The use of the Ilizarov method as a salvage

procedure in infected nonunion of the distal

femur with bone loss

A. Saridis, We reviewed 13 patients with infected nonunion of the distal femur and bone loss, who had
E. Panagiotopoulos, been treated by radical surgical debridement and the application of an Ilizarov external
M. Tyllianakis, fixator. All had severely restricted movement of the knee and a mean of 3.1 previous
C. Matzaroglou, operations. The mean length of the bony defect was 8.3 cm and no patient was able to bear
N. Vandoros, weight.
E. Lambiris The mean external fixation time was 309.8 days. According to Paleys grading system,
eight patients had an excellent clinical and radiological result and seven excellent and good
From The University functional results. Bony union, the ability to bear weight fully, and resolution of the
of Patras, Rio Patras, infection were achieved in all the patients. The external fixation time was increased when
Greece the definitive treatment started six months or more after the initial trauma, the patient had
been subjected to more than four previous operations and the initial operation had been
open reduction and internal fixation.

A. Saridis, MD, Consultant
Orthopaedic Surgeon

E. Panagiotopoulos, MD,
Associate Professor

M. Tyllianakis, MD,
Associate Professor

C. Matzaroglou, MD,
Consultant Orthopaedic
Surgeon

N. Vandoros, MD,
Consultant Orthopaedic
Surgeon

E. Lambiris, MD, Professor
Department of Orthopaedics
University of Patras, Rio
Patras 26504, Greece.
Correspondence should be
sent to Dr E.
Panagiotopoulos; e-mail:
ecpanagi@med.upatras.gr
2006 British Editorial
Society of Bone and
Joint Surgery
doi:10.1302/0301-620X.88B2.
16976 $2.00
J Bone Joint Surg [Br]
2006;88-B:232-7.
Received 14 July 2005;
Accepted 7 October 2005
In distal femoral fractures bone loss and asso-
ciated contamination in open high-energy frac-
tures, devitalisation of metaphyseal bone
fragments from extensive surgical exposure1
and the use of bulky internal fixation may
result in an infected nonunion. A short distal
fragment, poor bone quality and compromised
soft tissues are also associated with this out-
come. Union may be achieved with deformity
and leg-length discrepancy leading to post-
traumatic osteoarthritis and stiffness of the
knee.2-5 This may necessitate arthrodesis as a
salvage procedure,6 and in some cases amputa-
tion may be required.
Although the Ilizarov external device has
been described for use in the management of
infected nonunion in general,7-9 to our knowl-
edge there have been few reports regarding the
treatment of infected nonunion of the distal
femur with bone loss. In this retrospective
study we have analysed the results of treatment
of this condition using the Ilizarov external
device in an attempt to salvage the limb and to
maximise its function.
Patients and Methods
We reviewed the records of 13 adult patients
who were treated with the Ilizarov system for
infected nonunion of the distal femur and bone
loss between January 1993 and December
2001. The inclusion criteria were the presence
of such a problem with bone loss of 3 cm or
more at a maximum distance of 9 cm from the
knee.4
There were ten men and three women with a
mean age of 34.6 years (19 to 55). The non-
union resulted from open high-energy complex
fractures of the distal femur, type III B in ten
and type III C in three according to the classi-
fication of Gustilo and Anderson.10 Road-traf-
fic accidents were the cause in 11 patients and
gunshot injuries in the remaining two, with the
left side affected in seven. The mean interval
from the original injury to application of the
Ilizarov fixator was 12.3 months (3 to 33). The
initial treatment of the fracture consisted of
open reduction and internal fixation (ORIF) in
five patients, unilateral transarticular external
fixation in seven and hybrid external fixation
in one (Table I). A total of 40 operations had
been performed before the definitive treat-
ment, including internal and external fixation,
intramedullary nailing, bone grafting, wound
debridement and the use of myofasciocutane-
ous flaps. The mean number of previous surgi-
cal procedures was three (1 to 5). The mean
number of interventions for the five patients
who had initial ORIF was four operations and
for those with initial external fixation it was
two.
The mean length of the bone defect was 8.3
cm (3 to 18). In five patients the infection was
active but without drainage, and in the remain-
ing eight there was active purulent drainage.

232 THE JOURNAL OF BONE AND JOINT SURGERY

 THE USE OF THE ILIZAROV METHOD AS A SALVAGE PROCEDURE IN INFECTED NONUNION OF THE DISTAL FEMUR WITH BONE LOSS 233

Table I. Details of the patients at the time of the definitive treatment

Age Type of Time to application Number of previous Operations before application Infecting Bone defect
Case (yrs) fracture10 of fixator (mths) operations of the Ilizarov device* organism (cm)
1 55 III B 18 5 ORIF > TEF > IM nail + BG > Staph. aureus 9
WD > MFCF
2 29 III B 4 4 ORIF > BG > WD + MFCF > Staph. aureus 10
WD
3 24 III B 33 3 ORIF > BG > WD + BG E. coli 8
4 29 III C 11 3 ST > ORIF > WD Staph. aureus 18
5 40 III C 4 2 TEF > WD Staph. aureus 3
6 49 III B 29 4 TEF > BG > BG > WD Pseudomonas 7
7 30 III B 3 2 TEF > WD Pseudomonas 3
8 32 III B 4 2 TEF > WD Staph. aureus 7
9 50 III C 3 1 TEF Staph. aureus 6
10 20 III B 4 4 Hybrid EF > BG > WD > IM nail + BG Streptococcus B 4
11 19 III B 14 2 TEF > WD Staph. aureus 8
12 46 III B 10 4 ORIF > WD > TEF + MFCF > Pseudomonas 18
WD + BG
13 27 III B 24 4 TEF > WD > WD > WD + BG Pseudomonas 7
* ORIF, open reduction and internal fixation; IM, intramedullary; EF, external fixator; BG, bone grafting; WD, wound debridement; TEF, transarticular
external fixator; ST, skeletal traction; MFCF, musculofasciocutaneous flap

Table II. Details of treatment in the 13 patients

Docking Lengthening EFT

Case Technique of osteosynthesis* site Cross knee Additional procedure (cm) (days)
1 BFCD-CO AS No FM-2, RD-1 8 547
2 BFCC-DO delayed corticotomy (after 3 mths) AD Yes FM-2, RD-2, MFCF Re-application of frame 9 607
3 BFCC-DO AD No Iliac grafts 6 163
4 BFCD-CO AS Yes FM-2, RD-1 16 493
5 BFCC-DO AD No Quadricepsplasty 2 138
6 BFCD-CO AS Yes No 5 263
7 BFCC-DO delayed corticotomy (after 1 mth) AD No No 2 136
8 BFCC-DO AD No No 6 221
9 BFCC-DO AD No No 6 153
10 BFCC-DO AD No Iliac grafts, RD-1 4 227
11 BFCC-DO AD No FM-1 8 312
12 BFCC-DO AD No FM-2 Quadricepsplasty 17 548
13 BFCC-DO AD Yes No 6 220
* BFCC-DO, bifocal combined compression-distraction osteosynthesis; BFCD-CO, bifocal consecutive distraction-compression osteosynthesis
AD, acute docking; AS, acute shortening
FM, frame modification; RD, repeated debridement; MFCF, musculofasciocutaneous flap
EFT, external fixation time

Staphylococcus aureus was the responsible organism in
seven patients, Pseudomonas aeruginosa in four and
Escherichia coli and streptococcus type B in two patients
each (Table I).
At the time of definitive treatment all patients had severe
limitation of movement of the knee and in four, this joint
was completely stiff. In some cases it was impossible to be
certain whether the movement clinically detected was
occurring at the knee or at the adjacent nonunion or in
combination. Seven patients had moderate or severe post-
traumatic osteoarthritis of the knee with radiologically-
identified irregularity of the articular surface and a
decreased joint space. Moderate to severe disuse osteopenia
of the distal femoral fragment and extensive scarring were
present in every patient and none was able to bear weight.
Pre-operatively the location of sinus tracts, previous
scars, and the neurovascular condition were carefully eval-
uated. Imaging included plain radiography, tomography,
CT, MRI and scintigraphy, which were used to identify the
severity of the infection and the extent of infected and
necrotic bone. Intra-operatively, sinography with Methy-
lene Blue was used in eight patients with active infection to
delineate the extent of the infection and bone destruction.
Nevertheless, being in the metaphyseal area, the full extent
of the infection could be determined only by inspection at
operation.1
Operative technique. The infected area was exposed in all
13 patients by an extensive lateral approach through the
less damaged soft tissues in order to remove all loose
implants and all infected and non-viable tissue. Samples
thus obtained were sent for Gram staining, culture and sen-
sitivity (aerobic/anaerobic) and fungal and mycobacterial
cultures.
The bone ends were freshened, the medullary canal
reconstituted and open reduction of the main fragments
was performed to achieve the best possible contact between

VOL. 88-B, No. 2, FEBRUARY 2006

234 A. SARIDIS, E. PANAGIOTOPOULOS, M. TYLLIANAKIS, C. MATZAROGLOU, N. VANDOROS, E. LAMBIRIS

Fig. 1b

Diagrams of the Ilizarov external fixator showing

a) a standard bifocal circular frame for infected
nonunion of the distal femur and b) fixation of the
distal femur by olive wires.

Fig. 1a

them. Impaction of the shaft of the femur into the condylar
area was attempted to increase internal stability at the site
of the defect in each case.9,11
Acute docking was possible in ten patients, while acute
shortening of the site of the defect in order to improve soft-
tissue cover and circulation was performed in three with
bone loss of more than 6 cm. In our experience acute short-
ening for more than 6 cm of femoral bone loss compro-
mises the soft tissues and may give deleterious effects.
After the acute shortening or acute docking an Ilizarov
external device (Smith & Nephew Orthopaedics, Memphis,
Tennessee) was applied. Primary wound closure was per-
formed in eight cases followed by adequate drainage. Two
different techniques of bifocal osteosynthesis7,9 were used
(Table II). In ten patients bifocal combined compression-
distraction osteosynthesis was undertaken. In two of these
with extensive soft-tissue damage and aggressive purulent
infection, a second-stage corticotomy was performed at one
and three months post-operatively. In three patients bifocal
consecutive distraction-compression osteosynthesis was
carried out. Proximal, subtrochanteric, corticotomy was
performed in each to restore initial femoral length by filling
the bone defect.
A standard bifocal circular frame for infected nonunion
with bone loss at the distal femur consisted of three or four
rings and one proximal semicircular ring (Fig. 1). In four
patients with short distal fragments and considerable insta-
bility in the interfragmentary area, transarticular circular
ring fixation on the proximal tibia without hinges was used
to enhance stability and to protect the distal fixation.3,12
Additional surgical procedures after the application of the
Ilizarov frame were required in some patients to eradicate
the infection and encourage union. These included repeated
debridement in four, modification of the external fixator in
five, a myofasciocutaneous flap in one and freshening of the
bone ends and iliac bone grafting in two with sclerotic avas-
cular bone ends and no progress to union at the docking
site.
Post-operative care. All patients received a course of anti-
biotics for at least four to six weeks post-operatively
according to the microbiological results. The erythrocyte
sedimentation rate and the level of C-reactive protein were
monitored on a regular basis. Thromboprophylaxis with
low-molecular-weight heparin was given to each patient.
The latency period before commencing bone transportation
was three to five days and the rate of distraction at the site
of the corticotomy ranged from 0.5 to 1.25 mm per day
depending on the regenerative capacity, pain experienced at
the corticotomy and the reaction at the hip. When docking
was achieved, interfragmentary compression was contin-
ued at the rate of 0.5 mm per day for five to seven days and
once consolidation had commenced at a rate 0.25 to 0.5 mm

THE JOURNAL OF BONE AND JOINT SURGERY

 THE USE OF THE ILIZAROV METHOD AS A SALVAGE PROCEDURE IN INFECTED NONUNION OF THE DISTAL FEMUR WITH BONE LOSS 235

Fig. 2a Fig. 2b Fig. 2c Fig. 2d

Fig. 2e Fig. 2f

Figures 2a and 2b Pre-operative a) anteroposterior and b) lateral radiographs of a 24-year-old man with infected nonunion and bone loss (type III B)
of the left distal femur. The initial treatment had consisted of plating and immediate patellectomy because of a concomitant open severely comminuted
fracture of the patella. Figures 2c and 2d Anteroposterior radiographs of the proximal and distal femur, two months after the application of the Ilizarov
device showing proximal femoral lengthening and a good reduction of the site of nonunion. Figures 2e and 2f Anteroposterior and lateral radiographs
of the site of e) the fracture and f) the proximal femoral corticotomy 26 months after operation showing healing at the site of nonunion and at the length-
ening sites.

per week. Patients were encouraged to bear partial weight
progressively with crutches on the second day after surgery.
Quadriceps isometric exercises were started immediately
after the operation to maintain or increase movement of the
knee and muscle strength.
After the removal of the external fixator, the patients
were restricted to partial weight-bearing for four to six
weeks and no brace was used. Full weight-bearing was
allowed between the fourth and tenth week post-opera-
tively, based on clinical and radiological evidence of healing

VOL. 88-B, No. 2, FEBRUARY 2006

236 A. SARIDIS, E. PANAGIOTOPOULOS, M. TYLLIANAKIS, C. MATZAROGLOU, N. VANDOROS, E. LAMBIRIS
at the nonunion and at the site of lengthening. All patients
returned for routine, clinical and radiological follow-up
every month after operation until the Ilizarov device was
removed, and every three months in the following year.
Post-operative radiographs were evaluated for residual
malalignment and evidence of union. Radiological union
was defined as the absence of a radiolucent line at the site of
the nonunion and filling of the bone defect with new bone13
at a minimum of three cortices on standard anteroposterior
and lateral radiographs.14 The results were assessed using
the functional and radiological scoring system described by
Paley and Maar.15
Results
The mean follow-up after removal of the frame was 42.4
months (19 to 72). No patient was lost to follow-up.
The mean external fixator time was 309.8 days (136 to
607) which correlated with the final lengthening achieved
(Table II). The mean external fixator time when definitive
treatment with application of the Ilizarov frame began early
(< 6 months after the initial trauma) was 247 days, whereas
when there was a delay the mean time was 363 days, an
increase of 49.6%.
The number of previous operations was also of some pre-
dictive value. When the patient had been subjected to one
to three operations before initiation of definitive treatment,
the external fixator time was 228 days but for four and five
operations it was 367.8 days, an increase of 61.3%. Addi-
tionally, the nature of the previous operations was of con-
siderable predictive value in determining such time. In our
series, in five patients (cases 1, 2, 3, 4 and 12) with primary
ORIF, the mean time was prolonged to 471.6 days com-
pared with the total mean of 309.8 days, an increase of
52.4%, and reduced to 208.7 days in the remaining
patients with conventional external fixation as the initial
treatment, a decrease of 32.6%. In the ORIF group the end
functional result was also worse; in four patients it was fair
and in one poor.
Bone. Clinical and radiological union and elimination of
infection were achieved in all 13 patients. According to
Paley and Maars grading system,15 eight patients had
excellent, four good, and one a fair result. Five patients
developed mild axial deformity of more than 5, two at the
regenerate site, two at the site of nonunion and one at both
sites. Residual leg-length discrepancy greater than 2.5 cm
was found in only one patient.
Function. At the time of the latest follow-up, all 13 patients
were able to bear weight fully on the affected leg without
any walking aid or brace. They had no pain or only mild
pain when undertaking activities of daily living. The func-
tional result was excellent in three patients, good in four,
fair in four, and poor in two (Fig. 2). Although seven
patients had an obvious limp, they were still able to per-
form the normal activities of daily living. In three patients
(cases 5, 7, and 9) with an interval between injury and
application of the Ilizarov external fixator of less than four
months, a full range of movement of the knee was regained.
A reduction of knee flexion to less than 90 was observed in
the remaining ten patients. Five had satisfactory functional
movement. Improvement in knee function occurred in most
of these patients up to six months after removal of the
frame. In five patients who had initial ORIF, a large bone
defect and prolonged external fixation time (cases 1, 2, 3, 4
and 12), movement of the knee was restricted to 0 to 10.
Two of these patients underwent quadricepsplasty after the
removal of the Ilizarov device in an attempt to address the
stiffness, but had no significant improvement. No patient
required arthrodesis of the knee.
Complications. There were no neurovascular intra-opera-
tive complications and no patient developed a neurovascu-
lar deficit or a compartment syndrome. The most common
complication was pin-track infection especially in the distal
segment as a result of severe osteopenia and poor soft-tissue
cover. In nine patients with grade-3 infection according to
Paleys classification,16 the wires had to be removed and
replaced under local anaesthesia. Prevention of pin-track
infection was attempted by insertion of the wires through
healthy skin, re-tensioning of loose wires and the use of
counter-opposing olive wires, meticulous cleaning and,
occasionally, administration of oral antibiotics after tests
for culture and sensitivity. Continuous mild pain requiring
analgesia was felt in all patients during the distraction
phase. One non-displaced re-fracture at the docking site,
which occurred two months after premature removal of the
external fixator, was treated by re-application of the
Ilizarov frame for three months.
Discussion
Infected nonunion of the distal femur with associated bone
loss represents one of the most difficult problems in ortho-
paedic practice.17 Apart from high-energy trauma, predis-
posing factors include the proximity to the knee and the
distracting pull of gastrocnemius, previous surgical inter-
ventions with extensive dissection and inadequate initial
fixation.5 The presence of infection in this area makes treat-
ment even harder. Specific problems related to this condi-
tion include lost bone with poor soft-tissue cover and
scarring, deformity, shortening, post-traumatic osteoarthri-
tis and stiffness in the knee.3,5
In our study, the cause of all the fractures was high-
energy trauma with extensive disruption of the soft-tissue
envelope and blood supply. Despite the fact that our series
was relatively small it is noteworthy that the external fix-
ator time reflected the extent of bone restoration required,
the delay in commencement of definitive treatment, the
number of previous interventions and the nature of the ini-
tial stabilisation of the fracture.
Unilateral temporary transarticular external fixators can
be applied easily and rapidly and are popular for the initial
management of high-energy fractures of the femur. These
protect the remaining blood supply, avoid shortening and
gross malalignment, while allowing adequate access for
THE JOURNAL OF BONE AND JOINT SURGERY
 THE USE OF THE ILIZAROV METHOD AS A SALVAGE PROCEDURE IN INFECTED NONUNION OF THE DISTAL FEMUR WITH BONE LOSS
wound care. However, because of eccentric loading there is
an increased risk of angular and shear displacement at the
site of the fracture.18 Current studies report good and satis-
factory results with primary application of the Ilizarov
external fixator for severely comminuted and open frac-
tures of the distal femur with extensive soft-tissue
injury.12,19,20
The use of the Ilizarov external device for definitive fixa-
tion of fractures of the distal femur is the safest and most
effective method since it provides adequate fixation and sta-
bilisation at the site of the fracture, thus increasing osseous
healing and decreasing infection.12,19,20 In the presence of
infection aggressive debridement has been shown to be the
cornerstone of successful treatment.21 In our series the use
of the Ilizarov method offered limb salvage with no need
for arthrodesis or amputation.
No benefits in any form have been received or will be received from a commer-
cial party related directly or indirectly to the subject of this article.
References
1. Kretek C, Schandelmaier P, Tscherne H. Distal femoral fractures: transarticular
reconstruction, percutaneous plate osteosynthesis and retrograde nailing.
Unfallchirurg 1996;99:2-10 (in German).
2. Wu CC, Shih CH. Distal femoral non-union treated with interlocking nailing.
J Trauma 1991;31:1659-62.
3. Ali F, Saleh M. Treatment of distal femoral non-unions by external fixation with
simultaneous length and alignment correction. Injury 2002;33:127-34.
4. Bellabarba C, Ricci WM, Bolhofner BR. Indirect reduction and plating of distal
femoral non-unions. J Orthop Traums 2002;16:287-96.
5. Johnson KD, Hicken G. Distal femoral fractures. Orthop Clin North Am 1987;18:
115-32.
VOL. 88-B, No. 2, FEBRUARY 2006
237
6. Webb LX. Bone defect non-union of the lower extremity. Tech Orthop 2001;164:
387-97.
7. Ilizarov GA. Transosseous osteosynthesis. Theoretical and clinical aspects of the
regeneration and growth of tissue. Berlin: Springer-Verlag, 1992.
8. Green SA. Skeletal defects: a comparison of bone grafting and bone transport for
segmental skeletal defects. Clin Orthop 1994;301:111-17.
9. Paley D, Catagni MA, Argnani F, et al. Ilizarov treatment of tibial non-unions with
bone loss. Clin Orthop 1989;241:146-65.
10. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand
and twenty five open fractures of long bones: retrospective and prospective analyses.
J Bone Joint Surg [Am] 1976;58-A:453-8.
11. Song HR, Cho SH, Koo KH, et al. Tibial bone defects treated by internal bone trans-
port using the Ilizarov method. Int Orthop 1998;22:293-7.
12. Arazi M, Memik R, Ogun TC, Yel M. Ilizarov external fixation for severely commi-
nuted supracondylar and intercondylar fractures of the distal femur. J Bone Joint Surg
[Br] 2001;83-B:663-7.
13. Wang JW, Weng LH. Treatment of distal femoral non-union with internal fixation,
cortical allograft struts, and autogenous bone-grafting. J Bone Joint Surg [Am]
2003;85-A:436-40.
14. Banaszkiewicz PA, Sabboubeh A, McLeod I, Maffulli N. Femoral exchange nail-
ing for aseptic non-union: not the end to all problems. Injury 2003;34:349-56.
15. Paley D, Maar DC. Ilizarov bone transport treatment for tibial defects. J Orthop
Trauma 2000;14:76-85.
16. Paley D. Problems, obstacles and complications of limb lengthening by the Ilizarov
technique. Clin Orthop 1990;250:81-104.
17. Bolhofner BR, Carmen B, Clifford P. The result of open reduction and internal fix-
ation of distal femur fractures using a biologic (indirect) reduction technique. J Orthop
Trauma 1996;10:372-7.
18. Aronson J. Limb-lengthening, skeletal reconstruction, and bone transport with the
Ilizarov method. J Bone Joint Surg [Am] 1997;79-A:1243-58.
19. Hutson JJ Jr, Zych GA. Treatment of comminuted intraarticular distal femur frac-
tures with limited internal and external tensioned wire fixation. J Orthop Trauma
2000;14:405-13.
20. Ramesh LJ, Rajkumar SA, Rajendra R, et al. Ilizarov ring fixation and fibular strut
grafting for C3 distal femoral fractures. J Orthop Surg (Hong Kong) 2004;12:91-5.
21. Marsh DR, Shah S, Elliot J, Kurdy N. The Ilizarov method in non-union, malunion
and infection of fractures. J Bone Joint Surg [Br] 1997;79-B:273-9.