wall 1 2 tion in relation to the degree of mal-reduction:
Anterior column 1 2
Transverse 2 2 the 16 cases in which the femoral head was re-
T-shaped 4 4 centred under the roof but the acetabulum was
Transverse and posterior 12 17 29 imperfect, and which developed osteoarthritic
wall changes, achieved 6 very good (37.5%), 4 good
Posterior column and 3 3
posterior wall
(25%), 2 fair (12.56%) and 4 poor results
Anterior column and 5 6 (25%);
posterior hemitransverse in the 19 cases of loss of parallelism of the upper
Both-column 6 21 27 joint space in which osteoarthritis developed,
Total 43 54 97 there were 6 very good results (31.5%), 3 good
(15.7%),2 fair (10.7%) and 8 poor (42%). Note
an increase in the incidence of poor results from
25% to 42% compared to the previous category;
the 9 cases of residual head protrusion, which
25.4.2.3 Osteoarthritis After Imperfect Reduction are close to the technical failures, achieved 1
very good result (11 %), 2 good (22%), 2 fair
This is much easier to understand and to accept (22%) and 4 poor (44.4%). However, the in-
than osteoarthritis after perfect reduction. Table cidence of poor results is comparable to that
25.4 shows the numbers of cases of different types following a single loss of parallelism of the up-
of osteoarthritis at the last follow-up and their per joint space;
overall results. of the 4 technical failures, there is one - one
Of 54 cases, 13 were type I or II, and of these might say paradoxical - good result; the other
the long-term results are clinically very good in 3 cases have poor results;
eight cases and good in five. The 17 cases assessed of the 6 cases of secondary surgical congruence
as type III are responsible for four very good, six which led to osteoarthritis, 3 have a clinically
good, three fair and four poor results. The 24 good result (50%), 1 a fair (16.6%) and 2 poor
cases of type IV or V, however, account for only (33%).

Table 25.4

Quality of initial reduction Type of Overall long-term result Total

osteoarthritis
Very good Good Fair Poor

Perfect 3 4
II 3 1 4
III 4 4 3 2 13
IV 4 7 12
V 1 9 10
10 9 6 18 43
Imperfect 3 4
II 7 2 9
III 4 6 3 4 17
IV 2 1 4 7
V 2 15 17
Total 12 13 6 23 54
556 Late Complications of Operative Treatment Within Three Weeks of Injury

Table 25.5

Type of imperfection of Type of Overall result Total

reduction osteoarthritis
Very good Good Fair Poor

16 Head centred, II 2 2
acetabulum imperfect III 4 3 9
IV 1
V 3 4
19 Loss of parallelism I 2 3
II 4 5
III 2
IV 1 1
V 7 8
9 Residual head II
protrusion III 2 2
IV 2 3
V 2 3
4 Technical failure II
III
IV
V
6 Surgical secondary 1
congruence III 2
IV 1
V 2 2
Total 13 13 7 21 54

It is among these imperfect reductions that the
wear of the femoral head against a mal-reduced
fracture line must be an essential factor in pro-
moting the onset of osteoarthritis.
25.4.2.4 Conclusion
Obviously the quality of the reduction is the most
important prognostic factor in the prevention of
post-traumatic osteoarthrosis. It would have been
easy simply to agree with this, but it is better when
the figures are in fact seen to be in accordance with
this statement.
Figure 25.6 shows an analysis of the time of
onset of osteoarthritis after surgical treatment.
After perfect reduction, 500/0 of osteoarthritic
changes are diagnosed between 10 and 25 years
later. Among the imperfect reductions, 80% of the
osteoarthritic changes occurred within 10 years
after surgery. Table 25.6 reports the incidence of
osteoarthritis after perfect and imperfect reduc-
tions in relation to patient age at the time of
surgery.
Let us now analyse a few other factors.
(a) Pre-operative signs of osteoarthritis. Among
our 97 cases of post-traumatic osteoarthritis
after surgical treatment performed within 3
weeks of injury, we noticed that in seven there
were pre-operative signs of osteoarthritis of the
injured side.
In six cases we achieved a perfect reduction.
One patient died early, one was lost to follow-
up within 1 year, and one is a recent case. Of
the remaining three patients, two had progres-
sion of their pre-operative signs of osteoar-
thritis at 2 years and 4 years; one still had
clinically very good function while the other
will need a total hip replacement soon. Hip
function in the last of these three cases was
rated 6.6.6 at 22 years after surgery, with a C 2
collarette of osteophytes.
In one case the reduction was imperfect; the
patient was lost to follow-up after 1 year but at
that time the pre-existing osteoarthritis was
stable and function was rated 6.6.6.
(b) Osteoarthritis of the non-operated hip. At the
time of the last follow-up, the opposite hip,
Post-operative Ectopic Ossifications 557

%
30

'"
:e
€ 24 24
Perfect reductions
~ Imperfect reductions
'"0
[£20
'"01
()

'0
CD
.0
E
:J
C

S
.Q
ill
10
-=
'0
ill
OJ
.'c!!
ill
Fig. 25.6. Time of onset of osteoar- f:!
ill
0..
thritis after surgical treatment in rela-
tion to quality of reduction (perfect or 2.3%
imperfect)
1 2 3 4-6 7-9 10-15 15-20 20-25 25-30 years

which was normal at the time of injury, had unknown in 46 cases;

developed osteoarthritic changes in 10 cases. an imperfect reduction in 45 cases;
Of the seven cases in which we achieved a an intra-articular screw in 5 cases;
perfect reduction, osteoarthritis developed on a secondary displacement in 1 case.
the operated side in four, at 26, 25, 25 and 20
The last point to be discussed is the outcome of
years respectively. The three other patients have
an intra-operatively recognised lesion of the femo-
a very good result at 4, 6 and 11 years.
ral head. Table 25.7 reports the cases in which
Of the three patients with imperfect reduc-
osteoarthritis occurred when a lesion of the femo-
tions, one had a good result when she died 15
ral head was recognised at operation. The figures
years later, while the other two have developed
relating to cartilage abrasions are almost certainly
osteoarthritis on the operated side. One toler-
underestimated, for two reasons: such abrasions
ates this perfectly at 28 years (hip rated 5.6.6);
were not always mentioned in the operative reports,
the other needs a total hip replacement (hip
and in by no means all operations is the whole of
rated 4.5.4).
the femoral head scrutinised.
All together, the cause of these 97 cases of
post-traumatic osteoarthritis is:

Table 25.6

Patient age Total Perfect reduction Imperfect reduction Overall number Cases of
(years) number of of cases of osteoarthritis
operated Cases without Cases with Cases without Cases with osteoarthritis as percentage
cases osteoarthritis osteoarthritis osteoarthritis osteoarthritis of total

10-19 37 31 2 2 2 4 10.81170
20-29 126 95 4 20 7 11 8.7%
30-39 131 88 10 21 12 22 16.71170
40-49 106 68 11 15 12 23 21.6%
50-59 93 50 9 19 15 24 25.81170
60-69 60 36 4 15 5 9 15.01170
70-79 15 7 2 5 3 20.01170
80-89
Total 569 376 42 97 54 96
558 Late Complications of Operative Treatment Within Three Weeks of Injury

Table 25.7

Lesions of the femoral head first Perfect reduction Imperfect reduction

recognised intra-operatively
Cases without Cases with Cases without Cases with
osteoarthritis osteoarthritis osteoarthritis osteoarthritis

Segmental fracture of the head 6 3

Localised impaction 4 4
Cartilaginous abrasion 4 0 1
Sub-chondral bone lesion 4 1 0
Wear of the head on fracture line 1
Other type of head lesion 0
Total 18 5 7 4

25.5 Post-operative Ectopic Ossification 6: normal;

5: 90 0 flexion;
Of 569 hips operated on within 3 weeks of injury, 4: between 70 0 and 90 0 ;
139 developed ectopic ossifications (24.4070). 3: less than 70 0 , no fixed deformation;
2: ankylosis in a position considered good;
1: ankylosis in a dysfunctional position.
For instance, an ectopic ossification classified IV-6
means an apparently ankylosing form according to
25.5.1 Clinical and Radiological Presentation
the Brooker classification but associated with nor-
mal mobility of the hip. On the other hand, IV-2
The site of these ossifications varied:
would mean a Brooker type IV ossification accom-
127 developed outside the pelvis; panied by ankylosis of the hip in a position con-
3 were inside the pelvis; sidered functional.
9 developed outside and inside the pelvis. Table 25.8 shows the association between
Brooker ectopic ossification type and hip mobility
The extent of these ectopic ossifications varied
greatly from one case to another. Like many other in all our operated cases in which ectopic ossifica-
authors, we have decided to use BROOKER'S classi- tion occurred, i.e. 182 instances when we include
those following delayed reconstructions.
fication, which distinguishes four types:
Ectopic bone formation appears early on radio-
graphy, and maturity is reached 6 months to 1 year
I: islands of bone less than 1 cm in diameter;
II: larger islands of bone, leaving at least 1 cm after operation. On three occasions we have seen
spontaneous regression of the bone formation and
free space between the two bones of the hip;
improvement in the range of motion.
III: the free space between the ossification and the
pelvis or the femur is less than 1 cm;
IV: apparent ankylosis of the joint by a bony Table 25.8
bridge uniting the pelvis and the femur. Brooker D' Aubigne hip mobility grade Total
ossification type
However, we have found this classification inade- 2 3 4 5 6
quate as the extent of bony formation does not par-
allel the residual mobility of the joint concerned. 2 18 21
II 1 10 7 62 80
So we decided to characterise an ectopic formation III 2 3 7 10 28 50
outside the pelvis by two numbers: the first refers IV 2 3 2 7 2 15 31
to the Brooker classification, the second one is the Total 2 5 6 26 20 123 182 a
d'Aubigne digit for the mobility of the hip joint.
This is scored as follows: a Out of a total of 891 operated cases.
Post-operative Ectopic Ossification 559

25.5.2 Aetiology
The aetiology remains unknown. The development
of any degree of ectopic bone formation is un-
predictable. It appears to be unrelated to the dif-
ficulty of the operation. Stripping of the gluteal
muscles from the external iliac fossa has been ac-
companied by the highest incidence; we have seen
very few ossifications in the internal iliac fossa,
despite the frequency of complete stripping of the
iliac muscle. Intra-pelvic formations were extremely
infrequent relative to the number of occasions
when we have had to elevate the obturator internus
muscle.
Ectopic bone formations have appeared after all
methods of approach. Table 25.9 shows the number
of ectopic ossifications observed after the different
types of approach in the 569 cases operated on
within 3 weeks of injury.
49 extended ilio-femoral approaches were fol-
lowed by ectopic ossifications in 28 cases
(57070), of which 12 were type IV and 5 type III
(together 34%);
after 315 Kocher-Langenbeck approaches there
were 83 formations (26.3%), more than a third
of which were of the severe forms III and IV;
145 standard ilio-inguinal approaches gave rise
to formations in 7 cases (4.8%), and not one
was an ankylosing form. In these 7 cases, we
cannot be sure, a posteriori, that the external
iliac fossa was not partially elevated for some
reason not mentioned in the operative reports.
It must be said that quite often during the
ilio-inguinal approach we elevate a little part of
the external iliac fossa along the iliac crest
and/or at the level of the anterior interspinous
notch, to put a clamp astride the bone in order
to mobilise the anterior column, and these
elevations, the extent of which is difficult to
assess, may be responsible for the rare cases of
ectopic ossifications associated with a standard
ilio-inguinal approach. However, if we take just
the 13 cases of an ilio-inguinal approach com-
bined with elevation of the external iliac fossa,
the rate of ectopic ossification is high, 6 cases
out of 13, and 4 of them grade III.
Double approaches were also accompanied by a
high incidence of ectopic bone formations of
variable severity. Of the 22 double approaches
performed in one or two stages there were 10 in-
stances of ectopic ossification (45.4%).
25.5.3 Treatment
Two conditions must be fulfilled before any opera-
tive procedure aiming to remove ectopic bone
comes into question:
the ectopic ossification must have considerably
reduced the mobility of the hip, i.e. it is a type
III or IV with a corresponding mobility grade
of 1- 3 or perhaps 4;
the new bone formation must be mature or
quiet.
Judgement of the latter is based on two things: the
radiographic appearance and isotopic bone scan-

Table 25.9

Operative approach Cases without Cases with ossification Overall number Number of Total number
ossification (Brooker type) of ossifications ossifications as of approaches
percentage of
2 3 4 total

Kocher-Langenbeck 232 12 38 25 8 83 26.3070 315

Extended ilio-femoral 21 10 5 12 28 57.1070 49
Ilio-inguinal 138 3 2 7 4.9070 145
Ilio-inguinal with 7 4 6 46.1070 13
stripping of the ex-
ternal iliac fossa
Ilio-femoral 11 1 1 8.3070 12
Double approach 12 4 3 3 10 45.4070 22
Other 9 2 4 30.7070 13
Total 430 19 53 40 27 139 24.4070 569
560 Late Complications of Operative Treatment Within Three Weeks of Injury
ning. A quiet ectopic bone has a stable appearance
on radiographs taken over several months and its
limits are perfectly delineated; the radio-isotope
uptake is just a little above that of the opposite
side.
Ectopic bone formation generally begins during
the first 2 - 3 weeks after surgery; if it results in a
reduction in mobility we either stop or reduce the
rehabilitation programme. Between 3 and 6 months
a first bone scan is performed to serve as a baseline,
another one is performed at 1 year and, if
necessary, a third at 18 months. Comparison of the
results helps one to plan the date of surgical
removal. In the absence of reliable means of
preventing recurrence, we never remove ectopic
bone earlier than 18 months to 2 years after the pre-
vious surgery.
Nowadays the patient is prepared for the exci-
sion by the administration of indomethacin star-
ting 2 days before the removal, and then given 7 Oy
radiation on the day after surgery. For 4 years we
administered 4 Oy for 3 days consecutively follow-
ing surgery.
To assess the exact extent and site of ectopic
bone formations, CT scanning is very useful in ad-
dition to the plain radiographs.
The operation is generally conducted through
the approach by which the acetabular fracture was
fixed. It has happened, however, that we use a
Kocher-Langenbeck incision for removal of ectopic
bone following a lateral approach, as this avoids a
new total elevation of the gluteal muscles from the
iliac wing.
Usually the ectopic bone appears to develop be-
tween the muscle and the surface of the capsule.
Sometimes, however, it is so closely attached to the
hip capsule that we have to open it during the pro-
cedure.
If the ectopic bone is not attached to the bone
it can be removed by sharp dissection, passing out-
side its "neo-capsule", and that is regarded as the
ideal. Most of the time, however, the bony forma-
tions are attached to one or both of the bones of
the hip and the architecture is so complex that it is
removed in pieces. Sometimes it proves impossible
to excise all the ectopic bone, but we remove as
much as we can in order to restore hip mobility as
nearly as possible to normal. The areas of at-
tachments to the pelvis or the femur are covered
with bone wax and several suction drains left for
several days.
The patient awakes on the hip continuous
passive motion unit and keeps it night and day for
12-15 days. Active exercises are performed two or
three times a day from the 2nd or 3rd post-
operative day.
Antibiotic therapy is begun 2 days before sur-
gery and continues until 2 days after the removal of
the last suction drain.
25.5.4 Prevention
Up until a few years ago, and over more than 25
years of experience, we had nothing to offer to-
wards prevention apart from recommending sur-
gery that caused as little trauma as possible to the
muscles. Diphosphonates were a source of great
hope for us, but treatment using them was a com-
plete failure.
Since October 1986, as recommended by other
authors, we have used indomethacin as a prophylac-
tic agent against ectopic bone formation. In the
USA, radiotherapy was tried again to prevent ossifi-
cation and appeared to be convincingly effective. So,
since 1987 we combine as often as we can these two
agents to prevent post-operative ossification, and
patients undergoing surgery by the Kocher-Langen-
beck or a lateral approach are given indomethacin
25 mg three times a day 24 h after surgery for 4
weeks post-operatively if they have no gastric con-
tra-indications. They are sent to the radiotherapist
to receive 7 Oy the day after surgery (for a while we
used 120y over 3 days post-operatively).
In 1980 we began to use the Kinetic continuous
passive motion (CPM) unit intensively immediately
post-operatively. This technique was considered by
some authors to be responsible for ectopic-bone for-
mation, but it seems to us that CPM changes
nothing in this area: it neither increases nor reduces
the rate of post-operative ossification, as our figures
demonstrate clearly. The figures showing the great
problem that this post-operative complication repre-
sents are reported in two tables.
Table 25.10 shows the number of cases of ec-
topic bone formation following the different main
surgical approaches in patients operated on within
3 weeks of injury in the period before we tried to
use prophylactic or preventive treatment, i.e. from
the beginning of our experience up to 1986. The
high incidence of ectopic bone formation after a
Kocher-Langenbeck, extended ilio-femoral or dou-
ble approach is striking. Another striking feature is
the difference in the incidence after a pure ilio-
inguinal approach (5%) and after an ilio-inguinal
approach combined with elevation of the external
iliac fossa (54070).
Post-operative Ectopic Ossification 561

Table 25.10

Operative approach Cases without ectopic ossification prophylaxis (up to 1986)

Number of Cases with ectopic ossification (Brooker type) Cases without

operated cases ossification
II III IV Total Percentage of
operated cases

Kocher-Langenbeck 281 11 37 25 6 79 28070 202

Kocher-Langenbeck/ 100% 0
triradiate
Extended ilio-femoral 26 6 5 6 18 69% 8

Ilio-inguinal 138 3 2 7 5% 131

Ilio-inguinal with 11 4 6 55% 5
stripping of the ex-
ternal iliac fossa
Ilio-femoral 12 1 8% 11
Double approach 22 4 3 3 10 45% 12
Ollier 2 1 50% 1
Others 6 0 0% 6
Total 499 18 50 39 16 123 24% 376

Table 25.11 shows the incidence of post-opera- tion has been attempted, i.e. since 1986. All pa-
tive ectopic bone formation among the patients tients also had the benefit of CPM post-operative-
operated on within 3 weeks of injury since preven- ly. Various protocols of preventive treatment were

Table 25.11

Type of prophylaxis/ Cases with ectopic ossification prophylaxis (since 1986)

operative approach
Number of Cases with ectopic ossification (Brooker type) Cases without
operated cases ossification
II III IV Total Percentage of
operated cases

Diphosphonates 10 mg
Kocher-Langenbeck 2 1 50
Extended ilio-femoral 2 2 2 100
Diphosphonates 20 mg
Kocher-Langenbeck 1 100
Extended ilio-femoral 7 3 3 6 86
Indomethacin 25 mg
Kocher-Langenbeck 19 2 11 17
Extended ilio-femoral 4 2 50 2
Ilio-inguinal 5 0 0 5
Ilio-inguinal with stripping 2 0 0 2
of external iliac fossa
Indomethacin + radiotherapy 12 Gy
Kocher-Langenbeck 7 0 7
Extended ilio-femoral 8 0 8
Indomethacin + radiotherapy 7 Gy
Kocher-Langenbeck 0
Ilio-inguinal with stripping of 0
external iliac fossa
Total 59 2 5 0 7 14 45
562 Late Complications of Operative Treatment Within Three Weeks of Injury
used. The table demonstrates that diphosphonates
were absolutely useless if not actually harmful (the
same goes for their effect in preventing ectopic
ossification after total hip replacement).
Indomethacin was the first efficient drug we
ever found for preventing ectopic bone formations.
Using this, we saw bone formation in four out of
29 cases (13.70/0), of which only one is type IV, the
three others being type I or II. However, the really
impressive results come from the combined use of
indomethacin and radiotherapy. Since we have
begun this form of preventive treatment we have
operated upon 17 acetabular fractures (eight
through a Kocher-Langenbeck approach, eight
through the extended ilio-femoral and one through
the ilio-inguinal with stripping of the external iliac
fossa) within 3 weeks of injury, without seeing even
one case of ectopic bone formation. Of course the
series is rather short and there are known potential
risks of radiotherapy, but the latter are considered
to be very low at the dosage in question here. Con-
sequently, for the first time, we have a prevention
technique which seems to be effective. This holds a
great deal of promise!
To emphasise these results, the figures for all
cases operated upon, whatever the delay from trau-
ma to surgery, will be summarised here.
Before 1986, of 499 cases operated upon with-
out treatment to prevent ectopic bone formation,
ectopic ossifications developed in 123 (18 type I, 50
type II, 39 type III, 16 type IV), i.e. 24.6%, and
44% of them were type III or IV.
Since we have used some form of treatment to
prevent ectopic ossification (112 cases), the results
have been as follows:
(a) Diphosphonates (34 cases): 24 ectopic ossifica-
tions (3 type I, 12 type II, 2 type III, 7 type IV),
i.e. 70.54%;
(b) Indomethacin (49 cases, comprising 25 Kocher-
Langenbeck, 8 ilio-inguinal, 1 Smith-Petersen,
and 15 extended ilio-femoral approaches): 5 ec-
topic ossifications (1 type I, 3 type II, 1 type
IV), i.e. 10.2%;
(c) Indomethacin and radiotherapy (29 cases, com-
prising 12 Kocher-Langenbeck and 17 extended
ilio-femoral approaches): no ectopic bone for-
mation, i.e. 0%.
The number of cases treated with indomethacin
and radiotherapy is still insufficient to warrant
definitive conclusions, but it is extremely encourag-
ing, especially since the figures for prevention of
ectopic ossifications after total hip replacement are
similar.
25.5.5 Results of Surgical Excision
of Ectopic Bone
Among the patients operated on within 3 weeks of
injury, ectopic ossification was removed in 14 cases.
The results were as follows:
4 type III: 3 excellent, 1 fair result (mobility
5);
10 type IV: 7 excellent, 2 mediocre, 1 poor re-
sult;
Thus, out of 14 cases, 10 results are excellent
(71 %), 3 mediocre and 1 poor because of infection.
Eight further patients are on the waiting list for
surgical excision.
25.5.6 Ectopic Ossification
and Cranio-cerebral Trauma
Here are the figures from our 849 operated cases:
no cranio-cerebral injury, no ossification: 489
cases (57.6%);
associated cranio-cerebral injury, no ossifica-
tion: 169 cases (19.90%);
associated cranio-cerebral injury and ossifica-
tions: 59 cases (6.94%);
ossifications, no cranio-cerebral injury: 129
cases (15 type I, 56 type II, 33 type III, 25 type
IV; 15.20%).
Consequently:
ectopic ossification without cranio-cerebral in-
jury occurred in 20.87% of cases;
ectopic ossification after associated cranial in-
jury occurred in 25.8% of cases.
Thus, an associated cranio-cerebral injury (45 short
coma, 8 long coma, 6 no coma) increases the in-
cidence of ectopic bone formation from 20.8% to
25.8%, which is less than is usually taught.
25.5.7 Ectopic Ossification and Type of Fracture
Table 25.12 shows the occurrence of ectopic ossifi-
cations according to anatomical type of fracture.
Obviously, they occur more frequently after the
types requiring either the Kocher-Langenbeck or
the lateral approach.
Post-operative Ectopic Ossification 563

Table 25.12

Type of fracture Ectopic ossifications Total number of Total number of Percentage with
(Brooker type) ectopic ossifications operated cases ectopic ossifications

II III IV

Posterior wall 6 20 11 4 41 208 20070

Posterior column 2 4 7 28 25070
Anterior wall 16
Anterior column 1 1 30 3070
Transverse 2 3 2 8 52 15070
T-shaped 2 7 2 2 13 59 22070
Transverse and posterior wall 5 21 14 11 51 175 29070
Posterior column and posterior wall 5 3 1 9 30 30070
Anterior column and posterior 4 2 7 55 13070
hemitransverse
Both-column 4 18 18 14 54 196 28070
Total 22 82 52 35 191 849 22070
26 Clinical and Radiological Results of Operation
Within Three Weeks of Injury

By 1 May 1990 we had operated upon 569 patients Very good:

with acetabular injuries within 3 weeks of injury, The hip is rated clinically 6.6.6 for a total of 18,
and of these, 544 had a minimum follow-up period or 17 (i.e. 6.6.5, 5.6.6, 6.5.6). These cases are not
of 1 year. However, 13 patients had died soon after included among the excellent results because
operation, and 39 had been lost to follow-up less they have either an unexplained slight func-
than 1 year after surgery, so 492 hips were available tional deficiency or some additional radiologi-
for evaluation. cal features, but at a grading which does not
The follow-up periods ranged from 1 to 33 alter the clinical result, or only very slightly.
years. Of these 492 hips, 105 were lost to follow-up These radiological features may be of four
during the 1987 - 1990 period, but they had been kinds: Heterotopic ossification; slight mal-
followed up over a period long enough to be includ- union of the acetabulum; osteoarthritic changes
ed in this chapter, as 27 of them were examined without any clinical symptoms; intra-operative-
after 1 (4 cases) or 2 years (23 cases), 25 after 3 - 5 ly achieved surgical secondary congruence.
years, 25 were followed up for 6-10 years, 23 for
Good:
11-15 years and 5 for 16-19 years.
The patient has normal activity despite a clini-
Fifty-three other patients died from other causes
cal imperfection or a radiological abnormality
during the period 1958 -1990, but they had also
been followed up for periods which justified in-
cluding their results with those of the 334 patients Table 26.1
who have been examined by myself in the course of
Pain Range of motion Ambulation
the last 3 years (follow-up periods for these 53 pa-
tients: 26 patients 1 - 5 years, 12 patients 6 - 10 6 None Flexion greater Normal
years, 8 patients 11 - 15 years, 4 patients 17 - 19 than 90°
years, and for the last 3 patients, 21, 22 and 33 5 Slight or intermittent; Flexion 70° - 90° No cane;
years, respectively). normal activity slight limp
after long
distance
working
26.1 Clinical Results 4 Pain after ambula- Flexion 50° - 70° Limp;
tion; easy walk of long distance
112 hour or more with cane or
The overall clinical results have been assessed for crutch
pain, mobility and gait, according to the method 3 Moderately severe; Flexion 30° - 50° Significant
walking no more than limp;
(slightly modified) of D'AuBIGNE and POSTEL
20 min cane per-
(1954; see Table 26.1) and on the anatomical and manently
radiological results as described in Chap. 23. Other 2 Severe; Flexion < 30° Very limited;
features such as sequelae of nerve complications ambulation limited to two canes
which may influence the gait are included in our 10min
Severe; Very restricted Bedridden
assessment when necessary. For overall assessment, prevents ambulation
the results are graded as:
If there is no contracture, just take flexion into account.
Excellent: If there is a contracture, subtract: 1 point for 20° flexion or ex-
The hip is clinically normal, rated 6.6.6, for a ternal rotation contracture, 2 points for 10° abduction, adduc-
total of 18, and radiologically perfect. tion, internal rotation contracture.
566 Clinical and Radiological Results of Operation Within Three Weeks of Injury

Table 26.2

Date of operation Clinical result at last follow-up Total

Excellent Very good Good Fair Poor

1953
1954 1
1955 2 3
1956
1957 1
Subtotal 0 0 2 4 7
1958 1 3
1959 2 3 6
1960 4 1 1 6
1961 2 3 2 2 9
1962 15 3 4 2 25
Subtotal 24 10 4 5 6 49
1963 11 2 4 3 6 26
1964 13 6 6 2 3 30
1965 12 5 3 8 29
1966 26 4 3 6 40
1967 19 1 5 26
Subtotal 81 18 16 8 28 151
1968 20 3 2 3 6 34
1969 8 2 3 2 16
1970 6 5 12
1971 14 3 3 20
1972 7 3 12
Subtotal 55 11 7 7 14 94
1973 4 1 7
1974 7 2 11
1975 4 2 6
1976 5 8
1977 18 1 20
Subtotal 38 4 4 0 6 52
1978 6 2 8
1979 8 2 13
1980 3 6
1981 10 11
1982 8 9
Subtotal 35 4 2 2 4 47
1983 7 10
1984 13 14
1985 5 2 9
1986 13 3 17
1987 17 1 3 21
Subtotal 55 6 2 6 2 71
1988 14 16
1989 5 5
Subtotal 19 0 0 21
Total 307 54 36 30 65 492
62.40OJo 10.98OJo 7.32OJo 6.10% 13.21OJo l00OJo
Clinical Results 567

Table 26.3

Follow-up year Clinical result at last follow-up Total

Excellent Very good Good Fair Poor

1 20 2 6 28
2 47 2 3 5 11 68
3 21 2 5 3 32
4 30 5 2 2 6 45
5 15 4 2 21
Subtotal 133 15 6 14 26 194
6 16 2 1 20
7 13 2 16
8 8 3 13
9 8 3 4 16
10 10 2 1 2 15
Subtotal 55 10 4 2 9 80
II 16 2 2 2 3 25
12 12 I 5 18
13 7 5 3 2 18
14 8 2 2 1 13
15 8 I 4 13
Subtotal 51 9 9 3 15 87
16 8 3 2 3 17
17 4 4 1 10
18 4 4
19 6 2 1 11
20 7 2 2 3 15
Subtotal 29 10 4 6 8 57
21 6 2 9
22 7 2 3 14
23 9 3 12
24 5 2 1 10
25 6 2 3 14
Subtotal 33 5 12 4 5 59
26 3 3 7
27 1
28 2
29 3
30 1 1
Subtotal 6 5 0 2 14
31 0
32 0
33 1
Subtotal 0 0 0 0
Total 307 54 36 30 65 492
62.400/0 10.98% 7.32% 6.10% 13.21% 100%

73.37%

80.69%
568 Clinical and Radiological Results of Operation Within Three Weeks of Injury

of the hip, and the total of the three digits is 15
or 16.
Intermediate or fair:
The hip is painful, activity is diminished, there
are radiological blemishes, but the joint is
nevertheless tolerable and the total of the three
digits is 13 or 14.
Poor:
- All other cases. The total of the three digits is 12
or less.
Table 26.2 shows the progressive yearly accumula-
tion of results. Table 26.3 shows the accumulated
results according to length of follow-up. Overall we
have the following results:
Excellent 62.4070
Very good 11.0070
Good 7.3070
Intermediate 6.1070
Poor 13.2070
100070
Thus, we have 62.4070 excellent results, i.e. normal
hips, and another 11070 of hips that are very good
clinically (although 72070 of them have stable
osteoarthritic changes), i.e. 73.4070 of hips that have
at least a very good clinical result. In the first edi-
tion of this book we reported 70.9070 very good
results. The only slight improvement in this figure
despite greater surgical experience can be explained
by two factors: (1) we have been referred more and
more complex cases; (2) some results have deterio-
rated with time as osteoarthritis set in, and
whatever time had elapsed since surgery, we feel
obliged to blame them on the original acetabular
fracture.
The overall results may be considered under the
following headings:
26.1.1 Type of Fracture
In Table 26.4 it can be seen that despite a high rate
of perfect reductions of posterior wall fractures
(94070; see Chap. 23), only 79.5070 achieved at least
a very good result (97 cases of 117). This is due to
the occurrence of osteonecrosis in this group, and
also to the fact that some of the reconstructions of
comminuted fractures were very difficult to per-
form.
Among the associated fractures, the worst group
according to the results of this review are the asso-
ciated posterior column and posterior wall frac-
tures: nine cases of 17 have a fair to bad result, due
to: one ectopic ossification despite a perfectly
reduced acetabulum, five cases of post-traumatic
osteoarthrosis, one necrosis of the femoral head
and two necroses of the head and the posterior
wall. This small series does not allow conclusions;
moreover, at the previous review of our cases this
group had no poor results. The other associated
fracture which pays a heavy toll in complications is
transverse and posterior wall.
Among the simple fractures the least satisfacto-
ry results occurred among the anterior wall frac-
tures (67070 excellent results, which can be explained
by the fact that these fractures often involved old
people with osteopenic bones).

Table 26.4

Type of fracture Clinical result Total Percentage of

excellent results
Excellent Very good Good Fair Poor

Posterior wall 87 6 3 4 17 117 741170

Posterior column 9 1 11 81.821170
Anterior wall 6 1 9 66.671170
Anterior column 12 2 16 75.001170
Transverse 17 1 1 19 89.471170
T·shaped 20 3 3 26 76.921170
Transverse and posterior wall 49 16 10 9 17 101 48.511170
Posterior column and posterior wall 5 1 2 1 8 17 29.411170
Anterior column and posterior hemitransverse 26 5 4 3 3 41 63.411170
Both-column 76 21 14 11 13 135 56.301170
Total 307 54 36 30 65 492 62.401170
62.401170 10.981170 7.321170 6.101170 13.211170 1001170
Radiological Results 569

Table 26.5

Clinical result Patient age at operation (years) Total

10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89

Excellent 28 88 78 50 35 23 4 307
Very good 2 5 12 12 15 5 3 54
Good 7 6 7 7 8 36
Fair 1 4 9 7 4 4 30
Poor 2 3 13 18 20 8 65
Total 34 107 118 94 81 48 9 492
Percentage of operated cases 6.91070 21.75070 23.98070 19.11070 16.46070 9.76070 1.83070 0.20070 100070
Percentage of excellent and 88.24070 86.92070 76.27070 65.96070 61.73070 58.33070 77.78070 100.00070
very good results

26.1.2 Age of Patient their degree, and many of them are stable and
clinically silent.
The grades of results according to the patients' ages In Chap. 25, reporting the late complications of
are shown in Table 26.5. The percentage of excellent surgery, we mention 97 cases of post-traumatic
and very good results decreases regularly with age osteoarthritis. This increase in the number of cases
except in the 70- to 79-year-old group, when it goes of osteoarthritis at the final evaluation of patients
up to the level of that of the 30- to 39-year-olds. is due to the fact that, with the passing of time,
The group is too small for the figures to be signifi-
cant, but this result should encourage surgeons not
Table 26.6
to abandon people aged over 70 to conservative
methods; furthermore, the only patient aged over Late radiological results
80 has an excellent result (see Fig. 19.7).
Perfect
Perfect 210 42.68070
26.2 Radiological Results Head collarette 1 - 2 mm 59 11.99070
Head collarette 3 - 4 mm 33 6.71070
Subtotal 302 61.38070
Post-operative radiological results have been dis- Osteonecrosis
cussed in Chap. 23, but numerous additional Femoral head 5 1.02070
features have had to be taken into account in the Femoral head and posterior acetabulum 2 0.41070
later assessment and it is difficult to produce a Subtotal 7 1.42070
comprehensive classification. These late radiologi- Osteoarthritis
cal features are listed in Table 26.6, while Table 26.7 Various causes 116 23.58070
Stable pre-existing osteoarthritis 0.20070
relates quality of radiological appearance at last Neglected incarcerated fragment 1 0.20070
evaluation to type of fracture. Deterioration of pre-existing 2 0.41070
These two tables report the final radiological ap- osteoarthritis
pearances of many cases which have been reported Subtotal 120 24.39070
and discussed previously in the chapters relating to Mal-union
the early and late post-operative complications. Severe 5 1.02070
The apparently low prevalence of osteonecrosis at Slight 16 3.25070
Subtotal 21 4.27070
last follow-up is due to the fact that in most of the
cases of osteonecrosis discussed in Chap. 25, osteo- Post-septic arthritis 8 1.63070
arthritic changes have been added to the initial im- Arthrodesis 3 0.61070
age and the cases are now in this group. The only Ectopic ossification with perfect 15 3.05070
ones which remain classified here as osteonecrosis acetabular reconstruction
are those which at the last examination still Wear of the head 2 0.41070
presented a typical picture of osteonecrosis. Surgical secondary congruence 14 2.85070
The number of cases of osteoarthritis is con- Total 492 100070
siderable (24070). It comprises all cases whatever
Table 26.7 v.
--J
0

Late radiological result Type of fracture Total

Posterior Posterior Anterior Anterior Transverse T-shaped Transverse and Posterior column Anterior column Both-column
wall column wall column posterior wall and posterior and posterior
wall hemitransverse

Perfect
Perfect 56 8 5 9 7 16 31 4 16 58 210
Head collarette 1 - 2 mm 20 7 2 6 2 7 12 59
Head collarette 3 - 4 mm 7 2 13 3 6 33
Subtotal 83 9 6 11 16 19 50 6 26 76 302
Osteonecrosis
Femoral head 3 2 5
Femoral head and posterior 2 2
acetabulum
Subtotal 0 0 0 0 0 0 3 4 0 0 7
()
Osteoarthritis S·
Various causes 28 2 2 2 4 34 5 10 29 116 ri·
Stable pre-existing osteo-
e.
Il>
arthritis ::>
0-
N eglected incarcerated ~
Il>
fragment
Deterioration of pre-existing 2
o·0-
0-
(JQ
osteoarthritis ri·
Subtotal 29 0 2 3 2 4 34 6 10 30 120 e.
Mal-union ~
Severe 1 5 ~
Slight 1 1 3 3 2 5 16 0
....,
Subtotal 2 2 3 3 3 6 21 0
'0
(1)
Post-septic or infectious 3 3 8 ....
arthrosis o·~
::>
Arthrodesis 2 3
~
Ectopic ossification with 3 5 6 15 :;-
perfect acetabular recon- S·
....,
struction
i'f
(1)
(1)
Wear of the head 2
Surgical secondary congruence 13 14 ~
(1)

i2
Total 117 11 9 16 19 26 101 17 41 135 492 0
....,
g
c....
'<
Radiological Results
some cases added arthritis to the radiological
features that they already had. Thus, it appears to
us that:
in 46 cases the cause of osteoarthritis is un-
known (38.30/0 of 120 cases);
in 46 cases the reason was an initially imperfect
surgical reduction (38.3 %);
in 5 cases an intra-articular screw was responsi-
ble;
in 14 cases avascular bone necrosis developed
initially;
in 5 cases cartilage necrosis developed during
the first months after operation;
in 1 case, early secondary displacement oc-
curred;
in 2 cases the femoral head was found to be
seriously damaged at operation;
in 1 case there was deterioration of pre-existing
osteoarthritis.
The difference between the total of 97 mentioned in
Chap. 25 and the final total of 120 is represented
essentially by osteoarthritic changes superimposed
on the initial 19 cases of necrosis of all types.
Cases of post-infectious arthrosis have been ex-
cluded from the previous group because the cause
is known without discussion, and because they
need to be separated out to remind us of the risks
this complication represents and the care that must
be taken to avoid them.
The arthrodeses are the result either of new
surgical procedures or of spontaneous arthrodesis
(1 case) after a surgical failure.
In 15 cases, significant ectopic bone formation
is present despite a nice acetabular reconstruction,
either affecting only the radiological appearance
or, if it decreases mobility, the clinical result as well.
Not all patients have accepted the suggested
removal of their ectopic bone. The results of the
removal are nine excellent or very good (60%), one
good and five fair.
Of the 25 cases in which we were only able to
achieve surgical secondary congruence, 14 are re-
ported here since they have exactly the appearance
we describe for surgical secondary congruence, ex-
cept for three who have clinically silent osteoar-
thritic changes. The other 11 cases, which had less
satisfactory results, are included in the other
groups according to the amount of degradation in
the condition of the joint after the immediate post-
operative condition (6 are good, 2 fair, and 3 bad).
In the analysis of results, cases which on
radiography show a stable collarette of osteophytes
571
around the head, whether 1-2 mm (C1) or
3 - 4 mm (C2) in thickness, will be included in the
perfect results. These images were studied in the
previous chapter, and although they may be a
threat for the long-term prognosis, we have seen
them stable for such long periods that we cannot
include them under osteoarthritis. Nevertheless,
they have been entered separately in Tables
26.6-26.8.
Table 26.8 shows that of 366 perfect intra-
operative reductions, 283 have achieved a perfect
radiological result (77.3 %). The remaining 83 cases
have had their initial perfect result impaired by the
development of some later radiological deteriora-
tion. However, we shall see later that these addi-
tional radiological features do not always com-
promise the clinical results: osteoarthrosis of types
I and II or even III may be stable and silent for
many years; ectopic ossifications may not reduce
mobility.
It can also be seen that of the 302 cases with
perfect radiological results (out of 492 operated
cases - 60.6%), 283 had perfect reductions
(93.2%); this is another way of confirming that an-
atomical reduction gives the best chances of an ex-
cellent clinical and radiological long-term result.
The other 19 perfect radiological results come
from imperfect reductions, such as ten with the
head centred, six with loss of parallelism and two
with surgical congruence, which looked so nice on
radiographs after so many years that we were
unable to class them as other than excellent. This
apparent paradox is due to the fact that the reduc-
tions were judged immediately post-operatively
before the healing of some persistent gaps had
obliterated the imperfection. After a period, the hip
with a parallel normal joint space, with no osteo-
phytes or very few, and without irregularity of the
density of the head or the acetabular bone, be-
comes classed as radiologically normal or of very
good quality even if a slight alteration persists in
the outline of certain landmarks.
Table 26.9 compares the clinical long-term
results and the late radiological appearance. Out of
302 perfect radiological results (in 92 of which
there is a collarette of osteophytes around the
head), 293 correspond to an excellent clinical result.
Five other cases are very good clinically, with a
slight radiological abnormality, and could probably
be added to the 293. Thus, a very good clinical
result corresponds to a perfect radiographic ap-
pearance in 98.6% of cases. The other four cases
classed as perfect radiologically are only "good"
572 Clinical and Radiological Results of Operation Within Three Weeks of Injury

Table 26.8

Late radiological result Quality of intra-operative reduction Total

Perfect Head Loss of Residual central Technical Surgical secondary

centred parallelism head protrusion failure congruence

Perfect
Perfect 200 7 2 210
Head collarette 1 - 2 mm 55 1 3 59
Head collarette 3 - 4 mm 28 2 1 33
Subtotal 283 10 6 0 2 302
Osteonecrosis
Femoral head 4 5
Femoral head and posterior 2
acetabulum
Osteoarthritis
Various causes 58 19 19 9 4 7 116
Stable pre-existing osteoar-
thritis
Neglected incarcerated frag-
ment
Deterioration of pre-existing 2 2
osteoarthritis
Mal-union
Severe 1 2 2 5
Slight 9 6 16
Post-infectious arthrosis 4 2 8
Arthrodesis 2 3
Ectopic ossification with 14 15
perfect acetabular recon-
struction
Wear of the head 2
Surgical secondary congruence 14 14
Total 366 46 32 13 8 27 492

clinically because of the sciatic nerve palsy in one osteoarthritis in a fourth case remained stable. The
case, and the clinical rating in the other three 116 other cases of osteoarthritis have various
(5.5.5), which is without radiological explana- causes, but in some the cause is not apparent; it is
tion. possible that a slight error in reduction, not ob-
Of 307 excellent clinical results, 293 (95070) have vious on radiographs, is the responsible agent.
a perfect radiological appearance (87 have a "col- Regular post-operative CT scanning will help in
larette"); the other 14 cases have some radiological more accurate assessment of reductions.
blemishes: one pre-existing stable osteoarthritis, The overall clinical results in these 116 cases of
seven slight mal-unions, five ectopic ossifications radiological osteoarthritis ended up as:
and one surgical secondary congruence.
Thble 26.9 shows other interesting points, as 25 very good 21.5%
follows: the seven cases of osteonecrosis account 24 good 20.7%
for one good, one fair and five bad results. 17 intermediate 14.6%
Osteoarthritic changes of various types were 50 poor 43.10%
seen in 120 cases. In two cases there was pre-
existing osteoarthritis which deteriorated after the Most of the cases with poor results have already
fracture treatment, a missed incarcerated fragment been treated with a total hip arthroplasty. Arthro-
was the cause in a third case, and pre-existing plasties of the hip, whatever their type, make no ap-
Late Overall Clinical Results and Quality of Reduction 573

Table 26.9

Late radiological result Clinical result Total

Excellent Very good Good Fair Poor

Perfect
Perfect 206 2 2 210
Head collarette 1 - 2 mm 57 59
Head collarette 3 - 4 mm 30 2 33
Subtotal 293 5 4 0 0 302
Osteonecrosis
Femoral head 3 5
Femoral head and posterior acetabulum 2 2
Subtotal 0 0 5 7
Osteoarthritis
Various causes 25 24 17 50 116
Stable pre-existing osteoarthritis
Neglected incarcerated fragment
Deterioration of pre-existing osteoarthritis 1 2
Subtotal 26 25 17 51 120
Mal-union
Severe 2 2 5
Slight 7 7 16
Post-infectious arthrosis 4 4 8
Arthrodesis 3 3
Ectopic ossification with perfect acetabular 5 4 5 15
reconstruction
Wear of the head 2 2
Surgical secondary congruence 11 2 14
Total 307 54 36 30 65 492

Table 26.10

Quality of intra-operative reduction Late clinical result Total Percentage of all

operated cases
Excellent Very good Good Fair Poor

Perfect reduction 283 18 15 17 33 366 74.390/0

Head centred 15 10 6 5 10 46 9.39%
Loss of parallelism 6 12 4 2 8 32 6.50%
Residual central head protrusion 0 3 3 3 4 13 2.64%
Technical failure 0 0 2 1 7 10 2.03%
Surgical secondary congruence 3 11 6 2 3 25 5.08%
Total 307 54 36 30 65 492 100.00%

pearance in this book, except for some reconstruc- the files as closed and the outcome of the case as
tions performed later than 4 months after injury. poor.
We consider that an arthroplasty of any kind -
hemi-arthroplasty, total arthroplasty, whatever,
cemented or not - is not a treatment of an acetab- 26.3 Late Overall Clinical Results
ular fracture, at least not in the group operated on and Quality of Reduction
within 3 weeks of injury. For treatment of acetab-
ular fractures, as soon as the evolution of the case The overall results are related to the quality of
obliges us to insert a hip prosthesis, we consider reduction in Table 26.10.
574 Clinical and Radiological Results of Operation Within Three Weeks of Injury
26.3.1 Perfect Reductions
Out of 366 patients with apparently perfect
reductions of the acetabulum and of the in-
nominate bone as a whole, and consequently
also of the femoral head (i.e. anatomically
perfect reductions), we have 283 (77.3070) with
functionally excellent clinical results, i.e. normal
hips. In 18 other cases (5%) the results are also
functionally very good, but in 12 of them there
are some clinically silent radiological signs of
osteoarthritis which have been stable for years.
So, overall 82.3% of the perfect reductions have
led to a very good overall result today.
15 cases have a functionally good result (4.1 %).
One patient had localised head necrosis and two
have only a collarette of osteophytes (1 of
1 - 2 mm, 1 of 3 - 4 mm); there is no apparent
explanation for the relatively poor result. In 13
cases there are radiological signs of osteoar-
thritis with related pain, reduction of mobility
or impairment of gait.
17 patients have results graded fair (4.6%). Of
these, 5 patients have developed ectopic ossifica-
tions which reduce mobility. Two patients have
sequelae of infectious arthritis. Ten patients
have developed unexplained osteoarthritis ap-
parently responsible for clinical functional im-
pairment (in one case it was secondary to
femoral head necrosis).
33 patients have bad results (9.01 %), which are,
however, treatable by total hip replacement.
Most of them have been operated upon, but it
is our rule not to report those results here. The
reasons for these bad results were dealt within
the previous chapter: they are osteoarthritis of
unknown cause, sequelae of infectious arthritis,
and necrosis of the femoral head or of the head
and the acetabular wall.
26.3.2 Imperfect Reductions
We have been stringent in the assessment of the
quality of our operative reductions. There is no
question that the quality of reduction is incom-
parably superior in the vast majority of cases to
that which could be obtained by conservative
management even at its best. That conservative
treatment can result in good or even very good
clinical results, in spite of an imperfect reduction
and a sometimes astonishing radiographic ap-
pearance, cannot be denied. This explains without
a doubt why less than perfect surgical reductions
can also furnish good or very good clinical results.
It is our opinion that the reason why imperfect
reductions after conservative or operative treat-
ment may lead to very good or good functional
results is because the femoral head remains con-
gruent to a segment of the articular surface large
enough to sustain for a long time the significant in-
crease in intra-articular pressure in some positions
of the hip. Our problem today is that we have no
means of measuring post-traumatic or post-
surgical articular congruence. We have already said
that in all probability 3-D CT scanning will allow
these measurements. We must also recognise that
the fractured acetabulum has, in some cases, in-
credible abilities of remodelling, which may explain
the surprising good functional results after unsatis-
factory reductions.
For this review of cases, imperfect reductions
have been grouped under five headings, according
to which the results will now be analysed in detail.
(a) Recentring of the femoral head under the roof.
In 46 cases (9.3%) we succeeded in restoring the
centring and stabilising of the femoral head
under the anatomical roof, some parts of the
acetabular surface being unreduced and no lon-
ger congruent with the femoral head. Fifteen
cases achieved an excellent result (32.6%) and
10 others a very good result, with some radio-
logical signs of osteoarthritis in eight of them.
These 25 very good results represent 54.3% of
the cases with recentred heads and an imperfect
acetabular reconstruction. This figure demon-
strates that recentring the head must never be
considered a satisfactory surgical result even if
in the long term it gives some very good results,
for it falls far short of the 82.3% very good
results afforded by perfect reconstruction of the
acetabulum.
The same kind of surgical result led to six
good results (13% of this group), explained in
five cases by post-traumatic osteoarthritis im-
pairing the function, and in one by discrete
mal-union without, at the time of the review,
osteoarthritic changes.
Five cases have an intermediate result
(10.9% of this group), due to sequelae of septic
arthritis in one case, significant mal-union after
a secondary displacement in one, and osteoar-
thritis in three.
Ten patients have a poor result (21.8% of
this group), due to spontaneous ankylosis in
Late Overall Clinical Results and Quality of Reduction
one case, sequelae of infectious arthritis in one,
osteoarthrosis in four, necrosis of the head in
three, associated necrosis of head and posterior
wall in one.
(b) Loss of parallelism of the upper joint space.
This was present on the post-operative radio-
graphs in 32 cases (6.5070), the reason being an
imperfection in the reconstruction of the ace-
tabulum. Six patients (18.7%) achieved an ex-
cellent radiological and clinical result, and 12
others (37.5%) have a very good overall result,
but nine of them have radiological signs of
osteoarthritis. That is all together 18 cases
(56%) with very good overall results. The rate
of excellent results here is much lower than
among the recentred-head group, but the rate of
very good results is a little higher; however, the
group is smaller. It may perhaps be concluded
that a minor incongruence of the upper joint
space is not a serious disadvantage in relation to
the previous group. Furthermore, we have occa-
sionally been surprised to find this alteration of
joint space disappearing during the later fol-
low-up period, which explains the cases which
ended up with a perfect radiographic appear-
ance. These remodellings, which are difficult to
label when they accompany a very good func-
tional result, are exemplified in the following
two cases.
Figure 26.1 A shows the imperfect recon-
struction of a transverse and posterior wall
fracture associated with a posterior dislocation,
performed 18 days after injury. Figure 26B
shows the radiographic appearance 6 years after
surgery, with a very good functional result, and
the joint space has recovered a better parallel-
ism. Figure 26.1 C shows the appearance 25
years after surgery. The hip is still functionally
very good. The remodelling has affected both
sides of the joint, the acetabulum and the head,
as is demonstrated by the superposition of the
femoral head on the other side, and on the
whole, the joint space seems to be satisfactory.
Figure 26.2A shows a T-shaped fracture im-
perfectly reduced (Fig. 26.2 B) in spite of two
consecutive approaches. The imperfection is
difficult to classify: either a loss of parallelism
or a residual head protrusion. The appearance
2 years after surgery and 2 months after the
removal of an ectopic bone formation type IV-3
is shown in Fig. 26.2 C, while Fig. 26.2 D shows
the appearance 14 years after surgery and the
575
considerable remodelling, which is probably
responsible for the very good function. Fig-
ure 23.1 also demonstrates remodelling after an
imperfect reduction leaving a loss of parallel-
ism. With time, the loss of parallelism is filled
up with new bone formation and a new line ap-
pears parallel to the femoral head. We have not
yet had the opportunity to assess this in vivo.
Apart from these cases, loss of parallelism
of the upper joint space led to:
4 good results, due in all 4 cases to osteoar-
thritic changes;
2 intermediate results, with more advanced
osteoarthritis;
8 poor results requiring total hip replace-
ment.
(c) Femoral head protrusion. In 13 cases (2.6%), in
spite of our efforts, the post-operative radio-
graphs showed an indisputable degree of femo-
ral head protrusion. Not one case achieved an
excellent result. However, three cases (23 %)
yielded very good results in the long term. The
explanation for this is not clear; we can suggest
possible remodelling, or possibly sufficient re-
maining congruence between the head and
some part of the acetabulum to stand up to the
intra-articular pressure, which is impossible to
assess precisely and which could facilitate the
remodelling. Two of these patients have osteo-
arthritic changes.
Three patients have fair results; one is the
consequence of an infectious arthritis, the two
others of osteoarthritis.
In four cases the results are poor (fair and
poor represent 53% of this group), due to four
cases of significant post-traumatic osteoarthro-
sis.
(d) Technical failures. In ten cases we were unable
to perform the reconstruction: these are techni-
cal failures (2%). Not one case achieved a very
good functional result. Two cases have good
results, one with a significant mal-union with-
out arthritis (Fig. 26.3) and the other develop-
ing arthritic changes at 6 years.
Figure 26.3 shows an anterior column frac-
ture approached through an ilio-femoral
approach and not reduced in 1964 (Fig. 26.3B).
Figure 26.3C shows the appearance 14 years
later: the functional result is only good but no
arthrosis has set in. The amount of post-trau-
matic congruence was probably sufficient to
 576 Clinical and Radiological Results of Operation Within Three Weeks of Injury

A B

Fig. 26.1 A - C Imperfect reduction of an associated transverse

and posterior wall fracture (Kocher-Langenbeck approach).
A Antero-posterior radiograph 18 days after injury, B appear-
ance 6 years after operation, C appearance 25 years after opera-
c tion. The hip is rated 6.6.6 (see text)
 Late Overall Clinical Results and Quality of Reduction 577

A B

C D

Fig.26.2A-D. T-shaped fracture. A Antero-posterior radio- (type IV-4) and one plate, D appearance 14 years after operation.
graph before operation, B reduction was imperfect despite the Note the remodelling. The hip is rated 5.6.6 and has excellent
use of two consecutive approaches, C appearance 2 years after mobility
operation, 2 months after the removal of ectopic ossification
 578 Clinical and Radiological Results of Operation Within Three Weeks of Injury

A B

Fig.26.3A-C Anterior column fracture. A Antero-posterior

radiograph before operation, B the fracture was very badly
reduced (in 1964) through an ilio-femoral approach, C appear-
c ance 14 years later: there is no arthritis and the hip is rated 5.6.5
 Late Overall Clinical Results and Quality of Reduction 579

A B

Fig. 26.4. Both-column fracture. A Antero-posterior radiograph before operation, B appearance 6 years after operation: a very good
(if not excellent) result of surgical congruence

allow nearly normal function of the joint, with
only a slight increase in intra-articular pressure.
Additionally there are one intermediate re-
sult due to mal-union and seven poor results.
(e) Surgical congruence. In 25 cases we were unable
to reconstruct the acetabulum and the in-
nominate bone together, but we succeeded in
grouping the fragments of the articular surface
around a more or less centrally displaced head,
creating what we called surgical secondary con-
gruence.
In three cases the result of this is excellent;
apart from a slight central displacement of the
head and the acetabulum, the radiological and
functional results cannot be classed as other
than excellent (Fig. 26.4).
In 11 other cases the overall results are very
good. That means that altogether 14 of these 25
cases with surgical secondary congruence
achieved a very good result (56070).
Six patients (24%) achieved a good result:
one of these has a collarette of osteophytes
(3 -4 mm) and is functionally rated 5.6.5; three
have some signs of osteoarthritis; one has a
type II ectopic ossification, which does not ex-
plain the reduced hip mobility (rated 6.4.5); and
one has an explained loss of mobility (rated
6.4.5) with a type III ectopic ossification. Two
patients have an intermediate result, one due to
osteoarthritis and one unexplained, but func-
tionally rated 5.6.4.
Three had a poor result and underwent total
hip arthroplasty.
Overall, these patients with surgical second-
ary congruence ended up with 56% very good
functional results and 24% good results.
26.3.3 Conclusions
Despite the small numbers concerned in some
series, there is evidence that the main aim should be
to obtain a perfect fit of the head into the
acetabulum restored to its normal place, with which
82.3% very good results can be expected in the long
term. If an imperfection of reduction persists while
580 Clinical and Radiological Results of Operation Within Three Weeks of Injury
the head remains perfectly centred and stable under
the roof, the percentage of very good results drops
to 43.5070. If the imperfection of the acetabulum
generates a loss of the upper joint space, the rate of
very good results is 56%, but the series on which
this figure is based is small. If, however, any impair-
ment of quality of acetabular reconstruction coex-
ists with a protrusio acetabuli of the femoral head,
the clinical result is unlikely to be good; never-
theless, three cases out of 13 (26%) achieved a very
good functional result not always easy to explain.
Of the ten technical failures not one achieved a
very good result.
It is interesting to point out that of 25 cases with
surgical secondary congruence, 14 (56%) achieved
a very good result and six others (24%) a good re-
sult. Without going so far as to advise this as a goal
to aim at, creating surgical congruence is a possible
way to get out of surgical difficulty.
26.4 Summary of Results
26.4.1 Early Results
Of 569 fractures of the acetabulum which were
operated upon during the first 3 weeks following
injury, 418 had perfect reductions of both the
acetabulum and the innominate bone as a whole
(73.4%). It should be noted that the faults of
reduction in operative treatment are, in the im-
mense majority of cases, less significant than those
resulting from conservative treatment, and also
that a good proportion of them have been followed
by a very good or good clinical result. At the same
time, failure to achieve perfection must not be ac-
cepted complacently, especially when put into con-
text with the undoubted risks of operation. These
have been discussed in Chap. 24 but will be men-
tioned again briefly:
(a) Mortality (2.3%). Of the 13 post-operative
deaths, two were directly related to the opera-
tion on the hip and 11 were due to general com-
plications of surgery.
(b) Post-operative infection (4.2%). This was unac-
ceptably frequent during our early experience,
especially using the ilio-inguinal approach. The
rate of infection has fallen during the later part
of this series. Early vigorous treatment is im-
portant, and while there has been significant
functional loss in nine hips, all the other cases
have achieved good or very good results (13
cases).
(c) Post-operative sciatic palsy (6.3%). Variable in
distribution and severity, onset was immediate
in 34 cases and delayed in two cases. Before the
utilisation of transcondylar femoral traction
with the knee flexed, the incidence was 18.42%.
Subsequently the incidence has been 3.3%; the
palsy is in most cases due to compression or
stretching of the nerve with retractors and is
therefore avoidable. The prognosis is fairly
good: out of 34 cases with motor impairment
there were nine complete recoveries and 12 very
significant improvements. Thus, two cases out
of three improve sufficiently to allow normal
activity. Three patients need an appliance.
Besides the above, surgery has failed to fulfill its
aim in 26.6% of the cases: an imperfection which
varied considerably in degree:
in 10.7% of all cases the head was recentred and
stable under the roof while a part of the acetab-
ulum remained displaced;
in 6% of all cases there exists a loss of parallel-
ism of the upper joint space due to a persistent
slight central displacement of the femoral head;
in 3.4% of all cases a protrusion of the femoral
head persisted;
in 2.1 % of all cases we failed totally: these are
the "technical failures";
in 4.6% of cases surgical secondary congruence
was achieved.
26.4.2 Late Results
The results in the 492 hips followed up for at least
2 years are summarised as follows:
Clinically, 73.4% have at least a very good result
(80.3% for fractures of the posterior wall and
71.8% for both-column fractures), 7.3% have good
results, 6.1 % have intermediate results, and 13.2%
have poor results. Radiologically, 302 hips are
perfect, representing 61 % of our operations.
The radiologically normal hips correspond well
to clinically very good results and it is our conten-
tion that early operation leads to a three-out-of-
four chance of such a good outcome.
Late complications must not be ignored:
Pseudarthrosis (4 cases).
Avascular bone necrosis of all types (22 cases or
4.5%), including 19 cases of femoral head ne-
crosis. Three instances of osteonecrosis of the
acetabulum were associated with necrosis of the
Summary of Results
femoral head. In addition to these, there were 6
cases of what was considered to be pure carti-
lage necrosis.
Post-traumatic osteoarthritis (97 cases or
19.7070). Of these cases, some were purely radio-
logical and asymptomatic, while the others were
symptomatic. In 10.2%, the osteoarthritis ap-
peared after perfect reduction of the head and
of the acetabulum, and in 35.7%, it appeared
after imperfect reduction: a number of the latter
could perhaps have been avoided. The rate of
osteoarthritis is 19.7%, comprising both the
purely radiological cases with a good or very
good clinical result and the clinically symp-
tomatic. Between our reviews of 1978 and 1990
the rate increased from 8.12% to 19.7%, but
some "necroses" of 1978 were considered as
osteoarthritis in 1990, and these figures relate to
cases followed up over more than 30 years.
Ectopic bone formation (139 cases). This num-
ber includes the 22 severe cases, comprising the
14 which required surgical removal and even-
tually achieved very good or good results and
the 8 awaiting removal.
Poor and intermediate results are in most instances
related to one of the complications, whether early
(such as infection) or late (osteoarthritis and osteo-
necrosis), perhaps related to inferior reduction of
the joint surface. Forty-six hips with the head
recentred under the roof yielded very good results
in 25 cases.
The persistence of a loss of congruence of the
upper joint space, even when associated with im-
perfection of the acetabular reduction, does not
seem to have such a deleterious effect as might be
expected, as 18 out of 32 cases were very good
clinically. By contrast, if the fault of reduction
581
caused a medial protrusion of the head, the results
over the long term were very uncertain. A persisting
central protrusion was associated with only three
very good clinical results and three good results
(out of 13 cases).
Ten cases of seriously deficient reduction were
associated with two good results, so one in five
marked failures of reduction paradoxically ended
with a good clinical outcome.
In 25 cases we were able to restore surgical sec-
ondary congruence with a mal-united innominate
bone; in 14 cases a very good result was achieved,
and in six a good result.
26.5 Conclusions
One single factor appears paramount: the reloca-
tion of the head under a sector of roof of sufficient
extent must be adequate. This is the practical prere-
quisite for all good results. However, it must not be
taken that obtaining this result obviates the need
for good reduction of the columns supporting the
acetabulum.
26.6 Comment
Surgery of acetabular fractures should not be
undertaken lightly. Full study of the diagnostic
methods and operative techniques demands much
time and effort, and without adequate preparation
in this way it is probably better to pursue a conser-
vative course or to refer the patient to a centre
where more experience has been gathered.
27 Reassessment of Patients Treated Operatively
Within Three Weeks of Injury
Over 33 years we have tried to assess the overall
results of our patients as regularly as possible -
for many of them, nearly every other year. Never-
theless, as all over the world, I suppose, while some
patients come regularly, others forget, and too
many of them are lost and become impossible to
trace. It is strange and difficult to understand why
French national institutions, such as the Social
Security system, totally refuse to help in finding pa-
tients for follow-up, although they would most
probably be able to trace them.
At five different times, and each time over a
period of 2 or 3 years, we have tried to examine
systematically all operated patients with at least 1
year of follow-up, and published our statistics in
1966, 1968, 1971 and 1978. From 1987 to 1990 we
attempted to trace all patients operated upon; their
results were reported in Chap. 26.
In order to try to assess the stability of the
surgical results, we have, with the help of the com-
puter, studied separately the outcomes of four
groups of patients at each of the successive follow-
ups:
Those who in 1966 had more than 1 year of
follow-up: 111 patients.
Those operated on between 1966 and 1971: 110
patients.
Those operated on between 1971 and 1978: 87
patients.
Finally, the 170 new patients operated on be-
tween 1978 and 1990.
This makes a total of 478 patients, therefore (some
patients operated upon towards the end of each
review period are not included here, as they did not
have a full year's follow-up at closure of the statis-
tics, but they are included in the overall statistics in
Chap. 26).
27.1 Evolution in Patients Operated
on Before 1966
In 1966 111 patients with follow-up for more than
1 year were evaluated. These were the results:
63 were rated excellent;
20 were very good clinically but 5 of these cases
already had some radiological signs of osteoar-
thritis;
11 were good;
13 were fair;
4 were poor.
We shall follow the evolution of these five groups
of cases (Fig. 27.1).
(a) Of the 63 excellent results of 1966 (Fig. 27.1 A):
In 1971:
44 patients still had an excellent result and
15 had been lost or had died with an ex-
cellent result and a mean follow-up period
of 4.4 years: a total of 93070.
4 results had deteriorated (7%):
1 to fair (deteriorated further to poor in
1990);
3 to very good, purely radiological signs
of osteoarthritis having developed (in
1990 one was good, one fair, one iden-
tical).
In 1978:
Of the 44 excellent results of 1971:
33 remained excellent and 6 were lost at ex-
cellent (1 died at 13 years and 5 lost with a
mean follow-up of 5.9 years): a total of
88%.
5 (11.3 %) had deteriorated:
 584 Reassessment of Patients Treated Operatively Within Three Weeks of Injury

1990

1978

1971

1966

o 10 20 30 40 50 60 70
• Excellent c::::J Very good fEl] Good
IIIIIID Fair • Poor c::::J Lost excellent

1990 1978 1971 1966

Lost excellent 9 6 15 0
Poor 2 1 0 0
Fair 0 1 1 0
Good 1 1 0 0
Very ~OOd 1 2 3 0
Excel ent 20 33 44 63
A

1990

1978

197 1

1966

o 5 10 15 20

• Very good c:::J Good • Lost very good

1990 1978 197 1 1966

o
8
I Lost very good
Good
Very good
5
2
8 I
2
15
o
I
2
1
17 I
o
20

1990

1978

197 1

1966

o 2 4 6 8 10 12
_ Excellent c::J Very good IITITl Good
_Fair • Poor c::::J Lost good

1990 1978 1971 1966

Lost good 2 1 1 0 Fig. 27.1 A-C. Successive assessments of patients
Poor 1 0 0 0 operated on before 1966 and with at least I year's
Fair 0 0 1 0 follow-up . Evolution from 1966 to 1990 of the
Good 3 7 8 11
very~ood 1 0 0 0 A 63 excellent results, 8 20 very good results, and
Excel ent 0 0 1 0 C 11 good results recorded in 1966
c
Evolution in Patients Operated on Before 1966
2 to very good, purely radiological signs
of osteoarthritis having developed (in
1990 one had a good result, one a fair);
1 result was good (became poor in 1990);
1 was fair (patient died at 13 years fol-
low-up);
1 was poor (16 years post-operatively).
In 1990:
Of the 33 excellent cases of 1978:
20 were still excellent and 9 were lost at ex-
cellent with a mean follow-up of 16 years: a
total of 88070.
4 results had deteriorated:
1 patient had developed purely radiolog-
ical osteoarthritis; his result was rated
very good;
1 result was good at 25 years;
2 were poor at 25 years.
Overall, of the 63 patients rated excellent in
1966:
In 1990:
20 still had an excellent result;
30 had been lost or died at excellent;
13 had deteriorated (20%):
2 developed purely radiological osteoar-
thritis and were rated very good;
3 deteriorated to good;
3 deteriorated to fair;
5 deteriorated to poor.
(b) Of the 20 patients rated clinically very good in
1966 (Fig. 27.1 B):
In 1971:
2 had been lost and still rated very good
at 8 and 6 years, respectively;
17 were still rated clinically very good;
1 had deteriorated to good and died at
10 years.
In 1978:
Of the 17 cases rated very good in 1971:
2 were lost with the identical rating at 9 and
4 years;
15 remained at the identical rating: a total
of 100%, of which 6 cases had osteoar-
thritic changes.
In 1990:
Of the 15 cases still rated clinically very good
in 1978:
585
8 were still rated identically;
5 were lost at the same grading with a mean
follow-up of 15.4 years;
2 had deteriorated to good (one died at 21
years).
Overall over 24 years, 7 cases out of 20 had
deteriorated (35%).
(c) Of the 11 results rated good in 1966
(Fig. 27.1 C):
In 1971:
8 were still good and 1 patient was lost at
good at 6 years follow-up: a total of 81 %;
1 patient deteriorated to fair and died at 7
years;
1 patient improved to excellent and died at
13 years.
In 1978:
Of the 8 results still good in 1971:
- 7 were rated identically;
- 1 patient was lost at good at 13 years.
In 1990:
Of the 7 good results in 1978:
3 were still good (mean follow-up 24 years)
and 2 died rated good at 22 years and 15
years: a total of 71 %;
1 had deteriorated to poor at 13 years;
1 had improved to very good and remained
so at 25 years.
Overall:
- 2 results deteriorated: 18%;
- 2 results improved: 18%.
(d) Of the 13 fair results of 1966:
In 1971:
6 were still rated the same and 4 had been
lost at fair with a mean follow-up of 4.9
years: a total of 77%;
1 deteriorated to poor at 4 years;
2 improved to good.
In 1978:
Of the 6 fair results of 1971:
2 cases were lost at 8 and 5 years, and 4 re-
tained the same rating (including the 2 im-
proved cases of 1971);
2 cases deteriorated to poor at 13 and 19
years.
586 Reassessment of Patients Treated Operatively Within Three Weeks of Injury

In 1990: In 1978:
Of the 4 fair results of 1978: 6 remained at and 6 were lost at the same
- 2 cases were lost at fair at 16 and 19 years; rating (1 died at 3 years, 5 were lost with a
- 2 deteriorated to poor at 29 and 15 years. mean follow-up of 5.4 years): a total of
80%.
(e) Of the 4 poor results in 1966, 1 improved to fair
2 deteriorated:
in 1971 and remained so up to 1990.
1 went to good at 13 years;
- 1 went to poor at 9 years.

In 1990:
27.2 Evolution in Patients Operated Of the remaining 6 results rated clinically very
on 1966 -1971 good in 1978:
5 were lost (2 died at 13 years, and 3 were
In 1971, 110 patients operated on between 1966 and lost, 1 at 9 and 2 at 13 years);
1971 were evaluated. Of their results: 1 deteriorated to fair at 22 years.
75 were rated excellent;
14 were very good clinically, but 4 of these pa- (c) Of the 10 results rated good in 1971
tients had purely radiological signs of osteoar- (Fig. 27.2C):
thritis; In 1978:
10 were good; 6 were still good and 2 were lost at good
6 were fair; (mean follow-up 2.5 years): a total of 80%;
5 were poor. 2 had deteriorated;
(a) Of the 75 excellent results of 1971 (Fig. 27.2A): 1 to fair at 9 years;
- 1 to poor at 12 years.
In 1978:
45 remained excellent, 27 had been lost at In 1990:
excellent (4 patients died with a mean
Of the 6 good results in 1978:
follow-up of 5.4 years; 23 were lost to fur-
4 were still good;
ther follow-up with a mean follow-up of 4.2
2 had deteriorated:
years): a total of 96070.
3 had deteriorated: - 1 to fair at 9 years;
2 became poor at 9 and 12 years; - 1 to poor at 12 years;
1 patient had developed purely radiolog- (d) Of the 6 cases rated fair in 1971:
ical osteoarthritis and was rated very
good. In 1978:
In 1990: 1 was still fair;
Of the 45 excellent results in 1978: 4 had been lost (2 patients died with a mean
follow-up of 3 years, 2 were lost with a
28 remained excellent and 13 had been lost
mean follow-up of 3 years);
at excellent (3 died with a mean follow-up of
1 had deteriorated to poor.
13 years; 10 were lost to further follow-up
with a mean follow-up of 10.5 years): a total In 1990:
of 91 %. One result was still fair.
4 had deteriorated: (e) Of the 5 results rated poor in 1971:
2 patients to very good as they had de- In 1978:
veloped purely radiological osteoar-
thritis; - 3 had been lost (mean follow-up 3.2 years);
1 to good at 25 years; - 2 were still poor.
1 to poor at 20 years. In 1990:
(b) Of the 14 clinically very good results of 1971 1 case was poor but still tolerable;
(Fig. 27.2B): 1 case was lost at poor at 10 years.
Evolution in Patients Operated on 1966-1971 587

1990

1978

1971

o 20 40 60 80
_ Excellent EJ Very good f]']] Good
_ Poor o Lost excellent

1990 1978 1971

Lost excellent 13 27
Poo r 1 2
Good 1 0
Very~d 2 1
Excel ent 28 45 75 A

1990

1978

1971

o 2 4 6 8 10 12 14
_ Very good CJ Good lillill Fa ir
~ Poo r _ Lost very good

1990 1978 197 1

Lost very good 4 6
Poor
,
I
Fair 1
Good
Very good 0 6 14 B

1990

1978

197 1

o 2 4 6 8 10
_ Good o Fair mE Poo r _ Lost good

1990 1978 197 1

Fig. 27.2A-C. Successive assessments of patients Lost good 0 2
operated on 1966 - 1971. Evolution from 1971 to Poo r 1 1
Fai r I 1
1990 of the A 75 excellent results, B 14 very good Good 4 6 10
results, and C 10 good results recorded in 1971 C
588 Reassessment of Patients Treated Operatively Within Three Weeks of Injury

27.3 Evolution in Patients Operated (e) Of the 4 poor results of 1978:

on 1971-1978 In 1990:
2 patients had died from other causes at 7
In 1978, 87 patients operated on between 1971 and and 10 years;
1978 were evaluated. Of their results: 2 patients had had a total hip replacement.
57 were excellent;
18 were very good, but in 10 of these patients
there were purely radiological signs of osteoar- 27.4 Assessment of Patients Operated
thritis; on 1978 -1990
6 were good;
2 were fair;
Of 170 patients operated upon from 1978 to 1990,
4 were poor.
37 were either lost before the end of 1 year or are
recent, so 133 patients with more than 1 year of
(a) Of the 57 excellent results of 1978 (Fig. 27.3 A):
follow-up were assessed in 1990.
In 1990:
105 results were rated excellent (78.5%), in·
35 were still excellent and 15 had been lost
cluding 2 patients who died at excellent at 1 and
at excellent (3 died with a mean follow-up of
3 years, and 2 who were lost at excellent at 2
8 years, 12 were lost at a mean follow-up of
years.
7 years);
11 cases were very good clinically (8%; 1 died at
7 had deteriorated:
2 years, 10 were examined).
2 patients had developed radiological
- 4 cases were very good: 3%;
osteoarthritis at 16 years and were rated
- 8 cases were fair (2 heterotopic ossifications,
very good;
one intra-articular screw, 1 non-union, 4
3 went to good at 11, 17 and 19 years;
osteoarthritis);
2 went to poor at 11 and 17 years.
5 cases were poor (4 osteoarthritis, 3 of
which already have a total hip replacement).
(b) Of the 18 very good results of 1978
(Fig. 27.3 B):
In 1990:
6 were still rated the same at 11 and 17 years
27.5 Longitudinal Assessment of
and 1 was lost at very good at 11 years: a
total of 40070; All Excellent or Very Good Results
11 cases had deteriorated:
4 to good at 13-14 years; To try to appreciate in another way, and as precisely
3 to fair as possible, the stability of our results, each
4 to poor at 15 years; operated case was considered according to its first
overall grading and the period over which it re-
(c) Of the 6 good results of 1978 (Fig. 27.3 C): mained with the same result; if it deteriorated it was
In 1990: abandoned.
For instance, a case rated excellent initially and
3 were still the same (50%; 1 examined, 2
remaining so for 20 years is reported in our yearly
lost at good at 12 and 14 years);
statistic for 20 years, and if then it turned to good
3 had changed:
or poor, its further conditions are not considered in
1 had deteriorated to poor at 2 years;
this section.
- 2 had improved to very good with purely
We studied how long each case regarded as hav-
radiological osteoarthritis.
ing an excellent or very good result at its first
(d) Of the 2 fair results of 1978: overall grading continued to have the same result,
or, to put it another way, how long this initially ex-
In 1990: cellent or very good result lasted before
1 had improved to very good; deteriorating. Table 27.1 shows what we found. For
- 1 had deteriorated to poor at 18 years. each year of follow-up is shown the number of
Assessment of Patients Operated on 1978-1990 589

1990

1978

o 10 20 30 40 50 60
_ Excellent o Very good INill Good
IIIIiIIIIIIII Poor o Lost excellent

1990 1978
Lost excellent 15
Poor 2
Good 3
Very ~OOd 2
Excel ent 35 57 A

1990

1978
,
o 5 10 15 20
_ Very good c:::J Good EIJ Fair
IIIIiIIIIIIII Poor _ Lost very good

1990 1978
Lost very good 1
Poor 4
Fair 3
Good 4
Very good 6 18
B

1990

1978
;::::;::1
_ Good
o 2

I!IEl
3
Poor
4

_
5 6
Improved VG

Fig. 27.3 A-C. Successive assessments of patients

operated on 1971-1978. Evolution from 1978 to
1990 of the A 57 excellent results, B 18 very good
IImproved VG
1990
2
1978

IC
Poor 1
results, and C 6 good results recorded in 1978 Good 3 6

operated cases altogether having that length of
follow-up (e.g. there were 235 cases with follow-up
at 10 years), the number of these cases in which ini-
tially excellent or very good results were still the
same at that length of follow-up (e.g. 165 of the 235
at 10 years), the number of cases whose initially ex-
cellent results had deteriorated to good since the
previous follow-up examination (14 cases at 10
years), and those whose initially excellent results
had deteriorated from excellent to fair or to poor
since the previous follow-up examination (6 and 8
respectively at 10 years), The figures in the table are
complete for each length of follow-up, so a patient
followed for 30 years will be included in every col-
umn. Patients are, of course, not examined every
year after operation, so to date the deteriorations
of results we decided a priori that any deterioration
would be assumed to have occurred three-quarters
of the way between the date at which the result was
last known to be excellent and the date on which
the deterioration was noted. Table 27.1 thus shows
(in the bottom row) the likelihood of an initially ex-
cellent or very good result remaining stable with the
passing of time.
590 Reassessment of Patients Treated Operatively Within Three Weeks of Injury

Table 27.1

Year of follow-up

2 3 4 5 6 7 8 9 10 11 12 13 14 15

Number of 492 464 396 364 319 298 279 263 250 235 220 195 177 159 146
operated cases
Total of results 407 387 332 303 270 253 235 220 207 193 179 159 143 129 119
initially excellent/
very good
Results still excel- 372 352 297 268 235 218 201 188 176 165 153 135 123 111 102
lent!very good
Dropped from 16 16 16 16 16 16 16 16 15 14 13 11 9 8 7
excellent to good
Dropped from 6 6 6 6 6 6 6 6 6 6 5 5 5 5 5
excellent to fair
Dropped from 13 13 13 13 13 13 12 10 10 8 8 8 6 5 5
excellent to poor
Total of excellent/ 831170 831170 841170 831170 851170 851170 841170 841170 831170 821170 811170 821170 811170 811170 821170
very good results
as 1170 0 f cases
operated on

Year of follow-up

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Number of 133 116 106 102 90 75 65 51 39 29 15 8 7 5 2

operated cases
Total of results 106 92 83 74 65 54 46 37 28 22 12 5 4 3
initially excellent!
very good
Results still excel- 93 82 74 70 61 52 45 36 27 21 12 5 4 3
lent/very good
Dropped from 7 5 5 3 3
excellent to good
Dropped from 3 3 2 0 0 0 0 0
excellent to fair
Dropped from 3 2 2 0 0 0 0 0 0 0 0 0 0 0 0
excellent to poor
Total of excellent/ 801170 791170 781170 731170 721170 721170 711170 731170 721170 761170 801170 631170 571170 601170 501170
very good results
as 1170 0 f cases
operated on
28 Operative Treatment Between Three Weeks
and Four Months After Injury
We do persist in individualising this group because
the surgical treatment of these patients is much
more difficult than that of patients who are operat-
ed on within 3 weeks. The surgical treatment in-
creases in difficulty during the 3rd week after in-
jury; after 3 weeks and up to 4 months it is always
difficult, sometimes very difficult. The reason we
separate these cases from those operated on after 4
months is that usually remodelling of the innomi-
nate bone is not complete within this secondary
period (as it generally is after 4 months), and with
experience it is nearly always possible to find the
original fracture lines, although they are a little
more irregular, giving the best chance of restoring
the normal anatomy of both the innominate bone
and the acetabulum. However, the need to reach the
initial fracture lines as well and as easily as possible
necessitates a much more frequent use of extensile
approaches. So far we have operated on 157 frac-
tures during this post-injury period.
28.1 Condition of Fracture Healing
Callus formation at the level of the innominate
bone varies greatly, and even before the end of 4
months after injury we have been, in some cases,
obliged to perform what can only be called oste-
otomies to free the fragments and restore the con-
gruity of the joint.
Table 28.1 shows the conditions of the 157 frac-
tures at the time of operation, according to the ana-
tomical type of fracture.
in 95 cases out of 157 (60.5070) the initial frac-
ture lines were still visible on radiographs.
7 cases were of mal-union/non-union. In 23
cases there was already organized mal-union
(14.6%), while 28 cases (17.8%) exhibited an
unreduced posterior dislocation with either a
posterior wall fracture (22 cases) or associated
transverse and posterior wall fracture (6 cases).
A segmental fracture of the head was present in
2 of these cases.
- 3 missed incarcerated fragments had to be re-
moved.
Surgery was performed:
between 21 days and 1 month after injury in 46
cases (29.5%);
during the 2nd month after injury in 55 cases
(35.2%);
during the 3rd month in 33 cases (21.2%);
during the 4th month in 23 cases (14.6%).
28.2 Surgical Approach
Table 28.2 shows the approaches which were used
to reconstruct these 157 cases according to fracture
type:
The Kocher-Langenbeck approach predomi-
nates: 84 cases (53.5%). The reason for this is
that in this series the posterior column was so
frequently involved: there have been 32 posteri-
or wall fractures, 23 associated transverse and
posterior wall, 8 posterior column, 5 transverse,
7 T-shaped, 5 associated posterior wall and
posterior column, and 4 both-column fractures.
Up to 1975 all cases of these types were ap-
proached posteriorly.
42 fractures (26.7%) were approached through
the extended ilio-femoral incision, mostly both-
column fractures and associated transverse and
posterior wall fractures.
17 ilio-inguinal approaches (10.8%) were used,
mainly for the treatment of anterior fractures (6
cases) and both-column fractures (5 cases).
2 ilio-femoral approaches were used to treat 2
anterior wall fractures.
4 double approaches were used, 3 in one stage
and 1 in two stages, to treat 2 T-shaped and 2
both-column fractures.
Table 28.1 v.
tv
'"
Condition of fracture Type of fracture Total Total as 0,10 of
healing overall total
Posterior Posterior Anterior Anterior Transverse T-shaped Transverse and Posterior wall Anterior column Both- (157 cases)
wall column wall column posterior wall and posterior and posterior column
column hemitransverse

Fracture lines still 8 4 4 2 11 13 29 22 95 60.5%

visible
Mal-union 3 3 3 6 2 3 23 14.6%
Mal-union/non-union 2 7 4.5%
Unreduced posterior 22 6 28 17.8%
dislocation
Missed incarcerated 2 3 1.9%
fragments
Other 0.6%
Total 35 8 6 3 14 17 42 5 4 23 157 0
'0
."
...,
~
:t
."

::;l
."
~
Table 28.2 3
."
g
Approach used Type of fracture Total Total as % of OJ
."
overall total ~
."
Posterior Posterior Anterior Anterior Transverse T-shaped Transverse and Posterior wall Anterior column Both- (157 cases) ."
:J
wall column wall column posterior wall and posterior and posterior column ...,
::T
column hemitransverse @
."

Kocher-Langenbeck 32 8 5 7 23 5 4 84 53.5%
~
~
Ilio-inguinal 1a 3 3 2 5 17 10.8% "'"
II>
:J
Ilio-femoral 2 2 1.3% 0-

Extended 5 8 17 2 10 42 26.7% 61
c...,
ilio-femoral
~
0
Double approach 3 4 2.5% :J

Other 2 3 8 5070 ~
>
;::>
Total 35 8 6 3 14 17 42 5 4 23 157 ."
...,

a The apparently paradoxical choice of the ilio-inguinal approach to this posterior wall fracture is explained by the fact that this was a pure (and unique) superior fracture taking
g
E;
with it a part of the iliac wing (see Fig. 5.12) '<
Surgical Technique
8 various approaches were used to treat 2
posterior wall fractures, 1 anterior wall fracture,
3 transverse fractures, 1 anterior and posterior
hemitransverse fracture and 1 both-column
fracture.
As always, the first problem of surgical treatment
is the choice of approach. Since we have used the
extended ilio-femoral, the choice has appeared to
us very straightforward: all partial or total fractures
of one column are approached through their respec-
tive incision: the Kocher-Langenbeck for posterior
column fractures, and the ilio-inguinal for anterior
column fractures. All other fractures require the ex-
tended ilio-femoral approach.
28.3 Surgical Technique
28.3.1 Cases with Visible Fracture Lines
When the fracture lines are still visible radiological-
ly (95 cases), the technique has no particular new
features: reduction and fixation are performed us-
ing the same methods and obeying the same prin-
ciples as in fresh fractures. However, the first thing
to do is to remove the already formed callus, start-
ing from its junction with the intact parts of the
bone and working up to the fracture lines in order
to identify them properly. Then the callus is re-
moved from inside the fracture lines using a
rongeur or a curette; if possible, it is also removed
along the fracture lines on the other side of the
fracture. Once the fracture lines are perfectly clean,
even though they are more irregular and already
slightly remodelled, reconstruction and fixation are
carried out just as in fresh cases (Figs. 22.18,
28.1-28.6, 28.8 and 28.18).
A direct view of the inside of the joint, obtained
by a capsulotomy along the acetabular lip, is ex-
tremely helpful, and we use this as often as we can.
Up to 4 months after injury, even if the callus
bridges the different parts of the fractured acetabu-
lum, the fracture lines through the cartilage are per-
fectly clear. It is also necessary to emphasise that in
some cases post-traumatic osteopenia makes the
manipulation of the fragments more difficult and
more demanding of care (Fig. 28.7).
28.3.2 Mal-union
In some cases (23 out of 157) a true mal-union has
already built up. That means that finding the
593
original fracture lines is impossible (Fig. 28.8). In
such cases a true osteotomy is needed, removing a
wedge of bone in order to restore contact between
the osteotomised fragments and at the same time
the sphericity of the acetabulum. Usually the best
way is to get the most complete view possible of the
inside of the joint and to perform the osteotomy
from the fracture lines dividing the articular sur-
face. The shape of the wedge depends on the
respective displacement of the mal-united frag-
ments. It is uncommon to succeed straightaway in
restoring the sphericity of the acetabulum; usually
further corrections of the sides of the wedge are
necessary (Fig. 28.10). Several osteotomies con-
ducted in the same way may be needed in complex
cases (Figs. 28.8 and 28.9).
28.3.3 Non-union/Mal-union
In cases where there is both non-union and mal-
union (7 out of 157), areas of non-union are excised
and areas of mal-union are osteotomised to allow
the reconstruction. Several trials are regularly nec-
essary to obtain perfect reduction of the acetabu-
lum (Figs. 28.11 and 28.12).
Non-union/mal-union is a frequent occurrence
in transverse fractures, which heal rapidly anterior-
ly and remain ununited posteriorly. A posterior
column fracture too may unite to the intact part of
the bone by bony bridges and remain non-united
between them. Excision of these bony bridges, even
the one uniting the column to the sacrum, is com-
pulsory (Figs. 28.6 and 28.13).
28.3.4 Neglected Posterior Dislocations
of the Femoral Head
We have had 28 cases of neglected posterior dislo-
cation of the head (Figs. 28.14, 28.15 and 28.16).
Whatever the chosen approach - Kocher-
Langenbeck or extended ilio-femoral, according to
the associated fracture - the problem is first to
recognise among the new bony structures envelop-
ing the head what is callus formation and what are
real posterior wall fragments, which are either free
or connected to the capsule, and then to try as far
as possible to keep the capsular connections intact.
Once this "new" coverage of the femoral head
has been put aside, the femoral head appears and
its damage is recorded. The head has always con-
tracted multiple fibrous adhesions from its anterior
594 Operative Treatment Between Three Weeks and Four Months After Injury

Fig. 28.1A- F. T-shaped fracture, associated with a fracture line

running through the wing and involving the sacro-iliac joint,
approached laterally 30 days after injury. A Antero-posterior
radiograph, B obturator-oblique radiograph, and C iliac-oblique
radiograph taken before operation, D antero-posterior radio-
graph, E obturator-oblique radiograph and F iliac-oblique
B radiograph taken 13 years after operation
D E

Fig.2S.1D-F

F
596 Operative Treatment Between Three Weeks and Four Months After Injury

A B C

o
Fig. 2S.2A-G. T-shaped fracture operated on 40 days after in-
jury (posterior and anterior approaches consecutively). A Ante-
ro-posterior radiograph, B obturator-oblique radiograph, and
C iliac-oblique radiograph before operation, 0 antero-posterior
radiograph after operation, E Antero-posterior radiograph,
F obturator-oblique radiograph, G iliac-oblique radiograph,
taken 8 years after operation, showing evidence of osteoar-
throsis. Thirteen years after operation, a total hip replacement
E
was necessary
 Surgical Technique 597

F G

Fig. 28.2 F, G
598 Operative Treatment Between Three Weeks and Four Months After Injury

Fig. 28.3A-F. Pure anterior wall fracture, operated on 40 days

after injury. A Antero-posterior radiograph, B obturator-
oblique radiograph, and C iliac-oblique radiograph taken before
operation, D antero-posterior radiograph, E obturator-oblique
radiograph, and F iliac-oblique radiograph taken 13 years after
operation. The hip is rated 6.6.6
Surgical Technique 599

D E F

Fig. 28.3D-F
600 Operative Treatment Between Three Weeks and Four Months After Injury

A B c

o E F

Fig. 28.4A-F. Both-column fracture operated on 35 days after taken before operation, D antero-posterior radiograph, E ob-
injury (anterior approach). A Antero-posterior radiograph, turator-oblique radiograph, and F iliac-oblique radiograph
B obturator-oblique radiograph, and C iliac-oblique radiograph taken 4 years after operation
Surgical Technique 601

Fig. 28.5 A-F. Both-column fracture operated

on 30 days after injury (anterior approach).
A Antero-posterior radiograph,
B obturator-oblique radiograph, and C iliac-
oblique radiograph taken before operation,
D antero-posterior radiograph, E obturator-
oblique radiograph, and F iliac-oblique
radiograph taken 4 years after operation
602 Operative Treatment Between Three Weeks and Four Months After Injury

B c

Fig. 28.5 B, C
Surgical Technique 603

Fig. 28.5 D - F

E
 604 Operative Treatment Between Three Weeks and Four Months After Injury

Fig. 28.6A-H. A Diagram showing the discrete bars of bone

which may unite a displaced posterior column with other parts
of the pelvic bone, B Anteror-posterior radiograph, C obturator-
oblique radiograph, D iliac-oblique radiograph of a posterior
column fracture operated upon 45 days after injury, E antero-
posterior radiograph, F obturator-oblique radiograph, G iliac-
oblique radiograph taken 9 years after operation, H nineteen
years after operation osteoarthritis has set in

c o
Surgical Technique 605

F
 606 Operative Treatment Between Three Weeks and Four Months After Injury

A B

Fig. 28.7 A-C. Pure anterior wall fracture (ilio-inguinal

approach). The osteopenic conditions made operation very dif-
ficult. A Antero-posterior radiograph, B obturator-oblique
radiograph, C iliac-oblique radiograph 1 year after operation.
c See Fig. 7.6 for pre-operative condition. The hip is rated 6.6.6
Surgical Technique 607

aspect to the iliac bone and to the remaining intact

parts of the acetabulum. All these abnormal
fibrous links have to be cut or divided; this is made
possible by progressively rotating the leg externally
with the knee flexed, the patient lying prone and
being approached posteriorly (this manoeuvre 00

rotates the head internally). Finally, the intact part '"

of the acetabular cartilage is identified and all the _ 00
new tissue filling up the cavity removed.
When the head is free and the acetabulum emp-
ty, relocation of the head is attempted. This should
be performed with as little traction as possible. The
strong traction we used at the beginning of our I ...,
practice or additional levering action on the head
led to squeezing of the femoral head, necessitating
an immediate arthroplasty, and we wrongly
thought that this squeezing was the consequence of os
r- I r-
necrosis of the head, instead of realising that it was
its hypervascularisation that allowed this destruc-
tion.
To obtain this lowering of the head without trac-
tion, as many as necessary of the peri-articular
muscles (femoral insertions of the gluteal muscles,
ilio-psoas and adductors) are cut.
Once the head has been relocated, its rotation
against the intact part of the acetabulum must be
studied. If it seems to tend to redislocate, this is
because its anterior aspect has not been totally
freed, and/or because of stretching of the anterior 01
00
01
00

capsule. It is then necessary to divide the gluteus

minimus close to the femur to reach the anterior
capsule and to cut it along the acetabular lip.
The femoral head safely relocated, a posterior
wall is reconstructed using the fragments identified
in the new wall of the dislocation cavity, if
necessary with supplemented iliac bone grafts.
Screws and plates are always necessary. Post-
operative traction is used if the posterior wall frac-
I on
ture involved more than half of the articular sur-
face.
<::
o
.~
....
(l)
(l)

28.3.5 Incarcerated Fragments 0-

o '"uos
""'o
There were three cases with incarcerated fragments.
t
(l)

Two were excised through a Rueter type approach;

they were free. In one case an incarceration of part
of the anterior wall was removed through an ilio-in-
guinal incision.
608 Operative Treatment Between Three Weeks and Four Months After Injury

B o
Fig. 28.8A-D
 Surgical Technique 609

E f

Fig. 28.8 A-F. Associated transverse and posterior wall fracture radiograph taken before operation, D antero-posterior radio-
with central dislocation of the femoral head, operated on 90 graph, E obturator-oblique radiograph and F iliac-oblique ra-
days after injury (posterior approach). A Antero-posterior diographs, all taken 15 years after operation. The hip is rated
radiograph, B obturator-oblique radiograph and C iliac-oblique 6.6.6
 610 Operative Treatment Between Three Weeks and Four Months After Injury

A B

Fig. 28.9 A - C. Trans-tectal transverse fracture operated on 110

days after injury (extended ilio-femoral approach). A Antero-
posterior radiograph, B obturator-oblique radiograph, C iliac-
oblique radiograph, all taken 4 years after operation. See
c Fig. 9.12 for pre-operative state. The hip is rated 6.6.6
Surgical Technique 611

Fig. 28.10. Schemes of the trapezoid-

shaped slice of bone that has to be
removed in order to reconstruct a mal·
union of A a transverse fracture, and B
an associated transverse and posterior
wall fracture B
 612 Operative Treatment Between Three Weeks and Four Months After Injury
Fig.2S.11A-c' Surgical correction of bilateral mal-union.
A Antero-posterior radiograph before operation. On the pa-
tient's right side, associated transverse and posterior wall frac-
ture; note the wear of the supero-lateral part of the head. On the
left side, a T-shaped fracture with considerable wear of the
superior part of the head, B antero-posterior radiograph after
operation. The right side was operated on 70 days after injury,
the left side 90 days after injury, both through an extended ilio-
femoral approach. On the left side a Sugioka osteotomy was
performed in the same operation as reduction and internal fixa-
tion of the acetabulum to try to compensate for the extreme
wear of the head, Cantero-posterior radiograph 14 months after
operation. On the patient's right side, avascular necrosis of the
femoral head seems to have developed; on the left side there is
massive necrosis of the femoral head, and the posterior wall has
also disappeared due to associated necrosis or to wear by the
necrotic head. A total hip replacement was performed on the left
side at that time
Surgical Technique 613

Fig. 28.118
614 Operative Treatment Between Three Weeks and Four Months After Injury

Fig.2S.11C
 Surgical Technique 615

A B

Fig. 28.12A-C. Associated trans-tectal transverse and postero-

superior wall fracture extending to the iliac crest, operated on
120 days after injury (extended ilio-femoral approach).
A Antero-posterior radiograph, B obturator-oblique radio-
graph, and C iliac-oblique radiograph 8 years after operation.
See Fig. 12.12 for pre-operative state. The hip is rated 6.6.6
despite a considerable "collarette" of ectopic bone around the
head C
 616 Operative Treatment Between Three Weeks and Four Months After Injury

A B

Fig. 28.13 A - F. Mal-union/non-union of a posterior column

fracture. A bony bridge unites the posterior column to the
sacrum. A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac-oblique radiograph before operation,
D antero-posterior radiograph, E obturator-oblique radiograph,
and F iliac-oblique radiograph 3 years after operation. The hip
c is rated 6.6.6
 Surgical Technique 617

D E

Fig. 28.I3D-F
F·
618 Operative Treatment Between Three Weeks and Four Months After Injury

A 8

c D

Fig. 2S.14A- D. Extended posterior wall fracture with persistent achieved with screws only; a plaster cast was applied for 2
posterior dislocation, operated on 30 days after injury (Kocher- months, C, D antero-posterior radiographs 20 years (C) and 30
Langenbeck approach). A Antero-posterior radiograph before years (D) after operation
operation, B the same view 2 years after operation. Fixation was
Intra-operative Complications 619

28.3.6 Review of Surgical Techniques

Table 28.3 displays the types of operations used ac-

cording to the conditions of fracture healing at the
time of surgery. Out of 95 cases in which the frac-
ture lines were clearly visible on the radiographs,
complete excision of the callus followed by internal
fixation was performed in 89 cases, but in 1 case an
unforeseen osteotomy was necessary, followed by a
hemiarthroplasty. Five patients underwent dif-
ferent procedures (arthrolysis, cheilectomy, removal
of a segmental fracture of the femoral head).
Of the 23 mal-unions, 21 required an osteotomy,
and in one case we performed a curved supra-tectal
osteotomy to bring the roof back against the head,
as the acetabulum was otherwise congruent to the
head (Fig. 28.16). A total hip arthroplasty was per-
formed in one case.
The seven non-union/mal-unions needed oste-
otomy of the mal-united area and excision of the
non-united zone. In one case the condition of the
head was such that an immediate hemi-arthroplas-
ty was performed.
The 28 neglected posterior dislocations were
repositioned and the posterior wall was rebuilt with
(5 cases) or without (23 cases) adding some iliac
bone grafts. Of the two cases with an additional
fracture of the femoral head, a hemi-arthroplasty
was performed in one case, while in the other, the
segmental fracture of the head was fixed with two
screws embedded in the cartilage, and the posterior
wall rebuilt. At 7 months after surgery the patient A
broke her femoral neck as a consequence of evident
extended femoral head necrosis.
Additionally, we removed missed incarcerated
fragments in three cases, and intra-articular hard-
ware inserted elsewhere in one case.

28.4 Intra-operative Complications

The following intra-operative complications oc-

curred:
1 cardiac arrest during a Kocher-Langenbeck
operation; the patient recovered;
6 injuries to the superior gluteal artery in the
course of 3 lateral approaches and 3 Kocher-
Langenbeck approaches;
3 injuries to the superior gluteal veins during 3
Kocher-Langenbeck approaches;
1 injury to the external iliac vein during an ilio-
inguinal approach; it was easily sutured.
Fig. 28.15A-F. Associated transverse and posterior fracture
with posterior dislocation, operated on 120 days after injury.
A Antero-posterior radiograph, B obturator-oblique radio-
graph, and C iliac-oblique radiograph taken before operation,
D antero-posterior radiograph, E obturator-oblique radiograph,
and F iliac-oblique radiograph taken 7 years after operation
(clinical result still good)
620 Operative Treatment Between Three Weeks and Four Months After Injury

B C

Fig. 28.15B,C
 Intra-operative Complications 621

o
E

Fig. 28.1SD - F F
622 Operative Treatment Between Three Weeks and Four Months After Injury

A B

c D

Fig. 28.16A- D. Posterior dislocation of the head with an asso-
ciated posterior wall fragment, operated on 114 days after in-
jury. A Obturator-oblique radiograph before operation, B ante-
ro-posterior radiograph and C obturator-oblique radiograph 3
months after operation. The antero-posterior view shows a near-
ly satisfactory condition of the joint, but the obturator-oblique
view shows extreme narrowing of the anterior part of the joint
space. Is this pure cartilage necrosis? D Obturator-oblique
radiograph 2 years after operation, just before total hip replace-
ment was performed
Early Post-operative Complications
Besides these typical complications, in two cases we
found an infected haemarthrosis of the hip at
operation. In spite of wide debridement, the pa-
tients developed infectious arthritis during the fol-
lowing year; one hip was fused, the other has a
poor result and the patient is waiting for a total hip
replacement.
(e) Post-operative Sciatic Nerve Palsies.
28.5 Early Post-operative Complications
(a) Deaths. Three patients died in the early post-
operative period: one of septicaemia linked to
an infection of the pelvis due to a missed blad-
der wound, and one of pulmonary embolism on
the 3rd day. The third died 2 months post-
operatively from unknown causes.
(b) Thrombo-embolism. Four cases of phlebitis
and 1 of pulmonary embolism had to be
treated.
(c) Local complications.
3 localised skin necroses along a lateral
approach (out of 42);
9 haematomata requiring surgical evacua-
tion, 5 after posterior approaches (out of
84), 4 after lateral approaches (out of 42).
(d) Infections. Immediate deep infections occurred
in three cases (2070): one after a Kocher-Langen-
beck approach, one after an ilio-inguinal, and
one after a double approach in one stage along
the ilio-inguinal incision. The two last-men-
tioned instances occurred when the ilio-inguinal
operation was not covered by antibiotic pro-
phylaxis.
There were four delayed infections. One case
of infectious arthritis developed several months
after a Kocher-Langenbeck approach not ini-
tially covered by antibiotics. One delayed infec-
tion occurred after a lateral approach in which
the hip was not apparently involved. In fact, the
Table 28.4
Type of palsy Recovery
Complete Significant
Total lateral popliteal
Partial lateral popliteal 1 1
Patchy lesion, limited 2 2
Patchy lesion, extensive 2 1
Delayed post-operative palsy
Total 2 5
623
debridement led us to the hardware, which was
removed and healing occurred. The last two
were cases of septic arthritis which developed a
few months after surgery. These two cases had
been operated on elsewhere initially and were
sent to us to try to improve an imperfect reduc-
tion; in both cases intra-operative cultures were
positive.
There were 16 cases of these, 13 with immediate
onset (8.3%) and 3 delayed, appearing in the
first weeks following operation (2.2%). The 13
cases with immediate onset followed a Kocher-
Langenbeck approach (15.5%): two occurred
before the use of transcondylar traction and 11
in spite of it. Two cases were total lateral
popliteal palsies, three involved the popliteal
component partially, and eight were patchy le-
sions either limited (5 cases) or extensive (3
cases).
In five patients who had a post-operative
palsy, the nerve function had not been checked
before surgery.
Thus, the risks for the sciatic nerve remain
considerable in the period of 3 weeks to 4
months, despite traction with the knee flexed;
the care that must be taken in manipulating the
retractors can never be emphasised enough.
Three cases of delayed sciatic palsy appeared
in the first weeks following operation (1.9%),
one following a Kocher-Langenbeck and the
two others a lateral approach. Like those occur-
ring after the treatment of fresh fractures, they
remain unexplained.
It is interesting to note that of this group of
157 operative patients, 29 had a pre-operative
sciatic palsy (18.6%): four total palsies, three
lateral popliteal, one medial popliteal and 21
patchy lesions.
Table 28.4 shows the long-term functional
results in these cases of post-operative palsy.
Total
Partial Nil Unknown
3
2
5
3
2 3
7 16
 624 Operative Treatment Between Three Weeks and Four Months After Injury

A B

Fig. 28.17 A-G. Associated anterior column and posterior

hemitransverse fracture with a secondary split of the anterior
column through the anterior wall (extended ilio-femoral
approach). A Antero-posterior radiograph, B obturator-oblique
radiograph, and C iliac-oblique radiograph before operation,
D antero-posterior radiograph, E obturator-oblique radiograph,
and F iliac-oblique radiograph 3 years after operation. Hip func-
tion is normal, G during surgery a curved osteotomy was per-
formed above the acetabular lip, bringing back against the head
c the segment of the articular surface which was away from it
 625
Early Post-operative Complications

Fig.2S.17D-F
626 Operative Treatment Between Three Weeks and Four Months After Injury
(f) Other Palsies.
We observed two cases of total palsy of the
gluteal muscles, one after a Kocher-Langenbeck
and the other after a lateral approach. There
were three other cases after Kocher-Langenbeck
approaches of partial weakness of the gluteal
muscles. It is always difficult to say whether the
surgery is totally responsible for this or whether
the gluteal nerve was injured at the time of
trauma, as testing of the gluteal muscles is
usually extremely difficult pre-operatively.
28.6 Late Post-operative Complications
(a) Non-union. Non-union occurred in one case of
an anterior wall fracture approached anterior-
ly. The non-union became obvious when the
plate broke, permitting a slight secondary dis-
placement. The result after 11 years is clinical-
ly good with radiological signs of osteoarthri-
tis.
(b) Avascular bone necrosis. This occurred in 21
cases, an overall rate of 13.4%. As usual, its ex-
tent varied; there were:
2 superior segmental limited femoral head
necroses, after 1 posterior wall and 1 asso-
ciated transverse and posterior wall frac-
ture;
7 superior segmental extensive after 2
posterior wall, 1 transverse, and 4 associat-
Fig. 28.17G
ed transverse and posterior wall fractures;
3 massive femoral head necroses, after 1 T-
shaped and 2 associated transverse and
posterior wall fractures;
1 posterior wall necrosis;
8 mixed or associated necroses of the
femoral head and the posterior part of the
acetabulum, following 3 posterior wall, 1 T-
shaped and 4 associated transverse and
posterior wall fractures (Figs.28.19 and
28.20).
To these 21 cases must be added two cases of
cartilage necrosis occurring after one posterior
wall fracture and one associated transverse and
posterior wall fracture. The total of 23 cases of
avascular necrosis represents 14.611,10 of the
operated cases.
Of 33 patients with intra-operative reduction
of a posteriorly dislocated head, 14 developed
avascular bone necrosis (42.4%). The extent of
the necrosis was superior limited (2 cases),
superior extensive (2 cases), massive cephalic (2
cases), head and posterior wall (6 cases),
posterior wall (1 case), and cartilage (1 case).
In 22 of the 33 cases, the dislocations had
persisted since the accident, and avascular bone
necrosis developed in 11 (50%), the associated
fractures being posterior wall (5 cases), trans-
verse and posterior wall (5 cases), and T-shaped
(1 case). In the other 11 cases an attempt at
closed reduction was followed by redislocation
 Early Post-operative Complications 627

A
B

Fig. 2S.1SA-C. T-shaped fracture operated upon 30 days after

injury (Kocher-Langenbeck approach). A Antero-posterior
radiograph and B obturator-oblique radiograph after operation.
See Fig. to.! 5 for pre-operative state. C Four years later the
antero-posterior radiograph shows associated necrosis of the
head and the posterior wall. The patient needed a total hip re-
C
placement
 628 Operative Treatment Between Three Weeks and Four Months After Injury

A B

c o

Fig. 28.19A-D. Associated transverse and posterior wall frac-
ture with posterior dislocation, operated on 60 days after injury.
A Antero-posterior radiograph before operation, B obturator-
oblique radiogaph after operation. The head has been reposi-
tioned through a Kocher-Langenbeck approach. The transverse
fracture has been imperfectly reduced, Cantero-posterior
radiograph 3 months after operation. Associated avascular
necrosis of the head and of the repositioned posterior wall is ob-
vious, D antero-posterior radiograph of the opposite side. This
right hip dislocated posteriorly at the time of the accident to the
left hip. The patient developed avascular necrosis on this side,
identified 3 years after the accident
Late Post-operative Complications
which was neglected. Avascular bone necrosis
occurred in three of these cases: two posterior
wall, and one transverse and posterior wall frac-
ture.
Of 26 patients with pre-operative closed
reduction of a posterior dislocation, associated
with a posterior wall fracture (13 cases), a
transverse and posterior wall fracture (11 cases),
or a posterior column fracture (2 cases), five
developed avascular bone necrosis (19070). The
necrosis involved only the cartilage in one and
was classified as superior extensive in the other
four.
Of 98 patients with a central (78 cases) or
anterior dislocation (13 cases), or without dislo-
cation, four developed avascular bone necrosis
(4.1 %). The associated fractures were: trans-
verse (1 case), T-shaped (1 case), transverse and
posterior wall (2 cases). The necrosis was ex-
tended superior in one case, massive cephalic in
one, and involved the head and posterior wall in
two.
If the incidence of osteonecrosis in this
group of fractures operated on between 3 weeks
and 4 months is compared to the figures report-
ed for the cases operated on within 3 weeks of
injury, a considerable rise can be seen: the in-
cidence climbs from 3.9% (or 4.9070 if cartilage
necrosis is included) to 13.5%. The rate of asso-
ciated femoral head and posterior wall osteo-
necrosis climbs from 13.6% to 38% of all
necroses.
(c) Wear of the Head. Certain cases of wear of the
femoral head must be distinguished separately
and not included among the osteonecroses.
There were three of these; the wear occurred in
one case along a mal-reduced fracture, in the
second it was due to an intra-articular
spongiosa screw, and in the third the wear was
against the fracture lines after a secondary
displacement. At their last follow-up these cases
all had poor results, at 3 years, 17 years and
1 year, respectively.
(d) Post-traumatic Osteoarthritis. The 32 cases of
post-traumatic osteoarthritis (20.4%) developed
after 22 perfect reductions (3 posterior wall
fractures, 2 posterior column, 2 transverse, 5 T-
shaped, 6 associated transverse and posterior, 1
posterior column and posterior wall, 1 anterior
column and posterior hemitransverse, 2 both-
column) out of 96 (23 %), and 10 imperfect
reductions (1 posterior wall, 1 posterior col-
umn, 1 transverse, 1 T-shaped, 1 anterior wall
629
and 5 transverse and posterior wall) out of 41
(24%). The cause of these cases of osteoar-
thritis is not always easy to find:
in 18 cases, all following perfect reductions,
the cause is unknown; nevertheless, 11 of the
cases are of type III and 7 of type V, i.e. ad-
vanced osteoarthritis needing a total hip re-
placement;
in 9 cases, a fault in reduction may be
regarded as the cause of the osteoarthritis,
which is of type III in 3 cases, type IV in 1
and type V in 5;
in 3 cases avascular necrosis (1 femoral head,
1 cartilage, 1 head and acetabulum) was
followed by the occurrence of osteoarthritis,
type III in 2 cases and type V in 1;
in 2 cases, despite perfect reconstruction of
the acetabulum, one may assume that the
significant lesions of the cartilage and the
sub-chondral bone of the head discovered at
operation were the decisive factor in the oc-
currence of osteoarthritis.
It should be added that in six of these recon-
structed cases we found a lesion of the femoral
head intra-operatively:
1 segmental fracture which was removed led
to osteoarthritis at 7 years;
1 localised impaction led to osteoarthritis at
21 years;
1 appreciable wear of the head led to osteo-
arthritis at 4 years;
3 localised abrasions of the sub-chondral
bone led to osteoarthritis at 4 years in 1 case,
and to 2 good results at 1 and 4 years, respec-
tively.
Comparing these figures to the corresponding
ones for cases operated on within 3 weeks of in-
jury, it is noticeable that even after perfect
reduction, the incidence of post-operative os-
teoarthritis has increased from 10.2% to 23%;
after imperfect reduction, paradoxically, it is a
little lower: 24% instead of 35.7%.
(e) Ecotopic Ossification. Ectopic ossification
developed in 39 cases: an incidence of 25%.
They followed posterior approaches in 21 cases
and double approaches, including a Kocher-
Langenbeck, in 3, i.e. 24 instances out of 84
posterior approaches in all (28.6%). Ectopic os-
sification also developed after 14 extended ilio-
femoral approaches out of 42 (33.3%).
630 Operative Treatment Between Three Weeks and Four Months After Injury

Table 28.5

Treatment to prevent Brooker type Total no. of No. of cases Total Cases with ossification
ossification cases with without as 070 of total
II III IV ossification ossification

None 3 13 9 7 32 93 125 25.6070

Diphosphonate 10 mg
Diphosphonate 20 mg 2 2 6
3
3
3
9 ] 50070
Indomethacin 11 12 8.3070
Indomethacin + 12 Gy
Indomethacin + 7 Gy
7
1
7
1 ] 0070

Total 4 16 11 8 39 118 157

One anterior Hueter approach was also com- 3 died in the early post-operative period;
plicated by a type I ossification. 4 underwent immediate arthroplasty (2 hemi-
Of these 39 ectopic ossifications, four were and 2 total);
type I in Brooker's classification, 16 type II, 11 8 were lost to follow-up before the end of 1 year;
type III and eight type IV. Of the eight type IV, 3 were operated on too recently;
two have been excised and two are waiting for 1 died of another cause.
excision.
Therefore, 138 patients were available to follow-up.
In regard to prevention, among the patients The length of follow-up for these 138 patients is:
with ossification 32 (7 with type IV) had receiv-
ed no preventive treatment, six (2 with type III, 1- 2 years: 38;
1 with type IV) had received diphosphonate 3 - 4 years: 26;
20 mg, and one with a type II ossification had 5 years: 9;
received indomethacin. Table 28.5 shows the 5-10 years: 21;
figures for prevention in all 157 cases. 10-15 years: 24;
15-20 years: 8;
20 - 25 years: 10;
28.7 Results 26 years: 1;
30 years: 1.
Of the 157 cases operated upon between 3 weeks
and 4 months after injury, 19 patients were not Table 28.6 shows the final functional results in rela-
available for follow-up: tion to the quality of operative reduction.

Table 28.6

Final functional Quality of operative reduction Total Total as 070 of

result overall total
Perfect Imperfect reduction (138 cases)
reduction
Head centred Loss of Residual head Technical Surgical
parallelism protrusion failure secondary
congruence

Excellent
Very good IT] 12
4 2
3
1
3
55 ]
20'
54.3070

Good 9 3 2 14 10.1070
Fair 5 3 1 1 10 7.2070
Poor 23 6 3 3 2 2 39 28.3070
Total 97 17 11 4 2 7 138

• Eight cases with radiological signs of osteoarthrosis

Late Post-operative Complications 631

Out of 97 perfect reductions:

48 cases (50070) achieved an excellent result. If to
these are added 12 other cases with a very good 'if..g.tJ!-~r$.
0\ I.t') ~ N f'I"'I
clinical result (6 of which have some radiologi- ~~or--:oO
('f')..,...., ....... N
cal signs of osteoarthritis), 60 cases (61.9%)
have a very good clinical result. As a reminder:
the perfect reductions performed within 3 weeks
of injury led to 82% very good functional , S
~

-5 ::l
results. 0-
p:) 8
9 other perfect reductions achieved good results
and 5 were assessed as fair.
23 of the perfect reductions have already gone
to a poor result, and 17 of these have already
been subjected to a hip replacement (13 cement-
ed, 2 uncemented, 2 hemi-arthroplasties).

As to the imperfect reductions:

17 cases with a recentred head and imperfect
acetabulum had 4 excellent and 1 very good
functional results (29%), 3 good, 3 fair and 6
poor.
11 cases of loss of parallelism of the superior
joint space ended up with 5 clinically very good
(2 excellent, 3 very good; 45.5%), 2 good, 1 fair
and 3 poor results.
4 cases of persisting head protrusion. One had
of clinically very good result and three a poor.
2 technical failures have poor results.
7 patients in whom surgical secondary con-
gruence was achieved. Four had a very good
result, one a fair and two had poor results.
Although the group is small, these figures re-
main the best among the imperfect reductions.
Overall, the 138 patients ended up with 75 at
least very good functional results (54.3%), 14 I I I
good results (10.2%), 10 fair (7.2%) and 39
poor results (28.3%).
Table 28.7 shows final functional results in
relation to fracture type. Perhaps surprisingly,
both-column fractures have the best results. As-
sociated posterior wall and posterior column
fractures have much better results here than
among the patients operated on within 3 weeks;
however, probably no conclusions can be drawn
from this as the series is too small. Posterior
wall fractures, in most cases associated with a
persistent posterior dislocation of the head, pay
a heavy toll in osteonecrosis and consequently
their overall proportion of very good results ap-
pears low.
Table 28.8 displays the final functional
results in relation to radiological appearance at
632 Operative Treatment Between Three Weeks and Four Months After Injury

Table 28.8

Final radiological appearance Total

Perfect Osteophyte Osteophyte Osteo- Avascular Mal- Ectopic Surgical Other

appearance collarette collarette arthritis necrosis union ossification secondary
1-2mm 3-4mm over perfect congruence
reconstruction

Excellent 40 6 6 3 55
Very good 9 5 2 2 20
Good 11 14
Fair 8 2 10
Poor 23 8 7 39
Total 41 6 7 51 9 9 4 2 9 138

Table 28.9

Quality of Type of fracture Total

reduction
",=
'E" 'E" -0 E ; 'E"
"@
::l
-0 "@ ::l
-0 "'-
ro"@ ~u 0C:5 ::l .... ~
'"
i:; "
-o 0
.-
-
E
'"
....
u
....
.S 5'"
u
....
1;l
.... -0 "'"
.... '"
.... ....
01;;" u >
~ ::l
-0
0
C .... ·c
0
·c "> "
0. >
"
'" .-
0
!i
.- 0 o.... '"
._ 0 ~ ro u
~
C5
~
C5 ;::" ;::"
'"
'"
ro
....
ro
-5i '"
ro ~'"
.... 0
....

"'-0
0. ~ o..~
;:: -g § 1:
0
0.. 0.. ~ ~ f-< r.:. f-< 0. o '"
0.. ro ~ro.<:: ~

Perfect 27 7 2 2 8 10 30 4 2 12 104
Head centred 5 4 6 19
Loss of parallelism 3 5 13
Residual head 4
protrusion
Technical failure 3
Surgical secondary 2 5 8
congruence
Total 34 8 6 3 12 17 42 5 4 20 151 a

a Six cases are not included here: 4 in which immediate arthroplasty was performed and 2 in which other operations were carried
out (1 neurolysis, 1 arthrolysis)
Results 633

last follow-up. Fifty-one cases out of 138 (37070) 28.8 Conclusion

had radiological signs of osteoarthritis; nine
were clinically silent (6.5%) and the functional
result was very good. In the other 42 the
osteoarthritis was clinically present and respon-
sible for 11 good (8 %), eight intermediate
(5.9%) and 23 poor results (17%).
The quality of intra-operative reductions for
each type of fracture is displayed in Table 28.9,
while Table 28.10 relates the quality of intra-
operative reduction to the condition of the frac-
ture at the time of operation.
Table 28.10
Before dealing with the cases operated on more
than 4 months after injury, which offer the greatest
difficulties to surgeons, it seemed to us necessary to
report the results of surgical treatment carried out
between 3 weeks and 4 months after injury.
Apart from the increasing intra-operative dif-
ficulties - a consequence of the fast callus forma-
tion in most of the cases - the long-term results in
this group, even after perfect reduction, show that
the rate of very good functional results drops sig-
nificantly compared to that after operations per-
formed within 3 weeks of injury. This degradation
of the functional results should constitute a main
argument to promote early surgical treatment of
the fractured acetabulum.

Quality of reduction Condition of fracture healing Total

Fracture lines Non-union Mal-union/ Unreduced posterior Missed incarcerated

still visible non-union dislocation fragments

Perfect 64 14 5 18 3 104
Head centred 9 3 7 19
Loss of parallelism 9 2 13
Residual head 3 4
protrusion
Technical failure 2 3
Surgical secondary 6 2 8
congruence

Total 93 22 6 27 3 151 a

a See footnote to Table 28.9

29 Operative Treatment More Than Four Months After Injury
29.1 General Considerations and Condition
of Fracture Healing
In patients seen more than 4 months after injury,
remodelling of the fracture site is generally com-
plete. The initial landmarks on which perfect re-
construction depends have disappeared under the
callus formation, and consequently the difficulties
of open reduction and internal fixation are signifi-
cantly increased. Even today, in many places, re-
construction is abandoned in these cases in favour
of attempted salvage by total hip arthroplasty.
It must be said here that total hip replacement
as a treatment for mal-union, non-union or non-
union/mal-union is far from easy. If the total hip
is to be positioned at approximately its original an-
atomicallocation, much cement and/or bone graf-
ting is necessary. The irregular speed of healing of
the fracture lines of a broken pelvis means that one
may find areas of non-union even years later, and
these, together with the mal-united areas, make it
difficult to place the socket securely in a good posi-
tion.
If there is no other option but to insert a total
hip, the ideal is to osteotomise the mal-united sites
of the innominate bone in order to restore its anat-
omy as well as possible and then to fix the different
parts. To perform this, either screws and plates may
be used as usual, or a hemispherical metallic socket
can be placed, from the inside of which are inserted
the screws which fix the osteotomised parts and re-
establish the socket at its proper place. In Fig. 29.1
both means are used.
It was in 1962 that we began trying to restore the
acetabulum in fractures more than 4 months old.
After this delay, different kinds of lesions are
found, namely (Tables 29.1 and 29.2):
unreduced posterior dislocations of the femoral
head with an associated acetabular fracture;
mal-unions;
non-unions;
non-union/mal-unions;
missed incarcerated fragments requiring re-
moval.
However, it is the condition of the femoral head
and of the parts of the acetabular articular surface
which determines whether or not reconstruction
should be attempted; it is in fact the main prog-
nostic factor. There may be an associated femoral
head necrosis, and in our opinion, magnetic
resonance imaging has now become a prerequisite
before surgical treatment, at least in cases with
posterior dislocation of the head. It is the best tech-
nique available today to assess the vascularity of
the head, and it undoubtedly gives reliable infor-
mation more than 4 months after injury.
29.2 Preconditions for Surgery
For an acetabular fracture more than 4 months old
to be treated operatively, the following are pre-
requisites:
the femoral head must be of normal density.
However, localised surface damage or wear is
not considered a contra-indication (Fig. 29.2);
osteoarthritis should not have developed, al-
though a small collar of osteophytes does not
constitute a contra-indication;
on the standard radiographs and on the stan-
dard CT scan it is important that the com-
ponents of the sub-chondral bone surface be
clearly visible. It is impossible to reconstruct
unrecognisable topography (Fig. 29.3);
during the last year we have added: there must
be no avascular necrosis signals on magnetic
resonance imaging.
3-D CT reconstructions, although not a pre-
requisite, are, of course, extremely helpful.
636 Operative Treatment More Than Four Months After Injury

29.3 Time of Operation After Injury

c
E
We have attempted surgical reconstruction in 123
"u
"0 cases at variable times after injury:
.Q
(5
24 cases at 4-6 months (19.5%);
~ 40 cases at 6 months to 1 year (32.50,10);
38 cases at 1 - 2 years (30.8 %);
10 cases at 2-3 years (8.1%);
11 cases at more than 3 years (8.9%);
(these last 11 cases were 6 posterior wall fractures
- 4 of them with missed incarcerated fragments -
1 anterior wall fracture, 1 anterior column and
posterior hemitransverse, and 3 both-column frac-
tures).

29.4 Choice of Surgical Approach

The choice of approach for operating on these

cases is easy, as it is the same for all delayed recon-
structions, whatever they are:
the Kocher-Langenbeck approach allows treat-
ment of union problems in posterior wall and
posterior column fractures;
the ilio-inguinal approach allows correction of
healing problems in anterior wall and anterior
column fractures.
All other fractures with healing problems are
treated through an extended approach such as our
extended ilio-femoral.
or)

29.5 Surgical Techniques Employed

(Table 29.2)

29.5.1 Cases in Which Reconstruction

Was Impossible

In these instances of late presentation of acetabular

fractures, we always aimed initially to reconstruct
the acetabulum and reposition the femoral head.
However, in 22 cases it appeared impossible to
achieve this because of the damage to the femoral
head and we were obliged to combine the acetabu-
00
lar reconstruction, most often of the posterior wall,
C
tt-.:':::: with an arthroplasty - for a long time, a hemiar-
o \1
C..c throplasty of the JUDET type. In recent years we
-"
.~
.
~

""~
c u
o cu
....
<!)
have given up performing herni-arthroplasties and
combined the acetabular reconstruction with the
u-:: insertion of a total hip (3 cases).
Surgical Techniques Employed 637

Table 29.2

Surgical technique Condition of fracture healing Total

Unreduced Missed Non-union Mal-union Non-union/ Miscellaneous

posterior incarcerated mal-union
dislocation fragments

Relocation of head and 11 11

acetabular reconstruction
Excision and internal 7 7
fixation
Osteotomy and internal 26 14 40
fixation
Removal of intra -articular 16 16
fragment
Arthroplasty and 12 2 5 2 22
reconstruction
Arthrodesis 3 3 6
Other 4 2 7 7 21
Total 30 16 11 41 17 8 123

In 27 other cases of these 123 late-presenting 29.5.2 Cases in Which Reconstruction

fractures the articular and bony conditions found
at operation were such that our initial goal of re-
construction was impossible to achieve, and we had
to adjust our technique to the intra-operative find-
ings:
6 arthrodeses and 2 ilio-trochanteric coapta-
tions were performed;
in 3 cases fixation devices previously inserted
elsewhere were merely removed (an additional
incarcerated fragment was extracted in 2 cases)
and the patients were put on the continuous
passive motion machine night and day for 3
weeks;
in 6 cases we tried, often successfully, to im-
prove hip mobility by removing the edges of a
mal-united acetabulum which had resulted in a
central head protrusio;
in 1 instance a curved supra-tectal osteotomy
was performed to bring back the anatomical
roof against a centrally displaced femoral head,
thus creating a surgical secondary congruence
(Fig. 29.4);
in 9 other cases various methods were used
(Colonna's operation, removal of ectopic bone
associated with a capsulotomy, etc.).
These previous 49 cases will not be considered fur-
ther, but they have been included in the series in
this book because they helped us to a better under-
standing of the pathological anatomy of the frac-
tures and the outcome of unreduced acetabular
fractures.
Was Possible
The remaining 74 late-presenting cases will be
analysed more closely, including the 16 instances of
delayed removal of a missed incarcerated fragment.
29.5.2.1 Missed Incarcerated Fragments
The surgical treatment is generally easy, and if sur-
gery attempts and generally succeeds in relocating
the femoral head by removing the incarcerated frag-
ment, we do not do anything to the acetabular frac-
ture if it is slightly displaced or undisplaced.
There have been 16 cases in which fragments of
acetabular articular surface in the joint itself have
been missed. Months after injury, the hip becomes
progressively painful and stiff, and in most cases an
abduction and/or rotational deformity has
developed. Unless these fragments are removed,
osteoarthritic changes are inevitable, and I assume
that most such cases go unrecognised and are con-
sidered as cases of standard post-traumatic
osteoarthritis.
In our 16 cases, the fragment or fragments were
removed at the following intervals after injury: five
between 6 months and 1 year, six between 1 year
and 2 years, one between 2 and 3 years and four at
more than 3 years. In 13 out of the 16 cases, the
original injury was a posterior dislocation with an
unrecognised small fracture of the posterior wall,
in two cases there was a transverse fracture without
dislocation, and in one case an associated anterior
column and posterior hemitransverse fracture.
 638 Operative Treatment More Than Four Months After Injury

A B

Fig. 29.1 A-G. Associated transverse and posterior wall fracture

with posterior dislocation, operated on 5 months after injury.
A Antero-posterior radiograph, B obturator-oblique radio-
graph, and C iliac-oblique radiograph before operation, D CT
scan showing the considerable lesions to the head. MRI demon-
strated extensive supero-lateral avascular necrosis, E antero-
posterior radiograph, F obturator-oblique radiograph, and G
iliac-oblique radiograph after operation. The acetabulum has
been osteotomised, reconstructed and fixed by means of lag
screws and a screwed hemispherical metallic socket. A cemented
C total hip was then inserted
 Surgical Techniques Employed 639

E F

Fig. 29.1E-G G
640 Operative Treatment More Than Four Months After Injury

B c D

Fig. 29.2A-D. Posterior column fracture with localised surface terior radiograph, C obturator· oblique radiograph, and D iliac-
damage to the femoral head, operated on 120 days after injury. oblique radiograph 5 years after operation. The hip is rated 5.5.6
A Antero-posterior radiograph before operation, B antero-pos-
 Surgical Techniques Employed
Some fragments retain a capsular flap con-
nection which tethers them not far from the edge
of the acetabulum, and they therefore keep the
head in an inferiorly displaced position; they soon
become adherent to the articular cartilage of the
roof, whether they oppose to it their cartilage or
their bony aspect. These fragments have to be
removed through a Kocher-Langenbeck approach
(Fig. 29.5), as was done in six cases.
Fragments free from capsular attachments
migrate in the cotyloid fossa (10 cases) and may be
difficult to see on standard radiographs, but they
do not escape detection by tomography or CT scan-
ning. By friction, they cause wear of the cartilage
and even of the sub-chondral bone of the femoral
head. All but one of such fragments were removed
through a Smith-Petersen approach on the or-
thopaedic table. A T-shaped capsular incision com-
bined with longitudinal traction as allowed by the
table and additional lateral traction gave a suffi-
cient view of the cotyloid fossa to allow removal of
the fragment if it was free or fixed with fibrous
tissue, or to osteotomise its base if it had healed to
the fossa and/or the adjacent border of the anterior
wall.
Two of these ten incarcerated fragments were as-
sociated with a transverse fracture. One was re-
moved through a Smith-Petersen approach, the
other through the Kocher-Langenbeck as the low
transverse fracture was slightly displaced and the
missing piece of the posterior wall was clearly seen
on radiographs. In one case the incarcerated frag-
ment was observed to increase in size (Fig. 29.6)
and was removed 3 years after injury.
No intra-operative or early post-operative com-
plications were seen among these 16 cases. The
results were as follows:
- 2 cases were lost to follow-up within 1 year;
Of the 14 cases available for follow-up:
4 had an excellent result at a mean follow-up of
4 years (Fig. 29.7);
5 had a very good clinical result with radiologi-
cal signs of osteoarthritis, at an average of 5.2
years;
5 developed clinically evident osteoarthritis (in
1 case pre-existing at the time the fragment was
removed), giving 2 good results (at 4 and 17
years), 1 fair (at 8 years) and 2 poor results (at
10 and 20 years).
In all 16 hips function was seriously compromised
at the time of surgery. Of the 14 cases available to
641
follow-up, nine (64070) have a very good clinical
result with a mean follow-up of 4.6 years, while two
others are good at a mean follow-up of 11.5 years,
one is fair, and two are poor results.
29.5.2.2 Unreduced Posterior Dislocation
of the Femoral Head
Of 30 cases of unreduced posterior dislocation
treated more than 4 months after injury, in 19 in-
stances (included in the discussion in Sect. 29.5.1) a
hemi- or total arthroplasty was inserted. There re-
mained 11 instances in which the condition of the
femoral head appeared adequate (despite some
slight alterations of its anterior aspect left by the in-
sertions of the anterior tissue bindings) for an at-
tempt at preserving it and reconstructing the ace-
tabular fracture.
These 11 unreduced posterior dislocations ac-
companied:
a posterior wall fracture in 9 cases;
an associated transverse and posterior wall frac-
ture in 1 case;
an associated posterior wall and posterior col-
umn fracture in 1 case.
All but one were approached through the Kocher-
Langenbeck incision; the associated posterior col-
umn and posterior wall fracture was approached
through the extended ilio-femoral incision because
the fracture line separating the posterior column
was located high up.
The main problem in these cases, and the first to
be addressed, is the freeing of the posteriorly dislo-
cated head. The technique described in Sect. 28.3.4
in relation to fractures operated on between 3
weeks and 4 months after injury applies in every
respect to those treated more than 4 months after
injury. Extreme care has to be taken while freeing
the head not to damage the cartilage or aggravate
an already existing lesion of the head. In cases of
pure posterior wall fractures, once the head is
repositioned against the remaining intact part and
rotates perfectly, the posterior wall is reconstructed
using the original fragments found in the new wall
surrounding the head, and if a defect persists, it is
filled with an iliac crest bone graft. Isolated screws
and the usual buttress plate fix the reconstruction.
Where the dislocation accompanies a more com-
plex fracture, the osteotomy of the other compo-
nent has to be performed before the repositioning
and fixation of the displaced part. This was done
in the cases of the associated transverse and
642 Operative Treatment More Than Four Months After Injury

Fig. 29.3 A - C Both-column fracture seen 2

years after injury and obviously impossible to
reconstruct. A Antero-posterior radiograph,
B obturator-oblique radiograph, C iliac-oblique
radiograph
A
Surgical Techniques Employed 643

8 C

Fig. 29.3B,C
 644 Operative Treatment More Than Four Months After Injury

A B

Fig. 29.4A- H. Associated transverse and postero-superior wall

fracture extending to the iliac crest (the obturator ring is in one
piece). A Antero-posterior radiograph, B obturator-oblique ra-
diograph, and C iliac-oblique radiograph taken after imperfect
reduction performed elsewhere through a wrongly chosen ilio-
inguinal approach. The greater part of the roof is away from the
head. D Amero-posterior radiograph, E obturator-oblique ra-
diograph, and F iliac-oblique radiograph after operation. A
supra-articular curved osteotomy, above the roof and parallel to
the acetabular lip, was performed through an extended ilio-
femoral approach. After adjustment, the roof was brought back
against the head and fixed with a Vitallium plate. G Antero-
posterior radiograph and H obturator-oblique radiograph 6
c years after operation. The hip is rated 5.6.5
 Surgical Techniques Employed 645

D E

Fig. 29.4D - F F
646 Operative Treatment More Than Four Months After Injury

G H

Fig. 29.4G,H
Surgical Techniques Employed 647

A B

Fig. 29.5A,B. An incarcerated fragment, 180 days after injury. A Antero-posterior radiograph before operation, B antero-posterior
radiograph after operation
 648 Operative Treatment More Than Four Months After Injury

At

A4 AS Fig. 29.6AI-A5
 Surgical Thchniques Employed 649

Fig. 29.6 A-D. An incarcerated fragment in the

acetabular fossa that increased in size progressively.
A I-AS Serial antero-posterior radiograph showing an
incarcerated fragment in the acetabular fossa increas-
ing in size, B antero-posterior radiograph, C ob-
turator-oblique radiograph, D iliac-oblique radiograph
taken 6 years after operation

c D
650 Operative Treatment More Than Four Months After Injury

A B

Fig. 29.7 A-F. An incarcerated fragment, 9 months after injury. oblique radiograph, and F iliac-oblique radiograph taken 13
A Tomogram before the operation, B antero-posterior radio- years after operation. Recent radiographs show the same ap-
graph and C iliac-oblique radiograph taken immediately after pearance 25 years after operation
the operation. D Antero-posterior radiograph, and E obturator-
Surgical Techniques Employed 651

Fig. 29.7C-F

o F
652
posterior wall fractures, and the associated
posterior column and posterior wall fracture.
These 11 cases were operated upon at various in-
tervals after injury: two between 4 and 6 months,
eight between 6 months and 1 year, and one be-
tween 1 and 2 years.
The post-operative complications included one
case of post-operative sciatic palsy (partial poplite-
al) and two cases in which the head redislocated
without any apparent explanation. They were re-
duced surgically; one became infected, while the
other developed avascular necrosis of the head and
the posterior wall, requiring total hip replacement.
Avascular necroses developed in five cases.
Apart from the instance following redislocation,
mentioned above, we observed one case of limited
superior necrosis, one of extended superior necrosis
(Fig. 29.8), one case of massive head necrosis, one
case of cartilage necrosis, and one of associated
head and posterior wall necrosis. In two cases the
necrosis was obvious at 6 months, in two between
6 months and 1 year, and in the last case between
1 and 1.5 years.
Three patients developed osteoarthritic changes
without previous avascular necrosis; in one of them
this was deterioration of pre-operatively existing
signs of osteoarthritis.
Of these 11 patients, two were lost to follow-up
within 1 year (one broke his femoral neck on the
operated side at 4 months after surgery and a total
hip was inserted).
Two patients have good results at 6 and 14 years:
the first of them has slight deterioration of pre-
existing osteoarthritis, the second (in whom reposi-
tioning was performed 184 days after injury),
despite satisfactory radiographs, has some pain and
limps after walking a long distance (Fig. 29.9). In
one case the result is rated fair because of type IV
osteoarthritis. In six cases the results turned to
poor because of avascular necrosis (2 cases), infec-
tion (1 case) and osteoarthritis secondary to
avascular necrosis (3 cases).
Thus, the prognosis for a posteriorly dislocated
femoral head repositioned more than 4 months
after injury is poor: of the nine cases we were able
to follow-up, we saw good results in 22070, avascular
necrosis in 55% and osteoarthritis in 33% - in
addition to the fact that, of an overall 30 cases, we
considered that the condition of the head obliged
us to perform an immediate hemi- or total
arthroplasty in 19 (63%). All patients in whom
repositioning took place after more than 6 months
of dislocation had a poor result, and it is prob-
Operative Treatment More Than Four Months After Injury
ably better to recommend arthroplasty after this
date.
29.5.2.3 Non-union
Seven cases were operated on in which non-union
had followed conservative management comprising
several weeks of skeletal traction. The fractures
were two of the posterior column, three transverse,
one associated transverse and posterior wall, and
one both-column fracture (in which only the poste-
rior column fracture had not united). Since closure
of this statistical study we have seen one case of
non-union of a posterior wall fracture which was
successfully treated.
Non-unions of acetabular fractures are charac-
terised by a variable amount of pain when the hip
is actively mobilised and some limping. The passive
mobility of the hip may be normal in some sectors
of motion while being painful in others.
Radiographically the fracture line is clearly visible,
sometimes enlarged compared to its initial ap-
pearance; its edges are densified and may be hyper-
trophied in some cases.
The only way to stop progression to arthritic
changes is to operate to reduce and fix the non-
union. It is sometimes possible, when the edges of
non-union are hypertrophic, to perform a decor-
tication which will facilitate the healing process. In
any case, the fibrous tissue of the non-union site
must be totally excised. An extended ilio-femoral
approach and extended capsulotomy allowing per-
fect control of the excision is then performed from
the inside of the joint. The edges of the non-union
are "freshened up" and the non-union reduced us-
ing different tricks: for instance, two forceps main-
taining the two extremities of a transverse fracture.
We frequently use temporarily inserted 3.5-mm
screws with the heads left proud, onto which
clamps can be applied.
Once the reduction has been performed and
verified, intra-articularly if possible, fixation is
achieved as in fresh fractures, using 3.5- or 4.5-mm
lag screws transgressing the two cortices, and/or
3.5- or 4.5-mm screws (lagged or not) inserted
along the axis of the columns or between the two
tables of the iliac wing. Usually screws do not suf-
fice and we use buttress or compression plates
perfectly contoured to lie along the columns or the
iliac wing; they must always be fixed with 3.5-mm
screws.
The internal fixation must always be sound and
reliable, allowing immediate mobilisation and the
 Surgical Techniques Employed 653

A
B

D
Fig. 29.8A-D. Posterior wall fracture with posterior dislocation
of the femoral head, operated on 9 months after injury. A Ante-
ro-posterior radiograph before operation, B antero-posterior
radiograph, and C iliac-oblique radiograph after operation
showing the femoral head osteonecrosis, D the total hip replace-
ment (Legrange-Letournel prosthesis), undergone 3.5 years after
c reoositioninl! of the head
 654 Operative Treatment More Than Four Months After Injury

Fig. 29.9 A-H. Posterior wall fracture with

posterior dislocation of the femoral head,
operated on 184 days after injury. The fracture
was associated with an inter-trochanteric frac-
ture, which healed in varus. A Antero-posterior
radiograph, B obturator-oblique radiograph,
and C iliac-oblique radiograph before opera-
tion, D antero-posterior radiograph, E ob-
turator-oblique radiograph, and F iliac-oblique
radiograph 2.5 years after operation, G antero-
posterior radiograph and H iliac-oblique
radiograph 13 years after operation. The hip is
rated 5.6.5. The hip was repositioned 184 days
A after injury
 Surgical Techniques Employed 655

B c

Fig. 29.9B,C
 656 Operative Treatment More Than Four Months After Injury

E Fig. 29.9D - F
 Surgical Techniques Employed 657

G H

Fig. 29.9G,H
658 Operative Treatment More Than Four Months After Injury

Fig.29.10A-G. Transverse fracture showing non·

union 150 days after injury. A Antero·posterior
radiograph, B obturator-oblique radiograph, and C
iliac·oblique radiograph taken before operation (hip
function rated 2.5.3), D tomograms demonstrating
the non-union, E antero-posterior radiograph, F ob-
turator-oblique radiograph, and G iliac-oblique ra-
diograph 15 years after operation. The hip is rated
6.6.6
A
 Surgical Techniques Employed 659

B c

Fig. 29.10B-D D
 660 Operative Treatment More Than Four Months After Injury

E F

G Fig. 29.10 E-G

 Surgical Techniques Employed
use of continuous passive motion the 2nd post-
operative day. Walking without weight-bearing or
just with touch-down is allowed at about 8-15
days and full weight-bearing when radiological
healing is obvious.
These seven non-united fractures were ap-
proached through five Kocher-Langenbeck and two
extended ilio-femoral incisions. The immediate sur-
gical reduction was perfect in five cases, and imper-
fect in two. No avascular necrosis occurred, but two
patients developed osteoarthritis (one 10- 15 years
and the other 20- 25 years after surgery).
Healing was achieved in all seven cases. The
long-term results are as follows:
2 excellent at 8 and 15 years, respectively
(Fig. 29.10);
2 good at 5 and 12 years, respectively;
3 poor: 1 deteriorated at 2 years and needed a
total hip replacement, the other 2 were very
good for 20 years (Fig. 29.11) and 19 years,
respectively, and then deteriorated to poor. The
first of these patients has already undergone a
total hip replacement; the second is scheduled
to receive one.
A B
Fig. 29.11A,B. Pagetic non-union of the posterior column.
A Antero-posterior radiograph 6 months after operation: decor-
ticated and plated through a posterior approach, it healed. See
Fig. 6.12 for pre-operative condition, B antero-posterior radio-
graph 20 years later: the hip has deteriorated and needs a total
661
29.5.2.4 Mal-union
Twenty-six cases of true mal-union were recon-
structed more than 4 months after injury. The frac-
tures involved were four of the posterior wall, two
of the anterior column, four transverse, six T-
shaped, four transverse and posterior wall, two as-
sociated posterior wall and posterior column and
four both-column fractures.
Undoubtedly some mal-unions are tolerated and
cause no symptoms for years, and some will go on
like this, probably for ever. These are the ones
which create little post-traumatic incongruence, the
femoral head rolling under a sufficient part of the
articular surface (such as in low transverse frac-
tures, low anterior column fractures or small
posterior wall fractures) and, paradoxically, those
belonging to the most complex type of both-col-
umn fractures but in which post-traumatic second-
ary congruence is achieved (see Sect. 14.1.5). How-
ever, when a mal-union is accompanied by clinical
symptoms such as increasing pain or restriction of
hip motion, the only way to try to prevent progres-
sion towards osteoarthritis is to restore the normal
congruence of the hip joint surgically.
hip replacement. However, on the other side the patient under-
went total hip replacement 12 years earlier. Pagetic disease is to
blame for the present osteoarthritis, as it is for the earlier non-
union
 662
A complete radiological study including stan-
dard radiographs, tomograms and CT scanning is
necessary for the pre-operative planning. Some-
times arthrography combined with CT is useful. In
these cases 3-D reconstruction and the correspond-
ing plastic model are advisable in order to plan the
reconstruction better.
The ideal treatment is to restore perfect spherici-
ty to the acetabulum by means of one or more
osteotomies and to fix them with reliable, solid in-
ternal fixation. However, even if this is successfully
achieved, the restoration of congruity depends
upon one other essential factor: the condition of
the femoral head. If wear is present, involving only
the cartilage, or involving the sub-chondral bone as
well, re-establishing the parallelism of the joint
space may be impossible. What the consequences
of wear of the head will be on a perfectly restored
acetabulum is unpredictable; undoubtedly in some
instances, despite perfect restoration of the spheric-
ity of the acetabulum, verified as far as possible in-
tra-operatively, osteoarthritis sets in or continues to
progress, increasing in degree with time. Never-
theless, one thing remains true: the only technical
possibility available today is to restore the spherici-
ty of the acetabulum, since so far all attempts to
compensate for loss of substance of the femoral
head (e.g. bone bank grafting) have failed.
The osteotomies needed to correct a mal-union
consist in removing the callus which has filled up
the gap between the fragments, and in fact a wedge
of new bone has to be excised, the shape of which
depends on the nature of the displacement of the
fragments. Standard radiographic views, tomo-
grams and 3-D reconstruction help in determining
A B
Fig. 29.12A, B. Intra-operative views through an extended ilio-
femoral approach. A The osteotomy is performed with a car-
Operative Treatment More Than Four Months After Injury
the shape of the required wedge. From experience
we have learnt that the healing process is much
quicker at the bone level than at the cartilage level
- i.e. for a long time the track of the original frac-
ture line is easily recognisable from the inside of the
joint. Consequently the osteotomies are best per-
formed from the inside of the joint (Fig. 29.12).
These osteotomies may be conducted with a chisel
of the carpenter's type or with an oscillating saw.
The results of treatment of the 26 mal-united
fractures were as follows.
This series included only four mal-united poste-
rior wall fractures coexisting with a symptomatic
hip. We presumed that a possible instability of the
hip joint or decrease of the acetabular articular sur-
face was responsible for the clinical symptoms.
Through the Kocher-Langenbeck approach, the
displaced fragment was osteotornised, reshaped,
repositioned and fixed as usual with plates and
screws. The results were not rewarding. In two cases
there were operative signs of osteoarthritis and the
operation did not stop their progression; one was
rated fair at 6 years, the other underwent a Mac-
Murray osteotomy 1 year after operation, remained
good for 18 years, and needed a total hip replace-
ment at 21 years. The result in the third case
deteriorated to poor at 4 years, while one case has
a good result at 21 years in spite of clinical and
radiological signs of osteoarthritis.
Two anterior column mal-unions were treated,
one through the ilio-inguinal approach, one
through an extended ilio-femoral incision already
used by a previous surgeon. The wedge osteotomy
was performed without a direct intra-articular view
but with intra-operative radiographs. Fixation of
penter's chisel from the inside of the joint, B the result of the
osteotomy: the sphericity of the acetabulum is restored
Surgical Techniques Employed
the column was in the standard way. The result in
one case is rated only good with satisfactory
radiographs at 2 years; the other is rated very good
clinically with radiological type IV osteoarthritic
changes at 4 years (Fig. 29.13).
Four transverse fractures were treated through
the extended ilio-femoral approach, which allows a
controlling view of both extremities of the fracture
line and the intra-articular reduction. The access to
the quadrilateral surface through the greater sciatic
notch, in combination with the access to the inter-
nal iliac fossa provided by the elevation of the iliac
muscle, permits safe performance of the wedge os-
teotomy through the iliac wing, the pelvic brim and
the quadrilateral surface.
When the wedge has been excised, the transverse
fracture is reduced and fixed with two clamps, one
astride the greater sciatic notch, the other astride
the anterior edge of the bone. Fixation is achieved
using 3.5- or 4.5-mm lag screws, gaining purchase
on both cortices to maintain the posterior part of
the transverse fracture line, and 4.5- or 6.5-mm
screws inserted along the axis of one or both col-
umns. In my opinion, an additional buttress plate
applied along the posterior column is always neces-
sary.
Of these four cases:
2 have an excellent result at 4 and 10 years,
respectively (Fig. 29.14);
1 has a clinically good result at 10 years, with
radiological signs of osteoarthritis;
1 was lost to follow-up in less than 1 year.
Six mal-united T-shaped fractures were operated
upon: four through a Kocher-Langenbeck and two
through an extended ilio-femoral approach. In one
case we ignored the vertical stem line and, through
the Kocher-Langenbeck approach, only the trans-
verse fracture was corrected and fixed. The result at
11 years is fair, due to osteoarthritic changes. In the
cases treated through the extended approach, in ad-
dition to the wedge osteotomy of the transverse
component, the cotyloid fossa osteotomy was need-
ed to allow perfect reduction of the segments of the
two columns (the combination of longitudinal and
lateral traction allowing satisfactory access to the
cotyloid fossa). If necessary, the ischio-pubic ra-
mus is osteotomised close to the ischial tuberosity.
Once correction has been achieved, fixation is iden-
tical to that performed in a fresh fracture.
663
The results in these cases are:
1 excellent at 3 years (Fig. 29.15);
1 good at 11 years;
1 fair at 11 years with osteoarthritic changes
(Fig. 29.16);
3 poor because of osteoarthritis: 1 at 1 year, 1
at 3 years (this patient had pre-operative signs
of osteoarthritis), and 1 at 13 years.
Four mal-united associated transverse and posteri-
or wall fractures were operated upon through our
extended approach. After osteotomy of the
posterior wall fragment, the wedge osteotomy of
the transverse fracture was possible with a good
controlling intra-articular view and satisfactory
reduction and fixation were achieved.
The results in these cases were:
1 clinically very good at 7 years (Fig. 29.17);
1 good with radiological signs of osteoarthritis
at 1 year;
1 fair at 1 year because of osteoarthritis;
1 poor because of avascular necrosis of the head
and posterior acetabulum.
Two mal-united associated posterior column and
posterior wall fractures were approached through
the Kocher-Langenbeck incision. In these fractures,
after osteotomising the posterior wall fragment -
keeping its capsular attachments if possible - the
mal-union of the posterior column is exposed and
osteotomised through the posterior column and the
cotyloid fossa, the ischio-pubic ramus being
osteotomised close to the ischial tuberosity. Com-
bined longitudinal and lateral traction allow a good
view of the inside of the hip when the posterior
fragment is retracted. The posterior column is fixed
with a plate once its reduction appears perfect, and
the posterior fragment is repositioned after the nec-
essary reshaping and fixed with screws and an addi-
tional buttress plate.
One case has an excellent result at 5 years, the
other is very good with radiological signs of osteo-
arthritis at 11 years.
In four cases of both-column mal-union,
marked incongruence prompted us to attempt
restoration through the extended ilio-femoral ap-
proach. As will readily be understood, mal-union
of both-column fractures offers the greatest dif-
ficulties. Osteotomising all the fracture lines would
be extremely difficult and has never been attempted
to my knowledge, certainly not in our practice. In
these four cases we attempted to osteotomise some
of the fracture lines in order to try to restore joint
 664 Operative Treatment More Than Four Months After Injury

A 8

Fig. 29.13A-C. Mal-union of an intermediate anterior column

fracture. A Antero-posterior radiograph 7 months after injury.
Some wear of the femoral head is already visible, B antero-
posterior radiograph 13 days after operation (ilio-inguinal
approach). The restoration of the landmarks is satisfactory, but
the head has not recovered its normal position because of the
pre-operative wear, Cantero-posterior radiograph 4 years after
operation. The hip is rated 5.6.6, but type IV osteoarthritis is
present. At 6 months we removed the plate as CT scanning
C showed that the long screw was lying against the head
Overview of the 123 Cases Treated More Than Four Months After Injury
congruence around a femoral head that was still
centrally displaced. The head was left in contact
with the largest displaced part of the articular sur-
face and the rest of the displaced, non-congruent
parts were osteotomised, brought back against the
head and fixed. Thus we created surgical secondary
congruence, accepting mal-union of the in-
nominate bone.
As to the results:
1 case is recent (in fact it quickly showed a poor
result);
1 is good at 4 years;
1 is fair at 4 years (osteoarthritis);
1 was bad at 5 years and needed a total hip re-
placement, which was performed with the ace-
tabulum in a much better condition than it was
in at the time of the reconstruction.
29.5.2.5 Non-union/Mal-union
The speed of the healing process varies a lot, for
unknown reasons, in different parts of a broken
pelvic bone. This is why a combination of mal-
union and non-union is a frequent feature of late-
presented acetabular fractures. Even a single frac-
ture line such as the transverse fracture component
of a T-shaped fracture may unite posteriorly and
not unite anteriorly (see Fig. 29.15) or vice versa
(Fig. 29.18).
The combination of different degrees of union is
easily analysable from the standard radiographic
views and CT scans. Surgical treatment will require
the combination of the two techniques discussed
above: excision of a wedge at mal-union sites, and
of the fibrous tissue at non-union sites, always with
the same goal of restoring perfect sphericity to the
acetabulum.
Fourteen cases of non-union/mal-union were
operated upon, comprising: three posterior column
fractures, one anterior column, two T-shaped, two
transverse and posterior wall, and six both-column
fractures.
Mal-union/non-union of posterior column frac-
tures is peculiar (see Fig. 28.6). Excision of all the
bony bridges is the precondition for perfect restora-
tion of the anatomy.
The results vary according to the fracture type:
Of the 3 posterior column fractures, all 3
had poor results at 1 year due to osteoarthritic
changes (in 1 case there was a deterioration of
pre-operatively existing signs of osteoarthritis);
665
the anterior column fracture had a fair result at
2 years because of osteoarthritis;
of the 2 T-shaped fractures, one is recent and 1
has a very good result with radiological signs of
osteoarthritis at 10 years (Fig. 29.18).
of the 2 associated transverse and posterior wall
fractures, 1 has a good result with osteoarthritis
at 5 years and 1 a fair result at 3 years because
of osteoarthritis;
of the 6 both-column fractures:
1 case has been lost to follow-up;
3 have very good results, 1 at 10 years
without osteoarthritis (Fig. 29.19), one at 5
years without deterioration of the pre-exist-
ing osteoarthritis, and 1 at 9 years despite
radiological signs of osteoarthritic changes
(Fig. 29.20);
1 has a good result at 1 year;
1 has a poor result because of osteoarthritis
at 3 years.
29.6 Overview of the 123 Cases Treated
More Than Four Months After Injury
29.6.1 Reconstruction Impossible (49 Cases)
In 27 cases (21.90/0) the ideal of reconstruction was
regarded as unattainable. Instead, many other pro-
cedures were carried out, as detailed in Sect. 29.5.1.
Some were simple, such as cheilectomy of a case of
central head protrusion which was impairing func-
tion, the hip remaining very good 25 years after
operation; or the simple removal of intra-articular
hardware followed by the use of continuous passive
motion night and day for 3 weeks in order to try to
restore the altered cartilage. The case should also be
mentioned of a lady with a high anterior column
fracture which healed at the iliac wing level but not
along the track of the fracture line through the
psoas gutter. She complained of a click and pain at
about 60 0 hip flexion. One just had to hit on the
idea that the ilio-psoas tendon might be dropping
into the fracture line, causing the click and the
localised non-union. Through an ilio-femoral ap-
proach this diagnosis was verified and the non-
union was excised and filled up with graft. The
click and pain disappeared; the result is excellent at
10 years (Fig. 29.21).
In 22 other cases, the conditions of the head
and/or the acetabulum make necessary a hemiarth-
roplasty or a total hip replacement.
 666 Operative Treatment More Than Four Months After Injury

A B

C o

Fig. 29.14A-G. Mal-union of a juxta-tecta! transverse fracture,
operated on 8.5 months after injury. A Antero-posterior radio-
graph before operation. The clinical symptoms prompted us to
try to restore congruency, B antero-posterior radiograph 9 days
after operation. The joint was approached through an extended
ilio-femoral approach. The restoration of the landmarks and of
the head is satisfactory, C. D CT scans demonstrating the mal-
union of a typical antero-posteriorly directed fracture line and
the enlargement of the bone. E Antero-posterior radiograph,
F obturator-oblique radiograph, and G iliac-oblique radiograph
4 years after operation. The hip is rated 6.6.6. There is a col-
larette of osteophytes (1 mm thick) around the head
 Overview of the 123 Cases Treated More Than Four Months After Injury 667

E F

Fig. 29.14E- G G
668 Operative Treatment More Than Four Months After Injury

c D
Overview of the 123 Cases Treated More Than Four Months After Injury 669

29.15A-F. T-shaped fracture operated on 120 days after injury before operation. D Antero-posterior radiograph, E obturator-
(posterior approach). A Antero-posterior radiograph, B obtura· oblique radiograph, and F iliac-oblique radiograph taken 3 years
tor-oblique radiograph, and C iliac-oblique radiograph taken after operation
670 Operative Treatment More Than Four Months After Injury

A B

D E F

Fig. 29.16A-F. T·shaped fracture with a juxta-tecta! transverse radiograph taken before operation, D antero-posterior radio-
fracture component, operated on 15 months after injury graph, E obturator-oblique radiograph, and F iliac-oblique
(Kocher-Langenbeck approach). A Antero-posterior radio- radiograph taken 11 years after operation. The hip is rated 5.6.5
graph, B obturator-oblique radiograph, and C iliac-oblique
 Overview of the 123 Cases Treated More Than Four Months After Injury 671

A B

Fig. 29.17 A-H. Associated transverse and posterior wall fracture. A Antero-
posterior radiograph and B obturator-oblique radiograph 5 months after injury,
when the hip had a very poor function rated 4.4.4, C CT scan showing that the
head is not at all centred. D Antero-posterior radiograph, E obturator-oblique
radiograph, and F iliac-oblique radiograph 14 days after operation. The joint
space has not recovered its width, especially on the antero-posterior view,
G antero-posterior radiograph 2 years after operation. The hip is rated 6.6.6,
H antero-posterior radiograph 7 years after operation. Function is excellent.
Arthritic changes are present, but the enlargement of the joint space, for which
we have no explanation, has been stable for 9 years now (last review not shown
here)
C
 672 Operative Treatment More Than Four Months After Injury

o E

F Fig. 29.J7D-F
 Overview of the 123 Cases Treated More Than Four Months After Injury 673

G H

Fig. 29.17G,H
 674 Operative Treatment More Than Four Months After Injury

A B

Fig. 29.18A- F. T-shaped fracture with a juxta-tectal transverse

component. A Antero-posterior radiograph, B obturator-
oblique radiograph, and C iliac-oblique radiograph 6 months
after injury. The hip function was extremely bad and rated 4.2.4.
The transverse fracture had united anteriorly but not posterior-
ly. Through an extended ilio-femoral approach the anterior col-
umn mal-union was osteotomised and a wedge of bone excised,
D antero-posterior radiograph, E obturator-oblique radiograph
and F iliac-oblique radiograph 10 years after operation. The hip
is rated 6.6.6. Osteoarthritis is radiologically present but has
c been clinically silent and stable for the last 6 years
 Overview of the 123 Cases Treated More Than Four Months After Injury 675

D E

Fig. 29.18D-F F
 676 Operative Treatment More Than Four Months After Injury

Fig. 29.19A-F. Non-union/mal-union

of a both-column fracture accom-
panied by functional troubles in the
hip and significant shortening.
A Antero-posterior radiograph,
B obturator-oblique radiograph,
and C iliac-oblique radiograph before
operation. D Antero-posterior
radiograph, E obturator-oblique
radiograph, and F iliac-oblique
radiograph 10 years after treatment
through an extended ilio-femoral
approach. Function is very good; the
A hips are level with each other

B C
 Overview of the 123 Cases Treated More Than Four Months After Injury 677

Fig. 29.19D - F

E F
 678 Operative Treatment More Than Four Months After Injury

A 8

Fig. 29.20A-G. Mal-union/non-union of a both-column frac-

ture operated on 6 months after injury. There was 2 cm shorten-
ing. The patient had not been allowed to bear weight for 6
months preoperatively. A Antero-posterior radiograph, B ob-
turator-oblique radiograph, and C iliac-oblique radiograph
before operation. Note the sciatic buttress making a separate
fragment which is united to the sacrum. D Anteror-posterior
radiograph 17 days after operation through an extended ilio-
femoral approach. E Antero-posterior radiograph, F obturator-
oblique radiograph, and G iliac-oblique radiograph 9 years after
operation. Osteoarthritic changes have set in but the function is
c very good; the leg length discrepancy has been corrected
Overview of the 123 Cases Treated More Than Four Months After Injury 679

G
 680 Operative Treatment More Than Four Months After Injury

A B

Fig. 29.21 A-E. Non-union of the track of a pure high anterior

column fracture through the anterior wall of the acetabulum.
A Antero-posterior radiograph, B obturator-oblique radio-
graph, and C tomogram before operation. The tomogram shows
the non-union well. The ilio-psoas tendon dropped into this
fibrous gap at 60 0 hip flexion, producing a click and pain.
D Antero-posterior radiograph and E obturator-oblique radio-
graph 10 years after operation through an ilio-femoral ap-
proach, in which the non-union was excised and filled up with
c graft held by a short plate. Excellent to-year result
 Overview of the 123 Cases Treated More Than Four Months After Injury 681

D E

Fig. 29.21D,E
682 Operative Treatment More Than Four Months After Injury

29.6.2 Repositioning of Posteriorly Dislocated wounding of the nerve through a Kocher-Lan-

Femoral Head (11 Cases) genbeck approach and recovered partially. The
other followed an ilio-inguinal approach with
Repositioning of a persistently dislocated head was freeing of the internal iliac fossa; it is in the pro-
followed by reconstruction of the posterior wall. cess of recovering.
This satisfactory treatment did not yield satisfacto-
These 47 cases were operated on at the following in-
ry results: of the nine cases we were able to follow-
tervals after injury:
up, only two had good results, at 6 and 14 years
(one had pre-existing signs of osteoarthritis), one is 13 cases between 4 and 6 months;
rated fair because of osteoarthritis, 6 cases turned 15 cases between 6 months and 1 year;
to poor (2 cases of avascular necrosis, 3 of osteoar- 12 cases between 1 and 2 years;
thritic changes, 1 infection). 4 cases between 2 and 3 years;
3 cases after more than 3 years.
As to the late complications:
29.6.3 Missed Incarcerated Fragment (16 Cases)
(a) Avascular necrosis (2 cases)
A missed incarcerated fragment causing pain and 1 patient with superior segmental extended
hip contracture in 16 cases was removed through an necrosis has a fair result;
anterior or posterior approach. Of the 14 cases 1 patient with associated head and posterior
available to follow-up, four had an excellent result, wall necrosis needed a total hip replacement
five were very good with radiological signs of os- within 2 years.
teoarthritis and five developed clinical osteoar-
thritis leading to two good, one fair and two poor (b) Osteoarthritis (28 cases, i.e. 570/0):
results.
8 patients had pre-operative radiological
signs of osteoarthritis; they are considered
29.6.4 Mal-unions, Non-unions, as having deterioration of pre-existing os-
Mal-union/N on-unions teoarthritis which was clinically evident be-
tween 1 and 6 years post-operatively. At
In 47 cases of faulty union (not counting that of their last review there were 4 poor results, 1
the posterior wall fractures accompanying persis- fair (at 6 years), 2 good (at 4 and 10 years)
tent posterior femoral head dislocation) recon- and 1 still very good with radiological signs
struction of the acetabulum was attempted: more of osteoarthritis;
precisely, in 7 cases of non-union, 26 of mal-union 20 other patients developed osteoarthritis
and 14 of non-union/mal-union. The surgical ap- even after satisfactory reconstruction: 2
proaches used were: 25 Kocher-Langenbeck, 3 ilio- within 6 months after operation, 2 between
inguinal, 11 extended ilio-femoral and 8 extended 6 months and 1 year, 6 between 1 and 1.5
ilio-femoral with exposure of the internal iliac years, 8 between 3 and 6 years, 1 case after
fossa. 15 years, and the last one after 25 years. In
In these 47 operations there were a few intra- these last 2 cases, the part played by the
operative complications: acetabular fracture in the genesis of the
2 injuries to the superior gluteal artery and 1 to osteoarthritis is open to discussion. The re-
superior gluteal veins without sequelae; sults among these osteoarthritic patients
1 injury to the sciatic nerve, partial section of its are: 8 poor, 5 fair, 4 good and only 2 very
lateral part. good with purely radiological osteoarthritis
at 10 years (see Fig.29.18) and 4 years,
Post-operative complications were:
respectively. The last of these patients who
1 superficial infection treated surgically and had some wear of the head at surgery
which healed uneventfully; developed an abduction contracture which
1 case of pulmonary embolism; was treated surgically by an adductor
1 case of phlebitis; tenotomy. The result is very good clinically
2 post-operative sciatic palsies of the extensive with some radiological signs of osteoar-
patchy type. One followed intra-operative thritis 9 years later (see Fig. 29.17).
Overview of the 123 Cases Treated More Than Four Months After Injury 683

Of these 28 patients with osteoarthritis, 15 were ....o

found to have important lesions of the femoral
head intra-operatively:
4 had wear of the head along a fracture line;
- 11 had a variable area of exposed and worn
sub-chondral bone.
(c) Ectopic ossification (5 cases)
In the five cases with ectopic ossification, there
were two oftype I ossification (1-6,1-6), two of
type II (11-6, 11-5), and one of type III (III-5).
The ossification occurred after 4 out of 19
lateral approaches (36%) and 1 out of 25
Kocher-Langenbeck approaches (4070).
It might be useful to mention here that, overall,
of the 123 patients treated more than 4 months
after injury, 12 developed ectopic bone ossifi-
cation:
2 (11-5, 11-5) occurred after a total of 63
Kocher-Langenbeck approaches (3.2%);
5 (1-6, 1-6, 11-6, 11-6, III-5) occurred after 23 ...o
o
lateral approaches (21.8%); Il.

5 (11-4, 11-6, 11-4, 1-6, IV-1) occurred after

33 Smith-Petersen or other anterior ap-
proaches (15.6%).
The rate of occurrence of ectopic ossification is
thus much lower than among cases treated
within 3 weeks after injury.
Of the total of 123 patients:
102 received no preventive treatment and 8
of these developed ectopic ossification (5
type II, 2 type III, 1 type IV): a rate of "0
o
9.2%; o
OIl
10 were given 20 mg Didronel and 4 of these
developed ectopic ossification (2 type I, 2 $
type II): a rate of 40%;
7 were given indomethacin and 4 others in-
domethacin and radiotherapy; none of
these developed ectopic ossification.
(d) Long-term results
Two patients were lost within 1 year after
operation and two operations were recent, so 43
results are reported in detail in Thble 29.3.

.!
Overall among the mal-union and non-union '"
cases, we have:
7 excellent results (mean follow-up 7.4 ....0.o
years): 16.3070; ~
6 very good results, 5 of these patients hav- ~
ing radiological osteoarthritis (mean follow- "3 ~
up 8 years): 13.9%; ~ Eo-<
684
11 good results (mean follow-up 8.4 years):
25.6070;
6 fair results (mean follow-up 5.5 years):
14%;
13 poor results (mean follow-up 5.8 years):
30.2%.
29.7 Conclusion
Are such results rewarding enough or good enough
to justify carrying on in this direction? Of course
we answer "Yes". We believe that abandoning these
cases and just waiting for a total hip replacement
is not a solution, for several reasons. In the first
place the "total hip" approach to attempting
salvage of a broken acetabulum is often extremely
difficult and its outcome far from the best. Besides,
the long-term problems of a total hip prosthesis, of
whatever type, are unpredictable. Already we can
say that we have saved or postponed total hip re-
placement in 24 of the 47 patients with mal-union/
non-unions (51 %), 12 of the 16 with incarcerated
Operative Treatment More Than Four Months After Injury
fragments (75%), and 22% of those with unre-
duced posterior femoral head dislocations.
The overall results may appear poor or average
when one thinks of the price of surgery, and it is
true that up to today, even in the hands of special-
ists, these cases still involve difficult or very dif-
ficult problems which have not all been solved. I do
think that these cases must be treated by experi-
enced surgeons in the field of acetabular fractures,
but also that these surgeons must carryon improv-
ing the techniques of reduction and post-operative
care. It is wise not to begin one's experience by
operating on these cases, and not to try to reduce
and fix them before good practice has been ac-
quired in treating fresh fractures. And I do believe
that the reward of this surgery is the number of pa-
tients whom we save total hip replacement, or for
whom we at least delay it. Furthermore, in restoring
the acetabular anatomy we ensure that, even if
osteoarthritis sets in as it still does occasionally, the
insertion of a total hip prosthesis will be easier and
its prognosis considerably improved by the greater
reliability of the socket fixation.
30 Exercises in Radiographic Diagnosis
The following examples of fractures of the acetabu-
lum are from the authors' collection. This Appen-
dix is designed to improve the reader's ability to
define the morphology of the various fracture com-
plexes.
The radiographs are accompanied by descrip-
tions and illustrated by line drawings. In order to
derive the greatest benefit from these examples, the
reader should study the radiographs and reach a
conclusion before consulting the descriptions and
line drawings.
1 Associated posterior wall and posterior column
fractures. The main fracture line detaching the pos-
terior column passes inferiorly through the body of
the ischium and the obturator foramen remains in-
tact. A segment of the quadrilateral surface of the
ischium has been elevated separately.
2 Juxta-tectal transverse fracture. There is a frac-
ture of the opposite pubic rami.
3 Intermediate anterior column fracture associated
with an incomplete hemi-transverse posterior frac-
ture. The incomplete hemi-transverse component is
most clearly seen in the antero-posterior view. It
runs obliquely upwards and inwardly and does not
cut the posterior border of the pelvic bone.
4 Juxta-tectal transverse fracture associated with a
posterior wall fracture. Initially there was a posteri-
or dislocation of the hip and the views here show
the condition after reduction of this dislocation.
5 Both-column fracture. The iliac wing component
reaches the iliac crest. The anterior column compo-
nent extends in one piece from the anterior superior
iliac spine to the superior pubic ramus. The margin
of the greater sciatic notch present unusually, hav-
ing pivotted about 90 0 in both antero-posterior
and iliac wing views. The sacro-iliac joint is not dis-
turbed.
6 Posterior wall fracture with marginal impaction.
The impacted fragment is seen in the anterior
posterior view lying parallel to the supero-medial
margin of the femoral head.
7 Both-column fracture. The iliac wing component
extends to the iliac crest and isolates a triangular
fragment of the wing. The posterior column frag-
ment is in one piece and includes the inferior part
of the sacroiliac joint (more frequently this con-
stitutes a separated fragment. See Chap. 14.4, a).
No spur sign is visible, which is again unusual.
8 Low anterior column fracture associated with a
hemitransverse posterior fracture. The anterior col-
umn fragment is further sub-divided into three
parts: the anterior wall fragment itself, the body of
the pubis and a segment of the ischio-pubic ramus.
9 Posterior column fracture associated with an un-
displaced anterior hemitransverse component. This
association is similar but not identical to a "T-
shaped" fracture (see page 140).
10 Juxta-tectal transverse fracture. There is some
comminution in the posterior part of the fracture
line which cuts the edge of the pelvis bone just
above the ischial spine.
11 Vertical T-shaped fracture associated with a
posterior wall fracture. The posterior fragment ap-
pears superimposed on the inner part of the roof
overlying the upper part of the femoral neck.
12 Low fracture of the anterior column. There are
two supplementary fragments, one comprising a
posterior segment of the ilio-pectinealline and the
other comprising part of the quadrilateral surface
of the ischium which has been elevated and is
hinged posteriorly.
13 Infra-tectal transverse fracture. There is mini-
mal displacement.
686
14 Vertical T-shaped fracture. There is a double
fracture in the ischio-pubic ramus and additionally
a segment of the anterior wall has been separated.
Marginal impaction of the inner part of the roof is
seen on the antero-posterior view.
15 Posterior wall fracture associated with an undis-
placed low transverse fracture. The transverse frac-
ture component crosses the posterior border of the
pelvic bone inferior to the ischial spine.
16 Exceptional example not included in our classi-
fication. The head of the femur has dislocated cen-
trally, having fractured the inferior part of the floor
of the acetabulum, but the ilio-pectineal line and
the anterior and posterior walls remain intact. The
inferior part of the ilio-ischial line is seen to be
fractured.
17 Juxta-tectal transverse fracture, associated with
a pertrochanteric fracture of the femur. Additional-
ly, there is a small detached posterior wall frag-
ment. An undisplaced fracture line extending to the
iliac crest can be seen which does not involve the
roof of the acetabulum which remains intact. The
head of the femur is posteriorly dislocated and was
buttonholed through a deficiency in the posterior
capsule.
18 Both-column fracture. The iliac wing fracture
complex comprises a 'Y' configuration. The
posterior limb of this does not reach the iliac crest.
The fracture at the root of the superior pubic ramus
involves the roof of the obturator canal and enters
the hip joint.
19 Associated anterior wall and posterior hemi-
transverse fractures. The fracture line through the
ischio-pubic ramus has isolated the lower part of the
anterior column. Marginal impaction has occurred
at the inner part of the roof of the acetabulum.
20 Extended anterior column fracture. There is a
secondary fracture line which divides the wing seg-
ment of the anterior column and extends to the
anterior border of the bone between the anterior il-
iac spines. There is some comminution of the inner
table of the iliac wing and, from the quadrilateral
surface of the ischium, a posteriorly hinged plaque
of bone has been elevated. Inferiorly, the anterior
column is interrupted by fracture lines through the
root of the superior pubic ramus, the body of the
pubis and the ischio-pubic ramus. The posterior
Exercises in Radiographic Diagnosis
border of the acetabulum and the posterior border
of the pelvic bone are both intact.
21 Both-column fracture with iliac wing compo-
nent extending to the iliac crest. The anterior col-
umn is split through the anterior wall and there is
a detached fragment from the internal iliac fossa.
22 Associated posterior column and postero-supe-
rior wall fractures. There is an associated fracture
of the superior pubic ramus; this case was compli-
cated by a rupture of the urethra.
23 Juxta-tectal transverse fracture associated with
a fracture line extending from the retro-acetabular
surface to the iliac crest. The upper segment of the
articular surface is associated with the anterior
fragment of the wing, which itself is split secon-
darily by a fracture line reaching the interspinous
notch. This fracture should be compared with the
typical fracture line of a both-column fracture.
24 Undisplaced infra-tectal transverse fracture as-
sociated with a postero-superior wall fracture. The
femoral head is dislocated posteriorly.
25 Both-column fracture with iliac wing compo-
nent extending to the iliac crest. The spur sign is
obvious in the obturator-oblique view. The posteri-
or column, markedly displaced inwardly, was in-
carcerated in the true pelvis and rested against the
sacro-iliac joint. Two operative approaches were
needed in order to reduce this fracture.
26 T-shaped fracture. There is a marked marginal
impaction of the inner part of the roof.
27 Posterior column fracture. Typically, the tear-
drop is not involved.
28 Both-column fracture with iliac wing compo-
nent extending to the crest and involving the sacro-
iliac joint. The posterior column includes the in-
ferior half of the sacro-iliac joint. (See also Exam-
ple 7). Two operative approaches were needed for
reduction in this case.
29 Anterior T-shaped fracture. There was com-
minution of the anterior component of this frac-
ture. Two approaches were necessary for reduction,
the Kocher-Langenbeck being employed first.
Exercises in Radiographic Diagnosis
30 Anterior T-shaped fracture. In the obturator-
oblique view, the ilio-pectineal line seems un-
disturbed but in the antero-posterior view a frac-
ture through this line is obvious. A small segment
of the ilio-pectineal line is separated by a double
fracture of the pelvic rim.
31 High anterior column fracture. The detached
anterior column has been split into two main frag-
ments. The outer one includes the posterior superi-
or segment of the articular surface and anterior
pillar of the iliac wing while the inner one includes
the cotyloid fossa, the medial half of the posterior
wall of the acetabulum and the inner margin of the
ischium. The inner aspect of this inner fragment in-
cludes a segment of the ilio-pectinealline and most
of the quadrilateral surface. The inner fragment
should be compared with that seen in Example 16.
687
32 Juxta-tectal transverse fracture associated with
an extended posterior wall fracture. The posterior
wall is comminuted and the inferior fragment
thereof includes the upper pole of the ischial
tuberosity. The femoral head is dislocated
posteriorly.
33 Anterior column fracture associated with a
posterior hemi-transverse component. The hemi-
transverse component splits the ischial spine. The
articular segment of the anterior column is further
fractured at two points and there is an undisplaced
fracture line dividing the sacro-iliac joint, merging
with the iliac wing fracture line. There are two im-
pacted marginal fragments.
688 Exercises in Radiographic Diagnosis

Case 1. Associated posterior wall and posterior column frac·

tures.
The fracture line detaching the posterior column runs in·
feriorly through the ischium, leaving undisturbed the obturator
foramen. A split through the quadrilateral surface isolates a
plaque of bone
Exercises in Radiographic Diagnosis 689

Case 2. Juxta-tecta! transverse fracture

690 Exercises in Radiographic Diagnosis

Case 3. Intermediate anterior column fracture associated with

an incomplete hemitransverse posterior fracture.
The incomplete hemitransverse component is clearly seen on
the antero-posterior view, running obliquely upwards and in-
wards. It does not cut the posterior border of the pelvic bone
Exercises in Radiographic Diagnosis 691

Case 4. Juxta-tectal transverse fracture associated with a posteri-

or wall fracture.
Initially the head was posteriorly dislocated
692 Exercises in Radiographic Diagnosis

Case 5. Complete both-column fracture.

The wing fracture line reaches the iliac crest. The anterior
column is detached in a single piece from the anterior superior
iliac spine down to the superior pubic ramus. An uncommon
fragment comprising the angle of the greater sciatic notch and
the upper part of the posterior column has pivoted about 90 0
on the antero-posterior view; the sacro-iliac joint is undisturbed
Exercises in Radiographic Diagnosis 693

Case 6. Posterior wall fracture with marginal impaction.

The impacted fragment is parallel to the supero-medial part
of the head in the antero-posterior view
694 Exercises in Radiographic Diagnosis

Case 7. Complete both-column fracture.

The iliac wing fracture line extends to the iliac crest, isolat-
ing a triangular fragment of the wing. The posterior column
fragment brings with it, in a single piece, the inferior part of the
sacro-iliac joint, which most frequently constitutes a separate
fragment. See Chap. 14.4 (a). Exceptionally, there is no spur sign
Exercises in Radiographic Diagnosis 695

Case 8. Low anterior column fracture associated with a hemi-

transverse posterior fracture.
Two secondary fracture lines divide the anterior column
fragment into three parts: the anterior wall fragment, the body
of the pubis and a segment of the ischio-pubic ramus
696 Exercises in Radiographic Diagnosis

Case 9. Posterior column fracture associated with an undis-

placed anterior hemitransverse component.
This association is similar to a "T-shaped" fracture (see
p.140)
Exercises in Radiographic Diagnosis 697

Case 10. Juxta-tectal transverse fracture, with some comminu-

tion in the posterior part of the fracture line.
The fracture line is lower posteriorly and cuts the edge of the
pelvic bone just above the ischial spine
698 Exercises in Radiographic Diagnosis

Case 11. Vertical T-shaped fracture associated with a posterior

wall fracture.
The posterior fragment outline is superimposed on the inner
part of the roof and also overlies the upper part of the femoral
neck
Exercises in Radiographic Diagnosis 699

Case 12. Low fracture of the anterior column.

There are two supplementary fragments, one comprising a
posterior segment of the ilio-pectineal line, and the other asso-
ciated with a minor fracture of the quadrilateral surface elevat-
ing a posteriorly hinged plaque of bone
700 Exercises in Radiographic Diagnosis

Case 13. Infra-tecta! transverse fracture with little displacement

Exercises in Radiographic Diagnosis 701

Case 14. Vertical T-shaped fracture.

There are two fracture lines through the ischio-pubic ramus
and an isolated fragment of anterior wall. Marginal impaction
of the inner part of the roof is clearly seen on the antero-posteri-
or view
702 Exercises in Radiographic Diagnosis

Case 15. Posterior wall fracture associated with an undisplaced

low transverse fracture, cutting the posterior border of the pelvic
bone under the ischial spine
Exercises in Radiographic Diagnosis 703

Case 16. This is a very special case, unique in the series and not
in our classification.
The centrally dislocated head has fractured the inferior floor
on the acetabulum, sparing the borders and also the ilio-pec-
tineal line. The inferior part of the ilio-ischial line is involved
704 Exercises in Radiographic Diagnosis

Case 17. Juxta-tectal transverse fracture.

There is a small posterior wall fragment, and an undisplaced
fracture line of the iliac wing delineates a big fragment of wing
to which the whole roof is attached. See Chap. 12.1.3 (d). A per-
trochanteric fracture of the femur was also present, and the
head, posteriorly dislocated, was entrapped through a hole in
the posterior capsule
Exercises in Radiographic Diagnosis 705

Case 18. Complete both-column fracture.

The iliac wing component of this complex fracture com-
prises a 'V' configuration. The posterior limb does not reach the
iliac crest. In the root of the pubic ramus there is comminution
involving the roof of the obturator canal
706 Exercises in Radiographic Diagnosis

Case 19. Associated anterior wall and posterior hemitransverse

fractures.
A fracture line through the ischio-pubic ramus isolates the
inferior part of the anterior column. Marginal impaction has in-
volved the inner part of the roof of the acetabulum
Exercises in Radiographic Diagnosis 707

Case 20. Extended anterior column fracture.

A secondary fracture line divides the wing segment of the
column extending to the anterior border of the iliac wing be-
tween the anterior iliac spines. There is some comminution of
the inner table of the iliac wing and, from the quadrilateral sur-
face of the ischium, a posteriorly hinged plaque of bone has
been elevated. Inferiorly, the anterior column is interrupted by
fracture lines through the root of the superior pubic ramus,
body of the pubis, and through the ischio-pubic ramus. The
posterior border of the acetabulum and the posterior border of
the pelvic bone are each intact
708 Exercises in Radiographic Diagnosis

Case 21. Both-column fracture with iliac wing fracture line

extending to the iliac crest.
The anterior column is split through the anterior wall and
there is a fragment of the internal iliac fossa
Exercises in Radiographic Diagnosis 709

Case 22. Associated posterior column and postero-superior wall

fracture.
There is an associated fracture of the superior pubic ramus,
and the urethra was ruptured
710 Exercises in Radiographic Diagnosis
Exercises in Radiographic Diagnosis 711

Case 23. Juxta-tecta! transverse fracture associated with a frac-

ture line extending from the retro-acetabular surface to the crest.
All the upper part of the articular surface belongs to the
anterior fragment of the wing. which is also divided by a split
reaching the interspinous notch. Compare with the fracture line
of a both-column fracture
712 Exercises in Radiographic Diagnosis

Case 24. Associated undisplaced infra-tectal transverse and pos-

terior wall fractures.
The main transverse component is associated with a compo-
nent which detaches the outer half of the roof of the acetabu-
lum. The femoral head is dislocated posteriorly
Exercises in Radiographic Diagnosis 713

Case 25. Both-column fracture with iliac wing fracture extend-

ing to the iliac crest.
The spur sign is obvious on the obturator-oblique view. The
posterior column, markedly displaced inwards, was entrapped in
the true pelvis, resting against the sacro-iliac joint. Two opera-
tive approaches were needed
714 Exercises in Radiographic Diagnosis

Case 26. T-shaped fracture with extensive marginal impaction of

the inner part of the roof
Exercises in Radiographic Diagnosis 715

Case 27. Typical posterior column fracture, not involving the

teardrop
716 Exercises in Radiographic Diagnos

Case 28. Both-column fracture with iliac wing fracture line e)

tending to the crest and involving the sacro-iliac joint.
The posterior column takes with it the inferior half of th
sacro-iliac joint. See also Case 7. Two operative approaches weI
needed
Exercises in Radiographic Diagnosis 717

Case 29. Anterior T-shaped fracture, with several splits of the

anterior column component.
1\\'0 approaches (KOCHER-LANGENBECK first) were required
718 Exercises in Radiographic Diagnosis

Case 30. Anterior T-shaped fracture.

The ilio-pectineal line seems undisturbed on the obturator-
oblique view, but its fracture is obvious on the antero-posterior
view. A small segment of the ilio-pectineal line is delineated by
the two breaks of the pelvic brim
Exercises in Radiographic Diagnosis 719

Case 31. High anterior column fracture.

Special features are the detachment of two large fragments.
A postero-superior piece of the articular surface includes an ex-
tended portion of the anterior pillar of the iliac wing. The other
separates from the inner aspect of the bone, the acetabular
fossa, the medial half of the posterior wall, and the inner margin
of the ischium; its inner aspect comprises a segment of the ilio-
pectineal line and most of the quadrilateral surface. Compare
the latter fragment with that seen in Case 16
720 Exercises in Radiographic Diagnosis

Case 32. Juxta-tectal transverse fracture associated with an ex-

tended posterior wall fracture.
The posterior wall is in several pieces, the inferior one of
which includes the upper pole of the ischial tuberosity. The
femoral head is dislocated posteriorly
Exercises in Radiographic Diagnosis 721
722 Exercises In Radiographic Diagnosis

Case 33. Complete anterior column fracture associated with a

posterior hemitransverse component dividing the sciatic spine.
The articular segment of the anterior column is split in at
least two points. There is an un displaced fracture line dividing
the sacro-iliac joint and merging with the iliac wing fracture line.
There are two impacted marginal fragments
Bibliography
Adam P, Labbe JL, Alberge Y, Austry P, Delcroix P, Ficat RP
(1985) The role of computed tomography in the assessment
and treatment of acetabular fractures. Clin Radiol 36
(1):13-18
Aho AJ, Isberg UK, Katevuo VK (1986) Acetabular posterior
wall fracture. 38 cases followed for 5 years. Acta Orthop
Scand 57 (2): 101-105
Ahrer E (1966) 39 cases of fractures with dislocation of the hip
joint (1959-1964). Klin Med (Wien) 21:266-267
Alpers DD (1970) Migration of broken hip pin into urinary
bladder. JAMA 212:2123 - 2124
Aubart F, Fares J, Chaise F (1986) Acetabular fracture with in-
trapelvic luxation following an epileptic attack. Apropos of
2 cases including 1 bilateral case (in French). Rev Chir Or-
thop 71 (2):143 -145
Aufranc OE (1967) Fracture of acetabulum with dislocation of
the hip and sciatic palsy. JAMA 201:690-691
Aufranc OE et al. (1962a) Luxation-fracture compliquee de
I'acetabulum. JAMA 182:858-861
Aufranc OE et al. (1962b) Fracture de la paroi interne de
I'acetabulum. JAMA 180:317-319
Aufranc OE, Jones WN, Harris WM (1963) Fracture dislocation
of the hip. JAMA 186:1160-1163
Aufranc OE, Jones WN, Harris WM (1964) Fracture of the
acetabulum. JAMA 188:382-385
Babosha VA, Zhukov IUB, Lobanov GV, Pasternak VN (1989)
Pulsating intrapelvic hematoma as a complication of a
transacetabular fracture (in Russian). Klin Khir 12:53 - 54
Balmer PM, Isler B, Ganz R (1988) Results of surgically treated
acetabular fractures (in German). Unfallchirurg 91 (4):
149-153
Banallec Y (1956) Les paralysies sciatiques dans les fractures
luxations du cotyle. Thesis, University of Lille
Barbieri E (1967) Traumatic posterior luxation of the hip.
Minerva Ortop 18:851-856
Barbieri E (1976) Central luxation of the femur. Minerva Ortop
18:622-627
Barnes ST (1972) Pseudo tumor of the ischium. J Bone Joint
Surg [Am) 54 (3):645
Batory I, Muller HP (1983) Our indications for the treatment of
acetabular and pelvic fractures with acetabular participation
(in German). Z Orthop 121 (2):154-158
Berkebile RD (1965) The gull-wing sign. Value of the lateral view
of the pelvis in fracture dislocation of the acetabular rim
and posterior dislocation of the femoral head. Radiology
84:937-939
Berman AT, Metzger PC, Chinitz JL (1981) Central acetabular
fracture-dislocation secondary to an epileptic seizure in a
chronic renal patient. J Trauma 21 (1):66-67
Bernard R, Charleux H et al. (1962) Abord transiliaque de cer-
taines fractures recentes du cotyle. Ann Chir 88 (10/11):
367-368
Beyssac L, Meunier M (1963) Hip dislocation with femoral head
fractures (3 cases). Lyon Chir 59:733 -734
Blaquiere RM (1985) Computed tomography in acetabular trau-
ma. Clin Radiol 36 (1):5-11
Blumberg ML (1980) Computed tomography and acetabular
trauma. Comput Tomogr 4 (1):47 - 53
Bonnel F (1975) Biomechanical observations and clinical use of
fixateur externe in acetabular fractures (in German). Hefte
Unfallheilkd 124:50- 53
Bosse MJ, Poka A, Reinert CM, Ellwanger F, Slawson R,
McDevitt ER (1988) Heterotopic ossification as a complica-
tion of acetabular fracture. Prophylaxis with low-dose ir-
radiation. J Bone Joint Surg [Am) 70 (8):1231-1237
Bosse MJ, Poka A, Reinert CM, Brumback RJ, Bathon H,
Burgess AR (1988) Preoperative angiographic assessment of
the superior gluteal artery in acetabular fractures requiring
extensile surgical exposures. J Orthop Thmma 2 (4):303 - 307
Bouillet R (1971) Fracture de la branche ischio-pubienne avec
perte de substance chez un enfant de 12 ans. Arch Orthop
Belg 37 (3):278
Bozdech Z, Rozkydal Z (1979) Treatment of inveterate acetabu-
lar fractures by total hip prosthesia (in Czech). Acta Chir
Orthop Traumatol Cech 46 (2):108-116
Bray TJ, Esser M, Fulkerson L (1987) Osteotomy of the trochan-
ter in open reduction and internal fixation of acetabular
fractures. J Bone Joint Surg [Am) 69 (5):711-717
Brinkmann WH (1967) On the hip dislocation with fracture of
the acetabulum. Monatsschr Unfallheilkd 70:273 - 289
Briquet J (1964) Apropos of intrapelvic impacted fractures of
the cotyloid region. Helv Chir Acta 31:556-580
Brumback RJ, Holt ES, McBride MS, Poka A, Bathon GH,
Burgess AR (1990) Acetabular depression fracture accompa-
nying posterior fracture dislocation of the hip. J Orthop
Trauma 4 (1):42-48
Bruns J, Dahmen G (1986) The gluteal approach to the ischium.
Arch Orthop Trauma Surg 105 (6):369-372
Bucholz RW (1982) Injury to the acetabular triradiate physeal
cartilage. J Bone Joint Surg [Am) 64 (4):600-608
Burk DL Jr, Mears DC, Herbert DL, Straub WH, Cooperstein
LA, Beck EA (1984) Pelvic and acetabular fractures: exami-
nation by angled CT scanning. Radiology 153 (2):548
Burk DL Jr, Mears DC, Cooperstein LA, Herman GT, Udupa
JK (1986) Acetabular fractures - three-dimensional compu-
ter tomographic imaging and interactive surgical planning. J
Comput Tomogr 10 (1):1-10
Burk DL Jr, Mears DC, Kennedy WH, Cooperstein LA, Herbert
DL (1985) Three-dimensional computed tomography of
acetabular fractures. Radiology 155 (1):183-186
Campanacci M (1967) Lesioni tramatiche del bacino. In: Gaggi
A (ed) Proceedings of the 52nd Congress of the Societa
Italiana di Ortopedia, Rome, 13-15 Oct 1967. Stanya Artie,
Bologna
724
Carnes ale PG, Stewart Ml, Barnes SN (1975) Acetabular disrup-
tion and central fracture-dislocation of the hip. A long-term
study. 1 Bone loint Surg [Am] 57 (8):1054-1059
Cauchoix M, Truchet P (1951) Les fractures articulaires de la
hanche (col excepte). Rev Chir Orthop 37:266- 332
Caudle Rl, Crawford AH (1988) Avulsion fracture of the lateral
acetabular margin. A case report. 1 Bone loint Surg [Am]
70 (10):1568-1570
Coleman SS (1990) Avulsion fracture of the lateral acetabular
margin. A case report. 1 Bone loint Surg [Am] 72 (4):633
(letter)
Colinet 1, Lemaire R (1984) Etude retrospective d'une serie con-
tinue de 226 fractures du cotyle. Acta Orthop Belg 50
(3):406 - 415
Copin G, Grosse A, lenny G, Kempf L (1975) Treatment of
acetabular fractures using Hoffmann's "fixateur externe" (in
German). Hefte Unfallheilkd 124:92-93
Cormier 1M, Florent 1 (1967) Les hematomes retroperitoneaux
par blessures des vaisseaux iliaques au cours des fractures du
bassin. Mem Acad Chir 93:458-462
Cottalorda (1922) Les fractures par enfoncement de la cavite
cotyloide. Thesis, Montpellier
Cottalorda (1923) Recherches experimentales sur les fractures
enfoncement de la cavite cotyloide. Lyon Chir 20:32
Craig lB (1986) Superior fracture-dislocation of the hip. Arch
Orthop Trauma Surg 105 (2):130-132
Creyssel 1, Schnepp 1 (1961 a) Contribution it l'etude radio-
logique des fractures transcotyloidiennes du bassin. 1 Radiol
Electrol Med Nucl 42:691- 699
Creyssel 1, Schnepp 1 (1961 b) Fractures transcotylOidiennes du
bassin. Masson, Paris
Creyssel 1, Schnepp 1 (1963) Present status of the prognosis and
treatment of traumatic inward displacement of the acetabu-
lum. Scalpel 116:392 - 393
Danis A, Detrain M (1984) Fractures du cotyle. Reconstruction
intra-articulaire pardes greffons spongieux. Acta Orthop
Belg 50 (3):391- 396
Danis A et al. (1971) Acetabulum fractures. A new surgical pro-
cedure. Acta Orthop Belg 37:117 -140
D'Aubigne RM (1966) Surgical treatment of hip fractures with
dislocations (so-called acetabular explosive fractures). Klin
Med (Wien) 21:249-252
D'Aubigne RM (1968) Management of acetabular fractures in
multiple trauma. 1 Trauma 8:333 - 349
Daudle Rl, Crawford AH (1988) Avulsion fracture of the lateral
acetabular margin. A case report. 1 Bone loint Surg [Am]
70 (10):1568-1570
Dazy G (1954) De l'enfoncement du cotyle et de son traitement
par Ie procecte de l'arceau. Thesis, University of Paris
Decoulx P, Duquennoy 1 (1979) Traitement des fractures du
cotyle. Cahiers de Conferences d'Enseignement de la SOF-
COT. Expansion Scientifique, Paris
De Mesquita Montes 1 et al. (1982) Lesion traumatique rare de
la hanche chez l'enfant. Rev Orthop Trauma (Paris) 8 (2):163
Derian PS (1966) Intra-articular small bowel herniation com-
plicating central fracture dislocation of the hip. 1 Bone loint
Surg [Am] 48:1614-1621
DeSierra Cano L, Seral Ifligo F, de la Sierra Setien lA (1906)
Fractura del acetabulo, vol 1. Palma de Mallorca
Dossa 1 et al. (1972) Resultats eloignes du traitement des frac-
tures du cotyle. Montpellier Chir 18:291-310
Dunn AW, Russo CL (1973) Central acetabular fractures. 1
Trauma 13 (8):695 -705
Bibliography
Dunoyer 1 et al. (1973) Necrose de la tete femorale et fracture
du cotyle. Rhumatologie 25:141-142
Duparc 1, Ficat P (1960) Fractures articulaires de l'extremite
superieure de tibia. Rev Orthop 46:399-481
Duquennoy A et al. (1982) Fractures du cotyle - resultats it plus
de 5 ans. Rev Chir Orthop 68 [Suppl 2] :47 - 79
Duquennoy A, Tillie B, Delcourt lP (1984) Fractures du cotyle:
congruence articulaire et indications therapeutiques. Acta
Orthop Belg 50 (3):343 - 355
Duquennoy A, Tillie B, Fontaine C, Gougeon F, Mestdagh H
(1985) Fractures du cotyle et desjonctions sacro-iliaques.
Rev Chir Orthop 71 (5):311 - 318
Ebraheim NA, Savolaine ER, Hoeflinger Ml, lackson WT
(1989) Radiological diagnosis of screw penetration of the
hip joint in acetabular fracture reconstruction. 1 Orthop
Trauma 3 (3):196-201
Ecke H, Babayan R, Burri C, Enzenross HG, lungbluth KH,
Kraus 1, Pannike A, Ruter R, Ruf N, Tscherne H, Weiss H,
Weller S (1975) Treatment results in acetabular osteosyn-
thesis (in German). Hefte Unfallheilkd 124:90-92
Ehalt W (1969) Dislocation fractures of the hip. Langenbecks
Arch Chir 325:766 - 767
Eichenholtz SN, Stark RM (1964) Central acetabular fractures.
A review of 35 cases. 1 Bone loint Surg [Am] 46:695 -714
Elliott RB (1956) Central fractures of the acetabulum. Clin Or-
thop 7:189-202
Ender HG (1975) Types of acetabular fractures (in German).
Hefte Unfallheilkd 124:9-33
Epstein (1974) Posterior fracture dislocations of the hip. 1 Bone
loint Surg [Am] 56 (6):1103
Farcy lP, Moskovich R, Leonardi L (1988) Management of ace-
tabular fractures. Bull NY Acad Med 64 (4):327 - 334
Faubel W (1967) On the X-ray diagnosis of the hip joint injuries.
Hefte Unfallheilkd 91:28 - 30
Feischl P (1966) Central hip joint dislocations. Klin Med (Wien)
21:247-249
Fischer L, Bochu M, Carret lP, Walch G, Barral F, Beju 1 (1984)
Apport de la tomodensitometrie dans les fractures du cotyle.
Rev Chir Orthop 70 [Suppl 2]:84-89
Fishman EK, Magid D, Drebin RA, Brooker AF lr, Scott Wl lr,
Riley LH lr (1989) Advanced three-dimensional evaluation
of acetabular trauma: volumetric image processing. 1 Trau-
ma 29 (2):214-218
Fisk C (1964) Acetabulum fractures. Radiol Technol 35:
330-333
Frank lL, Reimer BL, Raves JJ (1989) Traumatic iliofemoral
arterial injury: an association with high anterior acetabular
fractures. 1 Vasc Surg 10 (2):198-201
Fries G (1966) On the treatment of old traumatic dislocations of
the hip and dislocation fractures in childhood. Arch Orthop
Unfallchir 59:229-239
Gay B, Romen W (1983) Macroscopic and microscopic findings
in the femur head following acetabular fractures (in Ger-
man). Unfallheilkunde 86 (5):201- 204
Gergen M (1965) Contribution to the problem of secondary
acetabulum formation in unreduced central and posterior
fracture luxations of the hip joint. Arch Orthop Unfallchir
57:64-68
Gibson A (1950) Posterior exposure of the hip joint. 1 Bone
loint Surg [Br] 32:183
Gibson A (1950) Posterior exposure of the hip joint. Instruc-
tional course lectures. Am Assoc Orthop Surg 10:195
Gignoux M et al. (1970) Interets des incidences de trois-quarts
Bibliography
dans la radiologie des fractures du cotyle. Cah Med Lyon
46:43-44
Giudici L et al. (1964) Le traitement sanglant des fractures du
cotyle avec enfoncement de l'hemibassin. Minerva Med
55:1023
Gosset J (1975) Chirurgie du cotyle it double equipe. Possibilites
d'un abord simultane des deux colonnes du cotyle dans les
fractures complexes. Chirurgie 101 :290
Goulet JA, Bray TJ (1989) Complex acetabular fractures. Clin
Orthop 240:9-20
Gradischnig R (1966) On results of follow-up examination of
posterior hip fractures with dislocation. Klin Med (Wien)
21:276-277
Gresser A (1965) Contribution to the problem of central hip lux-
ation. Bruns Beitr Klin Chir 210:310-317
Griffiths HJ, Standertskjold-Nordenstam CG, Burke J, Lamont
B, Kimmel J (1984) Computed tomography in the manage-
ment of acetabular fractures. Skeletal Radiol 11 (1):22-31
Gualtieri G, Gualtieri I, Mele C, Gaglilardi S, Bettini N, Gui L
(1986) The treatment of acetabular fractures. Ital J Orthop
Traumatol 12 (3):289-294
Guerzoni PL (1970) Long-term results of the surgical treatment
of acetabular fractures (in Italian). Chir Organi Mov 58 (4):
363-365
Guingand 0, Rigault P, Padovani JP, Finidori G, Touzet P,
Depotter J (1985) Luxation traumatique de la hanche et frac-
ture du cotyle chez l'enfant. Rev Chir Orthop 71 (8):
575 -585
Hamada G (1963) Epiphysiolyse de l'ischion. Clin Orthop
31:117-130
Harley JD, Mack LA, Winquist RA (1982) CT of acetabular
fractures: comparison with conventional radiography. AJR
138 (3):413 -417
Harnach Z, Strmiska J (1975) Our treatment results in acetabu-
lar fractures (in German). Hefte Unfallheilkd 124:246-247
Harris WH (1969) naumatic arthritis of the hip after disloca-
tion and acetabular fracture. Treatment by mold arthroplas-
ty. An end result study using a new method of results evalua-
tion. J Bone Joint Surg [Am) 51:737-755
Haw DW (1965) Complication following fracture dislocation of
the hip. Br Med J 5441:1111-1112
Healy et al. (1986) Heterotopic bone formation after hip surge-
ry: prevention with single-dose postoperative hip irradiation.
Radiology 168:851- 854
Heeg M, Oostvogel HJ, Klasen HJ (1987) Conservative treat-
ment of acetabular fractures: the role of the weight-bearing
dome and anatomic reduction in the ultimate results. J
Trauma 27 (5):555-559
Heeg M, Visser JD, Oostvogel HJ (1988) Injuries of the acetab-
ular triradiate cartilage and sacroiliac joint. J Bone Joint
Surg [Br) 70 (1):34- 37
Heeg M, Klasen HJ, Visser JD (1989) Acetabular fractures in
children and adolescents. J Bone Joint Surg [Br) 71
(3):418-421
Heeg M, Klasen HJ, Visser JD (1990) Operative treatment for
acetabular fractures. J Bone Joint Surg [Br) 71 (5):383 - 386
Henderson MR, Apyan PM (1988) Both-column acetabular
fracture in a multiple trauma patient (clinical conference). J
Tenn Med Assoc 81 (8):516-518
Hofmann AA, Dahl CP, Wyatt RW (1984) Experience with
acetabular fractures. J Trauma 24 (8):750-752
Holter D (1969) Fracture pelvis and fractura acetabuli. Acta Or-
thop Scand 40:670-671
725
Iumashev GS, Lavrov IN, Shestakov VE, Vinogradov EV,
Rabinovich LS (1984) Treatment of acetabular fractures
complicated by central hip dislocation (in Russian). Sov
Med 10:37-40
Jacchia GE (1970) Transcotyloid fractures. Arch Putti Chir
Organi Mov 25:1-15
Jacchia GE, Marchetti N (1970) Clinico-radiographic aspects
and therapeutic trends in sequelae of cotyloid fractures.
Arch Putti Chir Organi M0V 25:189-212
Jenny G, Kempf I, Grosse A, Jung F (1973) Fractures du cotyle.
Deux cas traites par fixateur externe. Rev Chir Orthop
59:219-224
Johansson H, Olerud S (1971) The operative treatment of ace-
tabular fractures (in Swedish). Lakartidningen 68 (17):
2023-2030
Johnson EE, Eckardt JJ, Letournel E (1987) Extrinsic femoral
artery occlusion following internal fixation of an acetabular
fracture. A case report. Clin Orthop 217:209-213
Judet R (1966) Les fractures du cotyle. Acta Orthop Belg
32:469-476
Judet R, Lagrange J (1958) La voie postero-externe de Gibson.
Pre sse Med 66 (3):263 - 264
Judet R, Letournel E (1961) Mecanisme et anatomie pathologi-
que des fractures du cotyle. Mem Acad Chir 19:593 - 603
Judet R, Letournel E (1962a) Anatomie pathologique et traite-
ment des fractures recentes du cotyle. Med Hyg 19:523-525
J udet R, Letournel E (1962 b) Les fractures isolees des colonnes
du cotyle. Rev Chir Orthop 48:300-312
Judet R, Letournel E (1962c) Traitement chirurgical des frac-
tures recentes du cotyle. A propos de 46 cas operes. Mem
Acad Chir 20:369-377
Judet R, Letournel E (1963) Les fractures associees des deux col-
onnes du cotyle. Presse Med 71:1291-1293
Judet R, Letournel E (1964) Les fractures du cotyle. Acta Or-
thop Belg 30:285 - 293
Judet R, Letournel E (1968) Fractures of the acetabulum:
classification and guiding rules for open reduction. Arch
Orthop 81:119-158
Judet R, Letournel E (1973) Surgical management of fractures
of the acetabulum. In: Tronzo RG (ed) Surgery of the hip
joint. Lea and Febiger, Philadelphia, pp 472-505
Judet R, Letournel E (1974) Les fractures du cotyle. Masson,
Paris
Judet R, Judet J, Letournel E (1962) Traitement chirurgical des
fractures n!centes du cotyle (it propos de 46 cas operes). Ann
Chir 88 (10/11):369-371
Judet R, Judet J, Letournel E (1964) Fractures of the
acetabulum. Classification and surgical approaches for open
reduction. Preliminary report. J Bone Joint Surg [Am)
46:1615-1636
Judet R, Letournel E, Vacher D (1968) Fragmentary incarcera-
tions in fractures of the acetabulum. Presse Med
76:411-414
Jungbluth KH (1975) Osteosynthesis of acetabular fractures (in
German). Hefte Unfallheilkd 124:87 - 90
Jungbluth KH, Sauer HD, Schottle H (1979) Results of the
surgical reconstruction of dislocated acetabular fractures
(the collective statistics of the International Work Shop
Group for Problems of Osteosynthesis, the German section)
(in German). Hefte Unfallheilkd 140:154-160
Kaulbach C, Heller M, Triebel HJ, Spielmann RP, Richartz-
Heller M (1989) Radiologic diagnosis of acetabular fractures
(in German). Radiologe 29 (10):501- 507
726
Kelly RP, Yarbrough SG (1971) Fracture luxation posterieure de
la tete femoral avec maintien en place du fragment median
de la tete (compliquee de fracture acetabulaire). J Trauma
11:97 -108
Kienzler G (1968) Complications and late sequelae of reposi-
tioned traumatic luxations and luxation fractures of the hip
and their therapeutic results. Arch Orthop Unfallchir
64:151-163
Kjaersgaard-Andersen P, Jensen J, Ovesen J (1988) Total hip re-
placement following traumatic dislocation of the hip and/or
acetabular fracture. A review of 33 cases (in Danish). Ugeskr
Laeger 150 (14):850-852
Kleiman SG, Stevens J, Kolb L, Pankovich A (1971) Late sciatic
nerve palsy following posterior fracture dislocation of the
hip. A case report. J Bone Joint Surg [Am] 53:781-782
Kleinsorge F, Berg-Schlosser V, Maroske D (1985) Advantages
of computerized tomography diagnosis in acetabular frac-
tures (in German). Chirurg 56 (7):449 - 453
Knight RA, Smith H (1958) Central fractures of the acetabulum.
J Bone Joint Surg [Am] 40 (1):1-16
Kuhlman JE, Fishman EK, Ney DR, Brooker AF Jr, Magid D
(1989) Nonunion of acetabular fractures: evaluation with in-
teractive multiplanar CT. J Orthop Trauma 3 (1):33 -40
Labbe JL, Adam P, Alberge Y, Ficat RP (1984) Place of com-
puted tomography in the diagnosis and treatment of acetab-
ular fractures (in French). Acta Orthop Belg 50 (3):310 - 323
Lansinger 0 (1977) Fractures of the acetabulum. Acta Orthop
Scand [Suppl]:165
Lansinger 0, Romanus B, Goldie IF (1979) Fracture mechanism
in central acetabular fractures. An experimental study. Arch
Orthop Trauma Surg 94 (3):209-212
Lansinger 0, Karlsson J, Berg U et al. (1984) Unstable fractures
of the pelvis treated with a trapezoid compression frame. Ac-
ta Orthop Scand 55 (3):325 - 329
Lardennois J (1959) 8 Cas d'enfoncement du cotyle traites par
appareil it arceau. Ann Chir 13 (15/16):853
Larson CB (1973) Fracture dislocation of the hip. Clin Orthop
92:147 -154
Latarjet A, Rouviere H, Delmas (1940) Anatomy of the pelvic
bone. In: Rouviere H (ed) Traite d'anatomie humaine. Mas-
son, Paris
Lautel J et al. (1975) Une complication rare des fractures du
cotyle: I'incarceration d'une anse grele dans Ie foyer de frac-
ture. J Chir (Paris) 111 (2):185
Lavrov IN, Shestakov VE, Rabinovich LS, Rokhkind ALU
(1981) Treatment of acetabular fractures (in Russian). Ortop
Travmatol Protez 9:19-21
Lazzari A, Bonetti C (1965) Osteosynthesis with metallic screws
in fractures of the posterior cotyloid edge. Arch Putti Chir
Organi Mov 20:309 - 318
Lehtonen R (1968) A study of traumatic dislocations of the hip
joint and fractures of the acetabulum. Ann Chir Gynecol Fr
57 [Suppl]:1-87
Lesourd G (1969) Luxation traumatique de la hanche droite avec
decollement epiphysaire de la tete femorale et fracture de la
partie posterieure du cotyle chez un enfant de 15 ans. Rev
Onir Orthop 55:61-64
Lessan D (1975) Results of acetabular fractures without dis-
placement (in German). Hefte Unfallheilkd 124:60-62
Letic S (1975) Our experiences in the treatment of dislocation
fractures of the acetabular roof (in German). Hefte Un-
fallheilkd 124:260- 263
Letournel E (1961) Les fractures du cotyle. Etude d'une serie de
75 cas. Thesis, Arnette, Paris
Bibliography
Letournel E (1961) Les fractures du cotyle. Etude d'une serie de
75 cas. J Chir (Paris) 82:47 - 87
Letournel E (1966) Die operative Versorgung der Huftgelenk-
pfannenbruche. Langenbecks Arch Klin Chir 316:422-437
Letournel E (1968) Anatomie pathologique et traitement des
fractures du cotyle. Masson, Paris, pp 105 - 176 (Actualites
orthopediques de I'H6pitai R. Poincare, vol 5)
Letournel E (1972) Les fractures anciennes du cotyle. Masson,
Paris, pp 75 -122. (Actualites orthopediques de I'H6pitai R.
Poincare, vol 10).
Letournel E (1979) The results of acetabular fractures treated
surgically: 21 years experience. In: The Hip Society (ed) The
hip, vol 7. Mosby, St. Louis
Letournel E (1981) Traitement chirurgical des traumatismes du
bassin en dehors des fractures isolees du cotyle. Rev Chir Or-
thop 67 (8):771
Letournel E (1981) Acetabulum fractures: classification and
management. Clin Orthop Rei Res 151 :81 - 106
Letournel E (1984) Resultats du traitement chirurgicale des frac-
tures du cotyle. Acta Orthop Belg 50 (3):423 -433
Letournel E (1984) La voie laterale dans Ie traitement des frac-
tures du cotyle. Acta Orthop Belg 50 (3):381-390
Letournel E (1987) Surgical treatment of acetabular fractures.
Hip :157-180
Letournel E (1989) Acetabular fractures and their reconstruc-
tion. Curr Orthop 3:233-243
Letorunel E (1990) Classification and evaluation of acetabular
fractures. Techniques in orthopedics, p 524. Aspen publisher
Letournel E (1990) Diagnosis and treatment of Nonunions and
Malunions of acetabular fractures. Orthop Clin North-Am
Vol 21:769-788
Levine (1943) Surgical treatment of central fracture dislocations
of the acetabulum. J Bone Joint Surg [Am] 25:900-901
Likhoded S (1965) Classification of traumatic central disloca-
tion of the hip. Ortop Travmatol Protez 26:79-81
Lovelock JE, Monaco LP (1983) Central acetabular fracture
dislocations: an unusual complication of seizures. Skeletal
Radiol 10 (2):91-94
Luchko GD, Ostashko VI (1983) Clinical picture and therapy of
acetabular fractures in patients with associated and multiple
injuries (in Russian). Vestn Khir 131 (8):112-115
Lujubosic NA (1967) Acetabular injuries in childhood (in
Czech). Acta Chir Orthop Traumatol Cech 34 (5):393 - 400
Lyddon DW, Hartman JT (1971) Traumatic dislocation of the
hip associated with an ipsilateral fracture of the femur. J
Bone Joint Surg [Am] 53 (5):1012-1016
MacCoy DM (1975) Acetabular fracture fixation. J Am Vet Med
Assoc 166 (4):309-312 (letter)
Mack LA, Harley JD, Winquist RA (1982) CT of acetabular
fractures: analysis of fracture patterns. AJR 138 (3):
407 -412
Mack LA, Duesdieker GA, Harley JD, Bach AW, Winquist RA
(1983) CT of acetabular fractures: postoperative appear-
ances. AJR 141 (5):891-894
Magid D, Fishman EK, Brooker AF Jr, Mandelbaum BR,
Siegelman SS (1986) Multiplanar computed tomography of
acetabular fractures. J Corn put Assist Tomogr 10
(5):778 - 783
Malgaigne JF (1847) Traite des fractures et luxations. Bailliere,
Paris
Mandelbaum BR, Magid D, Fishman EK, Burgess AR, Bosse
M, Brooker AF Jr (1987) Multiplanar computed tomogra-
phy - a multidimensional tool for evaluation and treatment
of acetabular fractures. J Comput Tomogr 11 (2): 167 - 173
Bibliography
Margulies JY, Rubinstein N, Fast A, Floman Y (1983) Osteo-
porosis and seizures leading to central acetabular fracture
dislocation. Isr J Med Sci 19 (1 ):85 - 87
Marincek B, Porcellini B, Robotti G (1984) Computed tomo-
graphic classification of acetabular fractures (in German).
Radiologe 24 (5):205-210
Martimbeau CL (1986) Effort to standardize surgical approach
in acetabular fracture. Orthop Rev 15 (1):39-43
Martinek H (1975) Fractures of the posterior acetabular wall (in
German). Hefte Unfallheilkd 124:76-79
Martinek H, Egkher E, Fasol P (1978) Long-term results after
conservative treatment of acetabular fractures (in German).
Unfallheilkunde 81 (1):1-5
Mast J, Jakob R, Ganz R (1989) Planning and reduction tech-
niques in fracture surgery. Springer, Berlin Heidelberg New
York
Matta JM, Merritt PO (1988) Displaced acetabular fractures.
Clin Orthop 230:83 - 97
Matta JM, Letournel E, Browner BD (1986) Surgical manage-
ment of acetabular fractures. Instr Course Lect 35:382 - 397
Matzen PF (1972) Luxation fracture of the hip. Zentralbl Chir
97:78-89
Mazas F (1969) Treatment of fracture-dislocations of the acetab-
ulum. Reconstr Surg Traumatoll1:165-182
Mazas F, d'Aubigne RM (1969) Management of the late effects
of fracture dislocations of the hip. Injury 1:27 - 38
McCardel BR, Dahners LE, Renner JB (1989) Kirschner wire
migration from the pelvis to the heart and a new method of
fixation of articular fracture fragments, acetabular recon-
struction. J Ortop Trauma 3 (3):257 - 259
Mears D, Rubash H (1988) Pelvic and acetabular fractures.
Slack, Thorofare, NJ
Meinhard BP, Misoul C, Joy D, Ghillani R (1987) Central
acetabular fracture with ipsilateral femoral-neck fracture
and intrapelvic dislocation of the femoral head without ma-
jor pelvic-column disruption. A case report. J Bone Joint
Surg [Am] 69 (4):612-615
Merle d'Aubigne R (1968) Management of acetabular fractures
in multiple trauma. J Trauma 8 (3):333-349
Michalska D (1966) Case of anterior superior hip dislocation
complicated by fracture of the acetabulum. Chir Narzadow
Ruchu Orthop Pol 31:29-32
Moore TM, Hill JV, Harvey JP Jr (1970) Central acetabular
fracture secondary to epileptic seizure. J Bone Joint Surg
[Am] 52 (7):1459-1462
Nikodinovski J, Gruev V, Trajkovski L, Majstorov AN, Krakovic
M, Balabanovski A, Tolovski D, Krstevski S (1982)
Therapeutic dilemmas in the care of acetabular fractures (in
Serbo-Croatian, Roman). Acta Chir Iugosl 29 [Suppl 2]:
203-210
O'Brien T, McGoldrick F (1990) Avulsion fracture of the lateral
acetabular margin. A case report. J Bone Joint Surg [Am]
72 (1):152-153
Olerud S (1973) A reduction and compression instrument for
acetabular fractures. Injury 5 (2): 167 -168
Olive RJ Jr, Marsh HO (1989) Occult central acetabular fracture
resulting in fracture-dislocation. A case report. Clin Orthop
248:240-245
Ollier L (1892) Traite de resections. Masson, Paris
Pantazopoulos T, Mousafiris C (1989) Surgical treatment of
central acetabular fractures. Clin Orthop 246:57 - 64
Parkinson JR, Evarts CM, Hubbard LF (1982) Radiation
therapy in the prevention of heterotopic ossification after
total hip arthroplasty. In: The Hip Society (ed) Proceedings
727
of the 10th Open Scientific Meeting of the Hip Society.
Mosby, St. Louis, pp 211 - 217
Patterson FLP, Morton KS (1972) Complications neurologiques
des fractures et des luxations pelviennes. J Trauma
12:1013-1023
Pavlansky R, Grepl J, Trkal B, Garche U (1978) Fractures in the
acetabular region - classification and Roentgen diagnosis
(in German). Beitr Orthop Traumatol 25 (4):199-207
Pecorelli F, Della Torre P (1989) Classification of acetabular
fractures (in Italian). Arch Putti Chir Organi Mov 37 (2):
417 -424
Pennal GF, Davidson J, Garside H, Plewes J (1980) Results of
treatment of acetabular fractures. Clin Orthop 151: 115 -123
Percudani W (1967) Osteosynthesis in fractures of acetabulum
back wall. Prognostic considerations. Arch Orthop
80:293-306
Persson BM (1965) Trochanteric side traction for acetabular
fractures - a pin-holding device. Acta Orthop Scand 36
(2):219-220
Perugia L (1970) The treatment of articular fractures of the hip
(fractures of the posterior wall of the acetabulum) (in
Italian). Minverva Ortop 20:217 - 231
Phen HT, Kumar NVG, Ireland J (1989) Bilateral central dislo-
cation of the hip following a cerebrovascular accident. J
Bone Joint Surg [Br] 71 (4):703
Phillips PJ (1967) Posterior fracture dislocation of the hip joint.
Australas Radiol 11 :61 - 62
Piazza P, Girelli G, Coran F, Lutman M (1986) Computerized
tomography in acetabular fractures (in Italian). Radiol Med
(Torino) 72 (6):480-484
Piganiol G, Trouilloud P, Arnould H (1985) Acetabular frac-
tures and the gluteus medius (in French). Chirurgie 111
(3):226-228
Poeplau P, Mentzel M, Schweiberer L (1975) After care of ace-
tabular fractures (in German). Hefte Unfallheilkd 124:
112-114
Raaymakers EL (1973) Acetabular fractures. Bull Soc Int Chir
32:405-406
Rafii M, Firooznia H, Golimbu C, Sokolow J (1982) Bilateral
acetabular stress fractures in a paraplegic patient. Arch Phys
Med Rehabil 63 (5):240-241
Rafii M, Firooznia H, Golimbu C, Waugh T Jr, Naidich D
(1983) The impact of CT in clinical management of pelvic
and acetabular fractures. Clin Orthop 178:228-235
Ramirez (1976) Traitement chirurgical de la fracture luxation de
la hanche. Rev Orthop Trauma 20 I B (2):173
Rau D, Gustilo RB (1975) Treatment of acetabular fractures.
Minn Med 58 (7):518-524
Re C (1970) The acute surgical pelvis with special reference to
the treatment of fractures of the acetabular base (in Italian).
Minerva Orthop 21 (3):232-237
Rehn J, Schramm W (1967) On dislocations and acetabular frac-
tures of the hip joint with special reference to late sequelae
(in German). Arch Orthop Unfallchir 62 (2):139-149
Reinert CM, Bosse MJ, Poka A, Schacherer T, Brumback RJ,
Burgess AR (1988) A modified extensile exposure for the
treatment of complex or malunited acetabular fractures. J
Bone Joint Surg [Am] 70 (3):329-337
Riachi E (1967) Clinical and radiological considerations of
various cases of traumatic fracture luxation of the hip. Rev
Med Moven-Orient 24:432-433
Rigault P (1962) Les enfoncements du cotyle. 63 cas. Thesis,
Paris
Rigault P, Hannouche R, Judet J (1968) Luxations traumatiques
728
de la hanche et fractures du cotyle chez I'enfant. Rev Chir
Orthop 54:361-382
Riska EB, von Bonsdorff H, Hakkinen S, Jaroma H, Kiviluoto
0, Paavilainen T: Surgical approach for reduction and inter-
nal fixation of central acetabular fractures. Injury 1:31 - 33
Ritter MA, Sieber JM (1985) Prophylactic indomethacin for the
prevention of heterotopic bone formation following total hip
arthroplasty. Clin Orthop 196:217 - 225
Rodrigues KF (1973) Injury of the acetabular epiphysis. Injury
4 (3):258-260
Rooser B, Bengtson S, Herrlin K (1989) A case of acetabular
fracture with hip tamponade. Relief from aspiration of hem-
arthrosis. Acta Orthop Scand 60:623 - 624
Routt ML lr, Swiontkowski MF (1990) Operative treatment of
complex acetabular fractures. Combined anterior and poste-
rior exposures during the same procedure. J Bone Joint Surg
[Am] 72:897 - 904
Rouviere H (ed) (1940) Traite d'anatomie humaine. Masson,
Paris
Rowe CR, Lowell JD (1961) Prognosis of fractures of the ace-
tabulum. J Bone Surg [Am] 43A:30-59
Rubenstein J, Kellam 1, McGonigal D (1983) Acetabular frac-
ture assessment with computerized tomography. J Can
Assoc Radiol 34:19-22
Russe 0 (1965) Hip joint dislocation with fracture. Langenbecks
Arch Klin Chir 313:550-554
Ruter A, Burri C (1979) Surgical treatment of acetabular frac-
tures (in German). Hefte Unfallheilkd 140:105 -121
Rutten M (1969) Experiences with fractures of the acetabulum.
Z Orthop 106:626 - 630
Sajko D, Barac M, Nikolic V, Ranic V (1977) Significance of
preoperative diagnosis of fractures in the acetabular region
(in Serbo-Croatian, Roman). Acta Chir Iugosl 24 [Suppl
2]:277-280
Saks Bl (1986) Normal acetabular anatomy for acetabular frac-
ture assessment: CT and plain film correlation. Radiology
159:139-145
Sarmiento A, Laird CA (1973) Posterior fracture dislocation of
the femoral head. Report of a case. Clin Orthop 92: 143 - 146
Sawaguchi T, Brown TD, Rubash HE, Mears DC (1984) Stabili-
ty of acetabular fractures after internal fixation. A cadaveric
study. Acta Orthop Scand 55:601-605
Schade R (1970) Hip joint dislocation fractures. Beitr Orthop
Traumatol 17:431-438
Schmelzeisen H (1974) Diagnosis and classification of
acetabular fractures (in German). Monatsschr Unfallheilkd
77 (8): 379- 387
Schmelzeisen H (1979) Results of the surgical treatment of ace-
tabular fractures (in German). Hefte Unfallheilkd 140:
161-165
Schmidt SA, Kjrersgaard-Anderson P, Pedersen NW, Kristensen
SS, Pederson P, Nielsen lB (1988) The use of indomethacin
to prevent the formation of heterotopic bone after total hip
replacement: a randomized, double-blind clinical trial. 1
Bone 10int Surg [Am] 70:834-838
Scholten ET, van der Lande BA, Willemse AP, Elzenga P (1986)
Computed tomography with multi planar reconstruction of
acetabular fractures. Diagn Imaging Clin Med 55
(4/5):203 - 209
Schramm W (1965) On late results of dislocations and fractures
of the acetabulum of the hip. Langenbecks Arch Klin Chir
313:554-558
Schweikert CH, Weigand H (1979) Results after conservative
and surgical therapy of acetabular fractures (in German).
Hefte Unfallheilkd 140:166-180
Bibliography
Scott WW Jr, Fishman EK, Magid D (1987) Acetabular frac-
tures: optimal imaging. Radiology 165 (2):537-539
Scott WW, Magid D, Fishman EK, Riley LH Jr, Brooker AF Jr,
10hnson CA (1987) Three-dimensional imaging of acetabu-
1ar trauma. J Orthop Trauma 1:227 - 232
Senegas 1 (1984) Long-term results of surgery of acetabular
fractures (in French). Acta Orthop Belg 50:366 - 380
Senegas 1, Liorzou G, Yates M (1980) Complex acetabular frac-
tures: a transtrochanteric lateral surgical approach. Clin Or-
thop 151:107-114
Shea KP, Kalamchi A, Thompson GH (1986) Acetabular
epiphysis-labrum entrapment following traumatic anterior
dislocation of the hip in children. 1 Pediatr Orthop
6:215-219
Shirkhoda A, Brashear HR, Staab EV (1980) Computed tomog-
raphy of acetabular fractures. Radiology 134 (3):683 - 688
Shtutin ALA, Proskura VB (1974) Acetabular fractures and
their treatment (in Russian). Ortop Travmatol Protez
10:44-45
Simonka lA, Endrodi 1, Frater L (1981) Difficulties in recognis-
ing the non-displaced acetabular fracture (in Hungarian).
Magy Traumatol Orthop Helyreallito Sebesz 24 (2):138-142
Slawson RG, Poka A, Bathon H, Salazar OM, Bromback Rl,
Burgess AR (1989) The role of post-operative radiation in
the prevention of heterotopic ossification in patients with
post-traumatic acetabular fracture. Int 1 Radiat Oncol BioI
Phys 17 (3):669-672
Smith-Petersen MN (1917) A new supra-articular subperiosteal
approach to the hip joint. Ann 1 Orthop Surg 15:592
Snook GA (1965) Interposition of the joint capsule in traumatic
posterior dislocation of the hip. J Trauma 5:358-361
Soler R (1979) Luxation centrale de la hanche chez I'enfant. Rev
Orthop Trauma 23 I B (2):235
Sommelet 1 (1966) Thoughts on the treatment of fracture dislo-
cations of the hip. 1 Chir (Paris) 92:473-480
Soreff 1 (1976) Fracture-dislocation of the hip with entrapment
of the femoral head by impaction against the acetabular rim.
Injury 8 (2):127 -128
Soukup B, Pilnacek 1 (1987) Fractures of the femoral head asso-
ciated with acetabular injury - Pipkin's fracture (in Cech).
Acta Chir Orthop Traumatol Czech 54 (2):160-166
Spencer RF (1989) Acetabular fractures in older patients. 1 Bone
10int Surg [Br] 71:774-776
Stankovic F, Kammerer H, Wurm G, Sattel W (1975) Our ex-
periences in the treatment of acetabular fractures (in Ger-
man). Hefte Unfallheilkd 124:257 - 259
Stewart Ml, Milford W (1954) Fracture dislocations of the hip:
an end result study. J Bone 10int Surg [Am] 36:315 - 342
Stryhal F (1967) Inveterate posterior traumatic hip dislocation.
Acta Chir Orthop Traumatol Cech 34:262-270
1lmton 1 (1916) Traite des fractures, vol 4. Bailliere, Paris
Tillie B, Fontaine C, Stahl PH et al. (1987) Apport de la
tomodensitometrie au diagnostic et au traitment des frac-
tures du cotyle. A propos de 88 cas. Rev Chir Orthop
73:15-24
Tondelli P, Morscher E, Riiedi T (1977) Internal fixation of
acetabular fractures: results after 1 to 6 years (in German).
Unfallheilkunde 80 (5):197-203
Toni A, Gulino G, Baldini N, Gulino F (1985) Clinical and
radiographic long term results of acetabular fractures asso-
ciated with dislocations of the hip. Ital 1 Orthop Traumatol
11 (4):443-454
Toombs BD, Rayschkolb EN, Gibbs Bl (1982) CT of acetabular
fractures: software and image display. AJR 139:416-417
(letter)
Bibliography
Truchet P (1962) Aspects radiologiques des fractures du cotyle.
Lyon Chir 58:423-425
Urist MR (1948) Fractures of the acetabulum. The nature of
traumatic lesions and 2 years end results. Ann Surg
127:1150-1164
Vacher D (1966) Les incarcerations fragmentaires au cours des
fractures du cotyle. Thesis, Paris
Vacher D, Letournel E, Judet R (1968) Les incarcerations fragen-
taires au cours des fractures du cotyle. Presse Med
9:411-414
Vailas JC, Hurwitz S, Wiesel SW (1989) Posterior acetabular
fracture-dislocations: fragment size, joint capsule, and sta-
bility. J Trauma 29: 1494 - 1496
Vanni E, Franceschi (1964) Traumatic dislocation and fracture
dislocation of the coxo-femoral head with special reference
to the results. Minerva Ortop 15:51-62
Vasilev T, Asparukhov A (1981) Characteristics of acetabular
fractures and their treatment (in Russian). Ortop Travmatol
Protez 3:21-24
Vidal J (1966) Les fractures transcotylOidiennes. La place de
l'osteosynthese. Montpellier Chir 12:41- 55
Virenque M, Pasquie P (1956) Les voies d'abord de la face pel-
vienne du cotyle. Toulouse Med 57 (1):11-27
Wadsworth PL, Henry WB (1974) Dorsal surgical approach to
acetabular fractures in the dog. J Am Vet Med Assoc 165
(10):908-910
Walker RH, Burton DS (1982) Computerized tomography in
assessment of acetabular fractures. J 1tauma 22:227 - 234
Waller A (1955) Dorsal acetabular fractures of the hip (dash-
board fractures). Acta Chir Scand Suppl 205
Walther C (1981) Recherches experimentales sur certaines frac-
tures de la cavite cotylOide. Bull Soc Anat Paris 66:561- 566
Wechselberger F (1975) Conservative treatment of acetabular
fractures and late results (in German). Hefte Unfallheilkd
124:55-57
Weise K, Weller S (1987) Conservative therapy of acetabular
729
fractures - indications and results (in German). Aktuel
Traumatol 17 (6):277 - 283
Weller S (1967) On the treatment of fractures in the region of the
acetabulum (in German). Hefte Unfallheilkd 91:30-32
Weller S, Schmelzeisen H (1978) Indications of and techniques
for the operative treatment of acetabular fractures (in Ger-
man). Unfallheilkunde 81 (4):264-274
Weller S, Schmelzeisen H (1978) Diagnosis and therapy of
acetabular fractures (in German). Beitr Orthop 1taumatol
25 (8):436-446
Wheaton LG, Hohn RB, Harrison JW (1973) Surgical treatment
of acetabular fractures in the dog. J Am Vet Med Assoc 162
(5):385 - 392
Wilchinsky ME, Pappas AM (1985) Unusual complications in
traumatic dislocation of the hip in children. J Pediatr Or-
thop 5:534-539
Wingstrand H, Egund N, Nilsson LT, Strbmqvist B (1988)
Acetabular fracture causing hip joint tamponade. A case
report. Acta Orthop Scand 59 (3):323 - 325
Winter WG, Combs CR, Schenkar DL (1984) Central fracture
of the acetabulum: adjunctive innominate osteotomy as an
aid in operative reduction. J Bone Joint Surg [Am]
66:130-132
Wittebol P (1970) Early results of operative treatment of central
hip dislocation. Ned Tijdschr Geneeskd 114:109-110
Yanagital Y (1969) Management of central dislocation of the
femur in 3 cases. Experience up to the time of establishment
of a therapeutic policy. Orthop Surg (Tokyo) 20:27 - 31
Ylinen P, Santavirta S, Slatis P (1989) Outcome of acetabular
fractures: a 7-year follow-up. J Trauma 29 (1):19-24
Zelev Z (1967) On the operative treatment of traumatic disloca-
tion of hip joint with fracture of the posterior rim of acetab-
ulum. Folia Med (Plovdiv) 9:304-308
Zeumer G (1966) On the treatment and prognosis of acetabulum
fractures. Zentralbl Chir 91: 126 - 136
Subject Index

Acetabulo-obturator line 32 posterior column and posterior mobility 558

Acetabulum wall 189 Delayed operative treatment 635
anatomy 17 transverse and posterior wall 201 non-union 652
classification of fractures 64 T-shaped 163 preconditions for surgery 635
fractures 32 Avascular necrosis 545 surgical approach 636
fracture reduction 411 aetiology 549 Diphosphonates 556
mechanics of injury 23 results 551 Dislocation of hip
posterior border 32 clinical presentation 333
roof 32 Bilateral acetabular fractures 507 neglected 591
teardrop 33 Both column fractures
vascular supply 22 anterior column component 256 Ectopic bone formation
Age distribution of patients 330 choice of surgical approach 401 aetiology 559
Anterior and posterior hemitransverse computed tomography imaging 267 classification 372, 558
fractures curved image 262, 284 cranio-cerebral trauma 562
choice of surgical approach 401 differential diagnosis 267 extended ilio-femoral approach
classification 231 fixation recommendations 496 394
computed tomography imaging 246 key to reconstruction 259 ilio-inguinal approach 384
differential diagnosis 245 morphology 253 Kocher-Langenbeck approach 372
epiphyseal separation 245 operative treatment technique 484 late post-operative complica-
morphology 231 posterior column component 254 tions 629
operative treatment technique 475 radiology 261 post-operative 558
posterior fracture characteristics 233 secondary congruence 258 prevention 560
radiology 234 spur sign 263, 290 results of surgical excision 562
roof segment 239 Brooker ectopic ossification classifica- treatment 559
Anterior column tion 372, 558 Elementary fractures
anatomy 18 classification 64
screw fixation 444 Capsular injuries 324 posterior column fractures 85
Anterior column fractures Cartilage necrosis 543 posterior wall fractures 67
classification 115 Children 343 Epiphyseal separations
computed tomography imaging 128 Classification of fractures of the posterior column 94, 343
fixation techniques 434 acetabulum 64 anterior and posterior hemitransverse
morphology 115 Columns of the acetabulum 18 fractures 245
operative treatment technique 431 Combined anterior and posterior Extended ilio-femoral surgical approach
radiology 119 approaches 385 application 391
Anterior wall fractures in both-column fractures 502 complications 394
computed tomographic imaging Complications ectopic bone formation 394
111 early post-operative 535 technique 386
morphology 103 late post-operative 542 Extra-acetabular fractures of the
operative treatment technique 428 Computed tomography pelvis 316
quadrilateral surface fragment 103 landmarks 47
radiology 104 technique 45 Femoral head injuries 323
Antero-posterior radiograph Conservative management Femoral nerve palsies 341
landmarks 31 indications 347
technique 30 methods 348 Gull sign 99
Antibiotic treatment 412 results 349
Anticoagulant treatment 413 Coumarin treatment 414 Head-roof congruency 521
Associated fractures Cranio-cerebral trauma 562 Heparin treatment 414
anterior with posterior Curved image 262, 284 Heterotopic bone formation (see Ec-
hemitransverse 231 topic bone formation)
both-column 253 D'Aubigne hip assessment Hip injuries associated with acetabular
classification 64 clinical assessment 565 fractures 326
732 Subject Index

Iliac-oblique radiograph Mortality, post-operative 535, 621 antibiotics 412

landmarks 43 anticoagulant treatment 413
technique 30 Neurological complications 537 Post-operative complications
Ilio-femoral surgical approach Non-union 595 avascular necrosis 545, 626
application 374 post-operative complication 652 cartilage necrosis 543
complications 375 Nutrient arteries of the acetabulum 21 death 535
technique 373 ectopic ossification 558, 629
Ilio-inguinal approach Obturator-oblique radiograph infection 535
application 381 landmarks 37 non-union 626
complications 383 technique 30 osteophytes 551
ectopic ossification 384 Oilier surgical approach 364 pseudarthrosis 541
technique 375 Operating table 403 sciatic nerve damage 537
Ilio-ischial line 35 Operative fracture management secondary displacement 539
Ilio-pectineal eminence 19 delayed 635 thrombo-embolism 540
Ilio-psoas tendon gutter 19 indications 359 wound complications 540
Indomethacin 560 learning curve 528 Pre-operative care 399
Innominate bone rationale 358 Pseudarthrosis 541
anatomy 17 timing of surgery 359
bone structure for screw fixa- Osteoarthritis Radiography
tion 402 after perfect reduction 554 antero-posterior view 30
trabeculae 20 post-operative 553 computed tomography 45
vascular supply 20 Osteophytes, post-operative 551 generalities 29
Infection, post-operative iliac-oblique view 30
after ilio-inguinal approach 383 Parallelism 521 obturator-oblique view 30
early complications 535, 621 Paralysed hips 507 tomography 49
treatment 537 Pathologic fractures 343 Radiological results
Instruments Pelvic bone (see also Innominate bone) early 522
ball spike 406 injuries associated with acetabular late 569
forceps 403 fractures 325 Radiology
retractors 406 Pelvic brim 36, 38 ilio-ischial line 35
Internal fixation screw insertion 436 pelvic brim 36
implants 407 Plates for osteosynthesis 409 radiological "U" 33
method 410 Posterior column roof of the acetabulum 32
screws 402, 409 anatomy 17 teardrop 33
Intra-articular bone fragments screw fixation 470 Radiotherapy
diagnosis 339 Posterior column fractures post-operative 560
incarceration 337 classification 85 Reduction assessment 521
treatment 507, 604 computed tomographic imaging 101 and osteoarthritis 554
epiphyseal separations 94 Claude Martimbeau technique 533
Kocher surgical approach 363 extended posterior column frac- immediate radiological assess-
Kocher-Langenbeck surgical tures 93 ment 522, 574
approach 364 gull sign 99 imperfect reduction 526
application 369 operative treatment technique 421 Joel Matta technique 522
complications 371 radiology 90 Reduction techniques
ectopic bone formation 372 transitional fractures 99 acetabular fractures 411
extension into tri-radiate Posterior column and posterior wall femoral head 411
approach 395 fractures Results, early
technique 365 classification 189 summary 580
computed tomography imaging 194 Results, late
Lateral cutaneous nerve of the distinction from posterior column clinical 573
thigh 382 fractures 191 mal-union 661
Langenbeck surgical approach 363 morphology 189 non-union 652
Learning curve 528 operative treatment technique 447 non-union/mal-union 665
Line radiology 189 radiological 569
acetabulo-obturator 32 Posterior wall fractures summary 580
ilio-ischial 35 classification 67 Retractor
computed tomographic imaging 78 sciatic nerve 406
Mal-union 595, 661 operative treatment technique 417 Retro-cotyloid surface 17
Marginal impaction postero-inferior fractures 72 Retro-peritoneal haematoma 325, 333
appearance on computed postero-superior fractures 69
tomography 49 radiology 68 Sciatic buttress 20
treatment 417 with incomplete transverse frac- Sciatic nerve lesions
Mechanism of injury ture 76 post-operative 537, 625
Morel-Lavalle lesion 337 Post-operative care 412 pre-operative 334
Subject Index
mechanism of injury 335
Screw fixation 402, 409
anterior column 444
insertion along pelvic brim 435, 441
positioning around the
acetabulum 420
posterior column 470
Secondary congruence
definition 258
surgical 521, 579
Secondary displacement 539
Smith-Petersen surgical approach 363
Spur sign 263, 290
Sub-cotyloid groove 17
Surgical approaches
choice of approach 399
extended ilio-femoral approach 386
ilio-femoral approach 373
ilio-inguinal 375
Kocher approach 363
Kocher-Langenbeck approach 364
Langenbeck approach 363
Oilier approach 364
Smith-Petersen approach 364
tri-radiate approach 395
Teardrop 33
Technique
anterior column fractures 431
anterior column screw 444
anterior wall fractures 428
approach, choice of 399
associated anterior and posterior
hemitransverse fractures 475
associated both-column frac-
tures 484
associated posterior column and
posterior wall fractures 447
associated transverse and posterior
wall fractures 455
bilateral fractures of the
acetabulum 507
combined approaches 385
extended ilio-femoral approach
386
ilio-femoral approach 373
ilio-inguinal approach 375
incarcerated bone fragments 507
internal fixation 410
Kocher-Langenbeck approach 364
paralysed hip 507
posterior approach (Kocher-
Langenbeck) 364
posterior column fractures 421
posterior wall fractures 417
sacro-iliac joint injury
screw fixation 402, 436
transverse fractures 442
tri-radiate surgical approach 395
T-shaped fractures 461
Thrombo-embolism
anticoagulant prophylaxis 413
post-operative complication 540
Tomography 49
Traction 348
Transitional fractures 315
Transverse fractures
choice of surgical approach 400
733
classification 141
computed tomography imaging 155
evaluation of displacement 150, 155
fixation techniques 443
mechanism of displacement 142
morphology 141
operative treatment technique 442
Transverse and posterior wall fractures
choice of surgical approach 400
classification 201
computed tomography imaging 221
direction of dislocation 213
large postero-superior fragment 213
morphology 202, 208
operative treatment technique 455
radiology 203, 211
Tri-radiate surgical approach 395
T-shaped fractures
choice of surgical approach 400
classification 163
computed tomography imaging 179
confirmation of the stem compo-
nent 169
displacement 167
morphology 163
operative treatment technique 461
radiology 167
Urinary tract injuries 328
Vascular injuries 324
Visceral injuries 328
Wound complications 540