Vous êtes sur la page 1sur 235


indd ii 1/30/13 3:01 PM

Articular Cartilage Injury
of the Knee
Basic Science to Surgical Repair


Stannard_9781604068580_FM.indd i 1/30/13 3:01 PM

Stannard_9781604068580_FM.indd ii 1/30/13 3:01 PM
Articular Cartilage Injury
of the Knee
Basic Science to Surgical Repair

James P. Stannard, MD
J. Vernon Luck Distinguished Professor and Chairman
Department of Orthopaedic Surgery
University of Missouri Hospital
Columbia, Missouri

James L. Cook, DVM, PhD

William and Kathryn Allen Distinguished Professor in Orthopedic Surgery
Comparative Orthopaedic Laboratory
University of Missouri
Columbia, Missouri

Jack Farr, MD
Director, OrthoIndy Sports Medicine Fellowship
Director, OrthoIndy Cartilage Restoration Center of Indiana
Indiana Orthopaedic Hospital
Voluntary Professor of Orthopedic Surgery, IUMC
Indianapolis, Indiana

New York Stuttgart

Stannard_9781604068580_FM.indd iii 1/30/13 3:01 PM

Thieme Medical Publishers, Inc.
333 Seventh Ave.
New York, NY 10001

Executive Editor: Kay Conerly

Managing Editor: Judith Tomat
Senior Vice President, Editorial and Electronic Product Development: Cornelia Schulze
Production Editor: Torsten Scheihagen
International Production Director: Andreas Schabert
Vice President, Finance and Accounts: Sarah Vanderbilt
President: Brian D. Scanlan
Compositor: Thomson Digital
Printer: Everbest Printing Co.

Library of Congress Cataloging-in-Publication Data

Stannard, James P.
Articular cartilage injury of the knee : basic science to surgical repair / James P. Stannard, James L.
Cook, Jack Farr.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-60406-858-0 (hardcover) ISBN 978-1-60406-859-7 (e-ISBN)
I. Cook, James L. (James Lee), 1965- II. Farr, Jack. III. Title.
[DNLM: 1. Knee Injuriessurgery. 2. Cartilage, Articularinjuries. 3. Cartilage, Articularsurgery.
4. Knee Jointsurgery. 5. Tissue Engineering. WE 870]

Copyright 2013 by Thieme Medical Publishers, Inc. This book, including all parts thereof, is legally protected
by copyright. Any use, exploitation, or commercialization outside the narrow limits set by copyright legislation
without the publishers consent is illegal and liable to prosecution. This applies in particular to photostat
reproduction, copying, mimeographing or duplication of any kind, translating, preparation of microfilms, and
electronic data processing and storage.

The chapters in this book have been published previously in The Journal of Knee Surgery.

Important note: Medical knowledge is ever-changing. As new research and clinical experience broaden our
knowledge, changes in treatment and drug therapy may be required. The authors and editors of the material
herein have consulted sources believed to be reliable in their efforts to provide information that is complete
and in accord with the standards accepted at the time of publication. However, in view of the possibility of
human error by the authors, editors, or publisher of the work herein or changes in medical knowledge, neither
the authors, editors, nor publisher, nor any other party who has been involved in the preparation of this work,
warrants that the information contained herein is in every respect accurate or complete, and they are not
responsible for any errors or omissions or for the results obtained from use of such information. Readers are
encouraged to confirm the information contained herein with other sources. For example, readers are advised
to check the product information sheet included in the package of each drug they plan to administer to be
certain that the information contained in this publication is accurate and that changes have not been made in
the recommended dose or in the contraindications for administration. This recommendation is of particular
importance in connection with new or infrequently used drugs.

Some of the product names, patents, and registered designs referred to in this book are in fact registered
trademarks or proprietary names even though specific reference to this fact is not always made in the
text. Therefore, the appearance of a name without designation as proprietary is not to be construed as a
representation by the publisher that it is in the public domain.

Printed in China


ISBN 978-1-604068580

Also available as an e-book:

eISBN 978-1-60406-859-7

Stannard_9781604068580_FM.indd iv 1/30/13 3:01 PM

Dedicated to the pioneers of cartilage surgery and the scientists
and surgeons who continue to put the science behind the surgery
in finding better solutions for our patients. May the strong body
of work that is reviewed in this book provide the foundation
and inspiration for the future.

Stannard_9781604068580_FM.indd v 1/30/13 3:01 PM

Stannard_9781604068580_FM.indd vi 1/30/13 3:01 PM

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

I. Diagnosis and Treatment Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Chapter 1 Staging and Comorbidities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Christian Lattermann and Matthew R. Luckett
Chapter 2 Magnetic Resonance Imaging of Cartilage Repair Techniques . . . 13
Catherine Hayter and Hollis Potter
Chapter 3 Evolving Biomarkers in Osteoarthritis. . . . . . . . . . . . . . . . . . . . . . . 32
Debabrata Patra and Linda J. Sandell
Chapter 4 Using Animal Models in Osteoarthritis Biomarker Research . . 42
Bridget C. Garner, Aaron M. Stoker, Keiichi Kuroki,
Richard Evans, Cristi Reeves Cook, and James L. Cook

II. Science and Techniques for Cartilage Repair . . . . . . . . . . . . . . . . . . . . . . . 59

Chapter 5 The CartilageBone Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Caroline D. Hoemann, Charles-Hubert Lafantaisie-Favreau,
Viorica Lascau-Coman, Gaoping Chen, and Jessica Guzmn-Morales
Chapter 6 Science and Animal Models of Marrow Stimulation
for Cartilage Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Lisa A. Fortier, Brian J. Cole, and C. Wayne McIlwraith
Chapter 7 Microfracture and Augments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Andreas H. Gomoll
Chapter 8 ACI and MACI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Elizaveta Kon, Giuseppe Filardo, Alessandro Di Martino,
and Maurilio Marcacci
Chapter 9 Particulated Articular Cartilage: CAIS and DeNovo NT . . . . . . . . 102
Jack Farr, Brian J. Cole, Seth Sherman, and Vasili Karas
Chapter 10 Nontraditional Modification to Articular Cartilage . . . . . . . . . . . 111
Vasili Karas, Neil Ghodadra, Ellen Kroin, and Brian J. Cole vii

Stannard_9781604068580_FM.indd vii 1/30/13 3:01 PM

Chapter 11 Osteochondral Autograft Transplantation/Mosaicplasty . . . . . . 118
Brett McCoy and Anthony Miniaci
Chapter 12 Improved Preservation of Fresh Osteochondral
Allografts for Clinical Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Aaron M. Stoker, Joeseph T. Garrity, Clark T. Hung,
James P. Stannard, and James L. Cook
Chapter 13 Osteochondral Allograft Transplantation in the Knee . . . . . . . . . 141
William Bugbee, Marco Cavallo, and Sandro Giannini
Chapter 14 The Use of Scaffolds in the Treatment of
Osteochondral Lesions in the Knee: Current Concepts
and Future Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Aad A. M. Dhollander, Victor R. Guevara Snchez,
Karl F. Almqvist, Ren Verdonk, Gust Verbruggen,
and Peter C. M. Verdonk
Chapter 15 Toward Engineering a Biological Joint Replacement . . . . . . . . . . 162
Grace D. OConnell, Eric G. Lima, Liming Bian,
Nadeen O. Chahine, Michael B. Albro, James L. Cook,
Gerard A. Ateshian, and Clark T. Hung

III. Postop Management and Outcomes Assessments . . . . . . . . . . . . . . . . . . 175

Chapter 16 Postoperative Management of Patients with
Articular Cartilage Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Jennifer Yasu Stone and Robert Schaal
Chapter 17 Biomechanical Outcomes of Cartilage Repair of the Knee . . . . . 184

Carmen E. Quatman, Joshua D. Harris, and Timothy E. Hewett

Chapter 18 Clinical Outcomes Assessment for Articular
Cartilage Restoration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Kai Mithoefer and Marco Acuna

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209


Stannard_9781604068580_FM.indd viii 1/30/13 3:01 PM


Articular cartilage health is critical to human composites with chapters on the bone carti-
movement. Because articular cartilage injury lage interface, autografts, and fresh allografts.
and degeneration are commonplace, effec- This section also includes a chapter on a novel
tive repair and regeneration strategies have osteochondral allograft preservation system
become an area of intense research with rapid that allows for superior maintenance of chon-
advancement. Unfortunately, the basic science drocyte viability such that the window of
and clinical data and articles in this area are time for implantation of grafts is more than
spread across many disciplines, sources, and doubled, and individual graft viability can be
journals. This book is our attempt to bring the assessed prior to clinical use.
worlds experts together to provide a contem- The final chapters concentrate on scaffolds
porary update on the topic of articular carti- for cartilage repair from the basic science level
lage repair of the knee in a single resource. to clinical application. We also feature a chapter
Our first goal is to marry basic science with on the challenges and possibilities of achieving
a comprehensive patient-based approach to a biologic knee replacement for human patients.
diagnosis and treatment. It is impossible to A chapter addresses biomechanical outcomes in
appropriately choose current options or apply cartilage replacement therapies while another
exciting new developments without a thorough concentrates on the clinical rehabilitation of
understanding of the basic science of articular these patients. Finally, we provide a chapter on
cartilage in health and disease. As such, we have assessment of outcomes after cartilage repair in
engaged basic science experts who also have the knee.
strong clinical backgrounds to bridge the divide Treatment of articular cartilage pathology of
that can limit effective clinical application of the knee is challenging. Our goal with this text
current and emerging treatment options. is to provide a comprehensive and up-to-date
The initial chapters of Articular Cartilage reference for surgeons and researchers work-
Injury of the Knee concentrate on staging and ing in this exciting and dynamic field. We hope
comorbidities, diagnostic imaging now and in you find it as useful and interesting as we have
the future, and the development and validation while editing the text.
of biomarkers for the early diagnosis, staging,
clinical decision making, and prognostication James P. Stannard, MD
of patients with articular cartilage pathology. James L. Cook, DVM, PhD
The subsequent chapters address the basic sci- Jack Farr, MD
ence and clinical aspects of marrow stimula-
tion, autologous chondrocyte implantation About the cover images
(ACI) and new developments with ACI, par- Top: Intraoperative photo of medial femoral
ticulated allograft cartilage therapy, and non- condylar articular cartilage lesion. Bottom:
traditional modifications of articular cartilage. Photo of lesion treated with investigational
The next section of the book concentrates on cartilage restoration technique. (Courtesy Jack
the topic of transplantation of bone cartilage Farr, MD)

Stannard_9781604068580_FM.indd ix 1/30/13 3:01 PM

Stannard_9781604068580_FM.indd x 1/30/13 3:01 PM
List of Contributors

Marco Acuna, MD Gaoping Chen, MD

Department of Clinical Services Department of Chemical Engineering
National Institute of Rehabilitation cole Polytechnique
Mexico City, Mexico Montreal, Quebec, Canada

Michael B. Albro, PhD

Brian J. Cole, MD, MBA
Department of Mechanical Engineering
Division of Sports Medicine
Columbia University
Department of Orthopedics
New York, New York
Rush University Medical Center
Karl F. Almqvist, MD, PhD Chicago, Illinois
Department of Orthopaedic Surgery and
Traumatology Cristi Reeves Cook, DVM, MS
Ghent University Hospital Comparative Orthopaedic Laboratory
Ghent, Belgium University of Missouri
Columbia, Missouri
Gerard A. Ateshian, PhD
Department of Mechanical Engineering
James L. Cook, DVM, PhD
Columbia University
William and Kathryn Allen Distinguished
New York, New York
Professor in Orthopedic Surgery
Liming Bian, PhD Comparative Orthopaedic Laboratory
Department of Mechanical and Automation University of Missouri
Engineering Columbia, Missouri
Biomedical Engineering Programme
The Chinese University of Hong Kong Aad A. M. Dhollander, MD, PT, PhD
Hong Kong Department of Orthopaedic Surgery and
William Bugbee, MD Laboratory of Connective Tissue Biology
Division of Orthopaedic Surgery Department of Rheumatology
Scripps Clinic Ghent University Hospital
La Jolla, California Ghent, Belgium
Marco Cavallo, MD
Il Clinic of Orthopaedics and Traumatology Alessandro Di Martino, MD
Rizzoli Orthopaedic Institute Biomechanics Laboratory-III Orthopaedic
Bologna, Italy Clinic
Rizzoli Orthopaedic Institute
Nadeen O. Chahine, PhD Bologna, Italy
Department of Bioengineering
The Feinstein Institute for Medical Research Richard Evans, PhD xi
Manhasset, New York Waukesha, Wisconsin

Stannard_9781604068580_FM.indd xi 1/30/13 3:01 PM

Jack Farr, MD Jessica Guzmn-Morales
Director, OrthoIndy Sports Medicine Department of Chemical Engineering
Fellowship cole Polytechnique
Director, OrthoIndy Cartilage Restoration Montreal, Quebec, Canada
Center of Indiana
Indiana Orthopaedic Hospital Joshua D. Harris, MD
Voluntary Professor of Orthopedic Surgery, Department of Orthopaedics
IUMC The Ohio State University
Indianapolis, Indiana Columbus, Ohio

Giuseppe Filardo, MD Catherine Hayter, MBBS

Biomechanics Laboratory-III Orthopaedic Department of Radiology and Imaging
Clinic New York, New York
Rizzoli Orthopaedic Institute
Bologna, Italy Timothy E. Hewett, PhD, FACSM
Sports Health and Performance Institute
Lisa A. Fortier, DVM, PhD Columbus, Ohio
Department of Clinical Science
Cornell University Caroline D. Hoemann, PhD
Ithaca, New York Department of Chemical Engineering
Institute of Biomedical Engineering
Bridget C. Garner, DVM, PhD Groupe de Recherche en Sciences et
Department of Veterinary Pathology Technologies Biomedicales
University of Georgia cole Polytechnique
Athens, Georgia Montreal, Quebec, Canada
List of Contributors

Joeseph T. Garrity, MS Clark T. Hung, MSE, PhD

Comparative Orthopaedic Laboratory Department of Biomedical Engineering
University of Missouri Columbia Columbia University
Columbia, Missouri New York, New York

Neil Ghodadra, MD Vasili Karas, BS

Division of Sports Medicine Division of Sports Medicine
Department of Orthopedics Department of Orthopedics
Southern California Orthopedic Institute Rush University Medical Center
Van Nuys, California Chicago, Illinois

Sandro Giannini, MD Elizaveta Kon, MD

Il Clinic of Orthopaedics and Traumatology Biomechanics Laboratory-III Orthopaedic Clinic
Rizzoli Orthopaedic Institute Rizzoli Orthopaedic Institute
Bologna, Italy Bologna, Italy

Andreas H. Gomoll, MD Ellen Kroin, BS

Harvard Medical School Division of Sports Medicine
Cartilage Repair Center Department of Orthopedics
Brigham and Womens Hospital Rush University Medical Center
Boston, Massachusetts Chicago, Illinois

Victor R. Guevara Snchez, MD Keiichi Kuroki, DVM, PhD

Ortopediay Tramatologia Comparative Orthopaedic Laboratory
Instituto Nacional de Rehabilitacion University of Missouri
Mexico City, Mexico Columbia, Missouri

Stannard_9781604068580_FM.indd xii 1/30/13 3:01 PM

Charles-Hubert Lafantaisie-Favreau, MSc Kai Mithoefer, MD
Institute of Biomedical Engineering Harvard Vanguard Orthopedics
cole Polytechnique and Sports Medicine
Montreal, Quebec, Canada Boston, Massachusetts

Viorica Lascau-Coman, MSc Grace D. OConnell, PhD

Department of Chemical Engineering Department of Biomedical Engineering
Groupe de Recherche en Sciences et Columbia University
Technologies Biomedicales New York, New York
cole Polytechnique
Montreal, Quebec, Canada Debabrata Patra, PhD
Department of Orthopaedic Surgery
Christian Lattermann, MD Washington University School of Medicine
Department of Orthopaedic Surgery and at Barnes-Jewish Hospital
Sports Medicine St. Louis, Missouri
University of Kentucky
Lexington, Kentucky Hollis Potter, MD
Department of Magnetic Resonance Imaging
Eric G. Lima, PhD Hospital for Special Surgery
Department of Mechanical Engineering New York, New York
Columbia University
New York, New York Carmen E. Quatman, MD
Sports Health and Performance Institute
Department of Orthopaedics
Matthew R. Luckett, MD
The Ohio State University
Department of Orthopaedic Surgery and
Columbus, Ohio

List of Contributors
Sports Medicine
University of Kentucky
Linda J. Sandell, PhD
Lexington, Kentucky
Department of Cell Biology and
Maurilio Marcacci, MD Washington University School of
Biomechanics Laboratory-III Orthopaedic Medicine at Barnes-Jewish Hospital
Clinic St. Louis, Missouri
Rizzoli Orthopaedic Institute
Bologna, Italy Robert Schaal, PT, BSPT
Department of Rehabilitation
C. Wayne McIlwraith, DVM, PhD Missouri Orthopedic Institute
Orthopaedic Research Center Columbia, Missouri
Department of Clinical Sciences
Colorado State University Seth Sherman, MD
Fort Collins, Colorado Division of Sports Medicine
Department of Orthopedics
Brett McCoy, MD Rush University Medical Center
Department of Orthopaedic Surgery Chicago, Illinois
The Cleveland Clinic Foundation
Garfield Heights, Ohio James P. Stannard, MD
J. Vernon Luck Distinguished Professor and
Anthony Miniaci, MD Chairman
Department of Orthopaedic Surgery Department of Orthopaedic Surgery
The Cleveland Clinic Foundation University of Missouri Hospital
Garfield Heights, Ohio Columbia, Missouri


Stannard_9781604068580_FM.indd xiii 1/30/13 3:01 PM

Aaron M. Stoker, MS, PhD Ren Verdonk, MD, PhD
Comparative Orthopaedic Laboratory Department of Orthopaedic Surgery and
Department of Veterinary Medicine and Traumatology
Surgery Ghent University Hospital
University of MissouriColumbia Ghent, Belgium
Columbia, Missouri
Peter C. M. Verdonk, MD, PhD
Jennifer Yasu Stone, PT, DPT, OCS Department of Orthopaedic Surgery
Department of Rehabilitation Stedelijk Ziekkenhuis Roeselare
Missouri Orthopedic Institute Roeselare, Belgium
Columbia, Missouri Department of Orthopaedic Surgery and
Gust Verbruggen, MD, PhD Ghent University Hospital
Laboratory of Connective Tissue Biology Ghent, Belgium
Department of Rheumatology
Ghent University Hospital
Ghent, Belgium
List of Contributors


Stannard_9781604068580_FM.indd xiv 1/30/13 3:01 PM

Diagnosis and Treatment

Stannard_9781604068580_Ch01.indd 1 1/30/13 2:01 PM

Stannard_9781604068580_Ch01.indd 2 1/30/13 2:01 PM
Staging and Comorbidities
Christian Lattermann and Matthew R. Luckett

Articular cartilage injuries are common.13 arthroscopic grading, and sizing as well
The spectrum of these injuries ranges from as assessment of the joint environment.
small, superficial defects (focal chondral Furthermore, specific comorbidities have to
defects) to complete degenerative delami- be taken into account prior to performing
nation of entire condyles with or without cartilage repair procedures as many of them
involvement of the subchondral bone and require additional staged or concomitant
adjacent structures (osteoarthritis). In an surgical procedures. In this chapter, we will
ideal world, focal chondral defects exist sequentially discuss the most pertinent fac-
in isolation, have clearly defined borders, tors that influence the decision-making pro-
are solitary defects, and are located in ide- cess in patients with symptomatic cartilage
ally accessible anatomic locations in young lesions of the knee.
patients that are physically active. These
types of lesions are the standard that is cur-
rently being used to enroll patients into Frequency and Prevalence
randomized clinical trials investigating the of Cartilage Injuries
efficiency of articular cartilage procedures.
Whereas these studies are important and Damage to articular cartilage is common
necessary to compare different techniques, and can result from acute traumatic injuries,
the reality is that most patients (95%) that are early posttraumatic degenerative changes,
presenting with clinically symptomatic carti- developmental factors affecting the subchon-
lage lesions do not fit these clear-cut criteria.4 dral bone such as osteochondritis dissecans
This presents a dilemma to the surgeon as the (OCD) lesions, or acquired metabolic factors
cartilage lesions most commonly treated are such as avascular necrosis (AVN).13
usually less clear cut and often involve best Articular cartilage lesions are frequently
clinical judgment to perform an adequate encountered in routine knee arthrosco-
assessment. This assessment process, or pies. Curl et al reported articular cartilage
staging, is necessary to guide both patient lesions in as many as 63% of over 35,000 knee
and physician toward a clinically feasible arthroscopies in the United States.3 This high
and satisfying solution for the knee cartilage incidence was corroborated by Hjelle et al in
injury patient. The staging process requires Norway and Widuchowski et al in Poland,
knowledge about frequency and prevalence who reported an incidence of 61 and 60%,
of cartilage defects, their clinical symptoms, respectively.1,2 The average age of patients

Stannard_9781604068580_Ch01.indd 3 1/30/13 2:01 PM

reported in these studies is high, and thus the osteochondral lesions, 68% of which were
percentage of treatable lesions in younger focal chondral lesions, 3% being OCD lesions
patients is likely much lower. In fact, upon and 29% being osteoarthritic lesions.8
further subanalysis of Curls data, 60% of the
reported lesions were grade III lesions and
thus were potentially treatable lesions. Only History and Physical
1,750 patients out of 31,516 were under the Examination
age of 40 and had Outerbridge grade III lesions.
Based upon this study, one can estimate that The clinical evaluation of patients with
5% of patients under 40 undergoing knee symptomatic cartilage lesions in the knee is
arthroscopies may present with a chondral difficult and follows the recommendations of
lesion that would be considered optimal a thorough history and physical exam of the
for current therapies. While these studies knee joint. No true evidence-based approach
provide some data regarding prevalence of is available to guide the clinician, but sev-
these types of lesions among patients, no eral factors that may be important should be
information is available regarding how many pointed out.
of the lesions are clinically symptomatic. Upon initial evaluation, it is important to
Interestingly, the mere presence of a lesion discover the history of symptoms that may be
does not seem to lead to an increase in the related to a cartilage lesion. Duration of symp-
I Diagnosis and Treatment Planning

osteoarthritic rate over time in large cross- toms has been associated with clinical out-
sectional studies, as the long-term natural come in patients undergoing microfracture.
history study conducted by Widuchowski et Mithoefer et al could show that patients with
al in 2010 suggests.5 Shelbourne et al found symptoms longer than 1 year had lower over-
that 123 out of 2,700 patients with anterior all subjective outcome results than patients
cruciate ligament (ACL) injuries and cartilage with more acute cartilage injuries.9 There is
lesions at the time of surgery showed lower a correlation of worse overall clinical out-
subjective Noyes scores 8 years after ACL comes after cartilage procedures in patients
reconstruction compared with the patients who receive workmens compensation.10,11
who did not have cartilage lesions at the History of smoking and family history of OA
time of surgery.6 Another study suggests that are often considered negative predictive fac-
the presence of cartilage lesions can lead tors for cartilage repair procedures; however,
to rapid progression of radiographic osteo- no clear evidence exists to actually link those
arthritis (OA), as documented by Messner two isolated factors to clinical outcomes.
and Maletius.7 These findings underline the History should include the documentation
importance of identifying the patient who of the body mass index (BMI). Whereas a BMI
has a clinically symptomatic cartilage lesion up to 35 does not seem to affect the overall
that may benefit from early treatment. outcomes in patients undergoing cell-based
cartilage procedures,12,13 a higher BMI clearly
affects the results of patients undergoing
Lesion Location and Size microfracture treatment.14 Similar consid-
eration needs to be given to the age factor.
The location of cartilage lesions is spread Several studies have shown that higher age
between the three compartments of the knee. influences clinical outcome negatively in
Lesions are most commonly found in the patients undergoing microfracture proce-
weight-bearing femoral condyle (43 to 58%). dures.14,15 The data for cell-based procedures
Patellar lesions are frequently encountered and are somewhat conflicting. A clear correlation
account for 11 to 36% of all lesions. Trochlear between age and clinical outcome has not
lesions overall are less frequent (6 to 16%).13 been shown. Basic science studies, however,
When analyzed for the lesion size, Hjelle suggest that chondrocytes from older donors
et al were able to show that the majority of ( 40 years of age) have a lower proteogly-
lesions (88%) were below 4 cm.1,2 can and collagen production and thus may
Widuchowski et al found that 60% of knees respond more slowly and less vigorously to
4 (average 39 years old) contained chondral/ the challenging intra-articular environment

Stannard_9781604068580_Ch01.indd 4 1/30/13 2:01 PM

after implantation.16 A little-researched topic well as ligamentous stability of the tibiofem-
that is of importance is the willingness to com- oral and patellofemoral joint.
ply with postoperative treatment protocols A crude visual gait analysis in the office
and rehabilitation procedures. Current proto- usually allows for detection of an antalgic
cols are not based upon evidence but rather gait, quadriceps avoidance gait, or a dynamic
on anecdotal experience or small case series varus or valgus thrust. Any of these findings,
by individual surgeons and rehabilitation spe- if present, can point the examiner toward
cialists.1719 Nevertheless, it is felt that adher- further underlying pathologies that may
ence to these basic protocols is important. A have a significant impact on the chosen treat-
history of noncompliance may therefore be a ment options. A varus thrust, for example,
warning sign to the cartilage surgeon poten- may point out an insufficiency of the lateral
tially indicating the patients lack of under- ligamentous structures (posterolateral cor-
standing or a significant difference in the goals ner, lateral collateral ligament [LCL]) and a
that the treatment is aiming to achieve. triple varus. A quadriceps avoidance gait may
indicate chronic anterior instability.
Knee joint effusions are generally felt to be
a significant sign for symptomatic cartilage
injuries. It is important to understand, how-
Pain assessment is an important part of the
ever, that intra-articular effusions can exist
preoperative exam. Localized pain may be
without pain and therefore can be present
able to pinpoint a specific area of articular
longer than the actual onset of pain.
cartilage damage or it may indicate injury
Range of motion assessment should be a

1 Staging and Comorbidities

to associated structures such as the menis-
routine part of the physical examination and
cus. The shorter the history of pain, the more
has to be assessed in comparison with the
reliably it can be considered to indicate the
uninjured side. Although small deficits in
affected area.
knee flexion can be observed with knee joint
No reliable data exist about the correlation
effusions, they are not normal in patients
of pain with a symptomatic cartilage lesion.
who have no effusion. An extension deficit is
However, the more chronic in nature the pain
an important finding as these are very diffi-
is, the less likely it is that a cartilage proce-
cult to correct and may indicate progression
dure alone is going to address the problem.
to OA already beyond the scope of cartilage
Most commonly utilized are visual analog
repair. Significant loss of motion is consid-
scales (VAS) or a Likert scale for pain.
ered a relative contraindication for cartilage
In absence of any clear evidence-based
repair procedures.
guidelines regarding pain, there are some
Mechanical symptoms, locking during the
pearls of wisdom that may help the less-
range of motion exam, or acute inability to
experienced cartilage surgeon. The ideal
flex or extend the knee joint may indicate an
patient should not report maximal pain
unstable meniscus or articular cartilage frag-
other than perhaps with heavy exertion.
ment or a loose body.
Likewise, patients with minimal or no pain
A clinical sign that utilizes this concept
are less likely to benefit from cartilage sur-
is the Wilson sign. This test was originally
gery. Typically, the patient reporting pain
performed to diagnose OCD lesions in the
in the midrange is considered an acceptable
medial femoral condyle. The knee is flexed
patient for treatment. It is also important
to 90 degrees. The tibia is forced into inter-
to assess pain with and without medication
nal rotation. Under gradual extension and
(particularly narcotic pain medication) in
external rotation of the tibia, the patient may
this context.
report pain when the lesion rotates into the
area of the soft spot of the medial femoral con-
Physical Examination dyle.20 This test can be modified by pushing
the thumb slightly into the soft spot. Another
The physical exam should evaluate the over- helpful test is the direct palpation of the
all dynamic and static alignment, antalgic medial and lateral patella facette. If palpation
gait, range of motion, muscle envelope, as is reproducing the patients pain, this can be 5

Stannard_9781604068580_Ch01.indd 5 1/30/13 2:01 PM

a sign for a clinically symptomatic lesion in well as the competency of the medial patel-
this area and will need to be correlated with lofemoral ligament (MPFL) and the lateral
the imaging results. Cartilage lesions do not retinaculum should be assessed. The patel-
typically hurt directly at the joint line. Direct lar apprehension test is helpful to rule out
palpation at the joint line is more likely asso- the previous patellar sub/dislocation. Q-angle
ciated with meniscal pathology. and patellar stability throughout the flexion
Ligamentous stability is a prerequisite for should be carefully evaluated.
cartilage procedures. It is therefore necessary
to perform a full ligament examination of the
knee joint. This usually includes varus and
valgus stress at 0 and 30 to test the collateral The Character of the Lesion
ligaments; the Lachman test; the pivot shift
exam, which evaluates ACL competency; the To assess the actual severity of a lesion,
posterior drawer test at 90 degrees of knee arthroscopic evaluation is imperative. The
flexion; and the posterior sag sign, which grading of the severity can be done using
evaluates posterior cruciate ligament (PCL) several different classification systems. The
sufficiency. In case of a potential posterolat- International Cartilage Research Society (ICRS)
eral corner injury, the dial test and the flex- has developed a universally accepted and
ion rotation drawer can be performed. Often comprehensive grading system that should
I Diagnosis and Treatment Planning

forgotten is the stability exam of the patella. be utilized to allow for the generalization of
The medial and lateral patella glide and tilt as arthroscopic findings (Fig. 1.1).

6 Fig. 1.1 ICRS grading scheme for cartilage defects. ICRS, International Cartilage Research Society.

Stannard_9781604068580_Ch01.indd 6 1/30/13 2:01 PM

approach. In other cases, it will be necessary
to perform either a biopsy with or without a
minor procedure such as a chondroplasty,
a partial meniscectomy, or a removal of a loose
body. The arthroscopy offers the unique
opportunity to assess and verify the loca-
tion, grading, and actual size of the lesion.
Additionally, it allows for assessment of the
entirety of a compartment, including the
status of the articular cartilage surfaces of
the tibial and femoral condyles surround-
ing a full-thickness lesion, as well as the
status of the meniscus, which often has been
treated in a prior procedure (the majority of
patients undergoing cartilage repair proce-
dures have had more than one previous sur-
gical procedure10,12). Particularly, globalized
Fig. 1.2 This is a Grade 4b lesion in a medial femoral findings such as compartment-wide grade 1
condyle after direct trauma. This patient was involved or 2 changes (ICRS) can elude radiographic
in a motor vehichle accident 3 months prior to this assessment but may indicate a more gener-
image and had a penetrating trauma to the knee. alized chondropenia in the affected compart-
ment. Development of osteophytes along the

1 Staging and Comorbidities

medial or lateral condylar rim is another sign
To restore articular cartilage, it is impor- for more generalized changes in the knee
tant to understand the reason for the initial that can easily be missed in X-ray and MRI
failure of the cartilage surface to maintain examination but may be a factor to be taken
its integrity. In a few cases, this can be asso- into account for the assessment of the future
ciated with an acute injury (Fig. 1.2). In success of a cartilage procedure. This arthro-
many cases, however, the underlying rea- scopic evaluation may also help to advise
son is more subtle. Even more importantly, the patient regarding the return to higher-
it is imperative to assess the true extent of level activities postsurgically. Figure 1.3
the chondral lesion. Diagnostic imaging has is an example of an isolated focal chondral
made incredible advances over the last dec- defect in an otherwise pristine knee joint
ade and is invaluable to characterize the (Fig. 1.3a). This is contrasted with an exam-
lesion and its surroundings better. Although ple of an isolated lesion in a knee joint
it is not the focus of this chapter (see chap- displaying grade 2 changes throughout the
ter 2), it needs to be understood that imag- entire compartment (Fig. 1.3b) indicating
ing provides information about the articular beginning chondropenia.
cartilage as well as the subchondral bone, As a final pearl regarding the arthroscopic
the synovial envelope, and the ligamentous examination, it should be noted that a video
structures of the knee joint. All of those need documentation of the lesion and the involved
to be assessed to create an overall picture compartment says more than an isolated
or character of a knee joint. An invalu- picture. In todays world, video documenta-
able tool to help synthesize all of the above- tion is easy, and it facilitates communication
mentioned aspects of information about the with colleagues and greatly improves the
patients knee is the arthroscopic evaluation surgeons recall of the character of a lesion
of the knee. For some procedures that allow in case of a likely time delay between the
for immediate point-of-care intervention, initial arthroscopy and the final restorative
such as the microfracture or the cartilage procedure. Another excellent alternative to
autologous implant system (CAISinvesti- improve communication is to combine images
gational and not currently available in the of defects taken with an intra-articular ruler
United States), this evaluation will be fol- and combine this with a map indicating the
lowed by an immediate final treatment size and location of the lesions. 7

Stannard_9781604068580_Ch01.indd 7 1/30/13 2:01 PM

Fig. 1.3 (a) An isolated Grade 3b defect in an receive an autologous chondrocytes implantation
otherwise pristine-appearing knee joint. This patient involving the majority of her condyle (2.2 4.8 cm).
went on to receive a microfracture and did well. Even though this is obvious on the video of this lesion,
I Diagnosis and Treatment Planning

(b) A similar-size Grade 3b lesion (indicated with the it is difficult to document this significant difference in
circle) surrounded by areas of Grade 2 lesions. This the character of this lesion in pictures.
patient failed an initial microfracture and went on to

Comorbidities is higher. For this reason, most surgeons will

consider the age of 50 a cutoff point for cell-
Prior to considering a cartilage repair pro- based procedures or allografts; however,
cedure, it is essential to perform a thorough some autologous procedures may be per-
analysis of comorbidities that potentially formed in patients up to the age of 60.2224 It
influence the success of the procedure or needs to be understood that the biological
may even be contraindications. age of the patient plays a larger role than the
Absolute contraindications for cartilage chronologic age. This may account for the
repair procedures are the documented pres- relatively soft recommendation of the age
ence of inflammatory arthritis (i.e., psoriatic, cutoff for these procedures.2527 Malalignment,
gouty, and rheumatoid) or established com- meniscus deficiency, or ligamentous instabil-
partmental OA with radiographic changes ity, even though they represent contraindica-
indicating joint space collapse (Kellgren tions to a cartilage repair procedure, can be
Lawrence IIIIV) or malignancy in the overcome by either a staged or a simultane-
involved limb. Uncorrected axial malalign- ous operation to correct the condition.
ment is an absolute contraindication for
tibiofemoral cartilage repair procedures, as
is chronic uncorrected ligamentous instabil- Axial Malalignment
ity. The same holds true for the patellofemo-
ral joint. Malalignment or instability in the Varus or valgus malalignment of the knee
patellofemoral joint is considered a contrain- is the major contributing factor to compart-
dication if it remains uncorrected; however, ment overload and thus has to be addressed
most cartilage surgeons will address obvi- when a cartilage repair procedure is consid-
ous patellar malalignment and instability in ered to address a cartilage defect in the over-
face of a cartilage repair in the patellofemo- loaded compartment.2527 When addressing
ral joint.21 Significant loss of range of motion a cartilage defect surgically, the goal is to
or arthrofibrosis is also considered to be an restore the normal load distribution that
absolute contraindication. allows the repair cartilage to adjust to physi-
Consensus exists that in the younger ologic rather than nonphysiologic loads.
8 patient the potential for a successful outcome The goal for axial alignment correction in

Stannard_9781604068580_Ch01.indd 8 1/30/13 2:01 PM

cartilage repair procedures is therefore Lavigne reported their results in a group of
not an overcorrection, as popularized by patients that was divided into ACI (patients
Coventry28 and others, but rather to correct with normal PF alignment) and ACI with AMZ
back to neutral alignment. It is imperative (patients with clinically present PF malalign-
that the origin of the malalignment be iden- ment).32 Interestingly, the group that did not
tified. Generally, varus alignment originates receive the AMZ because they did not have
in the proximal tibia and valgus alignment patellofemoral malalignment did worse than
in the distal femur. However, in some cases the group with patellofemoral malalignment
this may be different. It is therefore prudent requiring an AMZ. This study suggested that
to do a full axial alignment measurement there is either an additional effect of the
of the tibia and femur rather than just the anterior unloading of the patellofemoral joint
overall mechanical axis evaluation on the or perhaps some subtle patellofemoral mala-
long leg alignment full cassette X-ray. With lignment that was not detected as this study
todays hardware options, low-profile plates was published prior to the establishment of
can be utilized to perform well-controlled the tibial tubercletrochlear groove (TT-TG)
open wedge high tibial or distal femoral oste- measurements that are used today to deter-
otomies to address varus or valgus alignment mine patellofemoral alignment.33 The poten-
up to 10 degrees. Malalignment correction tial to unload the patellofemoral joint by
above 10 degrees may require additional doing an anteriorization of the tibial tubercle
bone grafting or alternate techniques. by less than 1 cm has been shown by Rue et
al, who concluded that the patellofemoral
contact pressures measured by Tekscan can

1 Staging and Comorbidities

Patellofemoral Malalignment be reduced by 20%.34 Overall, cartilage proce-
dures in the patellofemoral joint can be con-
Cartilage injuries in the patellofemo- sidered a valuable treatment option as long
ral joint are amongst the most difficult to as an adequate evaluation and concomitant
treat. Technically these lesions are easily treatment of an underlying PF malalignment
accessible, but the analysis of concomitant are performed.
pathologies is difficult. This fact explains the
initial disappointing results that Brittberg
et al reported. They saw five out of seven Meniscal Deficiency
patients undergoing autologous chondrocyte
implantation (ACI) of the patellofemoral joint The menisci are critical for load sharing and
fail.29 The authors recognized the importance shock absorption. They act as a transmission
of patellofemoral alignment and tracking at a within the knee linking the femoral condyle
later time point and advocated the combina- with the tibia. They also contribute to joint
tion of the ACI procedure with concomitant, lubrication and knee stability. Particularly,
or staged, unloading and normalization of the the medial meniscus has been shown to be
patellar tracking in the PF joint. As of 2011, the most important secondary stabilizer
cell-based cartilage procedures in patients against anterior translation of the knee.35
with PF malalignment were routinely com- This critical role is commonly impaired as
bined with an anteromedialization (AMZ) of meniscus injuries are the most common
the tibial tubercle.30 knee injury requiring arthroscopic surgery
Since, the clinical experience has been in the United States. Biochemical, biome-
promising. Brittberg et al reported 11 of 17 chanical, as well as clinical, radiographic, and
patients with good and excellent results at patient-related outcomes data have clearly
2 years and slightly better results (13/19) at established a direct relation of loss of menis-
9 years, indicating a long initial postopera- cus tissue to impairment of all these param-
tive recovery time with improvement over 1 eters.36 Impressive data have been published
year, postoperatively.27 In Minas and Bryants by Baratz et al, who showed an increase in
study of 45 patients, the authors performed contact pressures of 75% and an overall
an AMZ in over 60% and reported 71% good increase of 235% in peak-contact pressures
and excellent results.31 Henderson and after subtotal meniscectomy.37,38 Lee et al 9

Stannard_9781604068580_Ch01.indd 9 1/30/13 2:01 PM

showed that the periphery of the meniscus contribute to a significant increase in the
is more important for the overall pressure size of the cartilage lesion, as Murrell could
distribution in the compartment than the show.43 He evaluated patients for 2 months
central portion.39 These data are encourag- and 2 years after ACL tear prior to stabiliza-
ing and may indicate that patients after par- tion and found a six times larger loss of car-
tial meniscectomy still have a nearly normal tilage in patients with longer-standing ACL
pressure distribution in the joint. An isolated insufficiency. In patients who had a combi-
partial meniscectomy therefore may not nation of ACL injury and meniscal tear, this
pose a significant short-term risk for a car- rate increased to 18 times after 2 years. It
tilage repair procedure. However, long-term has been shown that knee ligament stabil-
data exist linking partial meniscectomies to ity is important to preserve meniscal integ-
the development of OA over a 15-year time rity. Particularly, the interaction of the
span. These data are even more compelling in medial meniscus and the ACL is important
conjunction with a ligamentous instability.40 as the lack of the medial meniscus may lead
Patients who have undergone a subtotal to early failure of the ACL graft due to the
or complete meniscectomy or have suffered meniscuss function as a secondary restraint
a nonrepairable radial tear have pathologic to anterior tibial translation.35 In patients
pressure distribution that is detrimental to with chronic ACL instability and pain, it is
the weight-bearing articular cartilage and important to evaluate the primary factor
I Diagnosis and Treatment Planning

any repair tissue. In these cases, a menis- pain or instability. Patients who have only
cus transplant may need to be considered. instability-related pain episodes may be
Although the indications for meniscal trans- served well with a correction of the instabil-
plant are still evolving, they are generally ity alone. Patients who have pain only may
considered in patients who are young, have benefit from an osteotomy. Lattermann and
unicompartmental pain, a history of previ- Jakob showed in a retrospective study that
ous meniscectomy, normal ligamentous sta- ACL-insufficient patients with varus align-
bility, and normal or correctable alignment. ment who predominantly have pain but no
Gomoll et al have published their series of instability may significantly improve after
seven patients undergoing cartilage resto- high tibial osteotomy and may not require
ration, high tibial osteotomy, and meniscus any other procedure. In these cases, a staged
transplantation. They reported encourag- approach may be beneficial.42
ing results in this small series with signifi-
cant improvement of the International Knee
Documentation Committee subjective score, Conclusion
Knee Injury and Osteoarthritis Outcome
Score, and Lysholm score after 24 months The careful evaluation of patients undergo-
(average) follow-up.41 As these patients are a ing cartilage repair procedures is of foremost
very challenging group, they can achieve sig- importance because these patients generally
nificant improvement if all three major fac- require very individualized care. Thorough
tors (axial alignment, focal chondral defect, examination and judgment of comorbidities
and meniscal deficiency) are addressed and their impact on the cartilage procedure
adequately. are imperative. Unfortunately, there are
no evidence-based guidelines or clear-cut
recommendations for the majority of the
Ligamentous Instability patients that are encountered in the practice
setting. However, with careful clinical deci-
A knee ligament insufficiency such as an sion making, evaluation of malalignment and
ACL insufficiency has been clearly linked to other comorbidities, and careful staging of
an increased risk of OA over time.40 Articular the lesion during arthroscopy the cartilage
cartilage lesions in ACL-injured patients surgeon can make good choices that will lead
are not uncommon. Not all of these lesions to good clinical outcomes as reported in the
are acute and clinically symptomatic42; literature. It is important to communicate the
10 however, ACL instability will over time complexity of the decision-making process to

Stannard_9781604068580_Ch01.indd 10 1/30/13 2:01 PM

the patient and make the patient aware that cohort study. J Bone Joint Surg Am 2005;87(9):
the proposed treatment is not a routine
15. Mithoefer K, McAdams T, Williams RJ, Kreuz PC,
straightforward, standardized procedure. Mandelbaum BR. Clinical efficacy of the microfrac-
ture technique for articular cartilage repair in the
knee: an evidence-based systematic analysis. Am J
References Sports Med 2009;37(10):20532063
16. Adkisson HD IV, Martin JA, Amendola RL, et al. The
1. Hjelle K, Solheim E, Strand T, Muri R, Brittberg M. potential of human allogeneic juvenile chondro-
Articular cartilage defects in 1,000 knee arthros- cytes for restoration of articular cartilage. Am J
copies. Arthroscopy 2002;18(7):730734 Sports Med 2010;38(7):13241333
2. Widuchowski W, Lukasik P, Kwiatkowski G, 17. Hambly K, Bobic V, Wondrasch B, Van Assche D,
et al. Isolated full thickness chondral injuries. Marlovits S. Autologous chondrocyte implantation
Prevalance and outcome of treatment. A retro- postoperative care and rehabilitation: science and
spective study of 5233 knee arthroscopies. Acta practice. Am J Sports Med 2006;34(6):10201038
Chir Orthop Traumatol Cech 2008;75(5):382386 18. Ebert JR, Robertson WB, Lloyd DG, Zheng MH, Wood
3. Curl WW, Krome J, Gordon ES, Rushing J, Smith DJ, Ackland T. Traditional vs accelerated approaches
BP, Poehling GG. Cartilage injuries: a review of to post-operative rehabilitation following matrix-
31,516 knee arthroscopies. Arthroscopy 1997; induced autologous chondrocyte implantation
13(4):456460 (MACI): comparison of clinical, biomechanical and
radiographic outcomes. Osteoarthritis Cartilage
4. Engen CNE, Aroen A. Knee cartilage defect patients 2008;16(10):11311140
enrolled in randomized controlled trials are not
representative of patients in orthopaedic practice. 19. Hirschmller A, Baur H, Braun S, Kreuz PC,
Cartilage 2010;1(4):312319 Sdkamp NP, Niemeyer P. Rehabilitation after
autologous chondrocyte implantation for isolated
5. Widuchowski W, Widuchowski J, Faltus R, et al. cartilage defects of the knee. Am J Sports Med
Long-term clinical and radiological assessment 2011; Epub ahead of print
of untreated severe cartilage damage in the knee:
a natural history study. Scand J Med Sci Sports 20. Wilson JN. A diagnostic sign in osteochondritis
dissecans of the knee. J Bone Joint Surg Am 1967;

1 Staging and Comorbidities

6. Shelbourne KD, Jari S, Gray T. Outcome of
untreated traumatic articular cartilage defects of 21. Farr J. Autologous chondrocyte implantation
the knee: a natural history study. J Bone Joint Surg improves patellofemoral cartilage treatment out-
Am 2003;85-A(Suppl 2):816 comes. Clin Orthop Relat Res 2007;463:187194
7. Messner K, Maletius W. The long-term prognosis 22. Hangody L, Kish G, Krpti Z, Udvarhelyi I, Szigeti I,
for severe damage to weight-bearing cartilage Bly M. Mosaicplasty for the treatment of articular
in the knee: a 14-year clinical and radiographic cartilage defects: application in clinical practice.
follow-up in 28 young athletes. Acta Orthop Scand Orthopedics 1998;21(7):751756
1996;67(2):165168 23. Steadman JR, Ramappa AJ, Maxwell RB, Briggs KK.
8. Widuchowski W, Widuchowski J, Trzaska T. An arthroscopic treatment regimen for osteoarthri-
Articular cartilage defects: study of 25,124 knee tis of the knee. Arthroscopy 2007;23(9):948955
arthroscopies. Knee 2007;14(3):177182 24. Braun S, Steadman JR, Rodkey WG, Briggs KK.
9. Mithoefer K, Williams RJ III, Warren RF, [Microfracture and specific rehabilitation for
Wickiewicz TL, Marx RG. High-impact athletics treating osteoarthritis of the knee. Indications,
after knee articular cartilage repair: a prospective surgical technique, and rehabilitation protocol.]
evaluation of the microfracture technique. Am J Z Rheumatol 2009;68(10):811818
Sports Med 2006;34(9):14131418 25. Ghodadra NB, Verma NN. Biologic joint reconstruc-
10. McNickle AG, Provencher MT, Cole BJ. Overview of tion: alternatives to arthroplasty. In: Cole BGA ed.
existing cartilage repair technology. Sports Med 1st ed. Thorofare, NJ: Slack Inc; 2009:1322
Arthrosc 2008;16(4):196201 26. Lindahl A, Brittberg M, Peterson L. Cartilage repair
11. Minas T, Gomoll AH, Solhpour S, Rosenberger with chondrocytes: clinical and cellular aspects.
R, Probst C, Bryant T. Autologous chondrocyte Novartis Found Symp 2003;249:175186, discus-
implantation for joint preservation in patients sion 186189, 234238, 239241
with early osteoarthritis. Clin Orthop Relat Res 27. Brittberg M, Tallheden T, Sjgren-Jansson B,
2010;468(1):147157 Lindahl A, Peterson L. Autologous chondrocytes
12. Zaslav K, Cole B, Brewster R, et al; STAR Study used for articular cartilage repair: an update. Clin
Principal Investigators. A prospective study of Orthop Relat Res 2001;(391, Suppl):S337S348
autologous chondrocyte implantation in patients 28. Coventry MB. Proximal tibial varus osteotomy for
with failed prior treatment for articular cartilage osteoarthritis of the lateral compartment of the
defect of the knee: results of the Study of the knee. J Bone Joint Surg Am 1987;69(1):3238
Treatment of Articular Repair (STAR) clinical trial. 29. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson
Am J Sports Med 2009;37(1):4255 O, Peterson L. Treatment of deep cartilage defects in
13. Rue JP, Yanke AB, Busam ML, McNickle AG, Cole the knee with autologous chondrocyte transplanta-
BJ. Prospective evaluation of concurrent menis- tion. N Engl J Med 1994;331(14):889895
cus transplantation and articular cartilage repair: 30. Gomoll AH, Minas T, Farr J, Cole BJ. Treatment of
minimum 2-year follow-up. Am J Sports Med chondral defects in the patellofemoral joint. J Knee
2008;36(9):17701778 Surg 2006;19(4):285295
14. Mithoefer K, Williams RJ III, Warren RF, et al. The 31. Minas T, Bryant T. The role of autologous chondro-
microfracture technique for the treatment of artic-
ular cartilage lesions in the knee. A prospective
cyte implantation in the patellofemoral joint. Clin
Orthop Relat Res 2005;(436):3039

Stannard_9781604068580_Ch01.indd 11 1/30/13 2:01 PM

32. Henderson IJ, Lavigne P. Periosteal autologous intraarticular contact areas and stress in the
chondrocyte implantation for patellar chondral human knee. A preliminary report. Am J Sports
defect in patients with normal and abnormal Med 1986; 14(4):270275
patellar tracking. Knee 2006;13(4):274279 39. Lee SJ, Aadalen KJ, Malaviya P, et al. Tibiofemoral
33. Schoettle PB, Zanetti M, Seifert B, Pfirrmann CW, contact mechanics after serial medial meniscec-
Fucentese SF, Romero J. The tibial tuberosity- tomies in the human cadaveric knee. Am J Sports
trochlear groove distance; a comparative study Med 2006;34(8):13341344
between CT and MRI scanning. Knee 2006;13(1): 40. Lohmander LS, Englund PM, Dahl LL, Roos EM.
2631 The long- term consequence of anterior cruciate
34. Rue JP, Colton A, Zare SM, et al. Trochlear contact ligament and meniscus injuries: osteoarthritis.
pressures after straight anteriorization of the tibial Am J Sports Med 2007;35(10):17561769
tuberosity. Am J Sports Med 2008;36(10):19531959 41. Gomoll AH, Kang RW, Chen AL, Cole BJ. Triad of
35. Allen CR, Wong EK, Livesay GA, Sakane M, Fu cartilage restoration for unicompartmental arthri-
FH, Woo SL. Importance of the medial meniscus tis treatment in young patients: meniscus allograft
in the anterior cruciate ligament-deficient knee. transplantation, cartilage repair and osteotomy.
J Orthop Res 2000;18(1):109115 J Knee Surg 2009;22(2):137141
36. Englund M, Guermazi A, Lohmander LS. The 42. Lattermann C, Jakob RP. High tibial osteotomy
meniscus in knee osteoarthritis. Rheum Dis Clin alone or combined with ligament reconstruc-
North Am 2009;35(3):579590 tion in anterior cruciate ligament-deficient
37. Baratz ME, Rehak DC, Fu FH, Rudert MJ. Peripheral knees. Knee Surg Sports Traumatol Arthrosc
tears of the meniscus. The effect of open ver- 1996;4(1):3238
sus arthroscopic repair on intraarticular contact 43. Jones HP, Appleyard RC, Mahajan S, Murrell GA.
stresses in the human knee. Am J Sports Med Meniscal and chondral loss in the anterior cruciate
1988;16(1):16 ligament injured knee. Sports Med 2003;33(14):
I Diagnosis and Treatment Planning

38. Baratz ME, Fu FH, Mengato R. Meniscal tears:

the effect of meniscectomy and of repair on


Stannard_9781604068580_Ch01.indd 12 1/30/13 2:01 PM

Magnetic Resonance Imaging
of Cartilage Repair Techniques
Catherine Hayter and Hollis Potter

The field of cartilage repair is expand- Basic Structure of Articular

ing rapidly and encompasses a wide range
of techniques, including microfracture,
first- and second-generation autologous
chondrocyte implantation, autologous oste- An understanding of the structure of articular
ochondral transplantation, and allograft cartilage is crucial to understanding the MR
transplantation. Although assessment of imaging appearance of normal and abnormal
patient outcome is clinically relevant, objec- cartilage, as well as the imaging appearance
tive evaluation of repair allows insight into following cartilage repair. Articular cartilage
the natural history of cartilage repair and is composed of chondrocytes that are embed-
may allow detection of early signs associ- ded in an organized extracellular matrix
ated with a poorer prognosis. Objective composed primarily of water (65 to 80%),
assessment has traditionally been per- collagen, and proteolgycan.3 The material
formed with second-look arthroscopy and properties of articular cartilage are imparted
histologic evaluation of biopsy specimens.1 mainly by the collagen and proteolgycan
Magnetic resonance (MR) imaging, however, components of the extracellular matrix.
offers a noninvasive method to assess car- Compressive strength is imparted by the
tilage repair. The information gained from proteolgycan molecules, which are com-
MR imaging is therefore complementary posed of negatively charged glycosamino-
to more subjective clinical outcome instru- glycans radiating from a protein core.4 These
ments that evaluate pain and function, and monomers bind to hyaluronic acid to form
plays a valuable role in patient follow-up large aggregates, which resist compression
after cartilage repair.2 because of their hydrophilic nature. The
This chapter discusses the MR imaging structural framework and tensile strength
techniques available for the assessment of articular cartilage is imparted by collagen.
of articular cartilage, including advanced Type II collagen fibers make up 95% of the
imaging techniques that allow assessment collagen in articular cartilage. These fibers
of cartilage biochemistry. The MR imaging have a long length-to-thickness ratio, thereby
appearance and assessment of microfracture, providing tensile stiffness and strength.5
autologous chondrocyte implantation, and Articular cartilage can be divided into four
osteochondral autograft and allograft trans- distinct zones on histology, which can be
plantation are reviewed. depicted with cartilage-sensitive MR pulse

Stannard_9781604068580_Ch02.indd 13 1/30/13 2:17 PM

sequences. The superficial zone and lamina of the articular cartilage, with consequent
splendens consist of highly organized colla- factitious thickening of the subchondral
gen fibers oriented parallel to the cartilage bone and thinning of the articular cartilage.7
surface, providing high tensile strength. The Several different cartilage-sensitive pulse
transitional zone has lower collagen content sequences are available for MR imaging; of
and consists of randomly oriented collagen these, fat-suppressed three-dimensional
fibers; this zone has a higher compressive (3D) gradient echo and FSE sequences are
strength than the superficial zone. The radial the most accurate and most widely used
zone consists of highly organized collagen techniques.
fibers oriented perpendicular to the cartilage Fat-suppressed 3D spoiled or T1-weighted
surface; this zone has the highest proteolgy- gradient echo images obtained with isotropic
can content and the lowest water content. voxels have the advantage of producing thin,
The deepest zone is the calcified cartilage contiguous slices that can be reformatted
layer. The tidemark represents the boundary in any plane. The sharp contrast boundary
between the uncalcified and calcified car- between the low-signal intensity bone and
tilage.6 At clinically relevant field strengths the high-signal intensity articular cartilage
using traditional MR sequences, the tidemark makes these sequences amenable to semiau-
is indistinguishable from the subchondral tomated segmentation algorithms, allowing
bone plate. for accurate assessment of cartilage thickness
I Diagnosis and Treatment Planning

The signal characteristics of articular car- and volume. This pulse sequence has been
tilage on MR imaging reflect both the mobil- used for longitudinal assessment of cartilage
ity of water and the degree of organization volume in osteoarthritis trials and for quan-
of the tissue. Therefore, in the radial zone, titative assessment of focal cartilage defects
where the collagen is highly ordered, lower and subsequent fill following repair.810
signal intensity is generated when compared This technique is, however, less sensitive to
with the transitional zone where the colla- partial-thickness cartilage defects than FSE
gen is more randomly oriented. The lamina sequences and requires longer scan times. It
splendens is also highly ordered and when is also limited by metal-induced susceptibil-
visualized on MR images appears of lower ity artifact, which may be a significant prob-
signal intensity. It is important to recognize lem when imaging in the presence of metallic
this normal gray-scale stratification when hardware or residual metallic debris follow-
performing MR imaging of cartilage, as loss ing arthroscopy.11
of the normal gray-scale stratification is one An intermediate echo time (TE) two-
of the earliest signs of articular cartilage dimensional FSE technique is one of the most
degeneration.6 popular pulse sequences for the assessment
of articular cartilage. This technique pro-
vides good contrast between the interme-
diate signal intensity of articular cartilage,
Morphologic Assessment the low signal intensity of fibrocartilage
of Cartilage and subchondral bone, and the high signal
intensity of synovial fluid. On intermediate
Cartilage-Sensitive Pulse Sequences TE FSE images, articular cartilage demon-
strates a normal gray-scale stratification,
Traditional T1- and T2-weighted techniques which corresponds to the cartilage zonal
are inadequate for the accurate assessment anatomy. Partial-thickness chondral lesions
of articular cartilage. T1-weighted images and chondral flaps are also well depicted
result in poor delineation between the inter- with this technique (Fig. 2.1).12 Use of an FSE
mediate signal intensity cartilage and the technique with a wide receiver bandwidth
low- to intermediate-signal intensity joint minimizes susceptibility artifact, allowing
fluid. Conventional spin-echo or heavily accurate assessment of cartilage in the pres-
T2-weighed fast spin echo (FSE) techniques ence of metallic hardware or debris.13
result in poor delineation between the Intermediate TE FSE sequences are subject
14 subchondral bone and the deep component to the magic angle effect, which must be

Stannard_9781604068580_Ch02.indd 14 1/30/13 2:17 PM

2 MRI of Cartilage Repair Techniques
Fig. 2.1 Sagittal (a) and coronal (b) fast spin echo a flap extending to subchondral bone (black arrow).
images in a 30-year-old man demonstrate delamina- The cartilage over the lateral femoral condyle demon-
tion of cartilage over the medial femoral condyle with strates normal gray-scale stratification (white arrow).

considered when interpreting these images. cartilage.7,16 Thus, routine assessment of

The signal intensity of ordered tissues, such articular cartilage with FSE techniques should
as cartilage and tendons, depends on the ori- include high-resolution, non-fat-suppressed
entation of the collagen fibers relative to the FSE images in at least two planes as well as a
external magnetic field (B0), which in a con- fat-suppressed FSE image in one plane.12
ventional MR machine runs parallel to the
long axis of the patients body. When highly
structured tissues are imaged at 55 degrees
to the external magnetic field using a short Evaluation of Cartilage
TE, there is a normal prolongation of T2 val- Biochemistry
ues, a phenomenon known as magic angle
effect.14,15 In the knee, magic angle effect com- Quantitative MR imaging techniques allow
monly manifests as increased signal intensity more sophisticated assessment of cartilage
in the cartilage over the anterior and poste- degeneration and cartilage repair. These
rior femoral condyles and in the submeniscal techniques detect changes in the ultrastruc-
zone of the posterior tibial condyles. ture of cartilage and provide an assessment
The application of fat suppression to FSE of cartilage biochemistry. They therefore
images allows the detection of bone mar- have the potential to detect changes in car-
row edema and increases the contrast differ- tilage biochemistry that may precede dis-
ences between cartilage, fluid, and synovium. cernible cartilage thinning on traditional MR
However, fat suppression results in a lower techniques. Quantitative MR imaging tech-
signal-to-noise ratio; therefore, it is more niques are classified into those that detect
difficult to achieve the high spatial resolu- alterations in collagen fiber orientation and
tion that is required to discern subtle fis- those that detect alterations in the proteogly-
sures and surface fibrillation in the articular can content. 15

Stannard_9781604068580_Ch02.indd 15 1/30/13 2:17 PM

Collagen imaging of cartilage (dGEMRIC), and T1rho
()-weighted MR imaging.
T2 mapping is the most commonly used
quantitative MR technique for the assess-
ment of collagen orientation. Additional Sodium Imaging
techniques such as diffusion tensor imaging Similar to hydrogen nuclei, sodium-23 is
have been shown to be sensitive to collagen also a suitable nucleus for MR imaging. The
orientation1719 but are not currently in wide- presence of the negatively charged glycosa-
spread clinical use. minoglycan molecules in articular cartilage
generates an attraction toward the positively
charged sodium-23, allowing for relative
T2 Mapping
measurement of fixed-charge density.27,28
The T2 (spin-spin) relaxation time reflects Sodium imaging has been used as an imag-
the loss of signal that occurs due to dephas- ing standard by which to assess proteogly-
ing of the excited nuclei after the disturbing can distribution and content. However, this
radio-frequency pulse is applied. T2 mapping technique is limited in its clinical application
is performed by acquiring several images at by the lower concentration of sodium-23
different TEs at the same slice location. The in articular cartilage relative to hydrogen,
T2 calculation is performed on a pixel-by- resulting in low signal-to-noise ratio and
I Diagnosis and Treatment Planning

pixel basis by fitting the signal intensity from requiring long scan times. Sodium imaging
each echo image and the corresponding TE to also requires the use of specialized coils and
an exponential decay equation. the ability to scan with multinuclear spec-
The T2 map of articular cartilage reflects troscopy software, which is not widely avail-
the collagen fiber orientation and the able across clinical systems.29,30
mobile water content,20 and is displayed
using a color-coded map. In the radial zone,
where the collagen is oriented perpendicu- dGEMRIC
lar to the subchondral plate, short T2 val- dGEMRIC also exploits the fixed-charge prop-
ues are obtained. In the transitional zone, erty of articular cartilage through the use
where there is a more random orientation of an injection of negatively charged gado-
of collagen, longer T2 values are obtained. linium contrast. Gadolinium is administered
Prolongation of T2 relaxation times has been intravenously, the patient performs 10 min-
shown to be associated with osteoarthritis utes of exercise, and, following a 90-minute
and breakdown in cartilage structure.2123 delay, T1-weighted maps are obtained, usu-
Quantitative T2 measurements demonstrate ally through the use of a specialized inversion
excellent inter- and intraobserver reliabil- recovery pulse sequence.31 The gadolinium
ity,2426 thereby offering a tool for reproduc- penetrates the articular cartilage, with the
ible assessment of cartilage status over time. amount of penetration being inversely pro-
portional to the glycosaminoglycan content.
Proteoglycan The gadolinium acts to shorten T1-relaxation
times, allowing for the generation of T1 maps.
Imaging strategies aimed at assessing the In areas with depleted glycosaminoglycan
proteoglycan component of the extracel- content, there will be an increased distribu-
lular matrix exploit the fixed-charge den- tion of gadolinium and therefore a higher
sity property of the articular cartilage. The T1 signal, which is reflected by a diminished
fixed-charge density in cartilage is largely relative glycosaminoglycan index.
due to the concentration and distribution of
the negatively charged glycosaminoglycan
T1rho () Mapping
chains within the proteoglycan macromole-
cules. Techniques used to assess proteoglycan T1rho () is a technique used to assess the
content and distribution include sodium-23 low-frequency interactions between hydro-
16 imaging, delayed gadolinium-enhanced MR gen in macromolecules and free water.

Stannard_9781604068580_Ch02.indd 16 1/30/13 2:17 PM

Termed spin-lattice relaxation in the rotating MR imaging is increasingly recognized as an
frame, this technique uses clusters of radio- alternative method of noninvasive evaluation
frequency pulses to lock magnetization in of the results of articular cartilage repair.41
the transverse plane and limit dephasing of The variables that should be assessed on
protons. After a specified period of time (the MR imaging following cartilage repair differ
spin-lock time), the magnetization vector is according to the repair technique2 and are
realigned with B0, and data are then acquired summarized in Table 2.1. The expected MR
using an FSE or a spiral sequence.32,33 Similar
to T2 mapping, T1rho is calculated on a pixel- Table 2.1 Diagnostic checklist for MR imaging of
by-pixel basis by fitting the signal intensity cartilage repair techniques
from each spin-lock image and the corre-
sponding spin-lock length to an exponential Microfracture
decay equation. Signal intensity of the repair cartilage
T1rho has been shown to reflect pro- Morphology of reparative tissue (flush, proud,
teoglycan content in articular cartilage. The Volume or percent of fill by reparative tissue
normalized T1rho rate is highly correlated Peripheral integration (fissures at repair
to fixed-charge density in sodium-23 imag- native cartilage interface)
ing, as well as to proteoglycan content and Underlying subchondral bone (extent of bone
distribution on histology.3436 Subjects with marrow edema)

2 MRI of Cartilage Repair Techniques

Overgrowth of subchondral bone
osteoarthritis have longer T1rho values than Assessment of host cartilage (adjacent/
asymptomatic controls, and T1rho may be opposing surfaces)
even more sensitive to early cartilage degen- Reactive synovitis
eration than T2 mapping alone.37 While Autologous chondrocyte implantation (ACI)
clinically feasible at 1.5T and 3T,37,38 T1rho is Signal intensity of the repair cartilage
largely applied at 3T and is a promising tech- Morphology of reparative tissue (flush, proud,
nique to detect changes in proteoglycan con- depressed)
Volume or percent of fill by reparative tissue
tent in early cartilage degeneration. Peripheral integration (fissures at repair
Due to the specialty coil requirements for native cartilage interface)
sodium imaging as well as the logistical con- Presence of delamination
straints of the contrast-enhanced dGEMRIC Periosteum overlying the defect (periosteal
technique, the authors favor the use of non-
Underlying subchondral bone (bone marrow
contrast T1rho at 3T for assessment of pro- edema)
teoglycan content. Assessment of host cartilage (adjacent/
opposing surfaces)
Reactive synovitis

Evaluation of Articular Osteochondral transplantation

Osseous phase
Cartilage Repair Techniques Presence/absence of displacement of plugs
Restoration of radius of curvature of joint
In part due to its limited vascular supply, surface
mature articular cartilage has little to no Peripheral integration of osseous compo-
capacity for spontaneous repair.39,40 A wide Presence of subchondral bone marrow
variety of cartilage repair techniques have edema
been described; results following these pro-
Articular phase
cedures have varied widely. Most of the Signal intensity of the repair cartilage
literature has relied on subjective clinical Morphology of reparative tissue (flush, proud,
assessment or conventional radiographs to depressed)
evaluate the success of repair techniques. The Volume or percent of fill by reparative tissue
Peripheral integration (fissures at repair
use of second-look arthroscopy with biopsy native cartilage interface)
remains the gold standard but is limited by
Other features
its invasive nature, potential for operative
Assessment of host cartilage (adjacent/
surgeon bias, and poor patient acceptance. opposing surfaces)
With advancements in imaging techniques, Reactive synovitis 17

Stannard_9781604068580_Ch02.indd 17 1/30/13 2:17 PM

imaging appearance following commonly Microfracture
used cartilage repair techniques is summa-
rized in Table 2.2. Microfracture is a cartilage repair technique
that is based on local bone marrow stimula-
tion and the release of multipotential stem
Table 2.2 Summary of imaging findings following cells,42 which, over time, differentiate into
cartilage repair repair tissue, predominantly composed of
Microfracture MR imaging assessment following micro-
Early postoperative period fracture should include the assessment of
Reparative cartilage is hyperintense to native (1) the signal intensity of the repair carti-
cartilage lage compared with native cartilage; (2) the
Moderate subchondral bone marrow edema
morphology of the repair, being either flush,
proud, or depressed with respect to the
Late postoperative period native cartilage; (3) the volume or percent
Signal intensity of reparative cartilage
decreases as it matures fill of the defect in two imaging planes;
Subchondral bone marrow edema pattern (4) the degree of peripheral integration to
dissipates the adjacent cartilage, including the absence/
Specific complication presence of fissures; (5) assessment of the
I Diagnosis and Treatment Planning

Overgrowth of subchondral bone underlying bone, including the presence of

Autologous chondrocyte implantation (ACI) bone marrow edema; (6) the surface geom-
etry of the bone, including the presence of
Early postoperative period
Reparative cartilage is hyperintense to native proud subchondral bone; (7) the status of
cartilage and periosteal cover the articular cartilage in the adjacent and
Subchondral bone marrow edema pattern opposing surfaces; and (8) the presence of a
Hypertrophy of periosteal cover reactive synovitis.2
Late postoperative period The MR imaging appearance of a chondral
Signal intensity of reparative cartilage defect treated by microfracture evolves over
decreases as it matures
time. In the early postoperative period, the
Subchondral bone marrow edema pattern
dissipates reparative tissue is less organized and has
increased water content when compared
Specific complications
Graft delamination with normal articular cartilage, and as such,
Graft (periosteal) hypertrophy it appears hyperintense compared with the
Osteochondral autografts native cartilage.2,45 Over time, there is mat-
uration of the signal characteristics of the
Early postoperative period
Subchondral bone marrow edema pattern repair tissue, which may appear hypointense
to the native cartilage,2,45 suggesting the pres-
Late postoperative period
ence of reparative fibrocartilage.4648
Subchondral bone marrow edema pattern
dissipates The percent fill of the defect increases
with time; however, the overall percent fill
Specific complications
Plug displacement/subsidence following microfracture may be less than
Poor osseous integration that seen following autologous chondrocyte
Osteochondral allograft (OCA) implantation.2 Fissures between the area of
Early postoperative period (03 mo)
microfracture and native cartilage appear
Graft bone marrow edema pattern commonly45 and do not necessarily correlate
to symptoms.2 By 1 to 2 years after surgery,
Late postoperative period (36 mo)
Graft bone marrow edema pattern dissipates the treated defect should be filled with tis-
Overgrowth of subchondral bone may occur sue that has a smooth, well-defined surface.
Specific complications
Adverse functional scores at 2 years have
Graft rejection (persistent bone marrow been shown to correlate with poor percent-
edema, collapse, fluid undermining graft, age fill, indicating a correlation between
global synovitis) objective MR imaging findings and subjective
18 Graft collapse clinical outcome.45

Stannard_9781604068580_Ch02.indd 18 1/30/13 2:17 PM

Bone marrow edema may be seen in the values of the repair tissue may never return to
subchondral bone at short-term follow- normal. At a follow-up time of 24 months after
up but usually appears mild2 and gradu- microfracture, T2 values have been shown
ally diminishes over time.44 In patients with to remain globally reduced compared with
failure of the microfracture procedure, the native cartilage.51,52 The T2 index, calculated
subchondral bone marrow edema does not by the T2 value of the repair tissue compared
diminish and may become more conspicuous with native cartilage, correlates with subjec-
over time. tive functional scores, such as the Lysholm
Because microfracture involves the release score and International Knee Documentation
of pluripotential stem cells, these cells may Committee (IKDC) form of subjective func-
also differentiate into bone. Overgrowth of tion,51 further supporting the use of quan-
the subchondral bone has been noted in 25 titative MR imaging as an objective tool to
to 49% of patients following microfracture2,45 evaluate cartilage repair procedures.
(Fig. 2.2). Osseous overgrowth may not be a
negative prognostic factor and does not cor-
relate with adverse clinical scores at short- Autologous Chondrocyte Implantation
term follow-up.45 Osseous overgrowth may,
however, result in a thinner layer of repara- Autologous chondrocyte implantation (ACI) is
tive tissue and inferior filling of the cartilage a two-stage surgical technique for the repair

2 MRI of Cartilage Repair Techniques

defect, which has been shown to correlate to of symptomatic deep chondral defects49
inferior functional outcomes.49 that entails growth of the patients native
Studies using quantitative MR imaging of chondrocytes in tissue culture followed by
microfracture have shown that at short-term reimplantation and coverage by a periosteal
follow-up ( 6 months) T2 values over the flap.53,54 Matrix-assisted chondrocyte trans-
repair site are prolonged, mimicking the signal plantation (MACT) is a modification of this
characteristics of the repair tissue observed procedure, which uses biomaterials seeded
on conventional MR imaging. Over time with chondrocytes as carriers and scaffolds
( 12 months), there is progressive shorten- for cell growth, without the use of a peri-
ing and maturation of T2 values,50 although T2 osteal cover.55

Fig. 2.2 Sagittal inversion recovery (a) and sagittal medial femoral condyle. Proud bone formation is
(b) and coronal (c) fast spin echo images in a 42-year- seen at the repair site (arrow), but there is good fill by
old man 5 months following microfracture of the hyperintense reparative tissue. 19

Stannard_9781604068580_Ch02.indd 19 1/30/13 2:17 PM

ACI is frequently used for larger carti- the extracellular matrix to produce hyaline-
lage defects than microfracture; areas up to like repair tissue.57,58 The signal characteris-
12 cm2 have previously been transplanted.42,54 tics of the repair tissue on MR imaging reflect
ACI has been shown to provide better defect its histology. In the immediate postoperative
fill at all time periods when compared with period the reparative tissue is disorganized
microfracture2; however, the technique is with increased water content and therefore
associated with some specific potential com- appears hyperintense on MR images.44,59,60
plications, including delamination of the This allows the reparative tissue to be readily
graft and periosteal hypertrophy,49 which can differentiated from the overlying periosteum,
be well demonstrated on MR imaging. which appears hypointense. At 3 to 6 months
MR imaging assessment following ACI following repair, there is a decline in the
should include the assessment of (1) the signal intensity of the reparative tissue as it
signal intensity of the repair cartilage com- becomes increasingly organized and inte-
pared with native cartilage; (2) the morphol- grated with the adjacent cartilage (Fig. 2.3).
ogy of the repair, being either flush, proud, Complete integration can take up to 2 years.
or depressed with respect to the native car- Following ACI there should ideally be
tilage; (3) the volume or percent fill of the complete fill of the defect with repair tis-
defect in two imaging planes; (4) the degree sue that restores the contour of the articular
of peripheral integration to the adjacent car- surface. Complete fill of the defect has been
I Diagnosis and Treatment Planning

tilage, including the absence/presence of fis- observed as early as 3 weeks postopera-

sures and the presence of delamination; (5) tively.61 However, filling defects at the repair
the periosteum overlying the defect, includ- site in the early postoperative period (3 to
ing the presence of periosteal hypertrophy; 6 months) are common and often demon-
(6) the subchondral bone, including the strate progressive fill on follow-up exami-
presence of a bone marrow edema pattern; nations (6 to 12 months).59,62 The interface
(7) the status of the articular cartilage in the with the adjacent cartilage is rarely smooth;
adjacent and opposing surfaces; and (8) the hyperintense fissures less than 2 mm wide
presence of a reactive synovitis.2 are common.2 The long-term significance
Previous studies have investigated the cor- of small fissures is unknown, but larger fis-
relation between MR imaging findings and sures may represent failure of the repair car-
clinical outcome measures following MACT. tilage to integrate with the native articular
A study correlating MR imaging findings to cartilage.
the Knee Injury and Osteoarthritis Outcome Delamination of the ACI graft is uncom-
Score (KOOS) and visual analog scale (VAS) mon, occurring in fewer than 5% of patients,62
score for pain and function demonstrated and mostly occurs within the first 6 to
a significant correlation for the volume of 9 months following surgery.44,49,58 A displaced
repair fill, signal intensity of the repair tis- delaminated graft appears as a defect within
sue, structure of the repair tissue (presence the repair site, and the displaced tissue may
or absence of fissures), and changes in the be seen as a loose body within the joint. In
subchondral bone. Interclass coefficients the case of in situ delamination, the appear-
demonstrated a strong agreement between ance is similar to a cartilage flap, with a thin
observers, confirming the reproducibility of rim of fluid intensity between the base of the
MR imaging in the longitudinal assessment repair tissue and the underlying subchondral
of cartilage repair.56 bone.44 It must, however, be remembered
A three-phase time course has been pro- that a fluid-like appearance to the graft in the
posed in the healing of ACI. In the first first 4 weeks is normal due to immaturity of
6 weeks, the proliferative phase, the graft the repair tissue. This should not be misinter-
site fills with a soft, primitive repair tissue. preted as graft delamination; closer inspec-
From 7 weeks to 6 months, the transition tion will reveal the low-signal periosteum
phase, there is expansion of the extracellu- overlying the implant.44
lar matrix, and the graft takes on the consist- Overgrowth of reparative tissue has been
ency of gelatin. The remodeling phase from reported following ACI, which may be due
20 6 months to 3 years involves remodeling of to periosteal hypertrophy and/or thickening

Stannard_9781604068580_Ch02.indd 20 1/30/13 2:17 PM

2 MRI of Cartilage Repair Techniques
Fig. 2.3 Sagittal inversion recovery (a) and axial and moderate fill over the patella (black arrow) with
(b, c) fast spin echo images in a 43-year-old man 6 hyperintense reparative tissue. The periosteal cover is
months following autologous chondrocyte implanta- well depicted over the trochlea. There is no bone mar-
tion. There is good fill over the trochlea (white arrow) row edema or reactive synovitis.

of the matrix.2,63 Although the majority period. The intensity and volume of the edema
of patients remain asymptomatic, some tend to diminish over time to regress to nor-
patients report a sensation of catching, which mal or a thin line of increased signal inten-
can be painful.49,58 This complication most sity deep to the subchondral bone. Persistent
commonly occurs at 3 to 9 months following or increased edema in the marrow beneath
surgery and may require arthroscopic resec- an ACI suggests poor integration of the graft
tion of the hypertrophic periosteal cover.54 In to the subchondral bone and warrants close
a study of 35 ACI procedures, graft hypertro- clinical follow-up.61
phy occurred in 63% of lesions, accounting for Quantitative MR techniques have been
moderate postoperative morbidity.2 On MR used to provide insight into the biochemis-
imaging, periosteal hypertrophy appears as try of repair tissue following ACI. In a cohort
focal areas of repair tissue that extend above study of 15 patients treated with MACT, mean
the expected contour of the articular surface. T2 values of the repair tissue were found to
The hypertrophied tissue may grow beyond be significantly higher than control sites at
the margins of the graft, appearing as a flap short-term follow-up, but showed no signifi-
of tissue extending over the adjacent native cant difference at long-term follow-up (19 to
articular cartilage.61 42 months).64 In a study of 11 patients evalu-
The subchondral bone plate may appear ated with dGEMRIC, investigators noted that
irregular following ACI and often demon- at less than 12 months follow-up the relative
strates a mild to moderate bone marrow glycosaminoglycan index in the ACI repair
edema pattern in the early postoperative tissue was lower than that of native cartilage. 21

Stannard_9781604068580_Ch02.indd 21 1/30/13 2:17 PM

At follow-up of greater than 12 months, the degree of peripheral integration to the adja-
glycosaminoglycan index was similar to that cent cartilage, including the absence/presence
of native cartilage,65 suggesting progressive of fissures and the presence of delamination.
maturation of the repair tissue. The studies Additional features to assess include (1) the
suggest that quantitative MR imaging can status of the articular cartilage in the adjacent
provide information about the structure of and opposing surfaces and (2) the presence of
the cartilage repair tissue. a reactive synovitis.
In a study of 20 patients comparing micro-
fracture and MACT, investigators noted that
Autologous Osteochondral Plugs
at 2 years following microfracture there was
a global reduction in T2 values compared Autologous osteochondral transfer involves
with control tissues; the repair tissue fol- the harvesting of osteochondral plugs from a
lowing microfracture also showed lack of non-weight-bearing portion of the knee and
the normal T2 stratification. In contrast, the transfer into cored holes created within the
MACT repair tissue demonstrated T2 values articular defect to be treated, using press fit
and T2 stratification similar to that of native fixation.67,68
cartilage,52 suggesting that ACI results in Assessment of the osseous phase of incor-
more hyaline-like and mature repair tissue poration involves assessment of plug position
at the site of repair when compared with the and restoration of the radius of curvature of
I Diagnosis and Treatment Planning

microfracture technique. the joint surface. Incongruity of the articular

surface may be due to technical problems
during graft placement or may occur later as
Osteochondral Transplantation a result of subsidence or degradation of the
osteochondral plugs.69 On MR imaging, an
Osteochondral transplantation may be per- incongruent graft repair site can be seen as a
formed using autogenous tissue, biphasic step-off at the subchondral bone or articular
synthetic copolymer plugs, or fresh cryopre- surface (Fig. 2.4). Evaluation of restoration
served allograft tissue. Although some sur- of the radius of curvature is clinically impor-
geons reserve these techniques for the repair tant because surface incongruities of as little
of defects associated with bone and cartilage as 0.5 mm have been shown to be associated
defects, such as osteonecrosis or osteochon- with significant increases ( 40%) in surface
dritis dissecans, others may use osteochondral contact pressures.70 This may explain why, in
transplantation for isolated cartilage defects animal studies, elevated plugs tend to dem-
with intact subchondral bone. onstrate poor integration with the surround-
MR imaging assessment following osteo- ing articular cartilage when compared with
chondral repair includes the assessment plugs that are placed flush.71,72
of the osseous and articular phases.2,50,66 Osseous integration of the plugs manifests
Assessment of the osseous phase includes as trabecular incorporation with the native
(1) the presence or absence of displacement bone and should be complete or partial in the
of the plug or the presence of subchondral majority of patients. When solid bony incor-
collapse in the setting of allograft transplan- poration occurs, normal fatty marrow signal
tation; (2) the restoration of the radius of is seen within and around the plugs (Fig. 2.5).
curvature of the subchondral bone; (3) the In animal studies, autologous grafts demon-
degree of osseous integration of the plug/ strate solid osseous incorporation between
allograft into the recipient site; (4) the pres- 6 and 14 weeks.73,74 Failure of osseous inte-
ence of a subchondral bone marrow edema gration manifests as hyperintense signal at
pattern. Assessment of the articular phase the native bonegraft interface and may be
includes (5) the signal intensity of the repair associated with a bone marrow edema pat-
cartilage compared with native cartilage; tern (Fig. 2.6).66
(6) the morphology of the repair, being either There are some potential pitfalls in the MR
flush, proud, or depressed with respect to the imaging assessment of the osseous phase,
native cartilage; (7) the volume or percent fill which should not be misinterpreted as a fail-
22 of the defect in two imaging planes: (8) the ure of osseous integration. The appearance of

Stannard_9781604068580_Ch02.indd 22 1/30/13 2:17 PM

2 MRI of Cartilage Repair Techniques
Fig. 2.4 Axial fast spin echo images in an 18-year- normal T2 stratification over the repair site but
old woman 4 months (a) and 16 months (b) follow- prolongation of T2 values at the repairnative tissue
ing autologous osteochondral repair over the lateral interface (thin black arrows). At 16 months (d) there
patellar facet. The plug is proud to the subchondral is more diffuse prolongation of T2 values over the
bone. On the initial scan there is a hyperintense repair site and adjacent cartilage (thick black arrows).
fissure at the medial interface (thin white arrow). (Adapted with permission from Nho SJ, Foo LF, Green
On the follow-up scan, the fissure has remodeled DM, et al. Magnetic resonance imaging and clini-
but there is progressive increased signal intensity cal evaluation of patellar resurfacing with press-fit
in the cartilage over the plug (thick white arrow). osteochondral autograft plugs. Am J Sports Med
Corresponding T2 map at 4 months (c) demonstrates 2008;36(6):11011109.)

hypointense signal at the osseous interface is signal intensity, or the presence of a persis-
indicative of sclerosis due to the tight fit cre- tent bone marrow edema pattern suggests
ated with the press-fit technique used for poor osseous integration of the graft and
plug placement. Over time, this tends to fill warrants close follow-up (Fig. 2.7).44
in with normal fatty marrow signal. Even in The degree of cartilage integration involves
well-incorporated repairs, a small area of fat assessment of the signal intensity of the
devoid of trabeculae may be identified deep repair cartilage and percent fill. As the carti-
to the base of the osteochondral plug, as the lage over the site of repair consists of normal
recipient tunnel is often made deeper than articular cartilage, the signal in the cartilage
the length of the plug.44 This should not be over the plugs should ideally mimic that
misinterpreted as failure of the repair. of native articular cartilage. The T2 values
Subchondral bone marrow edema is often over autologous plugs also tend to show
present initially but is expected to resolve as the expected normal T2 stratification, with
the graft incorporates with the subchondral shorter values in the radial zone and pro-
bone.69 Detection of cystic cavities, fluid-like longation of values in the more superficial 23

Stannard_9781604068580_Ch02.indd 23 1/30/13 2:17 PM

I Diagnosis and Treatment Planning

Fig. 2.5 Sagittal inversion recovery (a) and sagittal restoration of the radius of curvature of the joint
(b) and coronal (c) fast spin echo images in a 19-year- surface. Thin ( 2 mm) fissures are present at the
old man 6 months following autologous osteochon- area of peripheral integration with the native articular
dral repair over the medial femoral condyle. There is cartilage (arrows).
good osseous incorporation of the plugs and good

zones.50 It is, however, common to observe repair.76 These biphasic plugs have a scaffold
hyperintense signal in the cartilage overlying impregnated with growth factors to promote
the plugs and prolongation of T2 values when bone growth in the deeper components, as
compared with native articular cartilage.69 well as the growth of repair cartilage over the
While the osseous portion of the plug typi- superficial surface.
cally demonstrates excellent incorporation, The signal characteristic of synthetic
persistent fissures at the cartilaginous level plugs on MR imaging are distinctly differ-
between the graft and native articular carti- ent from those of autologous tissue. In the
lage are common.2,69,75 Second-look arthros- early postoperative period (6 to 12 months)
copies and biopsy specimens have shown it is common to see unfavorable MR imaging
that the cartilage transplanted into the findings, including depressed morphology
recipient site remains hyaline-like and that a of the plugs, resorption at the interface with
fibrocartilaginous bond, consisting of organ- the native bone, incomplete fill by reparative
ized scar tissue, forms between the cartilage tissue, and prolongation of T2 values over
plugs and the native articular cartilage.71 the repair site. These findings do not nec-
essarily reflect a failure of the repair tech-
nique but rather reflect the natural history
Synthetic Biphasic Copolymer Plugs
of the synthetic plugs. With longer follow-
Without the use of backfill, the defect cre- up ( 16 months), there is usually a marked
ated by harvest of an autologous osteochon- improvement in the MR imaging appearance
dral plug will typically fill in with reparative with a high percentage of lesion fill, good
fibrocartilage and fibrous tissue. Because of osseous incorporation, and restoration of
concern about donor-site morbidity, there the radius of curvature of the joint surface.
has been increased interest in the use of In addition, progressive shortening of T2 val-
synthetic biphasic copolymer plugs, which ues is observed over the plugs, which come
can be used for backfill of the donor site to approach relaxation times observed in
24 and may also be used for primary cartilage native articular cartilage.77

Stannard_9781604068580_Ch02.indd 24 1/30/13 2:17 PM

2 MRI of Cartilage Repair Techniques

Fig. 2.6 Sagittal inversion recovery (left) and fast fissures are seen at the area of peripheral integration
spin echo (right) images in a 16-year-old man following with the native cartilage (thin white arrows). Images
autologous osteochondral transplantation over the at 14 months postsurgery (c, d) demonstrate fluid
medial femoral condyle. At 2 months follow-up (a, b) signal undermining the posterior two plugs (thick white
there is incomplete osseous incorporation of the plugs arrows) with sclerosis in the subchondral bone (black
with a mild bone marrow edema pattern. Hyperintense arrows), indicative of failure of osseous incorporation.

Stannard_9781604068580_Ch02.indd 25 1/30/13 2:17 PM

I Diagnosis and Treatment Planning

Fig. 2.7 Sagittal inversion recovery (left) and sagittal the native bone (black arrows), and flattening of the
(middle) and coronal (right) fast spin echo images in subchondral surface. The articular cartilage is hyperin-
a 12-year-old girl following autologous osteochondral tense but is intact (white arrows). At 36 months postsur-
transplantation over the lateral femoral condyle. At gery (d, e, f) there is progressive subchondral collapse
29 months postsurgery (a, b, c) there is persistent bone (black arrows) with delamination of the overlying carti-
marrow edema, cyst formation at the interface with lage and a flap extending down to bone (white arrow).

The delayed biologic incorporation of syn-

Osteochondral Allograft Plugs
thetic plugs when compared with autologous
plugs must be recognized when interpreting Osteochondral allograft (OCA) involves the
MR images. Complete incorporation of syn- harvesting of an osteochondral plug from
thetic plugs may take more than 2 years.78 a cadaver and transplantation into a donor
Increased signal in the plugs should therefore site using press-fit technique, with or with-
not be mistaken for delayed biologic incorpo- out pin fixation.44 As with autologous osteo-
ration (Fig. 2.8) and unfavorable MR imaging chondral plugs, the MR imaging assessment
findings at early follow-up should be inter- of OCA plugs includes assessment of osseous
26 preted with caution. integration, articular integration, and the

Stannard_9781604068580_Ch02.indd 26 1/30/13 2:17 PM

2 MRI of Cartilage Repair Techniques

Fig. 2.8 Sagittal inversion recovery (left) and sagit- Incomplete osseous incorporation (white arrows)
tal (right) fast spin echo images in a 17-year-old man persists at 12 months postsurgery (c, d), but there is
following autologous osteochondral transfer into progressive low signal intensity of the overlying repair
the medial femoral condyle with backfill over the cartilage with good fill of the cartilage defect (black
donor site using synthetic biphasic copolymer plugs. arrow). Delayed biologic incorporation of synthetic
At 6 months postsurgery (a, b) there is incomplete plugs is common when compared with autologous
osseous and articular incorporation of the plugs. plugs, and may take up to 2 years. 27

Stannard_9781604068580_Ch02.indd 27 1/30/13 2:17 PM

Fig. 2.9 Sagittal inversion recovery (a) and sagittal white arrow) with incomplete osseous incorporation
I Diagnosis and Treatment Planning

(b) and coronal (c) fast spin echo images in a 22-year- (black arrows). There is overall good restoration of the
old woman 11 months following osteochondral allo- radius of curvature of the joint. Small fissures are seen
graft transplantation into the lateral femoral condyle. at the anterior margin of peripheral integration with
There is a mild bone marrow edema pattern (thick the native articular cartilage (thin white arrow).

restoration of the radius of curvature of the responses demonstrate a decreased rate of

joint surface. bony incorporation and an increased rate of
Osseous incorporation is assessed by eval- bone marrow edema pattern and surface col-
uating the graft marrow signal, grafthost lapse of the graft.75 When the patient rejects
interface, and graft congruity (Fig. 2.9). In the an OCA, MR imaging usually demonstrates
early postoperative period, mild or moderate signal abnormalities in the graft marrow or
bone marrow edema in the graft marrow and at the grafthost interface before changes in
at the grafthost interface is common. Bone the cartilage become evident. A reactive syn-
marrow edema decreases over time as graft ovitis also appears to correlate to an adverse
incorporation occurs. Osseous trabecular humoral response.75
incorporation manifests as normal fatty mar- While bony incorporation occurs in the
row signal extending within and around the majority of OCA transplants, fissures at the
plugs, and it has been shown to be complete site of peripheral integration are common
or partial in the majority of patients and to and histologically have been shown to rep-
correlate to clinical outcome measures, such resent fibrous tissue or fibrocartilage.66 The
as the Short Form-36 outcome score.79 A cartilage over the allograft should display
grafthost interface that displays high sig- the normal signal characteristics and gray-
nal on fluid-sensitive images and becomes scale stratification of articular cartilage.
thickened over time suggests the presence Progressive cartilage degeneration over the
of granulation tissue, edema, or fluid due to allograft and cartilage delamination are
incomplete incorporation and possible graft potential complications that are well demon-
instability.44 The presence of low signal inten- strated on MR imaging.
sity on all pulse sequences in the allograft
bone strongly suggests loss of bone viability,
which may lead to eventual implant failure.79 Conclusion
As the donor tissue is obtained from a for-
eign host, there is the potential for an adverse MR imaging is an increasingly used, nonin-
immunologic reaction to the OCA. Patients vasive method for the assessment of articu-
28 who express positive humoral immune lar cartilage defects and cartilage repair.

Stannard_9781604068580_Ch02.indd 28 1/30/13 2:17 PM

MR imaging may define the natural history 13. Potter HG, Foo LF. Magnetic resonance imag-
ing of joint arthroplasty. Orthop Clin North Am
of cartilage repair procedures and detect 2006;37(3):361373; vivii
complications, potentially obviating the need 14. Erickson SJ, Prost RW, Timins ME. The magic
for second-look arthroscopy. The contin- angle effect: background physics and clinical rel-
ued application of quantitative MR imaging evance. Radiology 1993;188(1):2325
15. Fullerton GD, Rahal A. Collagen structure: the
allows for assessment of tissue biochemistry. molecular source of the tendon magic angle effect.
The future study of these pulse sequences J Magn Reson Imaging 2007;25(2):345361
may provide indirect insight into repair tissue 16. Potter HG, Chong R. Magnetic resonance imaging
mechanical properties, thus providing nonin- assessment of chondral lesions and repair. J Bone
Joint Surg Am 2009;91(Suppl 1):126131
vasive insight into functional tissue capacity. 17. Glaser C. New techniques for cartilage imaging: T2
relaxation time and diffusion-weighted MR imag-
ing. Radiol Clin North Am 2005;43(4):641653, vii
18. de Visser SK, Bowden JC, Wentrup-Byrne E, et
References al. Anisotropy of collagen fibre alignment in
bovine cartilage: comparison of polarised light
1. Mainil-Varlet P, Aigner T, Brittberg M, et al; Inter- microscopy and spatially resolved diffusion-
national Cartilage Repair Society. Histological tensor measurements. Osteoarthritis Cartilage
assessment of cartilage repair: a report by the His- 2008;16(6):689697
tology Endpoint Committee of the International 19. Meder R, de Visser SK, Bowden JC, Bostrom T, Pope
Cartilage Repair Society (ICRS). J Bone Joint Surg JM. Diffusion tensor imaging of articular cartilage
Am 2003;85-A(Suppl 2):4557 as a measure of tissue microstructure. Osteoar-

2 MRI of Cartilage Repair Techniques

2. Brown WE, Potter HG, Marx RG, Wickiewicz TL, thritis Cartilage 2006;14(9):875881
Warren RF. Magnetic resonance imaging appear- 20. Mosher TJ, Dardzinski BJ. Cartilage MRI T2 relaxa-
ance of cartilage repair in the knee. Clin Orthop tion time mapping: overview and applications.
Relat Res 2004;(422):214223 Semin Musculoskelet Radiol 2004;8(4):355368
3. Buckwalter JA, Hunziker E, Rosenberg J. Articu- 21. Mosher TJ, Dardzinski BJ, Smith MB. Human
lar cartilage and knee joint function. In: Whit EJ articular cartilage: influence of aging and early
ed. Articular Cartilage: Composition, Structure, symptomatic degeneration on the spatial varia-
Response to Injury and Methods of Facilitation of tion of T2preliminary findings at 3T. Radiology
Repair. New York: Raven Press; 1990:1956 2000;214(1):259266
4. Rosenberg L, Hellmann W, Kleinschmidt AK. Elec- 22. David-Vaudey E, Ghosh S, Ries M, Majumdar S. T2
tron microscopic studies of proteoglycan aggre- relaxation time measurements in osteoarthritis.
gates from bovine articular cartilage. J Biol Chem Magn Reson Imaging 2004;22(5):673682
1975;250(5):18771883 23. Alhadlaq HA, Xia Y, Moody JB, Matyas JR. Detect-
5. Buckwalter JA, Mankin HJ. Articular cartilage: tis- ing structural changes in early experimental
sue design and chondrocyte-matrix interactions. osteoarthritis of tibial cartilage by microscopic
Instr Course Lect 1998;47:477486 magnetic resonance imaging and polarised light
6. Black BR, Chong R, Potter HG. Cartilage imag- microscopy. Ann Rheum Dis 2004;63(6):709717
ing in sports medicine. Sports Med Arthrosc 24. Glaser C, Mendlik T, Dinges J, et al. Global and
2009;17(1):6880 regional reproducibility of T2 relaxation time
7. Shindle MK, Foo LF, Kelly BT, et al. Magnetic reso- measurements in human patellar cartilage. Magn
nance imaging of cartilage in the athlete: current Reson Med 2006;56(3):527534
techniques and spectrum of disease. J Bone Joint 25. Raya JG, Dietrich O, Horng A, Weber J, Reiser
Surg Am 2006;88(Suppl 4):2746 MF, Glaser C. T2 measurement in articular car-
8. Eckstein F, Schnier M, Haubner M, et al. Accuracy tilage: impact of the fitting method on accu-
of cartilage volume and thickness measurements racy and precision at low SNR. Magn Reson Med
with magnetic resonance imaging. Clin Orthop 2010;63(1):181193
Relat Res 1998; (352):137148 26. Mosher TJ, Zhang Z, Reddy R, et al. Knee articu-
9. Eckstein F, Hudelmaier M, Wirth W, et al. Double lar cartilage damage in osteoarthritis: analy-
echo steady state magnetic resonance imaging of sis of MR image biomarker reproducibility in
knee articular cartilage at 3 Tesla: a pilot study ACRIN-PA 4001 multicenter trial. Radiology
for the Osteoarthritis Initiative. Ann Rheum Dis 2011;258(3):832842
2006;65(4):433441 27. Lesperance LM, Gray ML, Burstein D. Deter-
10. Glaser C, Tins BJ, Trumm CG, Richardson JB, Reiser mination of fixed charge density in cartilage
MF, McCall IW. Quantitative 3D MR evaluation of using nuclear magnetic resonance. J Orthop Res
autologous chondrocyte implantation in the knee: 1992;10(1):113
feasibility and initial results. Osteoarthritis Carti- 28. Shapiro EM, Borthakur A, Gougoutas A, Reddy
lage 2007;15(7):798807 R. 23Na MRI accurately measures fixed charge
11. Potter HG, Foo LF. Magnetic resonance imaging density in articular cartilage. Magn Reson Med
of articular cartilage: trauma, degeneration, and 2002;47(2):284291
repair. Am J Sports Med 2006;34(4):661677 29. Wheaton AJ, Borthakur A, Shapiro EM, et al. Pro-
12. Potter HG, Linklater JM, Allen AA, Hannafin JA, teoglycan loss in human knee cartilage: quantita-
Haas SB. Magnetic resonance imaging of articu- tion with sodium MR imagingfeasibility study.
lar cartilage in the knee. An evaluation with use Radiology 2004;231(3):900905
of fast-spin-echo imaging. J Bone Joint Surg Am
30. Borthakur A, Shapiro EM, Akella SV, Gougoutas A,
Kneeland JB, Reddy R. Quantifying sodium in the

Stannard_9781604068580_Ch02.indd 29 1/30/13 2:17 PM

human wrist in vivo by using MR imaging. Radiol- microfracture. Clin Orthop Relat Res 2003;(407):
ogy 2002;224(2):598602 215227
31. Bashir A, Gray ML, Burstein D. Gd-DTPA2- as a 48. Frisbie DD, Trotter GW, Powers BE, et al. Arthro-
measure of cartilage degradation. Magn Reson scopic subchondral bone plate microfracture tech-
Med 1996;36(5):665673 nique augments healing of large chondral defects
32. Wheaton AJ, Dodge GR, Borthakur A, Kneeland JB, in the radial carpal bone and medial femoral con-
Schumacher HR, Reddy R. Detection of changes dyle of horses. Vet Surg 1999;28(4):242255
in articular cartilage proteoglycan by T(1rho) 49. Peterson L, Minas T, Brittberg M, Nilsson A,
magnetic resonance imaging. J Orthop Res Sjgren-Jansson E, Lindahl A. Two- to 9-year
2005;23(1):102108 outcome after autologous chondrocyte trans-
33. Li X, Han ET, Ma CB, Link TM, Newitt DC, plantation of the knee. Clin Orthop Relat Res
Majumdar S. In vivo 3T spiral imaging based multi- 2000;(374):212234
slice T(1rho) mapping of knee cartilage in osteoar- 50. Potter HG, Chong R, Sneag DB. Magnetic resonance
thritis. Magn Reson Med 2005;54(4):929936 imaging of cartilage repair. Sports Med Arthrosc
34. Duvvuri U, Reddy R, Patel SD, Kaufman JH, Knee- 2008;16(4):236245
land JB, Leigh JS. T1rho-relaxation in articular 51. Domayer SE, Kutscha-Lissberg F, Welsch G, et al.
cartilage: effects of enzymatic degradation. Magn T2 mapping in the knee after microfracture at
Reson Med 1997;38(6):863867 3.0 T: correlation of global T2 values and clinical
35. Wheaton AJ, Casey FL, Gougoutas AJ, et al. Correla- outcomepreliminary results. Osteoarthritis Car-
tion of T1rho with fixed charge density in carti- tilage 2008;16(8):903908
lage. J Magn Reson Imaging 2004;20(3):519525 52. Welsch GH, Mamisch TC, Domayer SE, et al.
36. Regatte RR, Akella SV, Borthakur A, Kneeland Cartilage T2 assessment at 3-T MR imaging:
JB, Reddy R. Proteoglycan depletion-induced in vivo differentiation of normal hyaline carti-
changes in transverse relaxation maps of carti- lage from reparative tissue after two cartilage
repair proceduresinitial experience. Radiology
I Diagnosis and Treatment Planning

lage: comparison of T2 and T1rho. Acad Radiol

2002;9(12):13881394 2008;247(1):154161
37. Li X, Benjamin Ma C, Link TM, et al. In vivo T(1rho) 53. Brittberg M, Lindahl A, Nilsson A, Ohlsson C,
and T(2) mapping of articular cartilage in osteo- Isaksson O, Peterson L. Treatment of deep car-
arthritis of the knee using 3T MRI. Osteoarthritis tilage defects in the knee with autologous
Cartilage 2007;15(7):789797 chondrocyte transplantation. N Engl J Med
38. Regatte RR, Akella SV, Wheaton AJ, et al. 3D-T1rho- 1994;331(14):889895
relaxation mapping of articular cartilage: in vivo 54. Minas T. Autologous chondrocyte implantation
assessment of early degenerative changes in for focal chondral defects of the knee. Clin Orthop
symptomatic osteoarthritic subjects. Acad Radiol Relat Res 2001; (391, Suppl):S349S361
2004;11(7):741749 55. Marlovits S, Zeller P, Singer P, Resinger C, Vc-
39. Buckwalter JA. Articular cartilage: injuries and sei V. Cartilage repair: generations of autolo-
potential for healing. J Orthop Sports Phys Ther gous chondrocyte transplantation. Eur J Radiol
1998;28(4):192202 2006;57(1):2431
40. Curl WW, Krome J, Gordon ES, Rushing J, Smith BP, 56. Marlovits S, Singer P, Zeller P, Mandl I, Haller J,
Poehling GG. Cartilage injuries: a review of 31,516 Trattnig S. Magnetic resonance observation of car-
knee arthroscopies. Arthroscopy 1997;13(4): tilage repair tissue (MOCART) for the evaluation of
456460 autologous chondrocyte transplantation: deter-
41. Trattnig S, Winalski CS, Marlovits S, Jurvelin mination of interobserver variability and correla-
JS, Welsch GH, Potter HG. Magnetic resonance tion to clinical outcome after 2 years. Eur J Radiol
imaging of cartilage repair: a review. Cartilage 2006;57(1):1623
2011;2(1):526 57. Breinan HA, Minas T, Hsu HP, Nehrer S, Sledge CB,
42. Steadman JR, Rodkey WG, Rodrigo JJ. Microfrac- Spector M. Effect of cultured autologous chondro-
ture: surgical technique and rehabilitation to treat cytes on repair of chondral defects in a canine model.
chondral defects. Clin Orthop Relat Res 2001;(391, J Bone Joint Surg Am 1997;79(10):14391451
Suppl):S362S369 58. Minas T, Peterson L. Advanced techniques in
43. Minas T, Nehrer S. Current concepts in the treat- autologous chondrocyte transplantation. Clin
ment of articular cartilage defects. Orthopedics Sports Med 1999;18(1):1344; vvi
1997;20(6):525538 59. Trattnig S, Ba-Ssalamah A, Pinker K, Plank C,
44. Alparslan L, Winalski CS, Boutin RD, Minas T. Post- Vecsei V, Marlovits S. Matrix-based autologous
operative magnetic resonance imaging of articu- chondrocyte implantation for cartilage repair:
lar cartilage repair. Semin Musculoskelet Radiol noninvasive monitoring by high-resolution mag-
2001;5(4):345363 netic resonance imaging. Magn Reson Imaging
45. Mithoefer K, Williams RJ III, Warren RF, et al. 2005;23(7):779787
The microfracture technique for the treatment 60. Moriya T, Wada Y, Watanabe A, et al. Evaluation
of articular cartilage lesions in the knee. A pro- of reparative cartilage after autologous chondro-
spective cohort study. J Bone Joint Surg Am cyte implantation for osteochondritis dissecans:
2005;87(9):19111920 histology, biochemistry, and MR imaging. J Orthop
46. Frisbie DD, Morisset S, Ho CP, Rodkey WG, Sci 2007;12(3):265273
Steadman JR, McIlwraith CW. Effects of calcified 61. Alparslan L, Minas T, Winalski CS. Magnetic reso-
cartilage on healing of chondral defects treated nance imaging of autologous chondrocyte implan-
with microfracture in horses. Am J Sports Med tation. Semin Ultrasound CT MR 2001;22(4):
2006;34(11):18241831 341351
30 47. Frisbie DD, Oxford JT, Southwood L, et al. Early
events in cartilage repair after subchondral bone
62. Henderson IJ, Tuy B, Connell D, Oakes B, Hettwer
WH. Prospective clinical study of autologous

Stannard_9781604068580_Ch02.indd 30 1/30/13 2:17 PM

chondrocyte implantation and correlation with 71. Hangody L, Kish G, Krpti Z, Udvarhelyi I, Szigeti
MRI at three and 12 months. J Bone Joint Surg Br I, Bly M. Mosaicplasty for the treatment of articu-
2003;85(7):10601066 lar cartilage defects: application in clinical prac-
63. Tins BJ, McCall IW, Takahashi T, et al. Autologous tice. Orthopedics 1998;21(7):751756
chondrocyte implantation in knee joint: MR imag- 72. Pearce SG, Hurtig MB, Clarnette R, Kalra M, Cowan
ing and histologic features at 1-year follow-up. B, Miniaci A. An investigation of 2 techniques for
Radiology 2005;234(2):501508 optimizing joint surface congruency using multi-
64. Trattnig S, Mamisch TC, Welsch GH, et al. Quan- ple cylindrical osteochondral autografts. Arthros-
titative T2 mapping of matrix-associated autolo- copy 2001;17(1):5055
gous chondrocyte transplantation at 3 Tesla: 73. Lane JM, Brighton CT, Ottens HR, Lipton M. Joint
an in vivo cross-sectional study. Invest Radiol resurfacing in the rabbit using an autologous oste-
2007;42(6):442448 ochondral graft. J Bone Joint Surg Am 1977;59(2):
65. Gillis A, Bashir A, McKeon B, Scheller A, Gray ML, 218222
Burstein D. Magnetic resonance imaging of rela- 74. Dew TL, Martin RA. Functional, radiographic,
tive glycosaminoglycan distribution in patients and histologic assessment of healing of autog-
with autologous chondrocyte transplants. Invest enous osteochondral grafts and full-thickness
Radiol 2001;36(12):743748 cartilage defects in the talus of dogs. Am J Vet Res
66. Glenn RE Jr, McCarty EC, Potter HG, Juliao SF, 1992;53(11):21412152
Gordon JD, Spindler KP. Comparison of fresh 75. Sirlin CB, Brossmann J, Boutin RD, et al. Shell oste-
osteochondral autografts and allografts: a canine ochondral allografts of the knee: comparison of
model. Am J Sports Med 2006;34(7):10841093 MR imaging findings and immunologic responses.
67. Matsusue Y, Yamamuro T, Hama H. Arthroscopic Radiology 2001;219(1):3543
multiple osteochondral transplantation to the 76. Williams RJ, Gamradt SC. Articular cartilage repair
chondral defect in the knee associated with ante- using a resorbable matrix scaffold. Instr Course

2 MRI of Cartilage Repair Techniques

rior cruciate ligament disruption. Arthroscopy Lect 2008;57:563571
1993;9(3):318321 77. Bedi A, Foo LF, Williams RJ III, Potter HG; The Car-
68. Bobi V. Arthroscopic osteochondral autograft tilage Study Group. The maturation of synthetic
transplantation in anterior cruciate ligament recon- scaffolds for osteochondral donor sites of the
struction: a preliminary clinical study. Knee Surg knee: an MRI and T2 mapping analysis. Cartilage
Sports Traumatol Arthrosc 1996;3(4):262264 2010;1(1):2028
69. Nho SJ, Foo LF, Green DM, et al. Magnetic reso- 78. Foo LF, Jawetz ST, Williams RJ. MRI and quantita-
nance imaging and clinical evaluation of patellar tive T2 mapping of cartilage repair using synthetic
resurfacing with press-fit osteochondral autograft biphasic acellular scaffold. ISMRM annual meet-
plugs. Am J Sports Med 2008;36(6):11011109 ing (paper presentation, abstract 1030). Berlin,
70. Koh JL, Wirsing K, Lautenschlager E, Zhang LO. Germany; 2007
The effect of graft height mismatch on contact 79. Williams RJ III, Ranawat AS, Potter HG, Carter T,
pressure following osteochondral grafting: a bio- Warren RF. Fresh stored allografts for the treat-
mechanical study. Am J Sports Med 2004;32(2): ment of osteochondral defects of the knee. J Bone
317320 Joint Surg Am 2007;89(4):718726


Stannard_9781604068580_Ch02.indd 31 1/30/13 2:17 PM

Evolving Biomarkers
in Osteoarthritis
Debabrata Patra and Linda J. Sandell

Accurate and efficient diagnosis and prog- incomplete understanding of OA pathogen-

nosis are central to effective patient care. esis in conjunction with the heterogeneity
Localized pain, discomfort, or an annual of the human population.
check-up is usually the starting point for Like several other chronic diseases, OA is
patient examination by a physician. But characterized by a silent phase of the disease
the route of patient care taken by a physi- when the individual is not aware that the
cian depends on several factors derived disease process has begun. Fig. 3.1 is a rep-
from blood tests, radiographs, magnetic resentation of the phases of OA in humans
resonance imaging (MRI) scans, or a physi- (modified from Kraus 2010)1 where bio-
cal examination, all of which generate a set marker validation is desirable for phases 0
of biomarkers advising the physician on the and 1 but is currently hypothetical. This has
next step in patient care. Thus, conceptually led to a concentrated movement in search of
a biomarker is any set of parameters that biomarkers to detect early OA. During early
when out of the normal range would inform OA, though an individual may appear normal
the physician of abnormal body physiology on radiographs or MRI scans, abnormalities
or injury in the patient. In the field of osteo- in joint physiology may indicate OA initia-
arthritis (OA), however, a physician has very tion by elevations in the levels of cartilage
little on which to plan an effective, in-depth breakdown products or markers of inflam-
patient care stratagem to prevent OA pro- mation. These biochemical markers are the
gression. Patients are usually seen when the next wave in OA diagnosis, monitoring OA
disease has advanced to a stage at which progression, understanding changes in body
further progression cannot be prevented. physiology, and determining efficacy of phar-
At this stage, a physicians recommenda- maceutical interventions. An OA biochemical
tions are limited to palliative or salvage biomarker could be any molecule that would
procedures such as activity modification as act as a surrogate to inform the physician of
in rest and/or joint immobilization, anti- either disease initiation or progression and
inflammatories, analgesics, or total joint allow for accurate disease staging and prog-
arthroplasty. These limitations in therapeu- nosis for determining optimal preventive and
tic options reflect a lack of clinically relevant therapeutic strategies. Ideally biomarkers for
diagnostic or disease-staging methodolo- OA will be joint-specific and detectable dur-
gies and the unavailability of any disease- ing the silent, nonsymptomatic phase of OA.
modifying treatments, compounded by an If possible, biomarkers will also be able to

Stannard_9781604068580_Ch03.indd 32 1/30/13 2:22 PM

3 Evolving Biomarkers in Osteoarthritis
Fig. 3.1 A diagrammatic representation of the magnetic resonance (MR) imaging they are begin-
osteoarthritis (OA) disease process in humans ning to be detected in phase 2. Early OA detection
showing the various stages of the disease and an in phases 0 and 1 is desirable but not in practice due
accompanying diagnosing scheme. Patients usually to incomplete development of biomarkers for these
enter this scheme at phase 3, but with advances in stages.

distinguish age- and trauma-related changes to be of use in detecting the elusive, early
to allow a physician to assess the degree of OA phases. Thus, a primary goal of OA bio-
damage after injury. marker development is to catch the disease
To derive the maximum benefit from a in the early stages to prevent the ravages of
biomarker it would be advisable to arrive at late-stage OA. Several biomarkers are cur-
a consensus for a definition of the various rently in use but have varying degrees of
stages of the disease process. In OA, often acceptance for their utility in OA (Table 3.1).
this is lacking. Many individuals are often The OA Biomarkers Global Initiative has been
reported to have radiographic OA as defined organized by Osteoarthritis Research Society
by joint space narrowing. However, most of International (OARSI) in the recent past to
these individuals often do not have sympto- accelerate OA biomarker development.2 A
matic OA as defined by pain or discomfort. documentation by the OARSI Food and Drug
But joint space narrowing would be due to Administration (FDA) initiative provides
changes in articular cartilage structure that a summary and guide for application of in
would reflect changes in articular cartilage vitro, soluble, biochemical biomarkers for
composition. An individual at such a stage monitoring OA and pharmacological trials
of OA, radiographic but nonsymptomatic, and also provides a scheme for the classifica-
may not be denoted clinically as having OA tion of these biomarkers.3
and therefore would be bypassed for treat-
ment. This puts the burden of proof on the
biomarker as it should be validated as truly Biomarkers Derived from
representative of the disease process, should Type II Collagen
be an irrefutable proof of disease initiation,
and should be universally recognized as The fibrillar type II collagen (Col II) is the
such. A late-stage biomarker more repre- major collagen of articular cartilage. Col II is
sentative of disease progression is not likely synthesized as a procollagen with the amino 33

Stannard_9781604068580_Ch03.indd 33 1/30/13 2:22 PM

Table 3.1 A list of biochemical biomarkers derived from cartilage, bone, and synovial tissue mentioned in this
review that are under scrutiny for their usefulness in diagnosing, staging, and monitoring OA progression and
in monitoring pharmacological interventions

Potential for clinical

Biochemical biomarker Indicator of Tissue origin application

PIINP, PIIANP, NPII Type II collagen synthesis Cartilage Mild

PIICP (CPII) Type II collagen synthesis Cartilage Mild
CTX-II Type II collagen breakdown Cartilage, bone Exceptionally strong
TIINE, Coll 21, Type II collagen breakdown Cartilage Mild
Coll 21NO2
ARGS Aggrecan breakdown Cartilage Moderate
AGG1 (G11H11), Aggrecan breakdown/ Cartilage Mild
AGG2 (6D6-G2), CS846 turnover
Keratan sulfate Aggrecan breakdown Cartilage Moderate
COMP COMP breakdown Cartilage, meniscus, Mild
NTX-1, CTX-1, PINP Bone resorption Bone Moderate
I Diagnosis and Treatment Planning

Osteocalcin Bone synthesis Bone Moderate

HA, Glc-Gal-PYD Synovial tissue breakdown Synovium, though Moderate
HA could also be from
cartilage and
MMP-3, MMP-13 Joint tissue breakdown Cartilage, synovium Moderately strong

Note: The potential for clinical application of these biochemical biomarkers is subjective and based on their
popularity for use in following and monitoring OA progression in published studies, recent advances, and
their utility and success in published pharmacological trials (see text for details).

(N)-terminal (PIINP and PIIANP) and carboxy of the pathological state of the cartilage.
(C)-terminal (PIICP; also referred to as CPII) The NPII assay (designed to measure pep-
domains removed during assembly of the tides derived from the Col II N-propeptide
molecule. Therefore, enzyme-linked immu- with the antibody detecting different pep-
nosorbent assays (ELISA) to detect these in tides than the PIIANP assay) demonstrated
urine or serum using antibodies to PIINP/ that N-terminal procollagen type II derived
PIIANP/PIICP have been designed to indicate peptide levels are indeed higher in human
recent Col II synthetic activity or increased plasma and urine from patients with radio-
collagen synthesis as a sign of abnormal graphically confirmed OA,6 suggesting that
cartilage metabolism. In a 4-year study to detecting Col IIderived peptides indicative
investigate the prognostic value of PIICP, of synthesis is a viable option for biomark-
synovial fluid levels of PIICP were found to ers. However, these potential biomarkers
correlate well with radiographic progres- have not been validated for clinical use in OA
sion of knee OA.4 However, detection of diagnosis or monitoring treatment.
PIINP (produced primarily by mature chon- The biochemical biomarkers to date that
drocytes5) and PIIANP (identical to PIINP have received a lot of attention and have
except that it has an additional 69 amino shown promise in diagnosing OA effectively
acid, cysteine-rich domain and is produced are primarily those that detect Col II break-
by chondroprogenitor cells and dedifferen- down. Of these, C-terminal cross-linked
tiated, pathological chondrocytes) provides telopeptide of Col II (CTX-II), which can be
an additional dimension in that changes detected in urine in humans by an ELISA, is
34 in PIIANP:PIINP ratios could be a reflection the most characterized and appears to be the

Stannard_9781604068580_Ch03.indd 34 1/30/13 2:22 PM

most promising as a noninvasive biomarker Besides CTX-II, other Col II breakdown
for monitoring OA.7 It provides specific and products have also been used as biomark-
sensitive data regarding Col II breakdown ers in monitoring OA, such as TIINE, Coll
and has found acceptance in OA and rheu- 21, and Coll 21NO2 in both serum and
matoid arthritis (RA) as a biomarker for joint urine in immunoassays. Urinary TIINE is a
structure changes. It has been demonstrated Col II neoepitope and unlike CTX-II has the
to correlate well with power Doppler ultra- advantage that the exact nature of the immu-
sound synovitis and bone mineral density noreactive epitope is known and has been
loss that are early markers of inflammatory characterized by mass spectrometry.14 It is
arthritis.8 However, as several OA and RA produced by the action of several MMPs, such
patients demonstrate normal levels of uri- as MMP-13, and this study demonstrated that
nary (u) CTX-II, CTX-II alone may not be diag- an MMP-13 inhibitor can reduce the levels of
nostically useful. Furthermore, CTX-II was the TIINE neoepitope, suggesting its utility
detected in calcified cartilagebone interface as a biomarker in monitoring drug efficacy.
besides the articular cartilage matrix,9 sug- Assays that detect Coll 21 and its nitrated
gesting that the tissue origins of CTX-II are form, Coll 21NO2, provide information on
not completely understood. However, levels oxidative related helical unwinding or fur-

3 Evolving Biomarkers in Osteoarthritis

of CTX-II have been demonstrated to cor- ther breakdown of the triple helical region
relate well with total body burden of osteo- of Col II and have been found to be increased
phyte, a major pathological feature of OA, in patients with primary knee OA.15,16 These
suggesting that CTX-II can be a component of relatively new biomarkers have, however,
a biomarker panel for clinical use in OA.10 In found only limited success in monitoring
practice, CTX-II changes are often monitored pharmacological trials.
in combination with other biomarkers, and
its use has found success in several pharma-
cological trials, though in monitoring primar- Biomarkers Derived
ily hip and knee OA. As such, by combining from Aggrecan
markers of both Col II synthesis (via use of
PIIANP) and collagen breakdown (via use of Although monitoring Col IIderived degra-
CTX-II), it was shown that patients with knee dation products has been a primary focus of
OA who had the largest uncoupling between investigators, several aggrecan breakdown
Col II synthesis (low levels of PIIANP) and products informative of OA pathological
Col II degradation (high levels for CTX-II) had conditions may have potential for applica-
an eightfold more rapid progression of joint tion as biomarkers. Given the fact that loss
damage than other patients,11 suggesting of proteoglycan staining in the articular
that combining these two biomarkers would cartilage in models of OA is almost imme-
be effective in identifying subjects with high diate, aggrecan-derived biomarkers could
risk for progressive knee OA. CTX-II was also potentially be effective for early detection
used to monitor the beneficial aspects of of OA. In fact, aggrecan breakdown result-
orally treating knee OA patients with salmon ing in production of the ARGS neoepitope
calcitonin (sCT).12 In this study it was moni- sequence in aggrecanase-cleaved aggre-
tored in combination with other noteworthy can17 has recently been demonstrated to be
biomarker candidates for OA such as matrix detectable in human synovial fluid, serum,
metalloproteinase (MMP)-3 (considered a and urine, with the second-generation BC3-
significant predictor of joint space narrow- C2 antibody in a sensitive immunoassay18
ing), along with the collagenase MMP-13 opening up avenues for aggrecan-derived
(enzyme that degrades Col II), all of which neoepitopes being considered as worthy bio-
demonstrated significant decreases on sCT markers for OA diagnosis and as clinical end
intake. CTX-II alone has also found success points for disease-modifying OA drugs. The
in well-controlled studies to identify OA degenerative aggrecan breakdown products
patients with high cartilage turnover and AGG1 (G11H11) and AGG2 (6D6-G2) have
also in monitoring pharmacological interven- been used in combination with CTX-II to vali-
tion of OA with glucosamine sulfate.13 date elevated levels of synovial fluidderived 35

Stannard_9781604068580_Ch03.indd 35 1/30/13 2:22 PM

visfatin as indicative of degenerative cartilage and reducing pain, stiffness, and joint func-
changes during knee OA.19 Likewise, in female tion in patients with primary knee OA.26,27 A
patients with knee OA, AGG1 and AGG2 cor- significant drop in both uCTX-II and uNTX-I
related well with synovial fluid adiponectin, was observed in these patients with risedro-
though no correlation was observed with nate treatment (though found ineffective at
CTX-II.20 Immunoassays designed to detect reducing OA progression).
a complex of fibronectin (another cartilage Very few study designs have based their
component) and aggrecan in synovial fluids conclusions on data from a single bio-
have found applications in monitoring knee marker. In the vast majority of studies, com-
pain due to meniscal injury.21 Fluctuations in binatorial biomarker tracking has allowed
the serum levels of CS846, a derivative of the investigators to distinguish between dif-
chondroitin sulfate side chains of aggrecan, ferent stages of OA and definitions of OA.
indicative of both aggrecan synthesis and For example, in a recent study to test the
turnover, have been shown to be associated relationship between biomarkers and early
with joint space narrowing, though it did not radiographically defined knee OA (based on
show promise in a pharmacological trial.22 the Kellgren-Lawrence [K/L] grading sys-
But significant decreases in serum levels of tem), knee pain, and joint inflammation,
keratan sulfate (from the keratan sulfate side uCTX-II levels alone did not correlate well
chains of aggrecan) in patients with knee OA with pain, but the ratio of uCTX-II/serum
I Diagnosis and Treatment Planning

treated with chondroitin sulfate have been CPII and levels of the synovial marker hya-
demonstrated, suggesting the usefulness of luronic acid (HA) increased with onset of OA
keratan sulfate in monitoring OA pharmaco- irrespective of joint pain in patients with K/L
logical trials and in the utility of chondroitin grade 2 of OA.28 On the other hand, in the
sulfate in treating knee OA symptoms.23 same study, levels of uCTX-II alone as well
as that of uNTX-I along with CPII and HA all
increased significantly in patients if they
Bone, Synovium, and Other had knee pain irrespective of the K/L grade.
A combination of bone formation (PINP
Sources of Biochemical and osteocalcin) and CTX-I and NTX-I (both
Biomarkers indicators of bone resorption activity) was
used to classify patients with knee OA into
As OA is a disease of the joint as opposed to subgroups who lose cartilage at different
only the articular cartilage, bone and synovial rates over 2 years.29 This study showed that
tissue degenerative changes also contribute to higher bone remodeling is associated with
the arsenal of biochemical biomarkers for OA. reduced cartilage loss. Thus, within the OA
But bone markers have not gained the level of population it will be possible to subgroup
confidence that cartilage-derived biomark- individuals based on rates of cartilage loss
ers have achieved. This is due to the fact that when combining both bone resorption and
bone markers are not likely to be exclusive formation markers.
and selective indicators of OA and may not be Detection of the glycosylated analogue of
able to distinguish OA-related pathophysiol- pyridinoline, glucosyl galactosyl pyridinoline
ogy from other bone activities, such as those (Glc-Gal-PYD) in urine by high performance
resulting from postmenopausal osteoporosis, liquid chromatography reflects specifically
age-related skeletal turnover, or other bone synovial tissue degeneration.30 Glc-Gal-PYD
diseases, such as osteonecrosis.24,25 However, in combination with uCTX-II was found to
bone markers have been used effectively in have a strong association with disease sever-
combination with cartilage turnover markers ity and the presence of OA at the tibiofemo-
to assess OA and effectiveness of pharmaco- ral and patellofemoral joints in men.31 In
logical interventions in OA. The urinary bone the same study, changes in serum levels of
resorption marker N-telopeptide of type I CTX-I and osteocalcin, however, did not cor-
collagen (NTX-1) in combination with the relate well with the disease. Glc-Gal-PYD
uCTX-II was used to determine the efficacy and uCTX-II was also used to study the effi-
36 of risedronate in improving joint structure cacy of ibuprofen in reducing the symptoms

Stannard_9781604068580_Ch03.indd 36 1/30/13 2:22 PM

of knee OA.32 Several synovial fluidderived not exist currently. A critical limiting factor
biomarkers, such as interleukin (IL)-1, IL-6, that prevents this unification is the diurnal
IL-8, IL-11, leukemia inhibitory factor (LIF), variation among the biological samples that
cartilage oligomeric matrix protein (COMP), are collected. Serum and/or urine collected at
osteocalcin, and osteogenic protein-1 (OP-1), different times of the day or night, before or
have been analyzed in the synovial fluid from after fasting, with or without physical activ-
OA and RA patients.33 This study suggested ity, show significant differences in levels of
that elevated levels of IL-11, LIF, and OP-1 some biomarkers. Levels of serum COMP are
would be more appropriate as biomarkers found to be constant during normal daytime
indicating OA, while elevated levels of IL-1, activities but decrease during the night.34
IL-6, IL-8, LIF, and OP-1 would be more likely Thus, as long as clinical sampling is limited
to be indicative of RA. to daytime, no further standardization is
The enzymes that break down Col II and necessary for analyzing serum COMP levels.
aggrecan are themselves excellent indicators However, in patients with radiographic knee
of disease states. Native fibrillar Col II is bro- OA, uCTX-II concentrations were found to
ken down first by the combined action of sev- vary with morning activity, decreasing signif-
eral MMPs such as MMP-1, -8, -13, and -14 and icantly when sampled ~4 hours after arising

3 Evolving Biomarkers in Osteoarthritis

then broken down further by MMP-2, -3, and from bed in the morning as compared with
-9. Thus, the detection of these Col II modify- sampling before arising from bed.35 On the
ing enzymes by immunoassays can also serve other hand serum levels of HA increased sig-
as biomarkers. As mentioned above, in stud- nificantly when analyzed 1 hour after arising
ies to test the effectiveness of sCT in allevi- from bed as compared with sampling before
ating symptoms of OA, MMP-3 and MMP-13 arising from bed.36 Diurnal variation in CTX-I
along with CTX-II also served as biomarkers.12 levels is also well established, decreasing sig-
Significant reduction in levels of both MMP-3 nificantly after food intake when compared
and MMP-13 was observed, attesting to the with fasting levels.37 These studies suggest
utility of these enzymes as useful biomarkers that serum or urine sampling for any bio-
for future drug efficacy studies. Thus, a panel marker needs to be standardized and these
of biomarkers will allow for rapid diagnosis standardization protocols should be applied
and improved determination of OA. It should both for diagnosis and for determining the
be noted, however, that a vast majority of efficacy of drugs in clinical trials. It would
these studies were conducted with patients also be important to correlate the levels of
suffering from knee OA. Therefore biomark- biomarkers with different stages of the dis-
ers that have found acceptance as indicative ease to make biological sense. For example,
of OA are directly informative of knee OA as low levels of cartilage turnover markers in
per these studies. In some cases, they are also advanced stages of OA when there is little
indicative of hip OA. However, much work or no cartilage left need to be interpreted in
still needs to be done to validate these bio- combination with other diagnostic parame-
markers as indicative of generalized OA or ters before coming to a conclusion about the
their relationship to other types of OA such stage of the disease.
as hand OA.

Genotypes as Potential
Diurnal Variations Biomarkers for OA
in Biomarker Detection
In addition to biochemical biomarkers,
Most studies using biomarkers are conducted genetic biomarkers are also being developed
within the framework of several variables to identify people with a genetic disposition
that are often not identical across studies. to develop OA, significantly ahead of any
Therefore unifying the diagnostic potential of radiographic changes. OA is a complex dis-
a particular biomarker among different stud- ease and its etiology is poorly understood.
ies will need a standardization of the meth- Besides the age-related degenerative carti-
odology to conduct these studies that does lage changes, myriad factors come into play 37

Stannard_9781604068580_Ch03.indd 37 1/30/13 2:22 PM

whose contribution to OA are difficult to used primarily single nucleotide polymor-
quantify and enumerate. Genetic biomark- phism (SNP) in genes to study their associa-
ers are designed primarily to understand tion with OA. These studies have identified
the etiology of OA with the perk being that several genes, but chief among them is the
they may also shed light on the pathophysi- SNP at the104 position in the 5 UTR of the
ology of OA. The premise for this undertak- growth and differentiation factor 5 (GDF5)
ing is the simple observation that OA tends gene that has shown a strong association to
to run in families. Several approaches have OA.43 In humans, the nucleotide T at104, as
been taken to understand the heritabil- opposed to C, results in higher incidence of
ity of OA; chief among them are candidate OA and was first discovered in the Japanese
gene studies or the genome-wide linkage and Chinese population with knee and hip
scans (GWAS). Candidate gene studies differ OA. This SNP exerts allelic differences in
from GWAS studies in that one has advance transcriptional activity in chondrogenic cells
knowledge of the association of the gene with the OA-associated104T allele, result-
with cartilage and/or skeletal development; ing in reduced GDF5 transcriptional activity.
GWAS studies on the other hand were imple- In a mouse model of dominant-negative Gdf5
mented to identify hitherto nonrelated OA mutation, joint formation is impaired and
susceptibility genes. Both these approaches mice develop severe OA endorsing the asso-
have identified mutations and polymor- ciation between GDF5 and OA in humans.44
I Diagnosis and Treatment Planning

phisms in several ECM genes and signaling Currently, the GDF5 gene has acquired a
molecules that show strong association with reputation that is close to that of a global
OA. In the candidate gene studies, several OA gene as later studies have replicated
mutations in ECM genes resulting in vari- this association in European populations as
ous forms of skeletal dysplasias have been well.45 Besides GDF5, an aspartic acid repeat
studied that demonstrate that these diseases polymorphism (the D14 allele) in the asporin
are severe, early onset, hereditary forms of (ASPN) gene has also shown strong associa-
OA.38 Identification of the disease genes by tion with hip and knee OA in different eth-
linkage analysis has uncovered mutations in nic populations, and like the GDF5 gene, it
the COMP gene to be responsible for pseudo- negatively affects chondrogenesis, the D14
achondroplasia (PSACH);39 mutation in the allele resulting in a stronger inhibition of
type IX collagen (COL9A1)40 or the matrilin-3 TGF- induced expression of cartilage ECM
(MATN3) gene results in a dominant form genes such as aggrecan and type II colla-
of multiple epiphyseal dysplasia (MED);41 gen.46 Interestingly, an SNP in the IL-1 gene,
spondyloepiphyseal dysplasia (SED) con- a chemokine known to drive the pathobiol-
genita results from a mutation in the type II ogy of OA, is associated with severe erosive
collagen gene (COL2A1).42 Patients suffering hand OA.47 Thus, these studies hint at the
from these diseases have a short stature and possibility to screen individuals genetically
invariably suffer from a highly specific form and identify them to be at risk for OA even
of OA as they get older. For example, patients before incidents of radiographic OA or pain,
with SED congenita exhibit severe hip OA, effectively behaving as genetic biomarkers
with moderately severe spine and knee OA for OA.
but never hand OA. But patients with COMP- It is important to realize, however, that
related MED demonstrate severe hip and ethnic differences can still come into play,
hand OA but minimal knee OA. These dif- and different genetic biomarkers may need
ferences occur despite the Col II and COMP to be validated for different ethnic popula-
proteins being widespread in cartilage. Thus tions. Even though GWAS studies are often
these are examples of monogenic diseases replicated using different ethnic popula-
showing familial OA that are essentially bio- tions to increase the reliability of a disease-
markers for OA. associated SNP, there is still considerable
While candidate gene association stud- diversity among different ethnic populations
ies target only a small portion of the human and even among individuals of the same
genome, GWAS studies on the other hand nationality. In some replication studies, dif-
38 perform a genome-wide screen and have ferences were also noted between the sexes.

Stannard_9781604068580_Ch03.indd 38 1/30/13 2:22 PM

In the Rotterdam study, SNPs in GDF5 gene coupled with mass spectrometry has allowed
correlated with knee OA, but not with hip for the detection of altered lipid metabo-
OA, primarily in women; no associations lism in plasma in humans with OA.50 Thus,
were found in men.48 These differences may abnormal lipid profiling, through lipidom-
be explained by the inclusion criteria of the ics, is a worthy addition as a biomarker for
population used in the study (patients had a more accurate OA diagnosis. Likewise, the
less severe radiographic OA as compared use of metabolomics has identified abnor-
with previous studies); the differences, how- mal metabolite ratios of valine to histidine
ever, could also be explained by hormonal and of leucine (or isoleucine) to histidine in
differences between men and women. Thus, human sera as significantly associated with
a common genetic biomarker for both men knee OA.51 It seems likely that these thinking
and women may not be appropriate to diag- out of the box approaches to OA will allow
nose OA. Furthermore, the arcOGEN consor- for a better understanding of the disease
tium stage 1 study suggested that OA is a process with consequent identification and
highly polygenic disease with multiple genes validation of a specific set of biomarkers to
making contributions to the OA disease phe- diagnose OA earlier, and to allow for better
notype.49 Thus, a genetic biomarker based treatment options in OA.

3 Evolving Biomarkers in Osteoarthritis

on a single gene may not have the diagnos-
tic power to identify individuals of different
ethnic populations at risk for OA. The stand- Acknowledgments
ardization of inclusion criteria of the popula-
tion cohorts for GWAS studies, as well as a We thank Virginia Kraus and Bentham
standardization of the definitions of differ- Science (publishers of Current Drug Targets)
ent stages of OA, would help in unifying data for permission to modify their figure on
from different laboratories. Furthermore, which our Fig. 3.1 is based. This work was
like biochemical biomarkers, combinatorial supported by National Institutes of Health
genetic biomarker tracking might need to be grants RO1 AR50847, RO1 AR45550, and RO1
explored to increase the level of confidence AR36994 to L. J. Sandell.
in their usefulness for OA.

Future in OA Biomarker
Development 1. Kraus VB. Waiting for action on the osteoarthritis
front. Curr Drug Targets 2010;11(5):518520
2. Kraus VB, Nevitt M, Sandell LJ. Summary of the
Though a significant body of work exists on OA biomarkers workshop 2009biochemical bio-
markers: biology, validation, and clinical studies.
biomarker development in OA, none of these Osteoarthritis Cartilage 2010;18(6):742745
biomarkers can yet be considered to be the 3. Kraus VB, Burnett B, Coindreau J, et al; OARSI FDA
gold standard for diagnosing, staging, moni- Osteoarthritis Biomarkers Working Group. Appli-
cation of biomarkers in the development of drugs
toring treatment, or prognosticating for OA. intended for the treatment of osteoarthritis. Oste-
Part of this problem is due to the complex oarthritis Cartilage 2011;19(5):515542
nature of OA, the lack of complete under- 4. Sugiyama S, Itokazu M, Suzuki Y, Shimizu K.
standing of its pathophysiology, and the het- Procollagen II C propeptide level in the syno-
vial fluid as a predictor of radiographic progres-
erogeneity of the human population. With sion in early knee osteoarthritis. Ann Rheum Dis
the current crop of biomarkers, the pitfalls in 2003;62(1):2732
diagnosis may be avoided by tracking a panel 5. Sandell LJ, Morris N, Robbins JR, Goldring MB.
Alternatively spliced type II procollagen mRNAs
of biomarkers and/or by expressing them define distinct populations of cells during ver-
in ratios. In the recent past, several worthy tebral development: differential expression of
efforts have been undertaken to apply the the amino-propeptide. J Cell Biol 1991;114(6):
latest state-of-the-art proteomic technolo- 6. Nemirovskiy OV, Sunyer T, Aggarwal P, Abrams M,
gies to improve understanding of disease Hellio Le Graverand MP, Mathews WR. Discovery
mechanisms of OA while also generating and development of the N-terminal procollagen
type II (NPII) biomarker: a tool for measuring col-
new biomarkers for potential clinical appli- lagen type II synthesis. Osteoarthritis Cartilage
cation. The use of liquid chromatography 2008;16(12):14941500 39

Stannard_9781604068580_Ch03.indd 39 1/30/13 2:22 PM

7. Christgau S, Garnero P, Fledelius C, et al. Collagen 20. Hao D, Li M, Wu Z, Duan Y, Li D, Qiu G. Synovial
type II C-telopeptide fragments as an index of car- fluid level of adiponectin correlated with levels of
tilage degradation. Bone 2001;29(3):209215 aggrecan degradation markers in osteoarthritis.
8. Freeston JE, Garnero P, Wakefield RJ, Hensor EM, Rheumatol Int 2010; May 13 (Epub ahead of print)
Conaghan PG, Emery P. Urinary type II collagen 21. Scuderi GJ, Golish SR, Cook FF, Cuellar JM, Bowser
C-terminal peptide is associated with synovitis RP, Hanna LS. Identification of a novel fibronectin-
and predicts structural bone loss in very early aggrecan complex in the synovial fluid of knees
inflammatory arthritis. Ann Rheum Dis 2011; with painful meniscal injury. J Bone Joint Surg Am
70(2):331333 2011;93(4):336340
9. Bay-Jensen AC, Andersen TL, Charni-Ben Tabassi N, 22. Mazzuca SA, Poole AR, Brandt KD, Katz BP, Lane
et al. Biochemical markers of type II collagen break- KA, Lobanok T. Associations between joint space
down and synthesis are positioned at specific sites narrowing and molecular markers of collagen and
in human osteoarthritic knee cartilage. Osteoar- proteoglycan turnover in patients with knee oste-
thritis Cartilage 2008;16(5):615623 oarthritis. J Rheumatol 2006;33(6):11471151
10. Kraus VB, Kepler TB, Stabler T, Renner J, Jordan J. 23. Uebelhart D, Thonar EJ, Delmas PD, Chantraine A,
First qualification study of serum biomarkers as Vignon E. Effects of oral chondroitin sulfate on the
indicators of total body burden of osteoarthritis. progression of knee osteoarthritis: a pilot study.
PLoS ONE 2010;5(3):e9739 Osteoarthritis Cartilage 1998;6(Suppl A):3946
11. Garnero P, Ayral X, Rousseau JC, et al. Uncou- 24. Garnero P, Delmas PD. Noninvasive techniques
pling of type II collagen synthesis and degra- for assessing skeletal changes in inflammatory
dation predicts progression of joint damage in arthritis: bone biomarkers. Curr Opin Rheumatol
patients with knee osteoarthritis. Arthritis Rheum 2004;16(4):428434
2002;46(10):26132624 25. Marx RE, Cillo JE Jr, Ulloa JJ. Oral bisphos-
12. Manicourt DH, Azria M, Mindeholm L, Thonar EJ, phonate-induced osteonecrosis: risk factors,
Devogelaer JP. Oral salmon calcitonin reduces prediction of risk using serum CTX testing, pre-
I Diagnosis and Treatment Planning

Lequesnes algofunctional index scores and vention, and treatment. J Oral Maxillofac Surg
decreases urinary and serum levels of biomarkers 2007;65(12):23972410
of joint metabolism in knee osteoarthritis. Arthri- 26. Bingham CO III, Buckland-Wright JC, Garnero P,
tis Rheum 2006;54(10):32053211 et al. Risedronate decreases biochemical mark-
13. Christgau S, Henrotin Y, Tank LB, et al. Osteoar- ers of cartilage degradation but does not decrease
thritic patients with high cartilage turnover show symptoms or slow radiographic progression in
increased responsiveness to the cartilage pro- patients with medial compartment osteoarthritis
tecting effects of glucosamine sulphate. Clin Exp of the knee: results of the two-year multinational
Rheumatol 2004;22(1):3642 knee osteoarthritis structural arthritis study.
14. Nemirovskiy OV, Dufield DR, Sunyer T, Aggar- Arthritis Rheum 2006;54(11):34943507
wal P, Welsch DJ, Mathews WR. Discovery and 27. Spector TD, Conaghan PG, Buckland-Wright
development of a type II collagen neoepitope JC, et al. Effect of risedronate on joint struc-
(TIINE) biomarker for matrix metalloproteinase ture and symptoms of knee osteoarthritis:
activity: from in vitro to in vivo. Anal Biochem results of the BRISK randomized, controlled trial
2007;361(1):93101 [ISRCTN01928173]. Arthritis Res Ther 2005;7(3):
15. Deberg MA, Labasse AH, Collette J, Seidel L, R625R633
Reginster JY, Henrotin YE. One-year increase of 28. Ishijima M, Watari T, Naito K, et al. Relationships
Coll 21, a new marker of type II collagen deg- between biomarkers of cartilage, bone, synovial
radation, in urine is highly predictive of radio- metabolism and knee pain provide insights into
logical OA progression. Osteoarthritis Cartilage the origins of pain in early knee osteoarthritis.
2005;13(12):10591065 Arthritis Res Ther 2011;13(1):R22
16. Deberg M, Labasse A, Christgau S, et al. New serum 29. Berry PA, Maciewicz RA, Cicuttini FM, Jones MD,
biochemical markers (Coll 21 and Coll 21 NO2) Hellawell CJ, Wluka AE. Markers of bone for-
for studying oxidative-related type II collagen net- mation and resorption identify subgroups of
work degradation in patients with osteoarthritis patients with clinical knee osteoarthritis who
and rheumatoid arthritis. Osteoarthritis Cartilage have reduced rates of cartilage loss. J Rheumatol
2005;13(3):258265 2010;37(6):12521259
17. Sandy JD, Flannery CR, Neame PJ, Lohmander LS. 30. Gineyts E, Garnero P, Delmas PD. Urinary excre-
The structure of aggrecan fragments in human tion of glucosyl-galactosyl pyridinoline: a specific
synovial fluid. Evidence for the involvement in biochemical marker of synovium degradation.
osteoarthritis of a novel proteinase which cleaves Rheumatology (Oxford) 2001;40(3):315323
the Glu 373-Ala 374 bond of the interglobular 31. Jordan KM, Syddall HE, Garnero P, et al. Urinary
domain. J Clin Invest 1992;89(5):15121516 CTX-II and glucosyl-galactosyl-pyridinoline are
18. Swearingen CA, Carpenter JW, Siegel R, et al. associated with the presence and severity of radi-
Development of a novel clinical biomarker assay to ographic knee osteoarthritis in men. Ann Rheum
detect and quantify aggrecanase-generated aggre- Dis 2006;65(7):871877
can fragments in human synovial fluid, serum 32. Gineyts E, Mo JA, Ko A, et al. Effects of ibuprofen
and urine. Osteoarthritis Cartilage 2010;18(9): on molecular markers of cartilage and synovium
11501158 turnover in patients with knee osteoarthritis. Ann
19. Duan Y, Hao D, Li M, et al. Increased synovial fluid Rheum Dis 2004;63(7):857861
visfatin is positively linked to cartilage degrada- 33. Kokebie R, Aggarwal R, Lidder S, et al. The role of
tion biomarkers in osteoarthritis. Rheumatol Int synovial fluid markers of catabolism and anabo-
2011; January 19 (Epub ahead of print) lism in osteoarthritis, rheumatoid arthritis and

Stannard_9781604068580_Ch03.indd 40 1/30/13 2:22 PM

asymptomatic organ donors. Arthritis Res Ther spondyloepiphyseal dysplasia. Proc Natl Acad Sci
2011;13(2):R50 U S A 1990;87(10):38893893
34. Andersson ML, Petersson IF, Karlsson KE, et al. 43. Miyamoto Y, Mabuchi A, Shi D, et al. A functional
Diurnal variation in serum levels of cartilage oli- polymorphism in the 50 UTR of GDF5 is associated
gomeric matrix protein in patients with knee oste- with susceptibility to osteoarthritis. Nat Genet
oarthritis or rheumatoid arthritis. Ann Rheum Dis 2007;39(4):529533
2006;65(11):14901494 44. Daans M, Luyten FP, Lories RJ. GDF5 deficiency in
35. Kong SY, Stabler TV, Criscione LG, Elliott AL, mice is associated with instability-driven joint
Jordan JM, Kraus VB. Diurnal variation of serum damage, gait and subchondral bone changes. Ann
and urine biomarkers in patients with radio- Rheum Dis 2011;70(1):208213
graphic knee osteoarthritis. Arthritis Rheum 45. Chapman K, Takahashi A, Meulenbelt I, et al. A
2006;54(8):24962504 meta-analysis of European and Asian cohorts
36. Criscione LG, Elliott AL, Stabler T, Jordan JM, Pieper reveals a global role of a functional SNP in the 5
CF, Kraus VB. Variation of serum hyaluronan with UTR of GDF5 with osteoarthritis susceptibility.
activity in individuals with knee osteoarthritis. Hum Mol Genet 2008;17(10):14971504
Osteoarthritis Cartilage 2005;13(9):837840 46. Kizawa H, Kou I, Iida A, et al. An aspartic acid
37. Qvist P, Christgau S, Pedersen BJ, Schlemmer A, repeat polymorphism in asporin inhibits chon-
Christiansen C. Circadian variation in the serum drogenesis and increases susceptibility to osteo-
concentration of C-terminal telopeptide of type I arthritis. Nat Genet 2005;37(2):138144
collagen (serum CTx): effects of gender, age, men- 47. Stern AG, de Carvalho MR, Buck GA, et al; I-NODAL
opausal status, posture, daylight, serum cortisol, Network. Association of erosive hand osteoarthri-
and fasting. Bone 2002;31(1):5761 tis with a single nucleotide polymorphism on the
38. Ikegawa S. Genetic analysis of skeletal dysplasia: gene encoding interleukin-1 beta. Osteoarthritis

3 Evolving Biomarkers in Osteoarthritis

recent advances and perspectives in the post- Cartilage 2003;11(6):394402
genome-sequence era. J Hum Genet 2006;51(7): 48. Vaes RB, Rivadeneira F, Kerkhof JM, et al. Genetic
581586 variation in the GDF5 region is associated with
39. Briggs MD, Hoffman SM, King LM, et al. Pseudoa- osteoarthritis, height, hip axis length and frac-
chondroplasia and multiple epiphyseal dysplasia ture risk: the Rotterdam study. Ann Rheum Dis
due to mutations in the cartilage oligomeric matrix 2009;68(11):17541760
protein gene. Nat Genet 1995;10(3):330336 49. Panoutsopoulou K, Southam L, Elliott KS, et al; arc-
40. Czarny-Ratajczak M, Lohiniva J, Rogala P, et al. A OGEN Consortium. Insights into the genetic archi-
mutation in COL9A1 causes multiple epiphyseal tecture of osteoarthritis from stage 1 of the arcOGEN
dysplasia: further evidence for locus heterogene- study. Ann Rheum Dis 2011;70(5):864867
ity. Am J Hum Genet 2001;69(5):969980 50. Castro-Perez JM, Kamphorst J, DeGroot J, et al.
41. Chapman KL, Mortier GR, Chapman K, Loughlin J, Comprehensive LC-MS E lipidomic analysis using
Grant ME, Briggs MD. Mutations in the region encod- a shotgun approach and its application to bio-
ing the von Willebrand factor A domain of mat- marker detection and identification in osteoarthri-
rilin-3 are associated with multiple epiphyseal tis patients. J Proteome Res 2010;9(5):23772389
dysplasia. Nat Genet 2001;28(4):393396 51. Zhai G, Wang-Sattler R, Hart DJ, et al. Serum
42. Tiller GE, Rimoin DL, Murray LW, Cohn DH. branched-chain amino acid to histidine ratio: a
Tandem duplication within a type II colla- novel metabolomic biomarker of knee osteoar-
gen gene (COL2A1) exon in an individual with thritis. Ann Rheum Dis 2010;69(6):12271231


Stannard_9781604068580_Ch03.indd 41 1/30/13 2:22 PM

Using Animal Models in
Osteoarthritis Biomarker Research
Bridget C. Garner, Aaron M. Stoker, Keiichi Kuroki, Richard Evans,
Cristi Reeves Cook, and James L. Cook

Osteoarthritis (OA) is characterized by mor- rabbit, and mouse, and for more details on
phologic, molecular, biochemical, and bio- each of these models the reader is directed to
mechanical abnormalities in the articular a series of initiatives.1927 Highlights of each
cartilage and other articular tissues. It has animal model, as well as a concise list of their
been estimated that over 27 million adults advantages and disadvantages (Table 4.1),
have a clinical diagnosis of OA; therefore, it are included here.
is the most common form of arthritis.1 OA Given that the dog is considered a nearly
was historically regarded as a noninflamma- ideal species for translational investigation
tory arthritis, but there is now cumulative of human OA and at least 20% of dogs over
evidence that inflammation has a promi- 1 year of age in the United States are affected
nent role in OA.25 Signals, including those by naturally occurring clinical OA,28 the dog is
associated with proinflammatory cytokines, the most used animal model for investigation
can be sent and received among articular of OA.20 Surgical transection of the cranial
tissues, thus creating a cross-talk envi- (anterior) cruciate ligament29 and meniscal
ronment. A complex interaction involving transection30 are useful methods of surgi-
cytokines,3,4,613 proteolytic enzymes,7,10,11,13,14 cal induction, but spontaneous or chemical
leukocytes,15 synoviocytes,16,17 and chon- induction models and additional surgical
drocytes3,1113,18 likely initiates and exacer- methods are also described.20 Surgical and
bates pathologic changes in osteoarthritic chemical induction methods are employed
joints. Even with this knowledge much of in rabbits and guinea pigs, whereas only sur-
the pathogenesis of OA remains unknown, gical induction methods have been reported
and diagnosis early in the disease course is in sheep and goats.22,23,27 The spontaneous
rare. Consequently the desire to develop an OA-prone Dunkin-Hartley guinea pig31 also
early OA diagnostic biomarker panel, a pro- offers a unique opportunity to study dis-
cess that could have profound ramifications ease mechanisms. A commonly used surgi-
on treatment and prevention of this disease, cal model of OA induction in the horse is the
has exploded. osteochondral fragment-exercise model,3234
Animal models are routinely used to study which, unlike most other surgical induction
human OA as these models provide significant models, does not induce joint instability. The
advantages over clinical human research. The medial meniscal tear model with or with-
most relevant models have been described in out concurrent exercise is most often used
the dog, horse, sheep, goat, guinea pig, rat, to induce OA in the rat.21,35,36 While mice can

Stannard_9781604068580_Ch04.indd 42 1/30/13 2:26 PM

Table 4.1 Advantages and disadvantages of each of the most relevant animal models for translational
investigation of human OA

Model(ref) Advantages Disadvantages

Dog20 Anatomically similar to humans Can be expensive to acquire if

research bred
Arthroscopy possible Larger facilities needed compared
with rodents
Clinical progression and treatment similarities Variability between breeds
Degenerative, trauma, and overuse etiologies occur

4 Animal Models in Osteoarthritis Biomarker Research

Genome sequence available
G. pig22 Histologically similar to human OA Not great for exercise model since
Husbandry practices well developed Not large enough for arthroscopy
Imaging such as CT and MRI can be used
Sedentary lifestyle ideal for some models
Similar risk factors for development of OA
Sufficient tissue to harvest
Horse24 Abundant tissue to collect Need special facilities
Arthroscopy possible Trained personnel needed for safety
Can control activity level
Progressive OA occurs
Mouse26 Easy to handle Difficult tissue and fluid collection
due to size
Genome sequence available Tissue sectioning requires trained
Histologically similar to human OA
Transgenic and knockout models available
Rabbit23 Animals readily available for low cost Gait impedes functional studies
Ease of handling Not large enough for arthroscopy
Lesions develop rapidly
More tissue than mice, rats, guinea pigs
Rat21 Animals readily available for low cost Iatrogenic injuries during surgery
Ease of handling Minimal tissue/body fluids for
Genome sequence available Not large enough for arthroscopy
Lesions develop rapidly Transgenic models not available
Sheep/goat27 Biomechanically similar to humans Genome not fully sequenced
Can house on pasture in some areas Need to validate oral treatments in
Easily managed and handled
Large enough for arthroscopy
Sufficient tissue for collection

Abbreviations: OA, osteoarthritis; G. pig, guinea pig; CT, computed tomography; MRI, magnetic resonance


Stannard_9781604068580_Ch04.indd 43 1/30/13 2:26 PM

be challenging to study for reasons related to turnover of the connective tissue matrix that
their small size, the availability of genetically occurs during growth and development, but
modified or knockout mice is invaluable to the unchecked production of MMPs is a com-
learning more about OA.26 As an example of monly documented cause of cartilage matrix
animal modeling, the authors will focus on degradation in OA in animals.48,49,54,6569
the dog to illustrate the level and limitations Increased collagen 2 denaturation and frag-
of modeling within a specific species. mentation are often present in OA articular
Biomarkers are objective indicators of nor- cartilage,48 and MMP13, which is increased
mal biologic processes, pathogenic processes, in OA chondrocytes, appears to be directly
or pharmacologic responses to therapeutic involved.50 Stimulation of canine chondro-
interventions,37 and they have been classified cytes with interleukin 1 (IL-1) to represent
under the BIPED scheme: Burden of disease, an in vitro OA environment led to an increase
Investigative, Prognostic, Efficacy of interven- in MMP13 versus controls.3 MMP3 has also
tion, and Diagnostic.38 Expansion of this clas- been shown to increase in cartilage and syn-
sification system is already under way as the ovial fluid from canines with experimentally
Osteoarthritic Research Society International induced OA,70 and cultured canine chondro-
Food and Drug Administration (OARSI FDA) cytes stimulated with IL-1 demonstrated
initiative Biomarkers Working Group recently increased matrix degradation and MMP3
elected to modify this acronym to BIPEDS to expression.71
I Diagnosis and Treatment Planning

include a safety category.39 Another method Keratan sulfate, a glycosaminoglycan con-

of biomarker classification involves whether stituent of the cartilage ECM, has been pri-
each is a direct or an indirect indicator of OA.40 marily investigated in animal models using
Direct biomarkers originate from the articular the dog. Synovial fluid keratan sulfate con-
tissues, whereas indirect biomarkers do not centrations have been shown to increase72,73
arise from the joint tissues but can ultimately or decrease74,75 in dogs with OA with no clear
affect the joints microenvironment.41 Both correlation to factors such as manner of dis-
types provide useful information; therefore, ease initiation (induced vs. spontaneous) or
the study of both direct and indirect biomark- chronicity of disease. Given the discordant
ers of OA is warranted. results, keratan sulfate may not be ideal for
Changes in direct markers reflect extracel- detecting or prognosticating OA, but it may
lular matrix (ECM) catabolism or anabolism prove useful for serial monitoring of known
and may reveal the status of joint degenera- OA individuals.75
tion. Several biomarker candidates causing COMP is one of the most comprehensively
or resulting from chondrocyte,13,4246 colla- studied diagnostic biomarker candidates in
gen,11,4751 or aggrecan5255 degradation have the synovial fluid, serum, and urine of vet-
been assessed, and these include matrix met- erinary patients. Elevations in COMP have
alloproteases (MMPs), keratan sulfate, carti- been reported in the serum and synovial fluid
lage oligomeric matrix protein (COMP), and of naturally occurring OA dogs compared
the collagenase-generated cleavage epitope with controls, but elevations due to induced
of type II collagen (C2C or Col23/4C). Studies synovitis were also reported, suggesting a
have shown strong correlations between lack of specificity (SP) for cartilage damage.46
these degradative markers and magnetic res- Another canine study supported these con-
onance imaging (MRI) findings.5660 Markers clusions by showing that a variety of articular
associated with bone turnover have also tissues, including ligament, tendon, menis-
been identified, but they are less commonly cus, and articular cartilage, produced COMP.76
considered to be diagnostic markers as these Regardless, COMP may be diagnostically use-
are often altered only during advanced stages ful as serum and synovial fluid levels have
of disease.6164 been shown to correlate with canine knee MRI
MMPs are zinc- and calcium-dependent grades of cartilage lesions.60 Urinary COMP
enzymes that degrade specific substrates was also predictive of the presence of osteo-
of the ECM and include stromelysins, colla- phytes in racehorses with carpal fractures.77
genases, gelatinases, and membrane metal- Serum COMP concentration was lower in
44 loproteases. MMPs have a role in the normal osteoarthritic horses versus normal.78

Stannard_9781604068580_Ch04.indd 44 1/30/13 2:26 PM

The C2C or Col23/4C has also been inves- cytokines, can be used to learn more about
tigated in several animal studies with some the processes preceding or leading to the
discordant results. Synovial fluid Col2 levels development of OA. Cytokine and chemokine
increased in dogs following induced anterior fluctuations within the synovial fluid of oste-
cruciate rupture.42,79 Serum increases were oarthritic patients have been documented,
also detected in the Pond-Nuki model, and but comprehensive assessment of the poten-
these increases correlated with a marker of tial clinical significance of those alterations
lipid peroxidation, suggesting a link between is largely lacking in the literature.9,16,89,90
oxidative stress and cartilage degeneration.80 Cytokines and chemokines have also dem-
In contrast, differences were not detected in onstrated roles in the pathogenesis of OA,
synovial fluid, serum, or urine concentrations including induction of proteinase expres-

4 Animal Models in Osteoarthritis Biomarker Research

in dogs with naturally occurring anterior cru- sion and inhibition of proteoglycan synthe-
ciate rupture compared with controls even sis.91,92 A more thorough exploration of these
though lameness, joint effusion, and osteo- roles may provide significant information
phytes were present.81 about the pathogenesis of OA and lead to
The pathophysiology of OA centers on the identification of early OA diagnostic bio-
an imbalance of cartilage degradation and markers. Following stimulation with IL-1,
synthesis; therefore, several anabolic mark- the up-regulation of cytokines, chemokines,
ers are also under investigation. Anabolic and MMPs was more rapid than the down-
markers reflect molecules that are present in regulation of matrix gene expression (COL2A1
small quantities or entirely absent in health, and aggrecan), suggesting that these types of
or they can exhibit an altered structure in molecules may be the first changes identi-
which epitopes usually hidden are revealed fiable in early OA.12 As a result, the authors
during OA.51,55,8286 Chondroitin sulfate were interested in learning more about the
(epitopes 3-B-3[] and 846) is one example cytokine and chemokine profiles of dogs
of an anabolic marker. with and without OA and the relationships of
Like keratan sulfate, chondroitin sulfate these profiles to MMP concentrations.
comprises a large portion of the proteogly- Our objectives for this study were (1) to
can aggrecan. A monoclonal antibody (3-B-3) delineate the temporal alterations of cytokine,
recognizes chondroitin sulfate epitopes with chemokine, and MMP concentrations in syn-
(3-B-3[]) or without (3-B-3[]) chondroi- ovial fluid, serum, and urine in induced and
tinase pretreatment. The 3-B-3() epitope is naturally occurring osteoarthritic dogs in
identified in the growth plate cartilage dur- comparison to healthy dogs and (2) to assess
ing normal growth and development and the diagnostic value of particular markers
in the early stages of OA, but it is absent in through evaluation of sensitivity (SN), SP,
healthy adult canine cartilage, suggesting and receiver operating characteristic (ROC)
anabolic production of an altered form of curve analysis. We hypothesized there would
chondroitin sulfate as an attempted repair be strong correlations between cytokine and
response.85,86 Furthermore, it is expressed chemokine fluctuations and the status of the
in superficial zone articular cartilage from joint with respect to OA.
destabilized canine knee joints in early stages
of disease, and it is detectable before a loss
of matrix and proteoglycans is identifiable Materials and Methods
by toluidine blue staining.86 Synovial fluid
values are increased in dogs with naturally Sample Collection
occurring OA and experimental OA as early as
4 weeks after surgical destabilization.73,79,85,87
Part 1: Model Dogs
In contrast to synovial fluid data, serum con-
centrations in dogs with hip dysplasia were All procedures were approved by the institu-
lower than in those without joint disease.88 tions animal care and use committee. In this
While direct markers may more closely study, 21 adult, intact female, purpose-bred
represent the status of joint degeneration, hound dogs  20 kg were included. No more
indirect markers, including but not limited to than 24 hours before surgery, blood was 45

Stannard_9781604068580_Ch04.indd 45 1/30/13 2:26 PM

drawn from the cephalic vein into serum sep- unilateral stifle OA (Pre-sx OA; n  10). These
arator tubes. The serum was harvested within dogs ranged from 3 to 8 years old (median
2 hours of collection, and the specimens were 4.5 years) and included five male castrated
stored in individual, airtight containers at and five female spayed dogs. Synovial fluid
80C. Urine was obtained from each dog was obtained from the affected stifle via rou-
by aseptic cystocentesis or by manual blad- tine aseptic arthrocentesis, and blood was
der expression if cystocentesis was not suc- collected via jugular venipuncture. Synovial
cessful. The urine specimens were kept on fluid samples were kept on ice until they
ice until processing for storage in individual, were aliquoted and frozen at 80C for sub-
airtight containers at 20C. Serum and urine sequent analysis. Clinical OA was confirmed
were collected again at 4, 8, and 12 weeks in each dog by a board-certified veterinary
after surgery using the same protocol. orthopaedic surgeon as determined by knee
On the day of surgery, each dog was pre- physical examination based on the presence
medicated, anesthetized, and aseptically pre- of effusion, periarticular fibrosis, pain upon
pared for arthroscopic surgery of the right flexion and extension, and a lameness evalu-
stifle (knee). Synovial fluid was collected via ation based on the visual examination of gait
aseptic arthrocentesis of the knee joint, and at a walk and trot. Radiographic evidence of
the samples were kept on ice until they were knee OA including signs of osteophytosis,
aliquoted and frozen at 80C for subsequent effusion, and sclerosis was confirmed by a
I Diagnosis and Treatment Planning

analysis. Using standard arthroscopic tech- board-certified veterinary radiologist.

nique and instrumentation for the canine All dogs underwent surgery for assessment,
knee,93 one of four surgical procedures was lavage, and stabilization of cruciate liga-
performed on each dog: transection of the ment deficiency and recovered uneventfully.
anterior cruciate ligament29 (ACL-T; n  5), Eight to 12 weeks later, the dogs returned
complete radial transection of the menis- for a postoperative recheck, and blood and
cus (MR)30 (n  5), creation of two 6.0- to synovial fluid were collected again to assess
8.0-mm-long full-thickness grooves (GRs) in changes in markers after surgical interven-
the cartilage of the weight-bearing portion of tion. Two female dogs did not return for
the medial femoral condyle (GR; n  6) using follow-up; therefore, their postoperative
a 3 mm OD (outside diameter) arthroscopic blood and synovial fluid samples could not be
curette, or manipulation of all the aforemen- obtained (Post-sx OA; n  8). These two dogs
tioned intra-articular structures without were subsequently excluded from paired sta-
insult using an arthroscopic probe (SHAM; tistical analysis.
n  5). Transections were visually confirmed,
and GRs were measured with a calibrated
probe. The nonoperated, contralateral hind Part 2b: Control Group
limb served as an internal control for each The normal control group comprised seven
dog, although synovial fluid was not collected medium and large breed adult dogs ranging
at the time of surgery (baseline). Twelve from 2 to 5 years old (median 2.75 years).
weeks later, synovial fluid was collected by There were three castrated males, two
arthrocentesis from both the operated and spayed females, and two intact males. These
contralateral control knees, and aliquots were dogs had no clinical history of joint trauma,
frozen at 80C. A second arthroscopic evalu- were not lame, and were deemed to be free
ation of the operated joint was performed and of clinical OA as determined by a board-
articular tissues were collected for histology certified veterinary orthopaedic surgeon.
as part of a concurrent study. Radiographic evaluation of the shoulders,
knees, and hips verified the absence of OA.
Blood and synovial fluid were collected in
Part 2a: Client-Owned OA Dogs
a similar manner to the OA dogs at a time
Informed client consent was obtained for each convenient to the clients, and synovial fluid
dog. Blood and synovial fluid were obtained samples were kept on ice until they were ali-
from 10 adult medium and large breed quoted and frozen at 80C for subsequent
46 dogs presenting for surgical intervention of analysis.

Stannard_9781604068580_Ch04.indd 46 1/30/13 2:26 PM

Multiplex Analysis fluid, serum, and urine) and between oper-
ated and contralateral limbs (synovial fluid
An aliquot (25 L) from each synovial fluid, only) were performed with the paired t-test.
serum, and urine sample was thawed. The Comparisons between surgery model groups
urine and synovial fluid samples were cen- or between OA and normal individuals were
trifuged at 14,000 rpm for 10 minutes to pel- performed with the unpaired t-test or the
let debris, and the supernatant was removed. MannWhitney rank sum test (SigmaStat
The synovial fluid was incubated with hyalu- 3.5; Systat Software, Inc., San Jose, CA).
ronidase (MP Biomedicals, LLC, Solon, Ohio) Significance was set at p 0.05. SN and SP
at 37C for 60 minutes to decrease viscosity. were calculated for select markers of interest
Each aliquot was subsequently analyzed in using the histopathological data as the refer-
duplicate using a multiplex canine cytokine

4 Animal Models in Osteoarthritis Biomarker Research

ence test for the model dogs and the clinical
and chemokine immunoassay (Millipore examination and radiographic data for the
Corp., St. Louis, MO) based on the xMAP hospital patients. When possible, Youdens
platform (Qiagen Inc., Valencia, CA) for IL-2, index was used to select optimal concentra-
IL-4, IL-7, IL-8, IL-15, IL-18, IP-10, interferon tion cut-offs that maximized SN and SP for
gamma (INF- ), tumor necrosis factor-alpha each marker. When this index did not pro-
), monocyte chemoattractant protein vide a balance between SN and SP, concen-
1 (MCP1), keratinocyte-derived chemoat- tration cut-offs that selectively led to higher
tractant (KC), and granulocyte-macrophage SN were chosen. The 95% confidence inter-
colony-stimulating factor (GM-CSF) accord- vals (CIs) were calculated using the Clopper
ing to the manufacturers directions. The syn- Pearson method.
ovial fluid and serum from the client-owned
OA dogs and their control group were also
analyzed for IL-6 and IL-10 using the same ROC Curves
assay. A multiplex human MMP immunoas-
ROC curves based on the logistic regression
say (R&D Systems, Minneapolis, MN) based
model were developed for markers of inter-
on the xMAP platform for four MMPs: MMP2,
est to the authors following initial statistical
MMP3, MMP9, and MMP13 was also used
analysis. These curves were created and area
to analyze the client-owned dogs and their
under the curve (AUC) calculated for assess-
control group in duplicate. This MMP assay
ment of diagnostic value of certain param-
had been previously shown within our labo-
eters using JMP 7.0.2 software (SAS Institute,
ratory to cross-react with samples of canine
Cary, NC).
origin.94 Briefly, for each of the xMAP assays
the synovial fluid, serum, and urine samples
were admixed with antichemokine, cytokine,
or MMP monoclonal antibodycharged,
small (5.6 micron), polystyrene microspheres
in a 96-well plate. Following an overnight Part 1: Induction of OA
incubation at 4C, a biotinylated polyclonal
secondary antibody was added, as well as Arthroscopic findings (Fig. 4.1a, b), lame-
streptavidin-phycoerythrin. The median ness scoring performed by a board-certified
fluorescence intensity was determined for veterinary surgeon, and histologic scoring
each sample. The urine creatinine concentra- performed by a board-certified veterinary
tion was measured with an in-house chem- pathologist confirmed the induction of clini-
istry analyzer (AU400; Olympus America cally significant OA in the ACL-T, GR, and
Inc., Irving, TX), and the urine cytokine and MR groups, whereas evidence of OA was
chemokine values were standardized to this not identified in the SHAM dogs. The ACL-T
concentration (pg/mg). and MR dogs exhibited the most severe joint
pathologic changes (although characteristi-
cally different from each other), but GR dogs
also had articular cartilage damage and syno-
Planned comparisons between preopera- vitis. Both investigators were blinded with
tive and postoperative samples (synovial respect to the group during evaluation. 47

Stannard_9781604068580_Ch04.indd 47 1/30/13 2:26 PM

Fig. 4.1 (a, b) Initial and 12-week postoperative transection; GR, groove model; MR, meniscal release;
arthroscopic views of one dog from each surgical SHAM, manipulation without insult.
induction model. ACL-T, anterior cruciate ligament

Part 1: Synovial Fluid not significantly higher in the ACL-T, GR, or

I Diagnosis and Treatment Planning

MR joints compared with the SHAM joints

Individuals without sufficient volume 12 weeks after surgery.
for analysis in operated (baseline and IL-8 (Fig. 4.3) was significantly increased
12 weeks) and contralateral hind limbs at 12 weeks in the GR dogs compared with
were excluded from paired statistical anal- baseline (p  0.024), and in the 12-week
ysis. There were no significant differences MR group compared with the SHAM dogs
between the baseline and the 12-week (p  0.006) at 12 weeks. KC (Fig. 4.4) was
SHAM samples or between operated and significantly decreased in the GR group
nonoperated hind limbs in the SHAM dogs operated and nonoperated limbs at 12 weeks
for any analytes. MCP1 (Fig. 4.2) was sig- (n  4) compared with baseline (p  0.011,
nificantly increased in ACL-T joints (n  4) p  0.017). There were no other significant dif-
12 weeks after surgery compared with base- ferences in cytokine or chemokine expression
line (p  0.036) and nonoperated joints at in ACL-T, GR, MR, and SHAM dogs between
12 weeks (p  0.018). MCP1 trended upward time zero samples and the 12- week non-
in the GR (n  5) and MR groups (n  3) at operated hindlimbs. Statistically significant
12 weeks compared with baseline and non- differences were not detected for any of the
operated limbs, but statistical significance remaining analytes.
was not reached (GR: p  0.24 and p  0.076, Comparing the synovial fluid concentra-
MR: p  0.11 and p  0.310). MCP1 was tions of SHAM joints with nonoperated hind

Fig. 4.2 Mean synovial fluid monocyte chemoat- Fig. 4.3 Mean synovial fluid interleukin-8 concentra-
tractant protein 1 concentrations (pg/mL) for surgery tions (pg/mL) for surgery model dogs at initiation and
model dogs at initiation and conclusion of the study. conclusion of the study. Bars represent standard error
48 Bars represent standard error of the mean. of the mean.

Stannard_9781604068580_Ch04.indd 48 1/30/13 2:26 PM

Fig. 4.4 Mean synovial fluid keratinocyte-derived Fig. 4.5 Mean synovial fluid cytokine, chemokine,

4 Animal Models in Osteoarthritis Biomarker Research

chemoattractant (KC) concentrations (pg/mL) for and MMP concentrations (pg/mL) for spontaneous
surgery model dogs at initiation and conclusion of the OA dogs at surgery and 8 to 12 weeks after surgical
study. Bars represent standard error of the mean. stabilization compared with normal dogs. Bars repre-
sent standard error of the mean. MMP, matrix metal-
loprotease; OA, osteoarthritis.
limbs and operated hind limbs revealed sig-
nificant concentration overlap (Table 4.2). samples (p 0.001). There was not a statis-
However, IL-8 had the highest SN and SP with tically significant difference in IL-8 between
an SN of 100% (95% CI, 76.8 to 100) and an the baseline and 4-week samples, nor
SP of 80% (95% CI, 28.4 to 99.5) at a concen- was there a difference between the 8- and
tration cut-off between 144 and 189 pg/mL. 12-week samples. Urine MCP1 was increased
MCP1 and KC were less sensitive at 78% in the 8-week postoperative SHAM group
(95% CI, 49.2 to 95.3), but MCP1 was the most compared with baseline (p  0.009). No
specific at 100% (95% CI, 47.8 to 100). other statistically significant changes were
detected in the serum or urine samples.
Part 1: Serum and Urine
Part 2: Synovial Fluid
IL-8 (data not shown) was significantly lower
in the serum of ACL-T dogs 12 weeks after MCP1 was significantly higher in the Pre-sx
surgery compared with those at baseline OA dogs and Post-sx OA dogs compared with
(p  0.049). IL-8 was significantly increased normal dogs (p 0.001, p  0.009), and there
in the 8- and 12-week postoperative urine was a significant decrease in MCP1 follow-
samples from each of the four treatment ing surgery compared with presurgery val-
groups compared with day 0 and the 4-week ues (p  0.016) (Fig. 4.5). IL-8 and KC were

Table 4.2 Median (range) synovial fluid concentrations from model dogs including the optimal concentration
cut-offs and associated SN and SP percentages (with 95% CI)

SHAM Non-op Op Cut-off b SN % SP %

Marker (n 5) (n 15a) (n 15a) (pg/mL) (95% CI) (95% CI)

IL-8 119 202 781 144189 100 80

(731,197) (49753) (1892,825) (76.8100.0) (28.499.5)
MCP1 506 278 879 620 78 100
(345619) (55933) (1681,981) (49.295.3) (47.8100)
KC 4209 3137 4041 3000 78 20
(2,9955,396) (2736,361) (1,6976,624) (49.295.3) (0.571.6)
One individual did not have sufficient postoperative synovial fluid volume and is therefore not included in the
non-op or op analyses.
Values below cut-off concentration denote SHAM dogs.
Abbreviations: SN, sensitivity; SP, specificity; CI, confidence interval; SHAM, values for time zero; non-op,
nonoperated hind limbs at 12 weeks after surgery; Op, operated hind limbs at 12 weeks after surgery; IL-8,
interleukin-8; MCP1, monocyte chemoattractant protein 1; KC, keratinocyte-derived chemoattractant. 49

Stannard_9781604068580_Ch04.indd 49 1/30/13 2:26 PM

Table 4.3 Median (range) synovial fluid concentrations including the optimal concentration cut-offs and
associated SN and SP percentages (with 95% CI) from clinical OA patients and normal dogs

Biomarker NL Pre-sx OA Cut-offa (pg/mL) SN % (95% CI) SP % (95% CI)

IL-8 0 (0) 438 (1124,004) 0112 100 (69100) 100 (59100)

MCP1 73 (091) 734 (2632,278) 91.5263.3 100 (69100) 100 (59100)
KC 201 (141377) 584 (1818,603) 275 90 (56100) 86 (42100)
MMP2 3.95 4.34 3.72b 80 (44.497.4) 57 (18.490.1)
(2.245.67)b (2.766.66)b
MMP3 2569 2150 2,451 40 (12.173.7) 42.8
(1,9502,700) (1,0975,081) (9.981.6)

Abbreviations: SN, sensitivity; SP, specificity; CI, confidence interval; OA, osteoarthritis; NL, normal dogs;
Pre-sx OA, dogs presenting for surgical stabilization of knee osteoarthritis before surgery; IL-8, interleukin-8;
MCP1, monocyte chemoattractant protein 1; KC, keratinocyte-derived chemoattractant; MMP2, matrix met-
alloprotease 2; MMP3, matrix metalloprotease 3.
Values below cut-off concentration denote normal dogs.
( 104).
I Diagnosis and Treatment Planning

significantly higher in the Pre-sx OA dogs Part 2: Serum

compared with normal dogs (p 0.001, p 
0.014) and in the Post-sx OA dogs compared MMP2 and MMP3 were highest in the
with normal dogs (p  0.002, p  0.029). Both normal dogs and lowest in the Pre-sx OA
analytes were numerically different (lower) dogs (Fig. 4.6). MMP2 was significantly
in Post-sx OA dogs compared with Pre-sx OA higher in normal dogs and Post-sx OA dogs
dogs (p  0.340, 0.115). IL-18 (INF- inducing compared with Pre-sx OA dogs (p  0.005,
factor) was significantly higher in the Pre-sx p  0.027), but there was not a significant dif-
OA dogs compared with Post-sx OA dogs ference between normal and Post-sx OA dogs
(p  0.016) and with normal dogs (p  0.002), (p  0.270). MMP3 was significantly higher in
but the remaining cytokines and chemokines normal dogs and Post-sx OA dogs compared
were below the limit of detection. MMP2 was with Pre-sx OA dogs (p  0.002, p  0.044),
highest in the Post-sx OA dogs, and this was and MMP3 was significantly higher in nor-
significantly higher than in the normal dogs mal dogs compared with Post-sx OA dogs
(p  0.010), whereas no significant difference (p  0.025). SN and SP with associated cut-
was found between the Pre-sx OA dogs and off values for MMP2 and MMP3 are shown
the normal dogs. MMP3 was highest in the
normal dogs and declined in the OA dogs after
surgery, but these changes did not reach sta-
tistical significance. MMP9 and MMP13 were
below the limit of detection of the assay.
Synovial fluid concentration data were
compared between normal and Pre-sx OA
dogs (Table 4.3). Concentration ranges did
not overlap for IL-8 and MCP1, but they did
for KC, MMP2, and MMP3. Optimal concen-
tration cut-offs were selected as previously
described, and at these particular cut-offs
IL-8 and MCP1 performed well individually
(SN and SP of 100% [SN 95% CI, 69 to 100; SP Fig. 4.6 Mean serum MMP2 and MMP3 concentra-
95% CI, 59 to 100]). KC also performed rea- tions (pg/mL) for spontaneous OA dogs at surgery
and 8 to 12 weeks after surgical stabilization com-
sonably well with an SN of 90% (95% CI, 56 to pared with normal dogs. Bars represent standard
100) and an SP of 86% (95% CI, 42 to 100). The error of the mean. MMP2, matrix metalloprotease 2;
50 MMPs were less strong differentiators. MMP3, matrix metalloprotease 3, OA, osteoarthritis.

Stannard_9781604068580_Ch04.indd 50 1/30/13 2:26 PM

Table 4.4 Median (range) serum concentrations including the optimal concentration cut-offs and associated
SN and SP percentages (with 95% CI) from clinical OA patients and normal dogs

Biomarker NL Pre-sx OA Cut-off a (pg/mL) SN % (95% CI) SP % (95% CI)

MMP2 2,174.4 1,684.9 2,000 100 56

(1,7713,002) (1,2811,945) (69.2100) (21.286.3)
MMP3 413.8 256.7 400 80 67
(2,93.5676) (93.5421.3) (44.497.5) (29.992.5)

Abbreviations: SN, sensitivity; SP, specificity; CI, confidence interval; OA, osteoarthritis; NL, normal dogs;
Pre-sx OA, dogs presenting for surgical stabilization of knee osteoarthritis before surgery; MMP2, matrix
metalloprotease 2; MMP3, matrix metalloprotease 3.

4 Animal Models in Osteoarthritis Biomarker Research

Values above cut-off concentration denote normal dogs.

in Table 4.4. Significant differences were not IL-8 was a strong discriminator between
detected between groups for serum IL-8, KC, normal and Pre-sx OA dogs, but the AUC for
MCP1, IL-18, IL-2, IL-7, or GM-CSF, and the MCP1 was slightly higher. Combining MMP2
remaining analytes were below the limit of and MMP3 slightly improved their individual
detection for the assay. performance, but their AUC remained lower
than MCP1, IL-8, and KC.
ROC Analysis

For the surgery model dogs (data not shown), Discussion

calculation of the AUC confirmed IL-8 was
the single marker with the strongest ability We have shown that changes in cytokine
to differentiate between SHAM and all OA and chemokine concentrations occur within
dogs. The combination of MCP1, IL-8, and KC canine synovial fluid following surgical
demonstrated similarly strong discrimina- destabilization of the knee joint and may be
tory ability, but the addition of MCP1 and KC useful for differentiating osteoarthritic ver-
did not greatly improve the performance of sus normal knees. Furthermore, we have con-
IL-8 alone. ROC curve analysis was also per- firmed changes in the same three cytokines
formed on the hospital patient data (Fig. 4.7). and chemokines occur in synovial fluid from
dogs with naturally occurring OA, and the
elevations in these markers decline once sur-
gical stabilization has occurred. These find-
ings suggest the incorporation of chemokines
into a diagnostic or even treatment efficacy
biomarker panel may prove useful, especially
given their apparent early involvement in
clinical OA.91,92
One of the significantly altered chemokines,
MCP1; also known as chemokine (C-C motif)
ligand 2, is responsible for selectively attract-
ing mononuclear cells such as monocytes
and memory T cells, but not neutrophils.9599
Arthropathies associated with monocytic
Fig. 4.7 ROC curve of MCP1, IL-8, KC, and MMP2 infiltrates such as rheumatoid arthritis have
and MMP3 from dogs with (Pre-sx OA) and without been historically linked to alterations in
(normal dog) spontaneously occurring OA. IL-8, inter- MCP1.17,100,101 MCP1 is elevated in synovial
leukin 8, KC, keratinocyte-derived chemoattractant; fluid and serum from RA sufferers, and the
MCP1, monocyte chemoattractant protein 1; MMP2,
matrix metalloprotease 2; MMP3, matrix metallopro-
synovial tissue macrophages are the domi-
tease 3; OA, osteoarthritis; ROC, receiver operating nant source of this cytokine.100 Evidence sug-
characteristic. gests MCP1 expression may be altered in OA 51

Stannard_9781604068580_Ch04.indd 51 1/30/13 2:26 PM

as well, although reports in the veterinary While MCP1 may be useful as a diagnostic
literature are far less numerous than those biomarker, it may also be helpful in evaluat-
in the human literature. Human OA articu- ing treatment efficacy. For example, synovial
lar chondrocytes express higher levels of fluid MCP1 was significantly lower in the
MCP1 mRNA than normal chondrocytes, and Post-sx OA dogs compared with their Pre-sx
OA synoviocytes also produce MCP1.91,102 OA values, whereas IL-8 and KC did not signif-
Stimulation with IL-1, a cytokine used in icantly decline after treatment. Furthermore,
in vitro models of OA, greatly increases the two client-owned dogs that developed minor
expression of MCP1.12,91 MCP1 can also subse- postoperative complications had the highest
quently augment MMP3 expression, inhibit MCP1 values of all the postoperative dogs.
proteoglycan synthesis, and enhance proteo- These findings suggest MCP1 has potential
glycan release from chondrocytes in vitro.91 for clinical use in both diagnostic and treat-
A canine study confirms that MCP1 serum ment monitoring applications.
levels are significantly higher in critically Neutrophil chemoattractants such as IL-8,
ill dogs compared with healthy dogs and also known as chemokine (C-X-C motif)
dogs recently undergoing surgery, includ- ligand 8 (CXCL8), and KC or CXCL115,104106 are
ing orthopedic procedures.103 As a difference both members of the CXC cytokine family,
was not noted between healthy dogs and and phylogenetic analysis has shown KC may
postoperative patients, it was concluded that be similar to growth regulated oncogene-
I Diagnosis and Treatment Planning

surgery alone would not significantly alter alpha (GRO

) found in humans.107 IL-8 and
serum levels of MCP1 or interfere with analy- KC have been evaluated in joint diseases,
sis in postoperative patients.103 Synovial fluid with the majority of literature focusing on
data were not available from those patients, humans. Specifically, IL-8 is increased in
but the lack of significant increase in our humans with rheumatoid arthritis, OA, and
SHAM dogs from baseline to 12 weeks sug- other arthritides,108,109 but in dogs it has been
gests that there is either a brief increase or linked with Lyme disease.110 The synovium
no significant change in synovial fluid MCP1 appears to play a significant role as synovial
following surgical manipulation of the joint. expression of IL-8 is high in humans with
MCP1 exhibits high SN and SP for the Lyme arthritis, and RA and OA synoviocytes
client-owned dogs compared with normal, constitutively express IL-8.102,111,112 This local-
and the concentration ranges did not overlap ization is helpful for the diagnostician as the
between these two groups. As a result, syn- degree of synovial expression appears to be
ovial fluid MCP1 was considered one of the reflected in the synovial fluid. For example,
best markers to distinguish between non-OA IL-8 mRNA expression is higher in canine
and OA dogs. When absolute synovial fluid OA knee synovial fluid than normal synovial
concentrations were analyzed, the client- fluid.90 Additionally, increased numbers of
owned OA dogs had very similar synovial functional receptors have been identified in
fluid MCP1 concentrations to the 12-week OA chondrocytes, and the ligandreceptor
postoperative model dogs. The most striking interactions in these tissues have been
difference in MCP1 concentrations between shown to induce matrix-degrading enzymes
the two segments of this study was the dis- such as MMP3.7 IL-8 and KC promote carti-
crepancy between the SHAM dogs at time lage hypertrophy, which can ultimately lead
zero and the control individuals. At baseline to dysregulated matrix repair and pathologic
the SHAM dogs had much higher synovial calcification in OA.113
fluid concentrations of MCP1 compared with Synovial fluid IL-8 performed well as an
both the nonoperated hind limbs 12 weeks individual marker of OA with high SN and
later and the normal dog group. At this time, SP in both portions of the study, and the IL-8
there is no clear explanation for this appar- concentration was increased in a majority of
ent elevation at the initiation of the surgery our surgery model dogs and spontaneous OA
model, but this may have contributed to dogs. Furthermore, the concentration ranges
the lack of significant differences between did not overlap between normal and Pre-sx
the SHAM and other surgery model groups OA dogs, and the optimal cut-off concentra-
52 12 weeks after surgery. tions were very similar between the model

Stannard_9781604068580_Ch04.indd 52 1/30/13 2:26 PM

dogs and the spontaneous OA dogs. This sug- was the only cytokine to show significant
gests the synovial fluid IL-8 cut-off concen- differences within the urine of OA dogs, but
tration identified here may be successfully since similar elevations were also found in
extrapolated to other populations to rule in the SHAM dogs the increase was not likely
or out OA. associated with OA.
Although synovial fluid KC was not as sen- In contrast to the cytokines and chemokines,
sitive or specific as IL-8 and MCP1 when used MMPs exhibited changes in the serum and
individually, it did promote an interesting the synovial fluid, although the serum con-
question. A significant decrease was noted centrations showed a clearer trend. Perhaps
in the GR model dogs at the time of sacrifice the lack of a clear trend in synovial fluid
compared with the initiation of the study, MMP2 and MMP3 was related to the variabil-

4 Animal Models in Osteoarthritis Biomarker Research

but there was no significant change between ity in the chronicity of disease of the hospital
baseline and the end point for ACL-T dogs. In patients. The concentrations of serum MMP2
contrast, the Pre-sx OA dogs had higher KC and MMP3 overlapped between the normal
synovial fluid concentrations than the client- and Pre-sx OA individuals, and the selected
owned dogs deemed free of disease. This cut-off concentrations had to sacrifice high
disagreement may suggest that the altera- SP to get a higher SN. However, serum MMP2
tion in KC is dependent on the type and/or and MMP3 were highest in the normal
chronicity of pathologic changes. All of the patients. Ling et al also showed serum MMP2
client-owned dogs had cruciate disease with was lower in individuals who subsequently
most similarities to those included in the developed OA and suggested altered ECM
ACL-T model group, but one limitation of the metabolism, including a decline in more con-
second half of the study was that the chronic- stitutively expressed MMPs such as MMP2,
ity of OA was variable or even unknown in played a role in the initiation and progression
the client-owned dogs. The inciting cause of of OA.117 Unfortunately, it cannot be ruled out
OA was not specifically selected for or against that serum MMP2 and MMP3 values in the
in the spontaneous OA dogs, but the result- control individuals in this study may have
ing cruciate-disease-heavy population was been elevated due to unknown concurrent
determined to be representative of sponta- processes not directly related to OA.
neous OA patients presenting to the hospital. To the authors knowledge, the combina-
Additional studies investigating dogs with tion of synovial fluid MCP1, IL8, and KC for
noncruciate disease knee instability may pro- use in an OA diagnostic biomarker panel is
vide insight as to whether or not KC may be unreported in dogs or humans. It has been
helpful in determining the tissue(s) affected. reported that human knee articular cartilage
Until that time, KC can be retained as a poten- produces MCP1, IL-8, and GRO
(the human
tial diagnostic biomarker of interest. counterpart to KC), but this production did
Unfortunately, the evaluation of cytokines not differ significantly between OA and other
and chemokines in the serum and urine was joint diseases.118 Further investigation is
less rewarding than in the synovial fluid. necessary to evaluate how reliably this par-
This is not entirely surprising since other ticular combination of markers differenti-
systemic conditions not related to OA can ates between types of arthritides in human
dramatically affect cytokine and chemokine and veterinary patients. Prospective studies
concentrations. Urinary IL-8 can increase to more closely follow the trends in these
in the presence of a urinary tract infection, cytokine, chemokine, and MMP fluctuations
likely associated with its involvement in the in earlier stages of OA are also warranted.
recruitment of neutrophils.114,115 However, This study has characterized the cytokine,
elevations in urinary IL-8 have been detected chemokine, and MMP changes that occur in
in individuals with inflammation of nonuri- dog models of OA and in dogs with spontane-
nary tract origin as well.116 Urinalyses were ously occurring OA. As many of the biomarker
not performed on the model dogs in this studies performed in dogs to date have focused
study to assess for urinary tract infection, solely on direct indicators of OA, the indirect
but such a condition must be considered as cytokine markers will be useful in further
a potential cause for the elevations seen. IL-8 understanding the mechanisms leading up 53

Stannard_9781604068580_Ch04.indd 53 1/30/13 2:26 PM

to the development of OA. More specifically, 4. Goldring MB. The role of cytokines as inflamma-
tory mediators in osteoarthritis: lessons from ani-
the evaluation of synovial fluid, serum, and mal models. Connect Tissue Res 1999;40(1):111
urine-derived chemokine, cytokine, and MMP 5. Scanzello CR, Plaas A, Crow MK. Innate immune
concentrations revealed potential additional system activation in osteoarthritis: is osteoar-
biomarker candidates for the early diagnosis thritis a chronic wound? Curr Opin Rheumatol
of OA in dogs. While synovial fluid IL-8 was 6. Borzi RM, Mazzetti I, Macor S, et al. Flow cyto-
the most sensitive marker for both portions of metric analysis of intracellular chemokines in
this study, the authors feel MCP1 and KC con- chondrocytes in vivo: constitutive expression
and enhancement in osteoarthritis and rheuma-
tribute additional information and should be toid arthritis. FEBS Lett 1999;455(3):238242
retained within the tentative biomarker panel 7. Borz RM, Mazzetti I, Cattini L, Uguccioni M,
until further investigations can be performed. Baggiolini M, Facchini A. Human chondrocytes
express functional chemokine receptors and
Although these markers were not as useful release matrix-degrading enzymes in response
in the serum, the authors suggest that use of to C-X-C and C-C chemokines. Arthritis Rheum
synovial fluid biomarkers has important clini- 2000;43(8):17341741
cal application based on the relative ease in 8. Fernandes JC, Martel-Pelletier J, Pelletier JP. The
role of cytokines in osteoarthritis pathophysiol-
obtaining samples, the associated costs, and ogy. Biorheology 2002;39(1-2):237246
the joint specific nature of these evaluations. 9. Hegemann N, Wondimu A, Kohn B, Brunnberg L,
MMP2 and MMP3 were not as sensitive or Schmidt MF. Cytokine profile in canine immune-
mediated polyarthritis and osteoarthritis. Vet
specific in the synovial fluid, but they may Comp Orthop Traumatol 2005;18(2):6772
I Diagnosis and Treatment Planning

prove useful in serum evaluations and should 10. Hulejov H, Baresov V, Klzl Z, Polansk M,
be studied further. Studies are currently Adam M, Senolt L. Increased level of cytokines
and matrix metalloproteinases in osteoarthritic
under way in our laboratory to determine the subchondral bone. Cytokine 2007;38(3):151156
specific tissue source(s) of these cytokines 11. Klatt AR, Paul-Klausch B, Klinger G, et al. A criti-
and to see if synovial fluid from other joints cal role for collagen II in cartilage matrix deg-
possesses a similar profile. radation: collagen II induces pro-inflammatory
cytokines and MMPs in primary human chon-
In conclusion, animal models are rou- drocytes. J Orthop Res 2009;27(1):6570
tinely used for the translational investigation 12. Sandell LJ, Xing X, Franz C, Davies S, Chang LW,
of human OA biomarkers. Each of the most Patra D. Exuberant expression of chemokine
genes by adult human articular chondrocytes
relevant models was mentioned here, and a in response to IL-1beta. Osteoarthritis Cartilage
brief list of their advantages and disadvan- 2008;16(12):15601571
tages was compiled. Dogs are often used in 13. Tetlow LC, Adlam DJ, Woolley DE. Matrix met-
alloproteinase and proinflammatory cytokine
studies such as the one described here, but production by chondrocytes of human osteoar-
species such as the guinea pig, the horse, thritic cartilage: associations with degenerative
and others have significantly contributed to changes. Arthritis Rheum 2001;44(3):585594
the level of OA research currently available. 14. Borden P, Solymar D, Sucharczuk A, Lindman B,
Cannon P, Heller RA. Cytokine control of intersti-
Given the various etiologies, risk factors, and tial collagenase and collagenase-3 gene expres-
presentations of OA, it is unlikely a single ani- sion in human chondrocytes. J Biol Chem 1996;
mal model will ever be sufficient to study all
15. Schrder JM, Persoon NL, Christophers E.
aspects of this disease. Lipopolysac-charide-stimulated human mono-
cytes secrete, apart from neutrophil-activating
peptide 1/interleukin 8, a second neutrophil-
References activating protein. NH2-terminal amino acid
sequence identity with melanoma growth stimu-
1. Helmick CG, Felson DT, Lawrence RC, et al; latory activity. J Exp Med 1990;171(4):10911100
National Arthritis Data Workgroup. Estimates of 16. Venn G, Nietfeld JJ, Duits AJ, et al. Elevated syn-
the prevalence of arthritis and other rheumatic ovial fluid levels of interleukin-6 and tumor
conditions in the United States. Part I. Arthritis necrosis factor associated with early experi-
Rheum 2008;58(1):1525 mental canine osteoarthritis. Arthritis Rheum
2. Ray A, Kuroki K, Cook JL, et al. Induction of matrix 1993;36(6):819826
metalloproteinase 1 gene expression is regu- 17. Villiger PM, Terkeltaub R, Lotz M. Production
lated by inflammation-responsive transcription of monocyte chemoattractant protein-1 by
factor SAF-1 in osteoarthritis. Arthritis Rheum inflamed synovial tissue and cultured synovio-
2003;48(1):134145 cytes. J Immunol 1992;149(2):722727
3. Kuroki K, Stoker AM, Cook JL. Effects of proin- 18. Tiku K, Thakker-Varia S, Ramachandrula A,
flammatory cytokines on canine articular chon- Tiku ML. Articular chondrocytes secrete IL-1,
drocytes in a three-dimensional culture. Am J Vet express membrane IL-1, and have IL-1 inhibitory
54 Res 2005;66(7):11871196 activity. Cell Immunol 1992;140(1):120

Stannard_9781604068580_Ch04.indd 54 1/30/13 2:26 PM

19. Aigner T, Cook JL, Gerwin N, et al. Histopathol- of triamcinolone acetonide on an in vivo equine
ogy atlas of animal model systemsoverview osteochondral fragment exercise model. Equine
of guiding principles. Osteoarthritis Cartilage Vet J 1997;29(5):349359
2010;18(Suppl 3):S2S6 35. Janusz MJ, Bendele AM, Brown KK, Taiwo YO,
20. Cook JL, Kuroki K, Visco D, Pelletier JP, Schulz L, Hsieh L, Heitmeyer SA. Induction of osteoarthri-
Lafeber FP. The OARSI histopathology initiative tis in the rat by surgical tear of the meniscus:
recommendations for histological assessments of Inhibition of joint damage by a matrix metal-
osteoarthritis in the dog. Osteoarthritis Cartilage loproteinase inhibitor. Osteoarthritis Cartilage
2010;18(Suppl 3):S66S79 2002;10(10):785791
21. Gerwin N, Bendele AM, Glasson S, 36. Moore EE, Bendele AM, Thompson DL, et al.
Carlson CS. The OARSI histopathology initiative Fibroblast growth factor-18 stimulates chon-
recommendations for histological assessments of drogenesis and cartilage repair in a rat model of
osteoarthritis in the rat. Osteoarthritis Cartilage injury-induced osteoarthritis. Osteoarthritis Car-
2010;18(Suppl 3):S24S34 tilage 2005;13(7):623631
37. De Gruttola VG, Clax P, DeMets DL, et al. Con-

4 Animal Models in Osteoarthritis Biomarker Research

22. Kraus VB, Huebner JL, DeGroot J, Bendele
A. The OARSI histopathology initiative siderations in the evaluation of surrogate end-
recommendations for histological assessments points in clinical trials: Summary of a National
of osteoarthritis in the guinea pig. Osteoarthritis Institutes of Health workshop. Control Clin Trials
Cartilage 2010;18(Suppl 3):S35S52 2001;22(5):485502
23. Laverty S, Girard CA, Williams JM, Hunziker EB, 38. Bauer DC, Hunter DJ, Abramson SB, et al; Osteo-
Pritzker KP. The OARSI histopathology initiative arthritis Biomarkers Network. Classification of
recommendations for histological assessments of osteoarthritis biomarkers: a proposed approach.
osteoarthritis in the rabbit. Osteoarthritis Carti- Osteoarthritis Cartilage 2006;14(8):723727
lage 2010;18(Suppl 3):S53S65 39. Kraus VB. Osteoarthritis year 2010 in review:
24. McIlwraith CW, Frisbie DD, Kawcak CE, Fuller biochemical markers. Osteoarthritis Cartilage
CJ, Hurtig M, Cruz A. The OARSI histopathology 2011;19(4):346353
initiativerecommendations for histological 40. McIlwraith CW. Use of synovial fluid and serum
assessments of osteoarthritis in the horse. Osteo- biomarkers in equine bone and joint disease: a
arthritis Cartilage 2010;18(Suppl 3):S93S105 review. Equine Vet J 2005;37(5):473482
25. Poole R, Blake S, Buschmann M, et al. Recommen- 41. Thonar EJ. Molecular markers of metabolic
dations for the use of preclinical models in the changes in osteoarthritis. Osteoarthritis Cartilage
study and treatment of osteoarthritis. Osteoar- 1999;7(3):338 339
thritis Cartilage 2010;18(Suppl 3):S10S16 42. Matyas JR, Atley L, Ionescu M, Eyre DR, Poole
26. Glasson SS, Chambers MG, Van Den Berg WB, AR. Analysis of cartilage biomarkers in the early
Little CB. The OARSI histopathology initiative phases of canine experimental osteoarthritis.
recommendations for histological assessments of Arthritis Rheum 2004;50(2):543552
osteoarthritis in the mouse. Osteoarthritis Carti- 43. Cawston T, Billington C, Cleaver C, et al. The regu-
lage 2010;18(Suppl 3):S17S23 lation of MMPs and TIMPs in cartilage turnover.
27. Little CB, Smith MM, Cake MA, Read RA, Murphy MJ, Ann N Y Acad Sci 1999;878:120129
Barry FP. The OARSI histopathology initiative 44. Bonassar LJ, Frank EH, Murray JC, et al. Changes
recommendations for histological assessments of in cartilage composition and physical properties
osteoarthritis in sheep and goats. Osteoarthritis due to stromelysin degradation. Arthritis Rheum
Cartilage 2010;18(Suppl 3):S80S92 1995;38(2):173183
28. Johnston SA. Osteoarthritis. Joint anatomy, physi- 45. Lohmander LS, Saxne T, Heinegrd DK. Release of
ology, and pathobiology. Vet Clin North Am Small cartilage oligomeric matrix protein (COMP) into
Anim Pract 1997;27(4):699723 joint fluid after knee injury and in osteoarthritis.
29. Pond MJ, Nuki G. Experimentally-induced osteo- Ann Rheum Dis 1994;53(1):813
arthritis in the dog. Ann Rheum Dis 1973;32(4): 46. Misumi K, Vilim V, Carter SD, Ichihashi K,
387388 Oka T, Sakamoto H. Concentrations of cartilage
30. Luther JK, Cook CR, Cook JL. Meniscal release oligomeric matrix protein in dogs with naturally
in cruciate ligament intact stifles causes lame- developing and experimentally induced arthrop-
ness and medial compartment cartilage pathol- athy. Am J Vet Res 2002;63(4):598603
ogy in dogs 12 weeks postoperatively. Vet Surg 47. Barksby HE, Milner JM, Patterson AM, et al.
2009;38(4):520529 Matrix metalloproteinase 10 promotion of
31. Bendele AM, Hulman JF. Spontaneous cartilage collagenolysis via procollagenase activation:
degeneration in guinea pigs. Arthritis Rheum implications for cartilage degradation in arthri-
1988;31(4):561565 tis. Arthritis Rheum 2006;54(10):32443253
32. Frisbie DD, Al-Sobayil F, Billinghurst RC, Kawcak CE, 48. Billinghurst RC, Dahlberg L, Ionescu M, et al.
McIlwraith CW. Changes in synovial fluid and Enhanced cleavage of type II collagen by colla-
serum biomarkers with exercise and early oste- genases in osteoarthritic articular cartilage. J Clin
oarthritis in horses. Osteoarthritis Cartilage Invest 1997;99(7):15341545
2008;16(10):11961204 49. Huebner JL, Otterness IG, Freund EM, Caterson B,
33. Frisbie DD, Kawcak CE, McIlwraith CW. Evalua- Kraus VB. Collagenase 1 and collagenase 3 expres-
tion of the effect of extracorporeal shock wave sion in a guinea pig model of osteoarthritis.
treatment on experimentally induced osteoar- Arthritis Rheum 1998;41(5):877890
thritis in middle carpal joints of horses. Am J Vet 50. Shlopov BV, Lie WR, Mainardi CL, Cole AA,
Res 2009;70(4):449454 Chubinskaya S, Hasty KA. Osteoarthritic lesions:
34. Frisbie DD, Kawcak CE, Trotter GW, involvement of three different collagenases.
Powers BE, Walton RM, McIlwraith CW. Effects Arthritis Rheum 1997;40(11):20652074

Stannard_9781604068580_Ch04.indd 55 1/30/13 2:26 PM

51. Conrozier T, Poole AR, Ferrand F, et al. Serum con- 65. Billinghurst RC, Buxton EM, Edwards MG,
centrations of type II collagen biomarkers (C2C, McGraw MS, McIlwraith CW. Use of an anti-
C1, 2C and CPII) suggest different pathophysiolo- neoepitope antibody for identification of type-II
gies in patients with hip osteoarthritis. Clin Exp collagen degradation in equine articular carti-
Rheumatol 2008;26(3):430435 lage. Am J Vet Res 2001;62(7):10311039
52. Arican M, Carter SD, Bennett D, May C. Meas- 66. van Meurs J, van Lent P, Stoop R, et al. Cleavage
urement of glycosaminoglycans and keratan of aggrecan at the Asn341-Phe342 site coincides
sulphate in canine arthropathies. Res Vet Sci with the initiation of collagen damage in murine
1994;56(3):290297 antigen-induced arthritis: a pivotal role for
53. Campion GV, McCrae F, Schnitzer TJ, Lenz ME, stromelysin 1 in matrix metalloproteinase activ-
Dieppe PA, Thonar EJ. Levels of keratan sul- ity. Arthritis Rheum 1999;42(10):20742084
fate in the serum and synovial fluid of patients 67. Freemont AJ, Hampson V, Tilman R, Goupille P,
with osteoarthritis of the knee. Arthritis Rheum Taiwo Y, Hoyland JA. Gene expression of matrix
1991;34(10):12541259 metalloproteinases 1, 3, and 9 by chondrocytes
54. Sondergaard BC, Schultz N, Madsen SH, Bay- in osteoarthritic human knee articular carti-
Jensen AC, Kassem M, Karsdal MA. MAPKs are lage is zone and grade specific. Ann Rheum Dis
essential upstream signaling pathways in proteo- 1997;56(9):542549
lytic cartilage degradation: divergence in path- 68. Karran EH, Young TJ, Markwell RE, Harper GP. In
ways leading to aggrecanase and MMP-mediated vivo model of cartilage degradationeffects of a
articular cartilage degradation. Osteoarthritis matrix metalloproteinase inhibitor. Ann Rheum
Cartilage 2010;18(3):279288 Dis 1995;54(8):662669
55. Poole AR, Ionescu M, Swan A, Dieppe PA. Changes 69. Mehraban F, Kuo SY, Riera H, Chang C, Moskowitz RW.
in cartilage metabolism in arthritis are reflected Prostromelysin and procollagenase genes are dif-
by altered serum and synovial fluid levels of the ferentially up-regulated in chondrocytes from
cartilage proteoglycan aggrecan. Implications for the knees of rabbits with experimental osteoar-
I Diagnosis and Treatment Planning

pathogenesis. J Clin Invest 1994;94(1):2533 thritis. Arthritis Rheum 1994;37(8):11891197

56. Chaganti RK, Kelman A, Lui L, et al; Study Of 70. Panula HE, Lohmander LS, Rnkk S, Agren U,
Osteoporotic Fractures Research Group (SOF). Helminen HJ, Kiviranta I. Elevated levels of syno-
Change in serum measurements of cartilage vial fluid PLA2, stromelysin (MMP-3) and TIMP
oligomeric matrix protein and association with in early osteoarthrosis after tibial valgus oste-
the development and worsening of radiographic otomy in young beagle dogs. Acta Orthop Scand
hip osteoarthritis. Osteoarthritis Cartilage 2008; 1998;69(2):152158
16(5):566571 71. Cook JL, Anderson CC, Kreeger JM, Tomlinson JL.
57. Hunter DJ, Li J, LaValley M, et al. Cartilage mark- Effects of human recombinant interleukin-1beta
ers and their association with cartilage loss on on canine articular chondrocytes in three-
magnetic resonance imaging in knee osteoarthri- dimensional culture. Am J Vet Res 2000;61(7):
tis: the Boston Osteoarthritis Knee Study. Arthri- 766770
tis Res Ther 2007;9(5):R108 72. Ratcliffe A, Beauvais PJ, Saed-Nejad F, Shurety W,
58. Bruyere O, Collette J, Kothari M, et al. Osteoarthri- Caterson B. Synovial fluid analyses detect and
tis, magnetic resonance imaging, and biochemi- differentiate proteoglycan metabolism in canine
cal markers: a one year prospective study. Ann experimental models of osteoarthritis and disuse
Rheum Dis 2006;65(8):10501054 atrophy. Agents Actions Suppl 1993;39:6367
59. King KB, Lindsey CT, Dunn TC, Ries MD, Stein- 73. Lindhorst E, Vail TP, Guilak F, et al. Longitudinal
bach LS, Majumdar S. A study of the relationship characterization of synovial fluid biomarkers in
between molecular biomarkers of joint degenera- the canine meniscectomy model of osteoarthri-
tion and the magnetic resonance-measured char- tis. J Orthop Res 2000;18(2):269280
acteristics of cartilage in 16 symptomatic knees. 74. Budsberg SC, Lenz ME, Thonar EJ. Serum and
Magn Reson Imaging 2004;22(8):11171123 synovial fluid concentrations of keratan sulfate
60. Qi C, Changlin H. Levels of biomarkers corre- and hyaluronan in dogs with induced stifle joint
late with magnetic resonance imaging progres- osteoarthritis following cranial cruciate ligament
sion of knee cartilage degeneration: a study on transection. Am J Vet Res 2006;67(3):429432
canine. Knee Surg Sports Traumatol Arthrosc 75. Innes JF, Sharif M, Barr AR. Relations between
2007;15(7):869878 biochemical markers of osteoarthritis and other
61. Schmidt-Rohlfing B, Thomsen M, Niedhart C, disease parameters in a population of dogs with
Wirtz DC, Schneider U. Correlation of bone and naturally acquired osteoarthritis of the genual
cartilage markers in the synovial fluid with the joint. Am J Vet Res 1998;59(12):15301536
degree of osteoarthritis. Rheumatol Int 2002; 76. Mller G, Michel A, Altenburg E. COMP (cartilage
21(5):193199 oligomeric matrix protein) is synthesized in liga-
62. Wong PK, Young L, Vaile JH, et al. Telopep- ment, tendon, meniscus, and articular cartilage.
tides as markers of bone turnover in rheuma- Connect Tissue Res 1998;39(4):233244
toid arthritis and osteoarthritis. Intern Med J 77. Arai K, Tagami M, Hatazoe T, et al. Analysis of
2004;34(9-10):539544 cartilage oligomeric matrix protein (COMP) in
63. Bruyere O, Collette JH, Ethgen O, et al. Biochemi- synovial fluid, serum and urine from 51 race-
cal markers of bone and cartilage remodeling in horses with carpal bone fracture. J Vet Med Sci
prediction of longterm progression of knee oste- 2008;70(9):915921
oarthritis. J Rheumatol 2003;30(5):10431050 78. Misumi K, Vilim V, Hatazoe T, et al. Serum
64. Arican M, Carter SD, Bennett D. Osteocalcin level of cartilage oligomeric matrix protein
in canine joint diseases. Br Vet J 1996;152(4): (COMP) in equine osteoarthritis. Equine Vet J
56 411423 2002;34(6):602608

Stannard_9781604068580_Ch04.indd 56 1/30/13 2:26 PM

79. Chu Q, Lopez M, Hayashi K, et al. Elevation of a 94. Breshears LA, Cook JL, Stoker AM, Fox DB, Luther
collagenase generated type II collagen neoepitope JK. The Effect of Uniaxial Cyclic Tensile Load on
and proteoglycan epitopes in synovial fluid fol- Gene Expression in Canine Cranial Cruciate Liga-
lowing induction of joint instability in the dog. mentocytes. Vet Surg 2010;39(4):433443
Osteoarthritis Cartilage 2002;10(8):662669 95. Yoshimura T, Robinson EA, Tanaka S, Appella E,
80. Goranov NV. Serum markers of lipid peroxida- Kuratsu J, Leonard EJ. Purification and amino
tion, antioxidant enzymatic defense, and colla- acid analysis of two human glioma-derived
gen degradation in an experimental (Pond-Nuki) monocyte chemoattractants. J Exp Med
canine model of osteoarthritis. Vet Clin Pathol 1989;169(4):14491459
2007;36(2):192195 96. Matsushima K, Larsen CG, DuBois GC, Oppen-
81. Hayashi K, Kim SY, Lansdowne JL, Kapatkin A, heim JJ. Purification and characterization of a
Djardin LM. Evaluation of a collagenase gen- novel monocyte chemotactic and activating fac-
erated osteoarthritis biomarker in naturally tor produced by a human myelomonocytic cell
occurring canine cruciate disease. Vet Surg line. J Exp Med 1989;169(4):14851490
2009;38(1):117121 97. Yoshimura T, Robinson EA, Tanaka S, Appella E,

4 Animal Models in Osteoarthritis Biomarker Research

82. Carney SL, Billingham ME, Caterson B, et al. Leonard EJ. Purification and amino acid analy-
Changes in proteoglycan turnover in experi- sis of two human monocyte chemoattractants
mental canine osteoarthritic cartilage. Matrix produced by phytohemagglutinin-stimulated
1992;12(2):137147 human blood mononuclear leukocytes. J Immu-
83. Caterson B, Mahmoodian F, Sorrell JM, et al. Mod- nol 1989;142(6):19561962
ulation of native chondroitin sulphate structure 98. Carr MW, Roth SJ, Luther E, Rose SS, Springer TA.
in tissue development and in disease. J Cell Sci Monocyte chemoattractant protein 1 acts as a
1990;97(Pt 3):411417 T-lymphocyte chemoattractant. Proc Natl Acad
84. Sorrell JM, Mahmoodian F, Schafer IA, Davis B, Sci U S A 1994;91(9):36523656
Caterson B. Identification of monoclonal anti- 99. Taub DD, Proost P, Murphy WJ, et al. Monocyte
bodies that recognize novel epitopes in native chemotactic protein-1 (MCP-1), -2, and -3 are
chondroitin/dermatan sulfate glycosaminoglycan chemotactic for human T lymphocytes. J Clin
chains: their use in mapping functionally distinct Invest 1995;95(3):13701376
domains of human skin. J Histochem Cytochem 100. Koch AE, Kunkel SL, Harlow LA, et al. Enhanced
1990;38(3):393402 production of monocyte chemoattractant pro-
85. Ratcliffe A, Shurety W, Caterson B. The quanti- tein-1 in rheumatoid arthritis. J Clin Invest
tation of a native chondroitin sulfate epitope in 1992;90(3):772779
synovial fluid lavages and articular cartilage from 101. Stankovic A, Slavic V, Stamenkovic B, Kamenov
canine experimental osteoarthritis and disuse B, Bojanovic M, Mitrovic DR. Serum and synovial
atrophy. Arthritis Rheum 1993;36(4):543551 fluid concentrations of CCL2 (MCP-1) chemokine
86. Visco DM, Johnstone B, Hill MA, Jolly GA, Caterson in patients suffering rheumatoid arthritis and
B. Immunohistochemical analysis of 3-B-(-) and osteoarthritis reflect disease activity. Bratisl Lek
7-D-4 epitope expression in canine osteoarthri- Listy (Tlacene Vyd) 2009;110(10):641646
tis. Arthritis Rheum 1993;36(12):17181725 102. Seitz M, Loetscher P, Dewald B, Towbin H, Ceska
87. Johnson KA, Hay CW, Chu Q, Roe SC, Caterson B. M, Baggiolini M. Production of interleukin-1
Cartilage-derived biomarkers of osteoarthritis receptor antagonist, inflammatory chemotac-
in synovial fluid of dogs with naturally acquired tic proteins, and prostaglandin E by rheumatoid
rupture of the cranial cruciate ligament. Am J Vet and osteoarthritic synoviocytesregulation by
Res 2002;63(6):775781 IFN-gamma and IL-4. J Immunol 1994;152(4):
88. Nganvongpanit K, Itthiarbha A, Ong-Chai S, 20602065
Kongtawelert P. Evaluation of serum chon- 103. Duffy AL, Olea-Popelka FJ, Eucher J, Rice DM, Dow
droitin sulfate and hyaluronan: biomarkers for SW. Serum concentrations of monocyte chem-
osteoarthritis in canine hip dysplasia. J Vet Sci oattractant protein-1 in healthy and critically ill
2008;9(3):317325 dogs. Vet Clin Pathol 2010;39(3):302305
89. Hay CW, Chu Q, Budsberg SC, Clayton MK, Johnson 104. Moser B, Clark-Lewis I, Zwahlen R, Baggiolini M.
KA. Synovial fluid interleukin 6, tumor necrosis Neutrophil-activating properties of the mela-
factor, and nitric oxide values in dogs with osteo- noma growth-stimulatory activity. J Exp Med
arthritis secondary to cranial cruciate ligament 1990;171(5):17971802
rupture. Am J Vet Res 1997;58(9):10271032 105. Lin M, Carlson E, Diaconu E, Pearlman E. CXCL1/KC
90. de Bruin T, de Rooster H, van Bree H, Cox E. Inter- and CXCL5/LIX are selectively produced by cor-
leukin-8 mRNA expression in synovial fluid of neal fibroblasts and mediate neutrophil infil-
canine stifle joints with osteoarthritis. Vet Immu- tration to the corneal stroma in LPS keratitis. J
nol Immunopathol 2005;108(3-4):387397 Leukoc Biol 2007;81(3):786792
91. Yuan GH, Masuko-Hongo K, Sakata M, et al. The role 106. Fujiwara K, Ohkawara S, Takagi K, Yoshinaga M,
of C-C chemokines and their receptors in osteoar- Matsukawa A. Involvement of CXC chemokine
thritis. Arthritis Rheum 2001;44(5):10561070 growth-related oncogene-alpha in monosodium
92. Rollins BJ. Monocyte chemoattractant protein 1: urate crystal-induced arthritis in rabbits. Lab
a potential regulator of monocyte recruitment in Invest 2002;82(10):12971304
inflammatory disease. Mol Med Today 1996;2(5): 107. Modi WS, Yoshimura T. Isolation of novel GRO
198204 genes and a phylogenetic analysis of the CXC
93. Mahn MM, Cook JL, Cook CR, Balke MT. Arthro- chemokine subfamily in mammals. Mol Biol Evol
scopic verification of ultrasonographic diag- 1999;16(2):180193
nosis of meniscal pathology in dogs. Vet Surg 108. Remick DG, DeForge LE, Sullivan JF, Showell HJ.
2005;34(4):318323 Profile of cytokines in synovial fluid specimens 57

Stannard_9781604068580_Ch04.indd 57 1/30/13 2:26 PM

from patients with arthritis. Interleukin 8 (IL-8) on Lyme borreliosis in dogs. Wien Klin Wochen-
and IL-6 correlate with inflammatory arthritides. schr 1998;110(24):874881
Immunol Invest 1992;21(4):321327 113. Merz D, Liu R, Johnson K, Terkeltaub R. IL-8/
109. Brennan FM, Zachariae CO, Chantry D, et al. CXCL8 and growth-related oncogene alpha/
Detection of interleukin 8 biological activity CXCL1 induce chondrocyte hypertrophic differ-
in synovial fluids from patients with rheuma- entiation. J Immunol 2003;171(8):44064415
toid arthritis and production of interleukin 8 114. Ko YC, Mukaida N, Ishiyama S, et al. Elevated
mRNA by isolated synovial cells. Eur J Immunol interleukin-8 levels in the urine of patients
1990;20(9):21412144 with urinary tract infections. Infect Immun
110. Straubinger RK, Straubinger AF, Hrter L, et al. 1993;61(4):13071314
Borrelia burgdorferi migrates into joint capsules 115. Zaki MelS. Interleukin 8 is a surrogate marker for
and causes an up-regulation of interleukin-8 rapid diagnosis of bacteriuria. Immunol Invest
in synovial membranes of dogs experimen- 2008;37(7):694703
tally infected with ticks. Infect Immun 1997; 116. Taha AS, Grant V, Kelly RW. Urinalysis for
65(4):12731285 interleukin-8 in the non-invasive diagnosis of
111. Straubinger RK, Straubinger AF, Summers BA, acute and chronic inflammatory diseases. Post-
Erb HN, Hrter L, Appel MJ. Borrelia burgdorferi grad Med J 2003;79(929):159163
induces the production and release of proinflam- 117. Ling SM, Patel DD, Garnero P, et al. Serum protein
matory cytokines in canine synovial explant cul- signatures detect early radiographic osteoarthri-
tures. Infect Immun 1998;66(1):247258 tis. Osteoarthritis Cartilage 2009;17(1):4348
112. Straubinger RK, Straubinger AF, Summers BA, 118. Pulsatelli L, Dolzani P, Piacentini A, et al.
Jacobson RH, Erb HN. Clinical manifestations, Chemokine production by human chondrocytes.
pathogenesis, and effect of antibiotic treatment J Rheumatol 1999;26(9):19922001
I Diagnosis and Treatment Planning


Stannard_9781604068580_Ch04.indd 58 1/30/13 2:26 PM

Science and Techniques for
Cartilage Repair

Stannard_9781604068580_Ch05.indd 59 1/30/13 2:35 PM

Stannard_9781604068580_Ch05.indd 60 1/30/13 2:35 PM
The CartilageBone Interface
Caroline D. Hoemann, Charles-Hubert Lafantaisie-Favreau,
Viorica Lascau-Coman, Gaoping Chen, and Jessica Guzmn-Morales

All CartilageBone Interfaces in the immature rabbit knee (Fig. 5.1ac).

Articular cartilage sits on top of the epiphy-
Are Derived from an Initially seal subchondral bone (Fig. 5.1b, e, h). The
Pure Cartilage Structure growth plate is a cartilage structure firmly
sandwiched between two layers of bone:
Long bones develop first from embryonic the epiphyseal and the metaphyseal bone
mesenchymal stem cells that coalesce to form (Fig. 5.1c, f, i).
a blastema, with a scant but uniform type I
collagen matrix.1 The blastema transforms at
early fetal stages into a cartilaginous struc-
ture, or cartilage anlage, with collagen type
Growth Plate CartilageBone
II as the main extracellular component.26 The Interface during Postnatal
cartilage anlage contains a mixture of fusi- Development
form and round chondroblast cells,7 which
upon terminal differentiation will hypertro- In the developing knee, epiphyseal bone will
phy (become unusually large) and begin to continue to expand into the cartilage anlage
express proteins that attract blood vessels until the cartilage interface forms a thin cal-
and facilitate biomineralization.8 cified layer that arrests vascular invasion.
Mineralized bone begins to form when Calcified cartilage forms at the base of the
the fetal cartilage undergoes focal hyper- articular cartilage and in certain growth plate
trophy, which launches a process of endo- reserve zones (Fig. 5.2), through mechanisms
chondral ossification (EO). The very first that are still not fully understood. Haines10
cartilagebone interfaces to form in the previously noticed that growth plate reserve
body are at the primary ossification centers zones fused to permanent epiphyseal lines
in the shaft of developing long bones. These develop a thin layer of calcified cartilage/
areas are marked by remodeling and vascu- tidemark (Fig. 5.2a, proximal trochlea) while
lar invasion in parallel with the deposition other reserve zones do not form tidemarks
of a mineralizing collagen type I matrix that (Fig. 5.2b, distal trochlea) and eventually
ensheathes and mechanically protects the close without leaving a scar.
blood vessels.4,9 After the secondary ossifi- Growth plate hypertrophic cartilage (HTC)
cation centers appear in the distal tibia and does not form a tidemark. This interface is
femur, three types of dynamic cartilage actually a mixture of cartilage and bone, by
bone interface are established, as illustrated definition of the primary spongiosa, where

Stannard_9781604068580_Ch05.indd 61 1/30/13 2:35 PM

II Science and Techniques for Cartilage Repair

Fig. 5.1 Three dynamic cartilagebone interfaces tape system (Leica Microsystems, Buffalo Grove, IL);
are established in the postnatal knee femoral end: the sections were predigested in hyaluronidase and
(1) articular cartilageepiphyseal bone, (2) growth protease to remove glycosaminoglycans and immu-
plateepiphyseal bone, and (3) growth plate nostained for collagen type II (monoclonal II6B3, DSHB,
metaphyseal bone. A macroscopic transverse view of USA) or collagen type I (monoclonal antibody I-H85,
a skeletally immature  4-month rabbit knee trochlea VWR, Canada), using secondary biotinylated goat anti-
shows the articular cartilage (AC, white arrowheads, mouse and avidin-alkaline phosphatase red substrate
a, b) and growth plate (GP, white arrows, a, c), and dis- detection with iron hematoxylin counterstain.91,92 In
tribution of collagen type II (d, e, f) and collagen type (a) and (b), the rough articular surface is a cutting
I (g, h, i) in serial sections from the area of the dashed artifact from the isomet diamond saw. The tear/crack
squares in a. The distal femur was fixed in 4% para- in the growth plate indicated by a white asterisk in (d)
formaldehyde/100 mM cacodylate, decalcified in EDTA is a cryosectioning artifact.92 Abbreviations: EO, endo-
with a Milestone microwave (Milestone, Shelton, CT), chondral ossification; BP, subchondral bone plate; AC,
cryoembedded and cryosectioned using the CryoJane articular cartilage; GP, growth plate.

new bone is deposited on the cartilage trabec- The growth cartilagemetaphyseal bone
ulae carved out by invading blood vessels and interface is a dynamic and ever-expanding
marrow (Figs. 5.1f, i, and 5.3). In trabecular front of HTC undergoing vascular invasion
bone maturing below the growth plate, an ini- and ossification. Interestingly, in newly
tially pure collagen type II-glycosaminoglycan formed endochondral bone, hypertrophic
(GAG) extracellular matrix is slowly incorpo- chondrocytes express zymogen forms of
rating collagen type I (EO, Fig. 5.3). Vascular enzymes capable of remodeling collagen
invasion of the hypertrophic zone spurs a matrix, including matrix metalloproteinases
continual endochondral expansion of the (MMP-13, MMP-9) and complement C1s.11,12
distal femur (arrows, Fig. 5.4a), in tandem In knockout mice for MMP-13 or MMP-9,
with appositional cartilage and bone growth conversion of collagen type II HTC to colla-
62 (Fig. 5.4b, c). gen type I trabecular bone is inhibited.1315

Stannard_9781604068580_Ch05.indd 62 1/30/13 2:35 PM

and can also promote vascular invasion and
osteogenesis.20,21 Apoptosis of hypertrophic
chondrocytes is also implicated as an impor-
tant driver of the endochondral growth

Articular CartilageBone
Interface during Postnatal
A distinct and more advanced EO process is
going on during postnatal articular cartilage
growth (Fig. 5.1e and h). In 3- to 6-month-old
rabbit articular cartilage, most chondrocytes
are no longer proliferating, and a tidemark
has formed at the base of the hypertrophic
zone.23 Bone is not being deposited along
cartilage trabeculae; it has developed layer
by layer to form a thick osteoid around blood
vessels subjacent to the calcified cartilage

5 The CartilageBone Interface

layer. Only small patches of cartilage persist
in the subchondral bone (EO, Fig. 5.1e). The
remnants of GAG and collagen type II in tra-
becular bone are the hallmarks of EO.
Neonatal articular cartilage is relatively
Fig. 5.2 The growth plateepiphyseal bone interface thick; it is filled with a system of endothe-
sometimes includes a layer of calcified cartilage and lial-lined canals distinct from the normal
a tidemark in the reserve zone (a, proximal trochlea), vasculature.7 Cartilage canals have been
and in other areas is devoid of calcified cartilage or
tidemark and fused to a more vascular bone (b, distal described in immature articular cartilage in a
trochlea). Representative decalcified transverse variety of large animals and in human (fetal
sections from  4-month-old rabbit trochlear growth ovine, 2-week-old calf, 2-year-old human).7,24
plates stained with hematoxylin and eosin are shown, Postnatal weight-bearing activity is asso-
from N  7 distinct New Zealand white rabbit femurs,
 4 months old. Abbreviations: TM, tidemark (white
ciated with regression of the canals and a
arrows); BV, blood vessels; CC, calcified cartilage; EO, thinning and anisotropic organization of the
endochondral ossification (cartilage remnant). articular cartilage layer. The articular layer
continues to grow postnatally through an
appositional or asymmetric layer-by-layer
Remodeling of the HTC front by osteoclasts, expansion, through cell division near the
chondroclasts, and bone-marrow-derived superficial zone6,25,26 (Fig. 5.4). In the deep
metalloproteinases drives the replacement zone near the articular cartilagebone inter-
of HTC with vascularized bone.1315 Growth face, chondrocytes terminally differentiate
plate cell proliferation and vascular invasion into hypertrophic chondrocytes, cease to pro-
can be diminished by nutritional deprivation, liferate, and express collagen type II, collagen
ischemia, or supraphysiologic loading.1618 type X, alkaline phosphatase, and osteopon-
Blood vessel invasion of the HTC layer is tin, a highly phosphorylated hydroxylapatite-
believed to be naturally driven by hyper- binding protein.2731 Like articular cartilage,
trophic chondrocyte secretion of angiogenic the growth plate hypertrophic zone also
factors,19 MMP-13,11,13 and gelatinases capa- contains collagen type X and alkaline phos-
ble of untethering matrix-bound vascular phatase, but a tidemark is notably absent.32,33
endothelial growth factor.15 Osteoclasts that The tidemark that forms at the base of mature
remodel the base of endochondral bone are articular cartilage develops slightly below the
also known to release angiogenic factors region of chondrocytes expressing collagen 63

Stannard_9781604068580_Ch05.indd 63 1/30/13 2:35 PM

II Science and Techniques for Cartilage Repair

Fig. 5.3 Newly synthesized endochondral metaphy- vessels. A 4-month-old rabbit knee femur end was
seal bone below the growth plate of a 4-month-old fixed in formalin, decalcified in 0.5N HCl/0.1% glutaral-
rabbit contains abundant sulfated glycosaminogly- dehyde, cut transversely in the trochlea, embedded in
cans (GAG, red Safranin O stain, a, b) and collagen paraffin, and stained with Safranin O/Fast green/Iron
type II (pink immunostain, c, d) in addition to collagen hematoxylin or immunostained for collagen type II as
type I (in this figure, the collagen type I matrix has previously described.60,72,91 Examples of cartilage rem-
been counterstained green by fast green in a, b, or nant present in the primary spongiosa formed by EO
blue by hematoxylin in c, d). In panel B, fast green are indicated in (b) and (d). Abbreviations: EO, endo-
counterstain also shows bone marrow and blood chondral ossification; HTC, hypertrophic cartilage.

type X.28 Mineral deposits form in the neo- thus normally vascularized, whereas the
natal calcified layer of the articular cartilage nonmineralized cartilage above the tide-
in line with the collagen fibers.34 Using fluo- mark is normally avascular. In a study by
rescent pulse labeling of the mineral phase, Bonde et al,37 blood vessels were observed
Oegema et al observed that the tidemark to only sporadically penetrate the tidemark
advances above the pulse-labeled minerali- into cartilage in normal patellar cadaveric
zation front in 4-month-old rabbit patella at subjects (less than one average blood ves-
a rate of 8 m/week, compared with 1 m/ sel per normal patella from subjects 75 to 89
week in the 7-month-old rabbit patella.26 years old) compared with an average of nine
Creeping advancement of the tidemark is tidemark-penetrating vessels per subject
associated with thinning of the articular car- in osteoarthritis (OA) femoral condyle sam-
tilage layer.23 ples. Pathologic blood vessel passage beyond
Vascular channels are branched structures the tidemark is associated with occasional
that supply the calcified cartilage of the thrombosis, a thicker calcified cartilage layer,
articular layer7,9,35 and have similarities with and tidemark duplication37 (Table 5.1).
vascular channels in the vertebral endplate The calcified cartilage layer is semiperme-
where bone abuts the cartilaginous nucleus able and permits passage of small molecules
64 pulposus.36 The calcified cartilage zone is ( 500 Da) from the subchondral bone to

Stannard_9781604068580_Ch05.indd 64 1/30/13 2:35 PM

Fig. 5.4 Illustration of femur growth and appo- proliferating chondroblasts deposit fibrillar collagen
sitional growth. (a) The distal femur grows (as type II above a previously existing layer of collagen
illustrated in faithful tracings of a transverse section type II; the proliferating chondrocytes are situated

5 The CartilageBone Interface

through the immature rabbit trochlea and proximal above invading blood vessels frequently capped with
condyles) through vascular invasion, which drives collagen type Ipositive mineralized bone. In intra-
endochondral ossification of the hypertrophic zone membranous bone growth (c), osteoblasts and blood
of the growth plates (arrows, thick dashed lines) while vessels are closely associated during inframembra-
blood vessels from the epiphyseal bone supply the nous generation of lamellar bone. The mechanisms of
articular cartilage hypertrophic zone (arrowheads). endochondral bone growth are not illustrated in this
During appositional growth of articular cartilage (b), diagram.

the articular cartilage layer.38,39 Conversely, trochlear articular cartilage in a 30-month-

immersion of a mouse distal femur end old rabbit knee (Fig. 5.5a). The tidemark at
in fluorescein allows full solute passage this stage is a strong hematoxylin-stained
through the articular cartilage and selective line (Fig. 5.5b). Some remnant or newly
fluorescein diffusion into chondrocytes in the duplicating tidemarks can be observed
calcified cartilage layer.39 Calcified cartilage within the mature calcified cartilage layer
permeability was measured as fivefold less (open arrowheads, Fig. 5.5b). In the skel-
than noncalcified cartilage in mature horse etally immature 4-month-old rabbit, a more
metacarpal tissues.38 It has been hypothe- irregular mineralization front is observed at
sized that venous congestion in the synovium the articular cartilagebone interface, which
and subchondral bone could play a role in corresponds to the subchondral bone and a
tidemark duplication.37 Thickening of the thin layer of calcified cartilage (Fig. 5.5c). At
calcified cartilage in OA could be expected to this stage a nascent tidemark can be visual-
reduce the flow of small solutes from the vas- ized using a hematoxylin-eosin stain, and
cularized subchondral bone to the deep zone the vascular bone channels are just below,
chondrocytes. with close communication between the
Once formed, the tidemark and calcified vasculature and calcified articular cartilage
cartilage layer persist as dynamic structures (Fig. 5.5d). By contrast, in the same 4-month-
that can change and remodel over time. Below old rabbit trochlear specimen, the growth
mature articular cartilage, the mineralization plate HTC contains a highly irregular and dis-
front is a relatively smooth and undulating continuous mineralization front at the grow-
plate-like surface, as illustrated in a micro- ing bonecartilage interface with no visible
computed tomography (micro-CT) 3D image tidemark (Fig. 5.5e, f). The mineral front at
of the calcified cartilage and bone below the the base of the growth plate corresponds 65

Stannard_9781604068580_Ch05.indd 65 1/30/13 2:35 PM

Table 5.1 Published histomorphometric and stereological measures of articular cartilage, tidemark, calcified
cartilage, bone plate thickness, and tidemark number or area in normal and OA human subjects and different
animal species

Articular Calcified
Cartilage Tidemark Cartilage Bone
N/OA Age Site (m) (number or area) (m) Plate

Lane and Bullough NH 2039 FH 1.2  0.1 193  28

Lane and Bullough NH 4059 FH 1.2  0.1 141 18
Lane and Bullough NH 6093 FH 1.8  0.4 119  24
Frisbie et al 200648 NH MFC 2,200 125 490
II Science and Techniques for Cartilage Repair

Frisbie et al 200648 NE MFC 2,000 210 375

Frisbie et al 2006 48
NO MFC 450 125 250
Frisbie et al 2006 48
NR MFC 200 100 250
Hunziker et al 200246 NH 2349 MFC 2,410 134 190
Wang et al 200943 NH 2045 MFCa 104  21
Bonde et al 200537 NH 6585 PAT 2.5 cm2 (1.83.9)b
Bonde et al 2005 37
OA 4786 MFC 7.7 cm (2.413.3)
2 c

Weight-bearing area.
Mean 0.5 (0 to 10) penetrating blood vessels in non-OA patellar tidemark.
Mean 9 penetrating blood vessels (2 to 47) in OA tidemark area; 3 out of 21 vessels had thrombosis.
Abbreviations: , not done; FH, femoral head; MFC, medial femoral condyle; NH, normal human; OA,
osteoarthritic human; NE, normal equine; NO, normal ovine; NR, normal rabbit; PAT, patella.

with the vascular bone and newly deposited proximal reserve zone and articular cartilage
collagen type I (black arrowheads, Fig. 5.5f). hypertrophic zone), and the tidemark serves
In the growth plate hypertrophic zone, cal- as a barrier to vascular invasion and calcifica-
cification of the collagen type II matrix is tion of hyaline cartilage.
much delayed compared with the articular
cartilage calcified layer. This is because after
birth the mammalian joints require a suitable Structure and Mineral
mechanically stable articular surface, while
growth plates in the long bones are continu-
Content of the Mature
ally expanding, even beyond sexual maturity. Articular CartilageBone
Cartilage calcification is therefore occurring Interface
only at the end-stage of cartilage growth.
After reaching skeletal maturity, growth Articular calcified cartilage is a mineralized
plates are completely resorbed and replaced layer in which extracellular matrix is chiefly
by collagen type Ipositive mineralized bone. composed of collagen type II, collagen type X,
To summarize, growth plates develop a and GAG19; the layer also contains extracel-
relatively stable reserve zoneepiphyseal lular alkaline phosphatase.31 Alkaline phos-
bone interface, with a purely collagen type phatase can generate free phosphate from
II GAG-rich cartilage phase and a mixture organophosphates such as -glycerol phos-
of collagen type I and collagen type II in the phate, for incorporation into hydroxylapatite
newly forming primary spongiosa. Calcified mineral (Ca-P).4042 In the calcified cartilage
cartilage becomes established at the edges layer of normal human femoral condyles,
66 of a permanent epiphyseal bone layer (i.e., chondrocytes are quiescent and present at

Stannard_9781604068580_Ch05.indd 66 1/30/13 2:35 PM

5 The CartilageBone Interface
Fig. 5.5 The mineralization front at the cartilage NRecon and CTAn software) from the 30-month-old
bone interface, in a skeletally aged 30-month-old rabbit (a), and the 4-month-old rabbit trochlea shown
rabbit trochlea articular cartilage-subchondral bone in Fig. 5.1c, e. B, d, and f show the histological
plate (a, b), and skeletally immature 4-month-old appearance of the cartilagebone interface from the
rabbit articular cartilage-epiphyseal bone interface 30-month-old (b), and 4-month-old rabbit trochlea
(c, d) and hypertrophic growth platemetaphyseal (d, f), hematoxylin & eosin stain. The small black
bone interface (e, f). (A), (c), and (e) show 3D recon- arrows show the tidemark. Open arrowheads show
structions from a microcomputed tomography scan vestigial or newly duplicating tidemarks (b) and the
(SkyScan 1172 instrument, 9.8 m/pixel resolution, solid black arrowheads show areas of bone osteoid.

a much lower density compared with hya- subchondral bone plate composed of lamel-
line cartilage (average of 51 cells/mm2 vs. lar bone, along with punctate regions where
152 cells/mm2).43 The calcified cartilage layer the calcified cartilage is in direct contact with
is flanked by an undulating tidemark, and an vascular channels.7,47 In any group of indi-
even more irregular cement line adjacent to viduals, the mean calcified cartilage thickness
the bone. Wang et al43 analyzed normal adult and mineral density will vary according to
human bone (20- to 45-year-old cadaveric) age, site in the joint, and mechanical loading
by histomorphometry and stereology to (Table 5.1).44,48,49 In a study of normal femo-
show that hyaline cartilage is interlocked ral head cadaveric specimens with no signs
tightly in a ravine-engomphosis structure of OA by Lane and Bullough,44 the calcified
with the calcified cartilage zone, which is cartilage thickness in the femoral head varied
then attached in a comb-anchor to bone.43 from 79 m to 243 m, with a thicker calci-
The surface roughness was determined to be fied cartilage in less stressed areas of the hip
1.14 (tidemark) and 1.99 (cement line).43 The joint. Mller-Gerbl et al45 performed a simi-
more irregular cement linebone interface is lar study in normal cadaveric femur heads
the end result of an inhomogeneous vascular and found the calcified cartilage thickness
invasion during development of the calcified varied from 20 m to 230 m, and that the
cartilage layer (white arrowheads, Fig. 5.1h). ratio of calcified cartilage to total cartilage
In normal human subjects, the mean cal- thickness was relatively constant. The calci-
cified cartilage thickness is variable, from fied cartilage layer shows gradual thinning
20 m to  250 m.4446 The calcified cartilage with age, along with tidemark duplication in
is tightly fused to the articular cartilage and subjects over 70 years old (see Table 5.1).44 67

Stannard_9781604068580_Ch05.indd 67 1/30/13 2:35 PM

Lane and Bullough concluded that the calci- remove hypomineralized calcified cartilage
fied layer is undergoing continual resorption from the image (i.e., areas between the black
and endochondral advancement over time.44 arrows and black arrowheads, Fig. 5.5d).
These observations are consistent with the In situ elemental analyses of the skel-
measured dwindling rate of advancement of etally mature normal human or OA human
the tidemark with age in rabbit patella.26 In cartilagebone interface has revealed the
31 normal human femoral condyles, an age- presence of calcium, phosphorus, potassium,
dependent loss in bone mass was measured sulfur, zinc, and strontium.52,53 In mature
in the subchondral bone plate.50 The bone vol- osteochondral samples, the mineralized sur-
ume fraction (bone volume/total volume%) face has a smoother texture and corresponds
of the bone plate region was observed to to the tidemark, with no visible difference by
decline from  36% for subjects in their 20s micro-CT between the calcified cartilage and
to  27% for those  80 years old.50 Bone loss osteoid immediately below (Fig. 5.5a, b).54 In
was attributed to thinning of the subchondral a normal human patella, mineralized carti-
trabeculae with age (as opposed to diminished lage showed a slightly but significantly higher
II Science and Techniques for Cartilage Repair

trabecular number), and this occurred at a calcium content than adjacent bone (25% vs.
relatively steady rate (Trabecular Thickness  23% w/v), and the mineral particles in bone
141 m  0.63  age).50 By contrast, in OA, and articular calcified cartilage were found to
a pathological increase in the calcified carti- align with the direction of collagen organiza-
lage layer thickness arises, along with abnor- tion.55 Using two-dimensional nuclear mag-
mal tidemark duplication37 (see Bonde et al, netic resonance (2D-NMR) spectroscopy and
Table 5.1), and this is frequently accompa- X-ray diffraction, Duer et al56 analyzed the
nied by subchondral bone plate thickening mineral component of pulverized calcified
and sclerosis.51 The consequence or implica- cartilage samples from skeletally mature
tion of tidemark duplication is not known, horse phalanx and distal radius. They con-
although Burr has proposed that microcracks cluded that the calcified cartilage mineral
at the bonecartilage interface may be impli- signal is similar to hydroxylapatite of bone,
cated in the etiology.51 but with smaller peaks indicating small crys-
The mineral component in the calcified tals or disorder in the mineral component.
cartilage layer is similar but distinct from They also found evidence that, unlike bone,
that found in bone, and notably influenced by GAG present in calcified cartilage provides a
the uniform presence of GAG. In a study by more hydrated matrix, anionic side-chains
Rey et al, pulverized calcifying cartilage from (carboxylate and sulfate) for binding calcium
2-month-old calves (collagen type II-positive in the mineral crystal surfaces, and hydroxyl
and type I-negative) had a very low mineral groups to H-bond with surface water, mineral
content (2.8% by weight), with an imma- hydroxyl, and phosphate ions.56 The spectra
ture, very poorly crystalline and low carbon- were also consistent with the presence of
ate apatite mineral [calcium/(phosphate Gla residues (-carboxyglutamic acid) in the
carbonate)], compared with bone (normally calcified cartilage layer; Gla-domain proteins
 0.20 carbonate/P).41 The calcified cartilage are found in a variety of mineral-binding pro-
mineral phase was also characterized by a teins such as osteocalcin.27
large proportion of nonapatite brushite- It is not well understood how the tide-
like phosphate.41 Unlike bone, this high mark is formed, and knowledge of its precise
nonmineral, labile phosphate content actu- composition is also limited. The tidemark is
ally increases over time.41 The low mineral a 5-m-thick structure that appears at the
content measured in immature calcified cartilagecalcified cartilage junction and
cartilage by Rey et al41 is consistent with the can be visualized with hematoxylin, a blue
quite irregular mineral surface of the epiphy- dye that is intensified by metal ions. The
seal growth plate shown in Fig. 5.5c. In this tidemark could potentially arise simply
sample, the mineral surface most probably by the accumulation and precipitation of
corresponds to mineralized collagen type I chondrocyte-derived extracellular matrix
because the threshold level used in this three- species and ions at the calcified cartilage
68 dimensional reconstruction model would front, due to the sharp decrease in tissue

Stannard_9781604068580_Ch05.indd 68 1/30/13 2:35 PM

permeability. The tidemark could serve to start with debridement of the surface of the
inhibit calcifying matrix vesicles released lesion to remove degenerated articular car-
from the bone from penetrating into hyaline tilage.62 Depending on the repair approach,
cartilage. Lectin staining has suggested that the debridement step may aim to retain the
the tidemark contains variously branched calcified cartilage layer for cell delivery6365
alkali-resistant glycans with
-galactosyl or to completely remove it, as during micro-
or N-acetyl-lactosamine termini.57 Zoeger fracture or marrow stimulation.6668 However,
et al53 detected a specific accumulation of if present, the tidemark and calcified car-
lead at the tidemark in normal cadaveric tilage are technically very challenging to
femoral head and patella. Oegema et al sug- debride with precision. Light curettage usu-
gested that the superficial articular layer ally leaves a thin layer of noncalcified deep
could be involved in a paracrine loop that zone articular cartilage, whereas shaving or
controls deep zone chondrocyte hypertrophy vigorous curettage often removes a consider-
and calcification, which could potentially able amount of subchondral bone plate with
explain thickening of the calcified cartilage the calcified cartilage.54,64,65 Vascular channels
in OA following loss of the superficial zone.26 containing erythrocytes terminate normally
Alternatively, microcracks in the calcified inside the calcified cartilage layer.7 Therefore,
layer could permit diffusion of bone-derived in a joint with only one tidemark, debride-
matrix vesicles farther into the deep zone, ment of the tidemark along with as little as
resulting in tidemark advancement. 50 m of the superficial mineralized layer is
Duplication of the tidemark in aging and expected to generate some bleeding at the

5 The CartilageBone Interface

OA is well documented.26,44,51,58 In aging sub- debrided surface, although bleeding from
jects, up to 5 duplicated tidemarks were these tiny capillaries may not be macroscopi-
observed in normal human subjects over cally visible. Skeletally immature animals
70 years old,44 and as many as 10 tidemarks have a greater ease of debridement and dif-
in primates over 20 years old.58 A significant ferent cell populations present in the epiphy-
correlation was observed between increasing sis compared with the adult knee (Figs. 5.1
tidemark duplication, mineral density, and and 5.5). In addition, the epiphyseal blood
carbonate content in primates.58 Repetitive vasculature in skeletally immature knees has
knee microtrauma in a rabbit model during active endothelial cell proliferation, whereas
9 weeks of loading was shown to lead to a adult vasculature has postmitotic endothelia,
mean 25% increase in the proximal tibial cal- and the subchondral bone no longer contains
cified cartilage layer thickness and tidemark osteoclasts.21 It is for these reasons that skel-
duplication, with no change in mean articu- etally immature animals are improper car-
lar cartilage thickness.59 Multiple tidemarks tilage repair models for adult knees.69 These
were observed to form at the cartilagebone same cautionary notes hold for rats and mice,
interface in tissues surrounding an osteo- whose growth plates never close.
chondral defect in rabbit trochlea 6 months Scarce information is available on tide-
postoperative,60 and in sheep above a metal mark regeneration. One may reasonably
implant placed in the subchondral bone.59 wonder whether tidemark regeneration
Tidemark duplication could be related to should be one of the goals of cartilage repair
uneven load-sharing following softening of a strategies. Frisbie et al showed that tidemark
focal area of damaged cartilage.60 could regenerate in equine microfractured
defects at 12 months postoperative only if
the calcified cartilage layer were completely
CartilageBone Interface in debrided.70 In lesions that retained calcified
Cartilage Repair cartilage and original tidemark, the new
repair tissue had poor tissue integration
In articular cartilage lesions, the tidemark with the base of the defect.70 Sheep femoral
is either fully retained (Outerbridge grade condyle microfracture defects treated or not
I to III partial-thickness lesions) or missing with a chitosan-based implant showed par-
to a variable extent (grade IV full-thickness tial tidemark regeneration at 6 months post-
lesions).61 Most cartilage repair procedures operative, at the base of hyaline-like cartilage 69

Stannard_9781604068580_Ch05.indd 69 1/30/13 2:35 PM

(HyC) repair with a collagen type II-positive can mature to acquire a stratified structure
and collagen type I-negative deep zone with vascular invasion and endochondral
fully integrated with bone.71,72 In an equine resorption at the base after 2 months in a
case study, a cartilage defect treated by rabbit model, leaving an articular layer of
deep debridement and a composite implant hyaline repair cartilage.
regenerated a tidemark at 12 months post- In various models of marrow stimulation
operative in repair cartilage with a deep in the rabbit, microdrill holes that are cre-
zone containing appropriate collagen fiber ated and left to bleed (as in clinical practice79)
organization.73 In a rabbit microdrill model of will spontaneously regenerate a fibrocarti-
cartilage repair, a tidemark was observed at lage repair tissue that contains both collagen
6 months postoperative where the bone and type I and collagen type II.21,60,78 This type of
cartilage tissue formed an integrated unit.60 spontaneous repair in a rabbit is illustrated
Several groups have shown that chondro- in Fig. 5.6, at 2.5 months postoperative.
genic foci will spontaneously form in drill In this rabbit, using a small arthrotomy, a
or microfracture holes generated in skel- 1.4-mm-diameter, 2-mm-deep microdrill
II Science and Techniques for Cartilage Repair

etally mature knee cartilage defects.7477 hole was created in the distal femoral knee
Subchondral cartilage repair tissue contains trochlea and allowed to bleed without fur-
cells with chondrocyte morphology that ther treatment. After 2.5 months postop-
normally progress to hypertrophy, vascular erative, the drilled defect is still undergoing
invasion, and replacement by bone.21,60,7478 EO and repair. The drill hole has spontane-
Chevrier et al75 concluded that chondrogenic ously regenerated fibrocartilage repair at
foci that appear near the top of the drill holes the top of the drill hole, which is anchored

Fig. 5.6 Spontaneous repair of a 1.4-mm-diameter, hypertrophic cartilage (HTC), and fibrocartilage
2-mm-deep osteochondral drill hole at 2.5 months (FC). Mineral formation below hypertrophic cartilage
postoperative in the knee trochlea of a skeletally aged during endochondral ossification is shown in (f),
(32 months) rabbit. Decalcified serial cryosections by a reconstructed 3D image from a micro-CT scan
through the drill hole were stained for SafraninO/Fast (SkyScan 1172, 9.8 m/pixel resolution, area corre-
green (a, e), or immunostained for collagen type II sponding to e) that was performed prior to decalcifi-
(b) or type I (c). The articular cartilage repair tissue cation. All protocols involving animals were approved
is characterized as fibrocartilage or fibrous because by Institutional Ethics Committees. Abbreviations: AC,
it contains mainly collagen type I with little collagen articular cartilage; FC, fibrocartilage; TM, tidemark;
type II and is depleted of GAG. (D) shows faithful trac- CC, calcified cartilage; EO, endochondral ossification;
ings of structures from (a), including bone-associated HTC, hypertrophic cartilage area; VB, vascular bone;
GAG as a marker of endochondral ossification (EO, arrowheads, bone mineralization front.
70 black), angiogenic bone marrow cavities (red),

Stannard_9781604068580_Ch05.indd 70 1/30/13 2:35 PM

to HTC that sits below the tidemark within was created in the left knee trochlea and
the flanking articular cartilage (Fig. 5.6ac). further treated by press-fitting a presolidi-
Endochondral vascular invasion and miner- fied chitosan-blood implant into the hole.80
alization are occurring at the base of the HTC Relative to the contralateral untreated drill
(black arrowheads, Fig. 5.6ae). Patches of hole, after 2.5 months of repair, a delayed
GAG and collagen type II in the new repair and altered EO process is seen in the treated
bone trabeculae reveal the tell-tale signs defect (Fig. 5.7). HyC repair tissue containing
of EO (Fig. 5.6ad). From the histology low levels of collagen type II and no collagen
drawing shown in Fig. 5.6d, we can appreci- type I is observed above the mineralization
ate that EO has been initiated at a previous front (Fig. 5.7b, c). The hyaline-like repair is
time point in this defect because the patches overlaid with undifferentiated mesenchyme
of GAG, which reveal remnants of hyaline surrounded by collagen type I (Fig. 5.7c). In
cartilage in the newly formed bone, occur this implant-treated defect, the mineraliza-
in areas up to 400 m below the blunted tion front has a more irregular appearance
mineralization front (Fig. 5.6df). The mor- and consists in bony vascular invasion of hya-
phology of the endochondral repair tissue line tissue (Fig. 5.7f). One can appreciate that
at this point, including cryosectioning tears, in this osteochondral defect the implant has
resembles that of the growth plate, and no delayed osteochondral ossification because
tidemark is visible. the mineralized GAG is only beginning to
In the same animal described above, a form at the repair cartilagebone interface
1.4-mm-diameter, 2-mm-deep microdrill hole (EO, Fig. 5.7a, d, e). Unlike the endochondral

5 The CartilageBone Interface

Fig. 5.7 Repair of a 1.4-mm-diameter, 2-mm-deep or collagen type I (c). (D) shows tracings of structures
osteochondral drill hole at 2.5 months postsurgery in in (a), including bone-associated GAG as a marker of
the knee trochlea of a skeletally mature (32 months) EO (black), angiogenic marrow cavities (red), HyC,
rabbit, where the drill hole was treated at surgery by area of HTC, and UM. Panel F shows a reconstructed
press-fitting a presolidified chitosan-NaCl/autologous 3D image from a micro-CT scan corresponding to the
whole blood clot implant into the hole.80 The defect area shown in (e). Black arrowheads: mineralization
was generated in the contralateral knee of the rab- front. The three white arrows in (f) show a bone-
bit defect shown in Fig. 5.6 under institutionally encased blood vessel very similar to a branched vas-
approved animal protocols. At 2.5 months postop- cular invasion histology image previously published
erative, femur ends were fixed, micro-CT scanned by Oegema et al.26 Abbreviations: HyC, hyaline-like
(Skyscan 1172, 9.8 m/pixel resolution), decalcified cartilage; HTC, hypertrophic cartilage area; EO, endo-
in EDTA, and cryosections stained for Safranin O/Fast chondral ossification; UM, undifferentiated mesen-
green (a, e), immunostained for collagen type II (b), chyme; arrowheads, mineralization front. 71

Stannard_9781604068580_Ch05.indd 71 1/30/13 2:35 PM

bone formed during spontaneous repair,
hypertrophic chondrocytes are scarcely pre-
sent at the advancing interface of new bone
and blood vessels (Fig. 5.8). Collagen type I of
newly formed bone is being deposited from
inside the invading bone marrow channels.
Given that the chondrocytes present in the
collagen type II repair matrix are not yet ter-
minally differentiated to hypertrophic cells,
the proximity of repair cells and invading
blood vasculature can still drive cell prolif-
eration and appositional growth of more col-
lagen type II hyaline-like matrix. At one edge
of the drill hole, the bone has regenerated to
the native tidemark level, and a new tide- Fig. 5.8 Developing cartilagebone interface in
II Science and Techniques for Cartilage Repair

the subchondral area of an osteochondral defect

mark can be observed (Fig. 5.9). treated with a presolidified chitosan-based implant
In some rabbit cartilage repair models at 2.5 months postoperative (from the same histo-
involving complete debridement of the logy image shown in Fig. 5.7c). The section was
calcified cartilage layer, subchondral bone immunostained for collagen type I (red stain) with
iron hematoxylin counterstain. Arrows show round
plate advancement beyond the native tide-
and crescent-shaped chondrocyte cells in hyaline-like
mark in flanking cartilage has been observed repair cartilage, and arrowheads show collagen type
after 3 to 9 months of repair.81,82 Bone plate I-expressing cells that cap the invading blood vessel
advancement could be a consequence of bone marrow cavities with new osteoid, similar to
delayed or failed tidemark regeneration dur- that described by Gilmore and Palfrey in neonatal
human lateral femoral articular cartilage.7 Both
ing bone marrow-driven EO below hyaline- round and cuboidal cells express collagen type I,
like repair tissue. as is seen in growth plate endochondral bone. Scale
In human cartilage repair, the extent of bar: 50 m.
tidemark formation in repair osteochon-
dral biopsies has been added to a new
histological scoring system generated by
the International Cartilage Repair Society
(ICRS II).83 The score uses a visual analog
scale (VAS) where the reader marks a line
on a 10-mm scale that is then converted to
a percentage between 0% (no tidemark) and
100%. In one randomized controlled clinical
trial comparing characterized chondrocyte
implantation (CCI) and microfracture (MFX),
osteochondral repair biopsies were analyzed
in a blinded fashion using the ICRS II scoring
system. At 12 months postoperative, both CCI
and MFX groups showed the same  16-point
mean clinical improvement from baseline
in overall Knee injury and Osteoarthritis
Outcome Score (KOOS)84 that progressed Fig. 5.9 Evidence of tidemark formation at 2.5
months postoperative at the edge of the repairing
to  20 CCI versus  15 MFX mean change
rabbit trochlear osteochondral hole at the level of the
from baseline KOOS at 5 years postoperative tidemark in an osteochondral drill hole treated with
(p  0.116).85 Biopsies collected at 12 months presolidified chitosan-blood implant. Hematoxylin-
postoperative from 48 out of 61 MFX-treated eosin stained EDTA-decalcified cryosection. The
patients showed a mean  18% tidemark for- arrows show the new tidemark (TM) and the arrow-
head shows a terminally differentiated hypertrophic
mation along the biopsy width compared chondrocyte inside the newly forming calcified carti-
with a mean  32% tidemark formation in lage layer. The image was taken from a field near the
72 biopsies from 38 out of 51 CCI-allocated upper right corner of Fig. 5.7a. Scale bar: 50 m.

Stannard_9781604068580_Ch05.indd 72 1/30/13 2:35 PM

patients (p  0.036).84 Given that cell therapy this important variable. To better evalu-
aims to retain the calcified cartilage layer at ate the progression and success of different
surgery,86 significantly greater tidemark pre- cartilage repair therapies, patient-reported
sent in CCI biopsies could be partly due to a outcomes90 need to be correlated with repair
lighter debridement of the initial tidemark in tissue architecture.72 A better understand-
the CCI lesions. ing of cartilage repair tissue maturity will be
Another randomized controlled clinical reached with new histological methods that
trial compared MFX to MFX and chitosan can distinguish between native and regen-
glycerol phosphate/blood implant (BST- erated tidemark, standardized measures of
CarGel), at 12 months postoperative.87,88 tidemark and calcified cartilage formation,
Blinded ICRS II scoring of osteochondral and further research on the mechanisms of
biopsies revealed more tidemark present cartilage calcification.
in nine biopsies from MFX-treated defects
compared with 12 implant-treated defects.88
Some biopsies from this study showed zonal Summary Points and Clinical
collagen organization resembling native Relevance
articular cartilage,89 and a hyaline-like deep
zone containing collagen type II and no col- 1. All cartilagebone interfaces develop
lagen type I.87,88 One MFX biopsy collected from an initially cartilaginous structure
at 12 months postoperatively consisted in a that undergoes coordinated invasion by
collagen type I /collagen type II fibrocar- blood vessels and osteoblasts. Formation

5 The CartilageBone Interface

tilage repair with an irregular bone interface, of a tidemark anatomically stabilizes
and 0% tidemark formation.72 the cartilagebone interface and arrests
To summarize, tidemark has been observed cartilage calcification and blood vessel
in some human cartilage repair osteochon- invasion. Vascularization of the calci-
dral biopsies 12 months following bone mar- fied cartilage layer and subchondral bone
row stimulation or cell therapy. The presence plate is an important feature of a healthy
of a tidemark could arise through hyaline cartilagebone interface.
cartilage regeneration via EO, or by incom- 2. The cartilagebone interface is a mineral-
plete debridement and persistence of the ized blood vessel boundary where collagen
native tidemark in the treated lesion. Finally, type II is integrated with collagen type I.
despite great care, it is also possible that 3. Animals that have permanently open
some biopsies with a complete tidemark may growth plates (mice and rats) and skele-
have been taken from outside the area of the tally immature animals with open growth
initial lesion.72 plates (rabbits less than 7 months old,
Cartilage repair is a complex process that and large animals less than  2 years
takes place over a long period of time. The old) are improper cartilage repair models
notion of cartilage repair as an isolated event for establishing the efficacy of therapies
should be discarded for the more comprehen- intended for use in adult human knees.
sive view of osteochondral repair, given the 4. Chronic medications (i.e., steroids), drugs
extensive cross-talk between cartilage repair (i.e., smoking), or surgical procedures
tissues, bone, and blood vessels in the devel- that produce chronic ischemia in the epi-
oping interface. New calcified cartilage layer/ physeal bone may contribute to articular
tidemark can be regenerated in a pure type II cartilage degeneration and/or suppress
collagen matrix containing GAG integrated cartilage regeneration. Conversely, treat-
to endochondral bone near the articular sur- ments that stimulate revascularization of
face. Residual cartilage and calcified cartilage subchondral bone damaged by drilling or
can block cell migration and vascular inva- microfracture have the potential to drive
sion during marrow-derived cartilage regen- epiphyseal endochondral repair.
eration; therefore, new tools or methods that 5. With increasing age, microtrauma, and
permit the surgeon to verify the presence advanced OA, the calcified cartilage layer
of residual cartilage and calcified cartilage either thins out or thickens and becomes
at the debridement step would help control more mineralized (Fig. 5.5a vs. 5.5c, 73

Stannard_9781604068580_Ch05.indd 73 1/30/13 2:35 PM

Table 5.1). Therefore, during clinical sur- References
gical procedures that aim to debride the
calcified cartilage layer, the potential 1. Sasano Y, Mizoguchi I, Kagayama M, Shum L,
Bringas P Jr, Slavkin HC. Distribution of type I
thickness and extent of mineralization collagen, type II collagen and PNA binding glyco-
of the calcified cartilage layer should be conjugates during chondrogenesis of three dis-
tinct embryonic cartilages. Anat Embryol (Berl)
taken into account. 1992;186(3):205213
6. Calcified cartilage and osteoid in the adult 2. Craig FM, Bentley G, Archer CW. The spatial and
subchondral bone have a similar mineral temporal pattern of collagens I and II and keratan
level, which means that during debride- sulphate in the developing chick metatarsophalan-
geal joint. Development 1987;99(3):383391
ment procedures the tidemark can be fully 3. Olsen BR, Reginato AM, Wang W. Bone develop-
removed only by cleanly and carefully ment. Annu Rev Cell Dev Biol 2000;16(10810706):
scraping off a specific mineralized depth 191220
from the entire lesional surface. 4. Sandberg M, Vuorio E. Localization of types I, II,
and III collagen mRNAs in developing human
7. The tidemark and calcified cartilage layer skeletal tissues by in situ hybridization. J Cell Biol
are technically very challenging to debride 1987;104(4):10771084
II Science and Techniques for Cartilage Repair

with precision. Light curettage usually 5. Shapiro F. Bone development and its relation to
fracture repair. The role of mesenchymal osteo-
leaves a thin layer of noncalcified deep blasts and surface osteoblasts. Eur Cell Mater
zone articular cartilage, whereas shav- 2008;15:5376
ing or vigorous curettage often removes a 6. Hunziker EB. The elusive path to cartilage regen-
eration. Adv Mater (Deerfield Beach Fla) 2009;
considerable amount of subchondral bone 21(3233):34193424
plate with the calcified cartilage. In a joint 7. Gilmore RS, Palfrey AJ. A histological study of
with only one tidemark, debridement of human femoral condylar articular cartilage. J Anat
the tidemark along with as little as 50 m 1987;155:7785
8. Leboy PS, Vaias L, Uschmann B, Golub E, Adams
of the superficial mineralized layer is SL, Pacifici M. Ascorbic acid induces alkaline phos-
expected to generate bleeding, which may phatase, type X collagen, and calcium deposi-
or may not be immediately visible. tion in cultured chick chondrocytes. J Biol Chem
8. Bone plate advancement could be a con-
9. Clark JM. The structure of vascular channels in the
sequence of delayed or failed tidemark subchondral plate. J Anat 1990;171:105115
regeneration during bone marrow-driven 10. Haines RW. The histology of epiphyseal union in
EO below hyaline-like repair tissue. mammals. J Anat 1975;120(Pt 1):125
9. When evaluating outcomes of cartilage 11. DAngelo M, Yan Z, Nooreyazdan M, et al. MMP-13
is induced during chondrocyte hypertrophy. J Cell
repair procedures, it is important to real- Biochem 2000;77(4):678693
ize that the presence of a tidemark could 12. Sakiyama H, Inaba N, Toyoguchi T, et al. Immunolo-
arise through true hyaline cartilage regen- calization of complement C1s and matrix metallo-
proteinase 9 (92kDa gelatinase/type IV collagenase)
eration or by incomplete debridement and in the primary ossification center of the human
persistence of the native tidemark. femur. Cell Tissue Res 1994;277(2):239245
13. Stickens D, Behonick DJ, Ortega N, et al. Altered
endochondral bone development in matrix
Acknowledgments metalloproteinase 13-deficient mice. Development
Funding for this work was from the Canadian 14. Little CB, Meeker CT, Hembry RM, et al. Matrix
metalloproteinases are not essential for aggrecan
Institutes of Health Research (CIHR, 185810- turnover during normal skeletal growth and devel-
BME), Natural Sciences and Engineering opment. Mol Cell Biol 2005;25(8):33883399
Research Council of Canada (NSERC, 15. Vu TH, Shipley JM, Bergers G, et al. MMP-9/
STGP 365025), the Canada Foundation for gelatinase B is a key regulator of growth plate
angiogenesis and apoptosis of hypertrophic chon-
Innovation (CFI), and Fonds de Recherche drocytes. Cell 1998;93(3):411422
Sant du Qubec (FRSQ, Groupe de Recherche 16. Reich A, Jaffe N, Tong A, et al. Weight loading
en Sciences et Technologies Biomdicales, young chicks inhibits bone elongation and pro-
motes growth plate ossification and vasculariza-
GRSTB). Salary support for C.D.H. was tion. J Appl Physiol 2005;98(6):23812389
from an FRSQ bourse Chercheur National/ 17. Farnum CE, Lee AO, OHara K, Wilsman NJ. Effect
National Researcher Career Award. We thank of short-term fasting on bone elongation rates: an
analysis of catch-up growth in young male rats.
Genevive Picard and Jun Sun for excellent Pediatr Res 2003;53(1):3341
technical contributions, and Julie Tremblay 18. Kim HK, Su P-H, Qiu Y-S. Histopathologic changes
74 for quality assurance. in growth-plate cartilage following ischemic

Stannard_9781604068580_Ch05.indd 74 1/30/13 2:35 PM

necrosis of the capital femoral epiphysis. An 35. Boyde A, Firth EC. Articular calcified cartilage
experimental investigation in immature pigs. J canals in the third metacarpal bone of 2-year-old
Bone Joint Surg Am 2001;83-A(5):688697 thoroughbred racehorses. J Anat 2004;205(6):
19. Alini M, Matsui Y, Dodge GR, Poole AR. The extracel- 491500
lular matrix of cartilage in the growth plate before 36. Kobayashi S, Baba H, Takeno K, et al. Fine structure
and during calcification: changes in composition of cartilage canal and vascular buds in the rab-
and degradation of type II collagen. Calcif Tissue Int bit vertebral endplate. Laboratory investigation. J
1992;50(4):327335 Neurosurg Spine 2008;9(1):96103
20. Cackowski FC, Anderson JL, Patrene KD, et al. Oste- 37. Bonde HV, Talman MLM, Kofoed H. The area of the
oclasts are important for bone angiogenesis. Blood tidemark in osteoarthritisa three-dimensional
2010;115(1):140149 stereological study in 21 patients. APMIS 2005;
21. Chen G, Sun J, Lascau-Coman V, Chevrier A, 113(5):349352
Marchand C, Hoemann CD. Acute osteoclast activ- 38. Arkill KP, Winlove CP. Solute transport in the deep
ity following subchondral drilling is promoted by and calcified zones of articular cartilage. Osteoar-
chitosan and associated with improved cartilage thritis Cartilage 2008;16(6):708714
tissue integration. Cartilage 2011;2:173185 39. Pan J, Zhou XZ, Li W, Novotny JE, Doty SB, Wang
22. Hashimoto S, Ochs RL, Rosen F, et al. Chondrocyte- LY. In situ measurement of transport between
derived apoptotic bodies and calcification of artic- subchondral bone and articular cartilage. J Orthop
ular cartilage. Proc Natl Acad Sci U S A 1998;95(6): Res 2009;27(10):13471352
30943099 40. Bellows CG, Aubin JE, Heersche JNM. Initiation
23. Hunziker EB, Kapfinger E, Geiss J. The structural and progression of mineralization of bone nodules
architecture of adult mammalian articular cartilage formed in vitro: the role of alkaline phosphatase
evolves by a synchronized process of tissue resorp- and organic phosphate. Bone Miner 1991;14(1):
tion and neoformation during postnatal develop- 2740
ment. Osteoarthritis Cartilage 2007;15(4):403413 41. Rey C, Beshah K, Griffin R, Glimcher MJ. Structural
24. Namba RS, Meuli M, Sullivan KM, Le AX, Adzick studies of the mineral phase of calcifying cartilage.
NS. Spontaneous repair of superficial defects in J Bone Miner Res 1991;6(5):515525

5 The CartilageBone Interface

articular cartilage in a fetal lamb model. J Bone 42. Hoemann CD, El-Gabalawy H, McKee MD. In vitro
Joint Surg Am 1998;80(1):410 osteogenesis assays: influence of the primary cell
25. Hayes AJ, MacPherson S, Morrison H, Dowthwaite source on alkaline phosphatase activity and miner-
G, Archer CW. The development of articular carti- alization. Pathol Biol (Paris) 2009;57(4):318323
lage: evidence for an appositional growth mecha- 43. Wang FY, Ying Z, Duan XJ, et al. Histomorphomet-
nism. Anat Embryol (Berl) 2001;203(6):469479 ric analysis of adult articular calcified cartilage
26. Oegema TR Jr, Carpenter RJ, Hofmeister F, Thomp- zone. J Struct Biol 2009;168(3):359365
son RC Jr. The interaction of the zone of calcified 44. Lane LB, Bullough PG. Age-related changes in the
cartilage and subchondral bone in osteoarthritis. thickness of the calcified zone and the number
Microsc Res Tech 1997;37(4):324332 of tidemarks in adult human articular cartilage. J
27. Heinegrd D, Oldberg A. Structure and biology of Bone Joint Surg Br 1980;62(3):372375
cartilage and bone matrix noncollagenous macro- 45. Mller-Gerbl M, Schulte E, Putz R. The thickness of
molecules. FASEB J 1989;3(9):20422051 the calcified layer of articular cartilage: a function
28. Gannon JM, Walker G, Fischer M, Carpenter R, of the load supported? J Anat 1987;154:103111
Thompson RC Jr, Oegema TR Jr. Localization of type 46. Hunziker EB, Quinn TM, Huselmann HJ. Quan-
X collagen in canine growth plate and adult canine titative structural organization of normal adult
articular cartilage. J Orthop Res 1991;9(4):485494 human articular cartilage. Osteoarthritis Cartilage
29. Jiang J, Leong NL, Mung JC, Hidaka C, Lu HH. Interac- 2002;10(7):564572
tion between zonal populations of articular chon- 47. Lyons TJ, McClure SF, Stoddart RW, McClure J.
drocytes suppresses chondrocyte mineralization The normal human chondro-osseous junctional
and this process is mediated by PTHrP. Osteoarthri- region: evidence for contact of uncalcified carti-
tis Cartilage 2008;16(1):7082 lage with subchondral bone and marrow spaces.
30. Sundaramurthy S, Mao JJ. Modulation of endo- BMC Musculoskelet Disord 2006;7:52
chondral development of the distal femoral 48. Frisbie DD, Cross MW, McIlwraith CW. A compar-
condyle by mechanical loading. J Orthop Res ative study of articular cartilage thickness in the
2006;24(2):229241 stifle of animal species used in human pre-clinical
31. Stephens M, Kwan APL, Bayliss MT, Archer CW. studies compared to articular cartilage thickness
Human articular surface chondrocytes initiate in the human knee. Vet Comp Orthop Traumatol
alkaline phosphatase and type X collagen synthe- 2006;19(3):142146
sis in suspension culture. J Cell Sci 1992;103(Pt 4): 49. Madry H, van Dijk CN, Mueller-Gerbl M. The basic
11111116 science of the subchondral bone. Knee Surg Sports
32. Ali SY, Sajdera SW, Anderson HC. Isolation and Traumatol Arthrosc 2010;18(4):419433
characterization of calcifying matrix vesicles 50. Koszyca B, Fazzalari NL, Vernon-Roberts B. Calci-
from epiphyseal cartilage. Proc Natl Acad Sci USA fied cartilage, subchondral and cancellous bone
1970;67(3):15131520 morphometry within the knee of normal subjects.
33. Shum L, Nuckolls G. The life cycle of chondro- Knee 1996;3:1522
cytes in the developing skeleton. Arthritis Res 51. Burr DB, Schaffler MB. The involvement of subchon-
2002;4(2):94106 dral mineralized tissues in osteoarthrosis: quan-
34. Robinson RA, Cameron DA. Electron microscopy titative microscopic evidence. Microsc Res Tech
of the primary spongiosa of the metaphysis at the 1997;37(4):343357
distal end of the femur in the newborn infant. J
Bone Joint Surg Am 1958;40-A(3):687697
52. Kaabar W, Iaklouk A, Bunk O, Baily M,
Farquharson MJ, Bradley D. Compositional and

Stannard_9781604068580_Ch05.indd 75 1/30/13 2:35 PM

structural studies of the bone-cartilage interface 67. Mithoefer K, Williams RJ III, Warren RF, et al.
using PIXE and SAXS techniques. Nuclear Instru- The microfracture technique for the treatment
ments & Methods in Physics Research Section of articular cartilage lesions in the knee. A pro-
Accelerators Spectrometers Detectors and Associ- spective cohort study. J Bone Joint Surg Am
ated Equipment 2010;619(13):7882 2005;87(9):19111920
53. Zoeger N, Roschger P, Hofstaetter JG, et al. Lead 68. Knutsen G, Engebretsen L, Ludvigsen TC, et al.
accumulation in tidemark of articular cartilage. Autologous chondrocyte implantation compared
Osteoarthritis Cartilage 2006;14(9):906913 with microfracture in the knee. A randomized
54. Marchand C, Chen HM, Buschmann MD, trial. J Bone Joint Surg Am 2004;86-A(3):455464
Hoemann CD. Standardized three-dimensional 69. Hurtig MB, Buschmann MD, Fortier LA, et al. Pre-
volumes of interest with adapted surfaces for clinical Studies for Cartilage Repair: Recommen-
more precise subchondral bone analyses by dations from the International Cartilage Repair
micro-computed tomography. Tissue Eng Part C Society. Cartilage. 2011;2(2):137152
Methods 2011;17(4):475484 70. Frisbie DD, Morisset S, Ho CP, Rodkey WG, Stead-
55. Zizak I, Roschger P, Paris O, et al. Characteristics man JR, McIlwraith CW. Effects of calcified car-
of mineral particles in the human bone/cartilage tilage on healing of chondral defects treated
interface. J Struct Biol 2003;141(3):208217 with microfracture in horses. Am J Sports Med
56. Duer MJ, Frisci T, Murray RC, Reid DG, Wise ER. 2006;34(11):18241831
The mineral phase of calcified cartilage: its molec- 71. Hoemann CD, Hurtig M, Rossomacha E, et al.
II Science and Techniques for Cartilage Repair

ular structure and interface with the organic Chitosan-glycerol phosphate/blood implants
matrix. Biophys J 2009;96(8):33723378 improve hyaline cartilage repair in ovine microf-
57. Lyons TJ, Stoddart RW, McClure SF, McClure J. racture defects. J Bone Joint Surg Am 2005;87(12):
The tidemark of the chondro-osseous junction 26712686
of the normal human knee joint. J Mol Histol 72. Hoemann CD, Kandel R, Roberts S, et al. Inter-
2005;36(3):207215 national Cartilage Repair Society (ICRS) Recom-
58. Miller LM, Novatt JT, Hamerman D, Carlson CS. mended Guidelines for Histological Endpoints for
Alterations in mineral composition observed in Cartilage Repair Studies in Animal Models and
osteoarthritic joints of cynomolgus monkeys. Clinical Trials. Cartilage 2011;2(2):153172
Bone 2004;35(2):498506 73. Kon E, Mutini A, Arcangeli E, et al. Novel nano-
59. Burr DB. Anatomy and physiology of the mineral- structured scaffold for osteochondral regenera-
ized tissues: role in the pathogenesis of osteoar- tion: pilot study in horses. J Tissue Eng Regen Med
throsis. Osteoarthritis Cartilage 2004;12(Suppl A): 2010;4(4):300308
S20S30 74. Shapiro F, Koide S, Glimcher MJ. Cell origin and
60. Marchand C, Chen G, Tran-Khanh N, et al. Micro- differentiation in the repair of full-thickness
drilled cartilage defects treated with thrombin- defects of articular cartilage. J Bone Joint Surg Am
solidified chitosan/blood implant regenerate a 1993;75(4):532553
more hyaline, stable, and structurally integrated 75. Chevrier A, Hoemann CD, Sun J, Buschmann MD.
osteochondral unit compared to drilled controls. Temporal and spatial modulation of chondrogenic
Tissue Eng Part A 2012;18(56):508519 foci in subchondral microdrill holes by chitosan-
61. Cameron ML, Briggs KK, Steadman JR. Reproduc- glycerol phosphate/blood implants. Osteoarthritis
ibility and reliability of the outerbridge classi- Cartilage 2011;19(1):136144
fication for grading chondral lesions of the knee 76. Hoemann CD, Chen G, Marchand C, et al.
arthroscopically. Am J Sports Med 2003;31(1): Scaffold-guided subchondral bone repair: impli-
8386 cation of neutrophils and alternatively activated
62. Mithoefer K, McAdams TR, Scopp JM, Mandelbaum arginase-1 macrophages. Am J Sports Med 2010;
BR. Emerging options for treatment of articular 38(9):18451856
cartilage injury in the athlete. Clin Sports Med 77. Chevrier A, Hoemann CD, Sun J, Buschmann MD.
2009;28(1):2540 Chitosan-glycerol phosphate/blood implants
63. Peterson L, Minas T, Brittberg M, Nilsson A, increase cell recruitment, transient vasculariza-
Sjgren-Jansson E, Lindahl A, Sjgren-Jansson E. tion and subchondral bone remodeling in drilled
Two- to 9-year outcome after autologous chon- cartilage defects. Osteoarthritis Cartilage 2007;
drocyte transplantation of the knee. Clin Orthop 15(3):316327
Relat Res 2000;374(374):212234 78. Chen H, Hoemann CD, Sun J, et al. Depth of
64. Drobnic M, Radosavljevic D, Cr A, Brittberg M, subchondral perforation influences the outcome
Strazar K. Debridement of cartilage lesions before of bone marrow stimulation cartilage repair. J
autologous chondrocyte implantation by open Orthop Res 2011;29(8):11781184
or transarthroscopic techniques: a comparative 79. Insall JN. Intra-articular surgery for degenera-
study using post-mortem materials. J Bone Joint tive arthritis of the knee. A report of the work of
Surg Br 2010;92(4):602608 the late K. H. Pridie. J Bone Joint Surg Br 1967;
65. Mika J, Clanton TO, Pretzel D, Schneider G, 49(2):211228
Ambrose CG, Kinne RW. Surgical preparation for 80. Guzmn-Morales J, Lafantaisie-Favreau C, Sun J,
articular cartilage regeneration without penetra- Rivard G, Hoemann CD. Analysis of the mid-term
tion of the subchondral bone plate: in vitro and effects of chitosan-NaCl/blood pre-solidified
in vivo studies in humans and sheep. Am J Sports implants in an in vivo osteochondral repair model.
Med 2011;39(3):624631 Paper presented at: International Cartilage Repair
66. Steadman JR, Rodkey WG, Singleton SB, Briggs KK. Society, 25 Sept, 2010; Barcelona, Spain.
Microfracture technique for full-thickness chon- 81. Qiu YS, Shahgaldi BF, Revell WJ, Heatley FW. Obser-
dral defects: technique and clinical results. Oper vations of subchondral plate advancement during
76 Tech Orthop 1997;7(4):300304 osteochondral repair: a histomorphometric and

Stannard_9781604068580_Ch05.indd 76 1/30/13 2:35 PM

mechanical study in the rabbit femoral condyle. histology scores in biopsies obtained from a ran-
Osteoarthritis Cartilage 2003;11(11):810820 domized controlled clinical trial comparing BST-
82. Chen HM, Chevrier A, Hoemann CD, Sun J, CarGel versus microfracture. Paper presented
Ouyang W, Buschmann MD. Characterization at: International Cartilage Repair Society, 27 Sept,
of subchondral bone repair for marrow-stim- 2010; Barcelona, Spain
ulated chondral defects and its relationship to 88. Mthot S, Hoemann CD, Rossomacha E, et al. ICRS
articular cartilage resurfacing. Am J Sports Med Histology Scores of Biopsies from an Interim Analy-
2011;39(8):17311740 sis of a Randomized Controlled Clinical Trial Show
Significant Improvement in Tissue Quality at 13
83. Mainil-Varlet P, Van Damme B, Nesic D,
Months for BST-CarGel versus Microfracture. Paper
Knutsen G, Kandel R, Roberts S. A new histology
presented at: International Cartilage Repair Society,
scoring system for the assessment of the quality of
27 Sept, 2010; Barcelona, Spain
human cartilage repair: ICRS II. Am J Sports Med
2010;38(5):880890 89. Changoor A, Nelea M, Mthot S, et al. Structural
characteristics of the collagen network in human
84. Saris DB, Vanlauwe J, Victor J, et al. Character- normal, degraded and repair articular carti-
ized chondrocyte implantation results in better lages observed in polarized light and scanning
structural repair when treating symptomatic car- electron microscopies. Osteoarthritis Cartilage
tilage defects of the knee in a randomized con- 2011;19(12):14581468
trolled trial versus microfracture. Am J Sports Med
2008;36(2):235246 90. Roos EM, Engelhart L, Ranstam J, et al. ICRS Recom-
mendation Document: Patient-Reported Outcome
85. Vanlauwe J, Saris DBF, Victor J, Almqvist KF, Instruments for Use in Patients with Articular Car-
Bellemans J, Luyten FP; TIG/ACT/01/2000&EXT tilage Defects. Cartilage 2011;2(2):122136
Study Group. Five-year outcome of character-
91. Lascau-Coman V, Buschmann MD, Hoemann CD.
ized chondrocyte implantation versus microf-
Rapid EDTA Microwave Decalcification of Rabbit
racture for symptomatic cartilage defects of the
Osteochondral Samples Preserves Enzyme Activ-
knee: early treatment matters. Am J Sports Med
ity and Antigen Epitopes. J Bone Miner Res 2008;
86. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson

5 The CartilageBone Interface

92. Chevrier A, Rossomacha E, Buschmann MD,
O, Peterson L. Treatment of deep cartilage defects Hoemann CD. Optimization of histoprocess-
in the knee with autologous chondrocyte trans- ing methods to detect glycosaminoglycan, col-
plantation. N Engl J Med 1994;331(14):889895 lagen type II, and collagen type I in decalcified
87. Hoemann CD, Tran-Khan N, Mthot S, et al. Cor- rabbit osteochondral sections. J Histotechnol
relation of tissue histomorphometry with ICRS 2005;28(3):165175


Stannard_9781604068580_Ch05.indd 77 1/30/13 2:35 PM

Science and Animal Models of
Marrow Stimulation for Cartilage
Lisa A. Fortier, Brian J. Cole, and C. Wayne McIlwraith

Microfracture perforation of the subchon- aspirate from the iliac crest and concentrated
dral bone for cartilage repair was originally by centrifugation.7 The two cellular popula-
described by Steadman in 1994.1 Microfracture tions were different with respect to cell sur-
involves penetration of the subchondral bone face markers. Neither cell type carried CD34
plate with an arthroscopic awl to allow bone or CD45 marker expression, suggesting that
marrow contents to fill the defect and form there were no hematopoietic cells in either
a superclot.2 In humans1,3,4 and nonhuman bone marrow aspirate concentrate or micro-
primates,5 microfracture results in increased fracture superclot. This result might suggest
tissue volume and improved patient comfort that neither cell source is derived from the
and function for an average of 2 to 3 years. bone marrow, but it must be interpreted
There are other described methods of bone with great caution because both cell sources
marrow stimulation such as drilling and abra- were cultured for at least two passages and
sion, but less research and clinical data are the cells were treated with trypsin before
available to critically evaluate the efficacy of flow cytometry analysis, both of which have
these techniques. been documented to alter cell surface protein
expression on stem cells.8,9 In a similar study
in which cells were derived from subchon-
The Superclot dral corticospongious bone and cultured over
time, the cells retained their multilineage
In theory, enhanced cartilage repair fol- potential to undergo trilineage differentia-
lowing microfracture is the result of the tion into cartilage, adipose, and bone phe-
superclot thought to be laden with bone notypes.10 Interestingly, MSC-based cartilage
marrow-derived mesenchymal stem cells studies continue to focus predominantly on
(MSCs) and growth factors.6 Although there the ability of the cells to differentiate into
have been several in vitro and in vivo animal and form neocartilage despite the growing
studies aimed at understanding how micro- evidence that MSCs function at least in part
fracture repair tissue remodels over time, it to modulate the local environment through
has never been well documented that the a paracrine effect and recruitment of other
superclot contains MSCs or growth factors. In progenitor cells and immunomodulation.11,12
a small study of 11 human patients with fem- Understanding the source and type of cells
oral condylar defects, superclot from micro- that populate microfracture defects is criti-
fracture was compared with bone marrow cally important. There are a number of studies

Stannard_9781604068580_Ch06.indd 78 1/30/13 2:40 PM

that have evaluated the various effects of or in older males, underscore the need for a
drugs, growth factors, devices, scaffolds, gene more refined understanding of the basic biol-
therapy, and rehabilitation on microfracture. ogy of microfracture.
Some of these modifying factors are being
promoted and even marketed predicated
on the concept that they enhance chemo- Animal Model Studies
taxis, adherence, and/or proliferation of bone
marrow-derived MSCs.1319 These cited stud- Animal model studies provide insight into
ies represent only a few of the many studies temporal changes following microfrac-
investigating the use of scaffolds, devices, ture (Fig. 6.1). Early animal model studies
and drugs, in vitro and in rabbit, canine, on microfracture repair were done in the
ovine, laprine, or equine animal models for horse.22,23 The horse model was also used

6 Science and Animal Models of Marrow Stimulation

augmentation of microfracture to enhance to validate the subjective clinical impres-
articular cartilage repair. This intense level of sion that removal of the calcified cartilage
investigation into scaffold/device-augmented layer was important to optimize volume
microfracture and its potential recruitment and attachment of repair tissue.24 Further
of MSCs lies in the thought that these tech- equine studies indicated that the volume
nologies could improve the clinical results of repair tissue did not change between 4
of microfracture alone and the relative ease and 12 months postmicrofracture in direct
and marketability of such technologies when weight-bearing sites (distal medial femoral
compared with cultured or manipulated condyle and distal radiocarpal bones), which
stem cell articular cartilage grafts. at a minimum suggests that the repair tis-
If the cell population of the subchondral sue did not deteriorate by 12 months post-
bone is truly different from that of bone operatively.22 Histologic assessment revealed
marrow aspirated from a bone marrow that there was more type II collagen present
space, then perhaps the results of in vitro at 12 months than at 4 months, suggesting
studies done on bone marrow aspirate or continued chondrogenic maturation of repair
metaphyseal-derived MSCs are not directly tissue to 12 months, but the aggrecan con-
applicable to microfracture, in which the cell tent remained far below normal.
is likely derived from the subchondral bone To provide information in a physiologic
plate in the area 2 to 4 mm underlying the and anatomic environment more closely
calcified cartilage layer.13,15 During the pro- related to the human, similar studies were
cess of maturation, the cell population in a performed in cynomolgus macaques.5 In
superclot might be composed of cells derived this study, repair tissue was studied at 6 and
from the bone marrow, subchondral bone, 12 weeks postmicrofracture and indicated
surrounding host cartilage, synovium, syno- that the repair tissue underwent progressive
vial fluid, or a combination thereof. Studies chondrogenic remodeling during this time
are routinely performed in vitro, and using period based on postmortem gross and his-
bone marrow-derived MSCs to investigate tologic assessments. It is interesting to note
a method to improve microfracture and that progressive maturation of microfracture
the results can change clinical practice. For repair tissue is not appreciated using arthros-
example, a recent study showed that chon- copy with validated categorical scoring sys-
drogenic differentiation of bone marrow- tems,25 which makes it difficult for a surgeon
derived MSCs is impaired by rheumatoid to make decisions regarding success based on
arthritis synovial fluid as compared with arthroscopic observation only.26 Noninvasive
synovial fluid from patients with osteoarthri- dGEMRIC and T2 mapping has been used to
tis or normal patients.20 Another study sug- evaluate repair tissue following microfrac-
gested that age in males, but not in females, ture at 24 and 48 weeks postoperatively in
negatively affects their ability to undergo a goat model.27 The achieved objective of the
chondrogenic differentiation.21 The potential study was to validate dGEMRIC and T2 map-
clinical ramifications of this study, where cli- ping as surrogate markers of biochemical and
nicians might presume failure of microfrac- histologic integrity of repair tissue. In addi-
ture in patients with rheumatoid arthritis tion, the study was the first to demonstrate 79

Stannard_9781604068580_Ch06.indd 79 1/30/13 2:40 PM

II Science and Techniques for Cartilage Repair

Fig. 6.1 Schematic representation of microfracture and type II collagen content (not depicted) progres-
maturation over time. (a) At time 0, the cartilage sively increase but remain low compared with normal
defect is debrided to include removal of calcified tissue. Chondrocyte cloning is evident in the adjacent
cartilage. Microfracture is performed to a depth host cartilage. The microfracture holes progressively
of 2 to 4 mm to penetrate the subchondral plate, heal during this time period. (c) At 12 months post-
thereby allowing bone marrow to gain access to the microfracture, the repair tissue has improved cellular
cartilage defect and form a superclot. (b) At 4 to organization and proteoglycan content but not type II
6 months postmicrofracture, there is progressive collagen. The microfracture holes are healed, and in
chondrogenic remodeling of the fibrocartilage repair some instances the subchondral bone is sclerotic and/
tissue filling the defect. The repair tissue is hypercel- or extends into the cartilage defect forming a central
80 lular. Proteoglycan (purple in base of repair tissue) osteophyte.

Stannard_9781604068580_Ch06.indd 80 1/30/13 2:40 PM

increased glycosaminoglycan and total col-
lagen content between 24 and 48 weeks
postmicrofracture measured with both R1
(1/s) and high-performance liquid chroma-
tography. Combined, these results suggest
that microfracture continues to mature for
the first 12 months after surgery, but the
lack of normal matrix molecules translates
to tissue with inferior biomechanical proper-
ties compared with normal cartilage, which
renders the repair tissue prone to injury and
deterioration. Based on animal model studies,

6 Science and Animal Models of Marrow Stimulation

it is unclear what biochemical or mechani-
cal changes happen beyond 12 months and
when, why, or how microfracture repair tis-
sue fails or not. In unpublished data by L.A.F,
2-year data are being analyzed in the horse.
Clinically, it may have less to do with the Fig. 6.2 Fast-spin echo magnetic resonance image
breakdown of microfracture repair tissue than (MRI) of a microfracture-treated defect on the lateral
with ability of the repair tissue to shield trochlear ridge of the femur, 12 months after surgery.
the subchondral bone from load that is theo- Subchondral bone sclerosis (white arrows) and protru-
sion of the subchondral plate into the cartilage defect
retically associated with the manifestation (black arrow) are evident.
of symptoms. If this theory is correct, then
methods to enhance or retain proteoglycan
content in the repair tissue would increase model in both the distal femur (Fig. 6.3)22,29
the compressive stiffness of the repair tissue and lateral trochlear ridge (Fig. 6.2)30 in
and should improve long-term results. ovine,14 and in nonhuman primates.5 Central
osteophyte formation clearly does not occur
in every case of microfracture, and there are
Central Osteophyte too few instances in the animal model studies
for robust observations into causality.
Formation and Subchondral Subchondral bone sclerosis has also been
Bone Sclerosis noted following microfracture in horses
when the repair tissue was assessed with
Microfracture has long been thought of as a radiographs or magnetic resonance imaging
cant hurt or burn no bridges type of pro- (MRI).26,30 Most animal studies evaluate repair
cedure. However, in more recent years, there with histology and not with radiographs or
is heightened awareness and concern about MRI, making assessment of subchondral bone
the formation of central/intralesional osteo- sclerosis difficult and subjective. Sclerosis of
phytes, which are protrusions of subchondral the subchondral bone has been postulated
bone extending above the level of the adja- as an initiating event in the development of
cent, normal subchondral plate (Fig. 6.2).28 osteoarthritis.3133 It should be restated that
Formation of central osteophytes is not spe- the animal model studies are limited to 1-year
cifically investigated a priori or mentioned duration so the long-term presence or conse-
in most animal studies despite being quite quences of this subchondral bone sclerosis on
obvious in figures contained in published the microfracture repair tissue or clinical out-
articles, irrespective of the animal model come of the patient are not evident.
studied. Figures presented in articles can be Microfracture by definition is fracturing
too high in magnification or focused on the of the subchondral bone, and the results of
repair-host tissue interface to appreciate cen- subchondral bone sclerosis or central osteo-
tral osteophyte formation. It should be noted phyte formation might be anticipated know-
that central osteophyte formation has been ing the natural course of healing following
observed in microfracture defects in the horse microfracture of cancellous subchondral bone. 81

Stannard_9781604068580_Ch06.indd 81 1/30/13 2:40 PM

remembered and investigated simultaneously
when developing technologies for augmenta-
tion of microfracture.

Subchondral Cystic
In animal models when the medial femoral
condyle is used as the treatment site, viola-
tion of the subchondral bone plate can result
in formation of subchondral bone cysts.3739
In preparation of a cartilage bed for microf-
racture, overexuberant debridement of the
calcified cartilage layer to include removal
II Science and Techniques for Cartilage Repair

of the subchondral bed can lead to subchon-

dral cyst formation.39 Precise attention to the
technical aspects of microfracture and the
use of skeletally mature animals (tidemark is
fully formed and the calcified cartilage layer
is visible) are crucial for successful modeling.
Radiolucent cyst-like areas in the medial
femoral condyle have been observed follow-
ing microfracture, but there was no evidence
of a cyst on histologic analysis.22 Although
Fig. 6.3 Histologic appearance of a microfracture- MRI was not performed, the authors were of
treated defect on the medial femoral condyle,
the opinion that the radiolucency represented
12 months after surgery. The fibrocartilage is well
adhered to the surrounding normal cartilage tissue bone edema.
and to the underlying, protruding new subchondral
bone. (Reproduced with permission from Frisbie,
et al. Vet Surg 1999;28(4):242255.) Microfracture Compared
with Microdrilling
Trabecular microfractures of the femoral head, In a rabbit study comparing microdrill-
spine, patella, and acetabulum have been ing with microfracture at a depth of 2 mm,
studied since the 1960s.34 These naturally microcomputed tomography imaging per-
occurring microfractures heal with woven formed 1 day postoperatively indicated that
bone microcallus. It is reasonable to presume microfracture led to more compaction of
that penetration of the subchondral plate with bone in the holes than did microdrilling.40
a microfracture awl to gain access to bone The authors concluded that this impaction of
marrow elements stimulates a similar bone bone might impede the ability of bone mar-
repair response. What circumstances lead to row to reach the articular defect and thereby
an overexuberant reaction with resultant cen- might negatively affect repair. Bleeding in
tral osteophyte formation is not clear. Bone only one of four microfracture holes was
repair/regeneration is complex and is influ- observed intraoperatively, but all defects
enced by many factors including age, mechan- were filled with a blood clot. The lack of
ical and cellular environments, bone mineral bleeding from the microfracture holes has not
content, and genetics.34,35 There are also dif- been reported, nor is it consistent with the
ferences in the response of cells to mechani- clinical experiences of the authors in humans
cal loading, and this too might influence cells or horses. Thus, it is likely a flaw of the rab-
in the superclot to differentiate down osteo- bit as an animal model or, more likely, as
genic or chondrogenic lineages.33,36 The ability the authors suggested, a result of the type of
82 of progenitor cells to differentiate into osteo- homemade microfracture awl specifically cre-
genic or chondrogenic cell lines should be ated for the study, which had a collar to limit

Stannard_9781604068580_Ch06.indd 82 1/30/13 2:40 PM

the depth of penetration to 2 mm. The col-
lar likely restricted movement of bone from Video 6.1
the microfracture holes, creating impaction
fractures in the subchondral bone. However, Continued impaction of bone from
impaction of subchondral bone surrounding the depth to the surface of the
the microfracture hole is seen using standard microfracture hole.
arthroscopic microfracture awls without a Online content including video
collar (Fig. 6.4; Videos 6.1, 6.2). Microdrilling sequences viewable at www.thieme-

6 Science and Animal Models of Marrow Stimulation

Video 6.2
Impacted bone surrounding the entire
circumference of the microfracture
Online content including video
sequences viewable at www.thieme-

might be as effective as microfracture but

obviously requires more surgical instrumen-
tation such as a drill compared with a hand-
held awl to generate a superclot.
In summary, basic science and animal
model studies indicate that microfracture
results in improved repair tissue that con-
tinues to mature and becomes more carti-
laginous for at least 1 year after surgery. The
superclot clearly remodels but does remain
quite inferior to normal articular cartilage in
matrix molecule composition and therefore
biomechanical function. Numerous studies
have been performed to augment microfrac-
ture even though we dont fully understand
the fundamental biology of microfracture
and therefore how to improve upon current
results. A potential detriment to the use of
microfracture is the formation of central/
intralesional osteophytes, which are unpre-
Fig. 6.4 Microcomputed tomography of normal dictable and have been associated with per-
equine lateral trochlear ridge subjected to micro- sistent or recurrent pain in human studies.
fracture. (a) Axial view of a microfracture hole dem- Microfracture remains a commonly per-
onstrating impaction of surrounding subchondral formed and investigated cartilage repair pro-
bone. (b) Sagittal view of a microfracture hole dem-
onstrating impaction of surrounding subchondral
cedure because it is easy to do and requires
minimal equipment, and clinical results in 83
bone. Minimum-intensity projection of micro-CT data
acquired with 4-m x-y-z voxel size. human patients are encouraging.

Stannard_9781604068580_Ch06.indd 83 1/30/13 2:40 PM

Acknowledgment from subchondral spongious bone marrow. J
Orthop Res 2007;25(10):12991307
14. Erggelet C, Neumann K, Endres M, Haberstroh K,
The authors thank Ms. Paula Sharp for her Sittinger M, Kaps C. Regeneration of ovine articu-
assistance in the generation of this chapter lar cartilage defects by cell-free polymer-based
implants. Biomaterials 2007;28(36):55705580
and Mark Riccio for generation of the micro-
15. Endres M, Andreas K, Kalwitz G, et al. Chemokine
CT images. profile of synovial fluid from normal, osteoar-
thritis and rheumatoid arthritis patients: CCL25,
CXCL10 and XCL1 recruit human subchondral
References mesenchymal progenitor cells. Osteoarthritis Car-
tilage 2010;18(11):14581466
1. Rodrigo JJ, Steadman JR, Silliman JF, Fulstone AH. 16. Kang SW, Bada LP, Kang CS, et al. Articular carti-
Improvement of full-thickness chondral defect lage regeneration with microfracture and hyalu-
healing in the human knee after debridement and ronic acid. Biotechnol Lett 2008;30(3):435439
microfracture using continuous passive motion. 17. Kramer J, Bhrnsen F, Lindner U, Behrens P,
Am J Knee Surg 1994;7:109116 Schlenke P, Rohwedel J. In vivo matrix-guided
2. Steadman JR, Rodkey WG, Singleton SB, Briggs KK. human mesenchymal stem cells. Cell Mol Life Sci
Microfracture technique for full-thickness chon- 2006;63(5):616626
II Science and Techniques for Cartilage Repair

dral defects: technique and clinical results. Oper 18. Lee CH, Cook JL, Mendelson A, Moioli EK, Yao H,
Tech Orthop 1997;7:300304 Mao JJ. Regeneration of the articular surface of
3. Knutsen G, Engebretsen L, Ludvigsen TC, et al. the rabbit synovial joint by cell homing: a proof of
Autologous chondrocyte implantation compared concept study. Lancet 2010;376(9739):440 448
with microfracture in the knee. A randomized 19. Kuo AC, Rodrigo JJ, Reddi AH, Curtiss S, Grotkopp E,
trial. J Bone Joint Surg Am 2004;86-A(3):455464 Chiu M. Microfracture and bone morphogenetic
4. Blevins FT, Steadman JR, Rodrigo JJ, Silliman J. protein 7 (BMP-7) synergistically stimulate artic-
Treatment of articular cartilage defects in ath- ular cartilage repair. Osteoarthritis Cartilage
letes: an analysis of functional outcome and lesion 2006;14(11):11261135
appearance. Orthopedics 1998;21(7):761767, 20. Krger JP, Endres M, Neumann K, Hupl T, Erggelet
discussion 767768 C, Kaps C. Chondrogenic differentiation of human
5. Gill TJ, McCulloch PC, Glasson SS, Blanchet T, subchondral progenitor cells is impaired by
Morris EA. Chondral defect repair after the micro- rheumatoid arthritis synovial fluid. J Orthop Res
fracture procedure: a nonhuman primate model. 2010;28(6):819827
Am J Sports Med 2005;33(5):680685 21. Payne KA, Didiano DM, Chu CR. Donor sex and age
6. Shapiro F, Koide S, Glimcher MJ. Cell origin and influence the chondrogenic potential of human
differentiation in the repair of full-thickness femoral bone marrow stem cells. Osteoarthritis
defects of articular cartilage. J Bone Joint Surg Am Cartilage 2010;18(5):705713
1993;75(4):532553 22. Frisbie DD, Trotter GW, Powers BE, et al. Arthro-
7. de Girolamo L, Bertolini G, Cervellin M, Sozzi G, scopic subchondral bone plate microfracture tech-
Volpi P. Treatment of chondral defects of the nique augments healing of large chondral defects
knee with one step matrix-assisted technique in the radial carpal bone and medial femoral con-
enhanced by autologous concentrated bone mar- dyle of horses. Vet Surg 1999;28(4):242255
row: in vitro characterisation of mesenchymal 23. Frisbie DD, Oxford JT, Southwood L, et al. Early
stem cells from iliac crest and subchondral bone. events in cartilage repair after subchondral bone
Injury 2010;41(11):11721177 microfracture. Clin Orthop Relat Res 2003(407):
8. Hackett CH, Flaminio MJ, Fortier LA. Analysis of 215227
CD14 expression levels in putative mesenchymal 24. Frisbie DD, Morisset S, Ho CP, Rodkey WG,
progenitor cells isolated from equine bone mar- Steadman JR, McIlwraith CW. Effects of calcified
row. Stem Cells Dev 2011;20(4):721735 cartilage on healing of chondral defects treated with
9. Radcliffe CH, Flaminio MJ, Fortier LA. Temporal microfracture in horses. Am J Sports Med 2006;
analysis of equine bone marrow aspirate dur- 34(11):18241831
ing establishment of putative mesenchymal pro- 25. van den Borne MP, Raijmakers NJ, Vanlauwe J, et al;
genitor cell populations. Stem Cells Dev 2010;19 International Cartilage Repair Society. Interna-
(2):269282 tional Cartilage Repair Society (ICRS) and Oswestry
10. Neumann K, Dehne T, Endres M, et al. Chondro- macroscopic cartilage evaluation scores validated
genic differentiation capacity of human mesen- for use in Autologous Chondrocyte Implantation
chymal progenitor cells derived from subchondral (ACI) and microfracture. Osteoarthritis Cartilage
cortico-spongious bone. J Orthop Res 2008;26 2007;15(12):13971402
(11):14491456 26. Fortier LA, McCarrel TM, Bradica G, et al. Evalua-
11. Bunnell BA, Betancourt AM, Sullivan DE. New tion of a biphasic graft for osteochondral repair in
concepts on the immune modulation mediated an equine model. Transactions of the 9th World
by mesenchymal stem cells. Stem Cell Res Ther Congress of the International Cartilage Repair
2010;1(5):34 Society. CD-ROM, 2010
12. Prockop DJ. Repair of tissues by adult stem/progen- 27. Watanabe A, Boesch C, Anderson SE, Brehm W,
itor cells (MSCs): controversies, myths, and chang- Mainil Varlet P. Ability of dGEMRIC and T2 map-
ing paradigms. Mol Ther 2009;17(6):939946 ping to evaluate cartilage repair after microfrac-
13. Endres M, Neumann K, Hupl T, et al. Synovial ture: a goat study. Osteoarthritis Cartilage 2009;17
fluid recruits human mesenchymal progenitors (10):13411349

Stannard_9781604068580_Ch06.indd 84 1/30/13 2:40 PM

28. Gomoll AH, Madry H, Knutsen G, et al. The subchon- 34. Fazzalari NL. Trabecular microfracture. Calcif
dral bone in articular cartilage repair: current Tissue Int 1993;53 (Suppl 01):S143S146, discus-
problems in the surgical management. Knee Surg sion S146S147
Sports Traumatol Arthrosc 2010;18(4):434447 35. Dimitriou R, Jones E, McGonagle D, Giannoudis PV.
29. McIlwraith CW, Frisbie DD, Rodkey WG, et al. Bone regeneration: current concepts and future
Evaluation of intra- articular mesenchymal stem directions. BMC Med 2011;9:66
cells to augment healing of microfractured chon- 36. Potier E, Noailly J, Ito K. Directing bone marrow-
dral defects. Arthroscopy 2011;27(11):15521561 derived stromal cell function with mechanics. J
30. Fortier LA, Potter HG, Rickey EJ, et al. Concentrated Biomech 2010;43(5):807817
bone marrow aspirate improves full-thickness 37. Kold SE, Hickman J, Melsen F. An experimental
cartilage repair compared with microfracture study of the healing process of equine chondral
in the equine model. J Bone Joint Surg Am 2010; and osteochondral defects. Equine Vet J 1986;
92(10):19271937 18(1):1824
31. Brandt KD, Mazzuca SA. Experience with a 38. Ray CS, Baxter GM, McIlwraith CW, et al. Develop-
placebo-controlled randomized clinical trial of ment of subchondral cystic lesions after articular

6 Science and Animal Models of Marrow Stimulation

a disease-modifying drug for osteoarthritis: the cartilage and subchondral bone damage in young
doxycycline trial. Rheum Dis Clin North Am 2006; horses. Eq Vet J 1996;28(3):225232
32(1):217234, xixii 39. McIlwraith CW, Fortier LA, Frisbie DD, Nixon AJ.
32. Burr DB, Radin EL. Microfractures and microc- Equine models of articular cartilage repair. Carti-
racks in subchondral bone: are they relevant to lage 2011;2(4):317326
osteoarthrosis? Rheum Dis Clin North Am 2003; 40. Chen H, Sun J, Hoemann CD, et al. Drilling and
29(4):675685 microfracture lead to different bone structure
33. Sanchez C, Pesesse L, Gabay O, et al. Regulation and necrosis during bone-marrow stimulation
of subchondral bone osteoblast metabolism by for cartilage repair. J Orthop Res 2009;27(11):
cyclic compression. Arthritis Rheum 2012;64(4): 14321438


Stannard_9781604068580_Ch06.indd 85 1/30/13 2:40 PM

Microfracture and Augments
Andreas H. Gomoll

Microfracture was originally developed The Microfracture Procedure

by Steadman in the 1980s in response to
perceived limitations of the then com- Indications
monly used marrow stimulation techniques
(MSTs), abrasion arthroplasty and subchon- The following indications are based on find-
dral (Pridie) drillingnamely, destabili- ings from multiple studies discussed fur-
zation of the subchondral plate and heat ther in the Results section. Microfracture
necrosis, respectively. Microfracture was is primarily indicated for the treatment of
quickly adopted for the treatment of carti- full-thickness articular cartilage defects
lage defects, first in the knee, followed by without significant bone loss (Outerbridge
the ankle, shoulder, elbow, and hip. The Grade 3 and 4; International Cartilage
technique aims to induce the formation of Research Society [ICRS] Grade 3) measuring
a reparative tissue by the creation of chan- less than 2 to 4 cm2 on the femoral condyles.
nels in the subchondral plate, allowing the Articular comorbidities such as malalignment
migration of mesenchymal stem cells (MSCs) and meniscal deficiency do not represent a
from the subchondral marrow space into the contraindication provided they are corrected
defect. Here, they differentiate and produce in a staged or concomitant fashion. Elevated
a fibrocartilaginous tissue to fill the defect. body mass index (BMI) over 30 kg/m2, defect
The underlying biology is discussed in size larger than 2 to 4 cm2, defect location in
greater detail by Dr. Fortier and co-authors the patellofemoral compartment or on the
in chapter 6. tibial plateau, and age older than 40 years are
This chapter will review the indications, associated with worse outcomes.
clinical application, rehabilitation, and out-
comes of the standard microfracture proce-
dure. In addition, it will present an overview Technique
of new technologies currently under devel-
opment that aim to augment microfracture Microfracture is generally performed as an
through the use of biomaterials and growth all-arthroscopic procedure utilizing stand-
factors in hopes of improving outcomes and ard anteromedial and anterolateral por-
broadening the indications. tals. Rarely, accessory portals may become

Stannard_9781604068580_Ch07.indd 86 1/30/13 2:41 PM

Fig. 7.1 Arthroscopic view of a femoral condyle Fig. 7.2 Same defect after debridement of degener-
cartilage defect. ated tissue and the layer of calcified cartilage with
creation of stable shoulders.

necessary for optimal access. When per- and destabilizing the subchondral plate. At
formed together with other intra-articular the end, the tourniquet should be deflated
procedures such as anterior cruciate ligament or the pump pressure lowered to observe fat

7 Microfracture and Augments

reconstruction or meniscal repair, microfrac- droplets and bone marrow from each hole;
ture should be performed last to preserve the otherwise, individual holes can be revisited
developing blood clot that could otherwise and deepened with the awl. The use of intra-
be irrigated away by the arthroscopic fluid. articular drains should be avoided because
The use of a tourniquet is optional. removal of the intra-articular hematoma
Once the entire joint has been carefully would be counterproductive to the formation
evaluated and any articular comorbidities of the desired marrow clot.
have been addressed, the cartilage defect
(Fig. 7.1) is prepared. First, all degenerated
cartilage is removed with a sharp curet, Rehabilitation
including any areas of surrounding cartilage
that is delaminated. Vertical shoulders of sta- The postoperative rehabilitation is a criti-
ble cartilage are thus created. Next, the layer cal and inherent component of the micro-
of calcified cartilage is removed with the fracture procedure, and its contribution to
curet; however, avoid excessive force that
can injure the subchondral plate (Fig. 7.2).
Generally, a motorized shaver can assist in
removing larger flaps but is inadequate to
appropriately prepare the defect by itself.
Microfracture awls are available with differ-
ent angled tips; depending on the location
of the defect, the awl providing perpendicu-
lar alignment of the tip to the defect surface
should be chosen. If the angle of placement
is too oblique, furrows are created rather
than holes, with increased damage to the
subchondral plate. Microfracture holes are
now created, starting at the periphery to
improve edge integration (Fig. 7.3). The
holes should be placed  3 to 4 mm apart
to prevent holes from becoming confluent Fig. 7.3 Same defect after microfracture.

Stannard_9781604068580_Ch07.indd 87 1/30/13 2:41 PM

the overall success cannot be overempha- defects larger than 2 to 4 cm2.7,912 The treat-
sized.13 Weightbearing restrictions are tai- ment of patellofemoral lesions with micro-
lored to the individual patient: small and fracture has been associated with worse
well-shouldered defects ( 1 cm2) require outcomes than the treatment of femoral con-
less protection than larger defects with dyles.11 Microfracture in patients older than
compromised shoulders. Generally, patients 35 to 40 years resulted in worse outcomes
are kept touch-down weightbearing on two than in younger patients.4,9,10,12,13 The influ-
crutches for 4 to 8 weeks. A continuous pas- ence of defect chronicity appears controver-
sive motion machine is started on postop- sial, with Steadman et al reporting no effect,
erative day 1 and continued for 6 weeks, 6 to while Mithoefer et al demonstrated better
8 hours per day, increasing range of motion outcomes with lesions less than 1 year old.4,6
(ROM) as tolerated. Quadriceps isometrics Finally, BMI over 25 to 30 kg/m2 appears neg-
and straight-leg raises can be started imme- atively correlated with outcomes.6,10
diately, adding resistance bands and mini- When comparing microfracture to other
squats at 2 months. More aggressive weight procedures, Knutsen et al demonstrated
II Science and Techniques for Cartilage Repair

training is delayed until 4 months post- overall comparable results to autologous

operatively. Impact sports, especially those chondrocyte implantation (ACI) for vari-
involving cutting or pivoting, should not be ous lesions sizes in an RCT,12 while Saris et
resumed until 6 to 9 months after the pro- al showed better histological and functional
cedure, and only once swelling has resolved outcomes with ACI.14,15 Coleman et al reported
and adequate muscle strength and proprio- a trial of microfracture versus ACI, showing
ception have returned. 44 versus 22% increased Cincinnati scores,
respectively. Magnetic resonance imaging
(MRI) scores were better for ACI; however,
Results this did not correlate with functional out-
comes.16 Basad et al specifically focused on
Several authors have reported on the out- size-related outcomes in an RCT of micro-
come of microfracture using both case series fracture versus ACI in defects larger than
and randomized controlled trials (RCTs). 4 cm2, which demonstrated better results
Steadmans group has the largest experi- for ACI.17 Kon et al presented results from a
ence with microfracture and has published cohort study comparing microfracture to ACI
extensively on this technique in various sub- with comparable results at 2 years, but worse
populations. They demonstrated symptom results for microfracture at 5 years.8 Gudas et
improvement in 80% of cases in a minimum al compared microfracture to osteochondral
7-year follow-up study of patients younger autograft transfer (OAT) in two RCTs. One
than 45 years without concomitant intra- demonstrated better arthroscopic, histo-
articular comorbidities; at 3 years post- logic, and MRI appearance, and higher return
operatively 16% rated themselves unchanged to play with OAT than microfracture (93 vs.
and 4% considered their symptoms worse 52%, respectively) in athletes.9 Gudas et als
than preoperatively.4 A more challenging second RCT randomized patients with osteo-
group of patients with degenerative defects chondritis dissecans lesions to OAT versus
reported improvements in pain and function; microfracture, showing better outcome with
13 of 81 (16%) patients required repeat sur- OAT at 4 years (83 vs. 63%, respectively).18
gery for lysis of adhesions, and 5 (6%) patients Several studies used MRI to evaluate the
underwent repeat microfracture or revision quality of the repair tissue at follow-up,
to arthroplasty at an average of 23 months reporting good and excellent fill in approxi-
postoperatively.5 In their experience, results mately half of patients or fewer.6,9,19,20 Poor
were not affected by lesion size or loca- fill on MRI correlated with worsening symp-
tion.4 Other groups have reported good and toms after an initial period of improvement.6
excellent results in 60 to 80% of patients4,69 When serial MRIs were performed, quality
but have recommended more narrow indi- improved in the early postoperative period
cations, most reporting worse outcomes in up to 2 years.6

Stannard_9781604068580_Ch07.indd 88 1/30/13 2:41 PM

Complications quality and functional outcomes has led to
the development of various biomaterials
Significant surgical complications are rare for implantation in conjunction with MSTs.
with this arthroscopic procedure. Depending Several of these treatment approaches will
on lesion size and location, patients can expe- be discussed briefly below.
rience catching until the defect has filled
with repair tissue. Over the mid to long term, Drilling
microfracture has been shown to result in
the formation of intralesional osteophytes, Pridie developed subchondral drilling as
subchondral sclerosis, and cysts in up to 50% of treatment for cartilage defects in the 1950s
patients (Fig. 7.4).6,9,20 The potential influence and reported patient satisfaction of 77%; 64%
of these subchondral changes on subsequent of the knees were rated as good.26 Utilizing an
revision surgery with ACI has been reported in open arthrotomy and postoperative immo-
several studies. Some investigators reported bilization to perform the drilling, however,
no negative influence of prior microfracture resulted in a high number of patients with
in subanalyses of studies designed for gen- stiffness, and heat necrosis of the subchondral
eral outcomes after ACI.21,22 Conversely, two bone was a concern. The rise of arthroscopic
publications specifically tailored to investi- instrumentation facilitated the development
gate this question reported failure rates of ACI of microfracture, addressing both the need
after prior marrow stimulation that were up for an open approach as well as any concern
to three times the failure rates seen in ACI in for heat necrosis.

7 Microfracture and Augments

not previously treated defects.23,24 More recently, several studies have pointed
to certain benefits of drilling over micro-
fracture, leading to a possible renaissance
Augmentation Techniques of this procedure. Since the current drilling
technique is performed arthroscopically in
Although microfracture provides good short- an aqueous environment, heat necrosis is of
term outcomes for many patients, it results lesser or no concern. Animal models have
in a fibrocartilaginous, rather than hyaline- demonstrated that drilling with actual drill
like, repair tissue.25 Mid- to long-term stud- bits (rather than smooth K-wires) results
ies have demonstrated gradual decreases in in better marrow clot formation. This has
functional outcomes after 24 to 36 months, been explained by the deeper channels cre-
potentially due to tissue degradation over ated with drilling rather than microfracture,
time.8,11 Increased interest in augmenting the which can access more of the subchondral
bodys own reparative response to improve marrow space. Also, while microfracture

Fig. 7.4 Intralesional osteophyte of the medial femoral condyle after failed microfracture depicted by computed
tomography arthrogram (a), arthroscopy (b), and surgical image (c) during revision with ACI.

Stannard_9781604068580_Ch07.indd 89 1/30/13 2:41 PM

compacts the bone around the hole and marrow stimulation (some studies use drill-
potentially seals off the marrow space, drill- ing rather than microfracture to generate the
ing removes bone and allows for more blood marrow clot).
flow into the defect. A rabbit model demon-
strated more complete fill and higher-quality
tissue after subchondral drilling than micro- Autologous Matrix-Induced Chondrogenesis
fracture.27,28 The same group also demon- (AMIC; Chondro-Gide, Geistlich Biomaterials,
strated better repair of the subchondral bone Switzerland)
with deeper drilling rather than the more
shallow microfracture, with less incidence of AMIC is a commonly used augmentation tech-
cystic and sclerotic abnormalities.28,29 Even nique in Europe that utilizes Chondro-Gide,
though many of the following techniques a type I/III collagen membrane to stabilize
have been tested with microfracture, drilling the clot in a marrow-stimulated defect. As
could be substituted, and additional studies such, the procedure is performed in an open
should confirm its benefits. fashion to size and secure the membrane
II Science and Techniques for Cartilage Repair

with either sutures or fibrin glue. Originally

introduced by Behrens, various authors have
Biomaterials reported on mid-term outcomes of this tech-
nique, generally showing improved pain and
Microfracture relies on the formation of a
function.3033 One group added platelet-rich
blood clot containing MSCs from the bone
plasma to AMIC for patellar defects in a five-
marrow to fill the defect. The reparative tis-
patient pilot project, demonstrating good
sue slowly matures with time but is vulner-
clinical outcomes but formation of intrale-
able in the early phase, requiring protected
sional osteophytes in three of five patients.31
weightbearing and ROM restrictions. Many
Due to reports on the benefits of drilling
groups are investigating modifications to
versus microfracture, Behrens most recently
the original microfracture technique, utiliz-
reported changing his technique to subchon-
ing biomaterials to augment the mechanical
dral drilling for marrow stimulation.27,28,30
stability of the early clot, hoping to retain
more of the early, MSC-rich blood in the
defect. Furthermore, bioactive materials BST-CarGel (Piramal Healthcare Inc.,
could improve cell proliferation, differen-
Laval, Quebec, Canada)
tiation, and matrix production, guiding the
tissue toward a more hyaline-like histologi- This technique uses chitosan mixed with
cal appearance with potentially better long- autologous blood to form a gel that is then
term stability. The former function uses the implanted in a marrow-stimulated carti-
biomaterial as a scaffold to maintain cells in lage defect.34 Chitosan is a polysaccharide
situ until the new repair tissue has achieved primarily composed of polyglucosamine; it
adequate mechanical stability; thereafter, the is thrombogenic, self-adhering, and com-
scaffold has fulfilled its purpose and should pletely resorbable. It has been the focus
resorb on its own. Several materials are being of extensive research for multiple tissue
investigated for this role, including the autol- engineering applications, including carti-
ogous fibrin clot forming in standard micro- lage repair. In preclinical animal models,
fracture, exogenous fibrin glue, alginate or chitosan improved the histological quality
agarose, collagen, hyaluronic acid, chitosan, of the repair tissue when compared with
and artificial polymers, such a polylactic and marrow-stimulated control defects.3537
polyglycolic acid and their modifications. The Preliminary data from an RCT were recently
second function as a bioactive substrate is presented: 81 patients were randomized to
far more complex and requires careful inves- microfracture with and without BST-CarGel,
tigation to ensure correct differentiation of whom 41 had completed 1-year follow-
signalsthat is, chondrogenic rather than up. Evaluation demonstrated no clinical dif-
osteogenic differentiation of the MSCs. The ferences at this early time point, but better
following section will review preclinical and MRI and histological appearance of the aug-
90 clinical data on biomaterial augmentation of mented group.38

Stannard_9781604068580_Ch07.indd 90 1/30/13 2:41 PM

Chondrotissue (Bio Tissue AG, been explored, but residence time in the
Freiburg, Germany) defect is limited. Other groups have investi-
gated repeated postoperative injections, viral
Chondrotissue is a nonwoven polyglycolic vectors, and matrix-bound proteins with var-
acid fleece infused with hyaluronic acid ying degrees of success.
that is implanted into cartilage defects after Gelse et al investigated the use of throm-
microfracture. Implantation of the scaffold in bospondin-1 and osteogenic protein-1
an ovine model demonstrated the formation (OP-1) after microfracture in a minipig
of cartilaginous repair tissue.39,40 Reports on model. Microfracture alone produced an
the clinical outcomes are limited at this point inferior fibrocartilaginous tissue. The addi-
with only a case report available.41 tion of OP-1 stimulated chondrogenesis but
also induced enchondral ossification, which
in turn was negated by treatment with
Gelrin C (Regentis Biomaterials,
thrombospondin-1. The authors concluded
Or-Akiva, Israel) that the combination of the two factors has
This scaffold is a biodegradable photopolym- the potential to improve tissue quality after
erized hydrogel of polyethylene glycol (PEG) microfracture and reduce ossification.49
diacrylate bound to fibrinogen. The scaffold Klinger et al found similar inhibitory effects
is injected into the previously microfractured on enchondral ossification in microfractured
defect as a gel that polymerizes in situ and defects with delivery of chondromodulin-I
completely degrades within 6 to 12 months. through viral vectors. Additional chon-

7 Microfracture and Augments

The major degradation products are polyeth- drogenic and antiangiogenic effects were
ylene glycolated peptides, amino acids, and observed in this minipig model.50 Sellers,
PEG, and have been shown to be nontoxic.42 Yang, Zhang, and Kuo and their colleagues
A clinical trial is currently under way. described improved tissue quality in micro-
fractured defects in a rabbit model treated
with BMP-2, -4, and -7, respectively.5154
Hyaluronan (HA) Morisset et al treated microfractured defects
in horses with injections of adenoviral vec-
The use of viscosupplementation for the
tors carrying the genes of interleukin-1
treatment of osteoarthritis is widespread in
receptor antagonist protein and insulin-like
clinical practice, although not without con-
growth factor-1, showing improved defect
troversy.43 Its application for microfracture
healing in comparison with saline injection
has been explored by several groups, demon-
controls.55 Feeley et al reported on the nega-
strating that postoperative injections of HA
tive influence of postoperative parathyroid
had anti-inflammatory effects and improved
hormone treatment on cartilage formation
repair tissue quality.4446
in a rabbit microfracture model and recom-
mended against its use.56
Bioactive (Growth) Factors
Stem Cells
Bioactive factors are proteins that can pro-
mote and inhibit cell differentiation, pro- There is considerable interest in the applica-
liferation, and matrix production in a tion of stem cells for cartilage repair. MSTs,
complex interaction. Marrow stimulation such as microfracture, attempt to recruit
in its original form relies on MSCs to differ- stem cells from the underlying marrow cav-
entiate into the correct (chondrogenic) cell ity. These techniques have been criticized
lineage, rather than into an osteogenic phe- due to the low concentration of MSCs in the
notype. Bioactive factors have been added subchondral bone marrow, and also due to
to marrow-stimulated defects in hopes of the resultant damage to the subchondral plate
improving the quality of the developing from the perforations. Techniques are being
reparative tissue.47,48 The delivery of bioactive investigated that obtain stem cells from other
factors remains challenging. Direct applica- areas such as the iliac crest or subcutaneous
tion of the growth factor in liquid form has fat, concentrate the cells, and then introduce 91

Stannard_9781604068580_Ch07.indd 91 1/30/13 2:41 PM

them either intraoperatively into the defect 2. Steadman JR, Rodkey WG, Rodrigo JJ. Microfrac-
ture: surgical technique and rehabilitation to treat
or postoperatively with injections. A com- chondral defects. Clin Orthop Relat Res 2001;(391,
prehensive review of this topic is beyond Suppl):S362S369
the scope of this chapter, but several authors 3. Hurst JM, Steadman JR, OBrien L, Rodkey WG,
have investigated their use in conjunction Briggs KK. Rehabilitation following microfracture
for chondral injury in the knee. Clin Sports Med
with MSTs, applied either during surgery or 2010;29(2):257265, viii
as injections in the postoperative period. 4. Steadman JR, Briggs KK, Rodrigo JJ, Kocher MS,
Saw et al reported on five patients under- Gill TJ, Rodkey WG. Outcomes of microfracture
for traumatic chondral defects of the knee: aver-
going second-look arthroscopy after previous age 11-year follow-up. Arthroscopy 2003;19(5):
drilling followed by 5 weekly injections of 477484
peripheral blood progenitor cells and hyalu- 5. Miller BS, Steadman JR, Briggs KK, Rodrigo JJ,
Rodkey WG. Patient satisfaction and outcome
ronic acid. The patients were part of a larger after microfracture of the degenerative knee. J
group of 180 undergoing this treatment. He Knee Surg 2004;17(1):1317
reported articular cartilage regeneration and 6. Mithoefer K, Williams RJ III, Warren RF, et al. The
histology demonstrating features of hyaline microfracture technique for the treatment of
II Science and Techniques for Cartilage Repair

articular cartilage lesions in the knee. A prospec-

cartilage.57 Gobbi et al reported on the use of tive cohort study. J Bone Joint Surg Am 2005;87
concentrated bone marrow aspirate without (9):19111920
marrow stimulation for the repair of large 7. Mithoefer K, Williams RJ III, Warren RF,
Wickiewicz TL, Marx RG. High-impact athletics
cartilage defects in 15 patients. The marrow after knee articular cartilage repair: a prospective
clot was implanted under a type I/III col- evaluation of the microfracture technique. Am J
lagen matrix. At 2-year follow-up, patients Sports Med 2006;34(9):14131418
8. Kon E, Gobbi A, Filardo G, Delcogliano M,
reported significant improvements in pain Zaffagnini S, Marcacci M. Arthroscopic second-
and functional scores.58 generation autologous chondrocyte implantation
compared with microfracture for chondral lesions
of the knee: prospective nonrandomized study at
5 years. Am J Sports Med 2009;37(1):3341
Conclusion 9. Gudas R, Kalesinskas RJ, Kimtys V, et al. A pro-
spective randomized clinical study of mosaic
osteochondral autologous transplantation versus
Microfracture is a widely accepted treatment microfracture for the treatment of osteochon-
option for full-thickness articular cartilage dral defects in the knee joint in young athletes.
defects. Its ready availability, low cost, and Arthroscopy 2005;21(9):10661075
minimal invasiveness make it attractive as 10. Asik M, Ciftci F, Sen C, Erdil M, Atalar A. The
microfracture technique for the treatment of
a first-line treatment option. Initially hailed full-thickness articular cartilage lesions of the
as a nonbridge burning procedure, it is now knee: midterm results. Arthroscopy 2008;24
being recognized as altering the subchondral
11. Kreuz PC, Steinwachs MR, Erggelet C, et al. Results
bone through sclerosis of the subchondral after microfracture of full-thickness chondral
plate, formation of subchondral cysts, and defects in different compartments in the knee.
intralesional osteophytes in over one-third Osteoarthritis Cartilage 2006;14(11):11191125
12. Knutsen G, Engebretsen L, Ludvigsen TC, et al.
of patients. Following strict indications, Autologous chondrocyte implantation compared
however, microfracture continues to pre- with microfracture in the knee. A randomized
sent a useful treatment option for the repair trial. J Bone Joint Surg Am 2004;86-A(3):455464
of cartilage defects smaller than 2 to 4 cm2, 13. Kreuz PC, Erggelet C, Steinwachs MR, et al. Is
microfracture of chondral defects in the knee
primarily in the femoral condyles in younger associated with different results in patients
patients. Broadening of these indications aged 40 years or younger? Arthroscopy 2006;22
and improved long-term outcomes might be
14. Saris DB, Vanlauwe J, Victor J, et al; TIG/ACT/
achieved through modification of the stand- 01/2000&EXT Study Group. Treatment of symp-
ard microfracture technique through the use tomatic cartilage defects of the knee: character-
of biomaterials and bioactive factors. ized chondrocyte implantation results in better
clinical outcome at 36 months in a randomized
trial compared to microfracture. Am J Sports Med
2009;37(Suppl 1):10S19S
References 15. Saris DB, Vanlauwe J, Victor J, et al. Character-
ized chondrocyte implantation results in better
1. Irrgang JJ, Pezzullo D. Rehabilitation following structural repair when treating symptomatic car-
surgical procedures to address articular cartilage tilage defects of the knee in a randomized con-
92 lesions in the knee. J Orthop Sports Phys Ther trolled trial versus microfracture. Am J Sports Med
1998;28(4):232240 2008;36(2):235246

Stannard_9781604068580_Ch07.indd 92 1/30/13 2:41 PM

16. Coleman SH, Malizia R, Macgillivray J, Warren RF. and its relationship to articular cartilage resurfac-
Treatment of isolated articular cartilage lesions ing. Am J Sports Med 2011;39(8):17311740
of the medial femoral condyle. A clinical nad MR 30. Benthien JP, Behrens P. The treatment of chondral
comparison of autologous chondrocyte implanta- and osteochondral defects of the knee with autol-
tion vs. microfracture. Ortop Traumatol Rehabil ogous matrix-induced chondrogenesis (AMIC):
2001;3(2):224226 method description and recent developments.
17. Basad E, Ishaque B, Bachmann G, Strz H, Knee Surg Sports Traumatol Arthrosc 2011;19(8):
Steinmeyer J. Matrix-induced autologous chon- 13161319
drocyte implantation versus microfracture in the 31. Dhollander AA, De Neve F, Almqvist KF, et al. Autol-
treatment of cartilage defects of the knee: a 2-year ogous matrix-induced chondrogenesis combined
randomised study. Knee Surg Sports Traumatol with platelet-rich plasma gel: technical descrip-
Arthrosc 2010;18(4):519527 tion and a five pilot patients report. Knee Surg
18. Gudas R, Simonaityte R, Cekanauskas E, Sports Traumatol Arthrosc 2011;19(4):536542
Tamosinas R. A prospective, randomized clini- 32. Benthien JP, Behrens P. Autologous matrix-
cal study of osteochondral autologous transplan- induced chondrogenesis (AMIC). A one-step pro-
tation versus microfracture for the treatment of cedure for retropatellar articular resurfacing. Acta
osteochondritis dissecans in the knee joint in chil- Orthop Belg 2010;76(2):260263
dren. J Pediatr Orthop 2009;29(7):741748
33. Gille J, Schuseil E, Wimmer J, Gellissen J, Schulz AP,
19. Von Keudell A, Atzwanger J, Forstner R, Resch H,
Behrens P. Mid-term results of Autologous Matrix-
Hoffelner T, Mayer M. Radiological evaluation of
Induced Chondrogenesis for treatment of focal
cartilage after microfracture treatment: a long-
cartilage defects in the knee. Knee Surg Sports
term follow-up study. Eur J Radiol 2011; June 16
Traumatol Arthrosc 2010;18(11):14561464
(Epub ahead of print)
20. Brown WE, Potter HG, Marx RG, Wickiewicz TL, 34. Kumar MN, Muzzarelli RA, Muzzarelli C, Sashiwa H,
Warren RF. Magnetic resonance imaging appear- Domb AJ. Chitosan chemistry and pharmaceutical
ance of cartilage repair in the knee. Clin Orthop perspectives. Chem Rev 2004;104(12):60176084
Relat Res 2004;(422):214223 35. Chevrier A, Hoemann CD, Sun J, Buschmann MD.
Chitosan-glycerol phosphate/blood implants

7 Microfracture and Augments

21. Zaslav K, Cole B, Brewster R, et al; STAR Study Prin-
cipal Investigators. A prospective study of autolo- increase cell recruitment, transient vasculariza-
gous chondrocyte implantation in patients with tion and subchondral bone remodeling in drilled
failed prior treatment for articular cartilage defect cartilage defects. Osteoarthritis Cartilage 2007;
of the knee: results of the Study of the Treatment 15(3):316327
of Articular Repair (STAR) clinical trial. Am J Sports 36. Hoemann CD, Sun J, McKee MD, et al. Chitosan-
Med 2009;37(1):4255 glycerol phosphate/blood implants elicit hyaline
22. McNickle AG, LHeureux DR, Yanke AB, Cole BJ. cartilage repair integrated with porous subchon-
Outcomes of autologous chondrocyte implanta- dral bone in microdrilled rabbit defects. Osteoar-
tion in a diverse patient population. Am J Sports thritis Cartilage 2007;15(1):7889
Med 2009;37(7):13441350 37. Hoemann CD, Hurtig M, Rossomacha E, et al.
23. Minas T, Gomoll AH, Rosenberger R, Royce RO, Chitosan-glycerol phosphate/blood implants
Bryant T. Increased failure rate of autologous improve hyaline cartilage repair in ovine microf-
chondrocyte implantation after previous treat- racture defects. J Bone Joint Surg Am 2005;87(12):
ment with marrow stimulation techniques. Am J 26712686
Sports Med 2009;37(5):902908 38. Stanish WD, Restrepo A, Macdonald PB, et al. An
24. Pestka JM, Bode G, Salzmann G, Sdkamp NP, International Randomized Clinical Trial Evaluat-
Niemeyer P. Clinical outcome of autologous chon- ing BST-CarGel: A New Gel Implant for Cartilage
drocyte implantation for failed microfracture Repair. Presented at: ISAKOS, 8th Biennial Con-
treatment of full-thickness cartilage defects of the gress; Rio de Janeiro, Brazil; 2011: #149
knee joint. Am J Sports Med 2011; November 5 39. Erggelet C, Neumann K, Endres M, Haberstroh K,
(Epub ahead of print) Sittinger M, Kaps C. Regeneration of ovine articu-
25. Hunziker EB. Articular cartilage repair: basic sci- lar cartilage defects by cell- free polymer-based
ence and clinical progress. A review of the cur- implants. Biomaterials 2007;28(36):5570 5580
rent status and prospects. Osteoarthritis Cartilage 40. Erggelet C, Endres M, Neumann K, et al. Forma-
2002;10(6):432463 tion of cartilage repair tissue in articular carti-
26. Insall JN. Intra-articular surgery for degenera- lage defects pretreated with microfracture and
tive arthritis of the knee. A report of the work covered with cell-free polymer-based implants. J
of the late K. H. Pridie. J Bone Joint Surg Br Orthop Res 2009;27(10):13531360
1967;49(2):211228 41. Patrascu JM, Freymann U, Kaps C, Poenaru DV.
27. Chen H, Hoemann CD, Sun J, et al. Depth of Repair of a post-traumatic cartilage defect with
subchondral perforation influences the outcome a cell-free polymer-based cartilage implant: a
of bone marrow stimulation cartilage repair. J follow-up at two years by MRI and histological
Orthop Res 2011;29(8):11781184 review. J Bone Joint Surg Br 2010;92(8):11601163
28. Chen H, Sun J, Hoemann CD, et al. Drilling and 42. Gonen-Wadmany M, Oss-Ronen L, Seliktar D.
microfracture lead to different bone structure Protein-polymer conjugates for forming photopo-
and necrosis during bone-marrow stimulation lymerizable biomimetic hydrogels for tissue engi-
for cartilage repair. J Orthop Res 2009;27(11): neering. Biomaterials 2007;28(26):38763886
14321438 43. Bellamy N, Campbell J, Robinson V, Gee T, Bourne R,
29. Chen H, Chevrier A, Hoemann CD, Sun J, Ouyang W, Wells G. Viscosupplementation for the treatment
Buschmann MD. Characterization of subchondral of osteoarthritis of the knee. Cochrane Database
bone repair for marrow-stimulated chondral defects Syst Rev 2006;(2):CD005321 93

Stannard_9781604068580_Ch07.indd 93 1/30/13 2:41 PM

44. Strauss E, Schachter A, Frenkel S, Rosen J. The effi- 52. Sellers RS, Zhang R, Glasson SS, et al. Repair of
cacy of intra-articular hyaluronan injection after articular cartilage defects one year after treatment
the microfracture technique for the treatment with recombinant human bone morphogenetic
of articular cartilage lesions. Am J Sports Med protein-2 (rhBMP-2). J Bone Joint Surg Am 2000;
2009;37(4):720726 82(2):151160
45. Legovi D, Zorihi S, Gulan G, et al. Microfrac- 53. Zhang X, Zheng Z, Liu P, et al. The synergistic
ture technique in combination with intraarticu- effects of microfracture, perforated decalcified
lar hyaluronic acid injection in articular cartilage cortical bone matrix and adenovirus-bone mor-
defect regeneration in rabbit model. Coll Antropol phogenetic protein-4 in cartilage defect repair.
2009;33(2):619623 Biomaterials 2008;29(35):46164629
46. Kang SW, Bada LP, Kang CS, et al. Articular carti- 54. Yang HS, La WG, Bhang SH, et al. Hyaline cartilage
lage regeneration with microfracture and hyalu- regeneration by combined therapy of microfracture
ronic acid. Biotechnol Lett 2008;30(3):435439 and long-term bone morphogenetic protein-2 deliv-
47. Angel MJ, Sgaglione NA, Grande DA. Clinical appli- ery. Tissue Eng Part A 2011;17(1314):18091818
cations of bioactive factors in sports medicine: 55. Morisset S, Frisbie DD, Robbins PD, Nixon AJ,
current concepts and future trends. Sports Med McIlwraith CW. IL-1ra/IGF-1 gene therapy modu-
Arthrosc 2006;14(3):138145 lates repair of microfractured chondral defects.
48. Frenkel SR, Di Cesare PE. Scaffolds for articular Clin Orthop Relat Res 2007;462:221228
cartilage repair. Ann Biomed Eng 2004;32(1): 56. Feeley BT, Doty SB, Devcic Z, Warren RF, Lane JM.
II Science and Techniques for Cartilage Repair

2634 Deleterious effects of intermittent recombinant

49. Gelse K, Klinger P, Koch M, et al. Thrombospondin-1 parathyroid hormone on cartilage formation in a
prevents excessive ossification in cartilage repair rabbit microfracture model: a preliminary study.
tissue induced by osteogenic protein-1. Tissue Eng HSS J 2010;6(1):7984
Part A 2011;17(1516):21012112 57. Saw K-Y, Anz A, Merican S, et al. Articular cartilage
50. Klinger P, Surmann-Schmitt C, Brem M, et al. regeneration with autologous peripheral blood
Chondromodulin 1 stabilizes the chondrocyte progenitor cells and hyaluronic acid after arthro-
phenotype and inhibits endochondral ossification scopic subchondral drilling: a report of 5 cases
of porcine cartilage repair tissue. Arthritis Rheum with histology. Arthroscopy 2011;27(4):493506
2011;63(9):27212731 58. Gobbi A, Karnatzikos G, Scotti C, Mahajan V,
51. Kuo AC, Rodrigo JJ, Reddi AH, Curtiss S, Grotkopp E, Mazzucco L, Grigolo B. One-Step Cartilage Repair
Chiu M. Microfracture and bone morphogenetic with Bone Marrow Aspirate Concentrated Cells
protein 7 (BMP-7) synergistically stimulate articu- and Collagen Matrix in Full-Thickness Knee Carti-
lar cartilage repair. Osteoarthritis Cartilage 2006; lage Lesions Results at 2-Year Follow-up. Cartilage
14(11):11261135 2011;2(3):286299


Stannard_9781604068580_Ch07.indd 94 1/30/13 2:41 PM

Elizaveta Kon, Giuseppe Filardo, Alessandro Di Martino, and
Maurilio Marcacci

The ultrastructure of articular cartilage is ago, showing good clinical results while at
unique: chondrocytes are sparsely distributed the same time overcoming most of the con-
within the surrounding matrix, maintaining cerns related to the first-generation ACI.12
minimal cell-to-cell contact. The interaction The use of cell-loaded scaffolds to regenerate
between cells, collagen framework, aggrecan, a cartilage-like tissue presents advantages
and fluid constitutes the complex ultrastruc- from both the biological and the surgical
ture of hyaline cartilage, making its replace- point of view, thus aiming to further opti-
ment or reproduction difficult.1 mize this regenerative surgical procedure.
New, ambitious regenerative procedures
are emerging as potential therapeutic options
for the treatment of chondral lesions, aiming Surgical Technique
to re-create a hyaline-like tissue, thus restor-
ing a biologically and biomechanically valid The surgical technique of both ACI and MACI
articular surface with durable clinical results. consists of two steps. The first one is an
Autologous chondrocyte implantation arthroscopic procedure in which a biopsy of
(ACI) was introduced in 1987 in Sweden, and healthy cartilage is harvested from a non
in 1994 Brittberg et al2 published the first weight-bearing site on the articular surface
clinical report showing satisfactory results for (usually intercondylar notch) for autologous
isolated femoral condyle lesions. Since then, chondrocyte cell culture, and the second step
several studies have followed, documenting consists of implanting the expanded chon-
both the production of a hyaline-like articu- drocytes (Fig. 8.1).
lar surface and good results in the majority The ACI procedure involves the implanta-
of the patients at medium-long follow-up. tion of a liquid cell culture, thus requiring
Treatment indications have been broad- the use of a flap to avoid leakage of chon-
ened,35 and this cell-based technique has drocytes from the defect area. An autologous
gained increasing interest worldwide.610 The periosteal patch has been traditionally used
development of bioengineering technology for its biological activity, but recently the use
further improved this regenerative treatment of a collagen xenograft membrane is becom-
approachessentially, transplanting biode- ing more popular. Through a parapatellar
gradable molecules that are used as tempo- arthrotomy, the flap is sutured to the defect
rary scaffolds for the growth of living cells.11 rim, and fibrin glue or sealant is applied to
Matrix-assisted ACI (MACI) techniques were create a watertight seal before the cultured
introduced in the clinical practice one decade cells are injected.10

Stannard_9781604068580_Ch08.indd 95 1/30/13 2:46 PM

Fig. 8.1 Schematic representation. (a) ACI, open approach; (b) MACI, mini-open approach; (c) MACI,
arthroscopic approach. Abbreviations: ACI, autologous chondrocyte implantation; MACI, matrix-assisted ACI.
II Science and Techniques for Cartilage Repair

The bioengineered MACI technology simpli- chondral nutrition and regenerative pro-
fied the second step, which differs depending cesses. Controlled mobilization exercises with
on the scaffold used. A mini-open approach reduced range of motion, early isometric and
can be used to prepare the lesion site, debrid- isotonic exercises, and controlled mechani-
ing the defect area down to the subchondral cal compression are performed. In the fourth
bone. Afterward, using a foil template reflect- week progressive touch-down weight bearing
ing the size and geometry of the defect, the with crutches is allowed and usually advanced
chondrocyte-loaded matrix is cut to size and to full weight bearing within 6 to 8 weeks after
fitted into the defect with the cell-loaded sur- surgery. Gait training in a swimming pool can
face facing the subchondral bone. In the case be prescribed to facilitate the recovery of nor-
of an arthroscopic approach, dedicated instru- mal gait phases. Subsequently, active func-
ments are used:13 a circular area with regular tional training can be started if there are no
margins for graft implantation is prepared symptoms of overloading, such as pain, effu-
with a specially designed cannulated low- sion, and tenderness. Proprioceptive, strength
profile drill. The delivery device is then filled and endurance exercises, and aerobic training
with the bioengineered tissue, which is trans- are then introduced, aiming to return to a cor-
ported and positioned in the prepared area. rect running pathway. The remainder of the
Depending on the adhesive characteris- rehabilitation is dedicated to the return to pre-
tics of the grafts, no fibrin glue or sutures vious sport activity, which is usually allowed
are needed, but for some biomaterials fibrin no earlier than 1 year after surgery. However,
glue is used to fix the implant, and a transos- time needed to recover may vary markedly
seous fixation technique has been proposed depending on the procedure used. The bio-
to ensure secure fixation of the graft even in engineered tissue significantly reduces the
defects without stable shoulders.12,14 inherent fragility of the culture implant dur-
ing the early postoperative stage and makes
an accelerated patient recovery possible. The
Rehabilitation Protocol arthroscopic approach results in lower surgi-
cal morbidity and may enable a further accel-
A similar rehabilitation protocol is used for eration of the functional recovery.15,16
both treatment approaches.
In the early stage (0 to 6 weeks), the reha-
bilitation strategies are focused on control- Results
ling pain, effusion, loss of motion, and muscle
atrophy, and on protecting the transplant by ACI
preventing weight bearing for  4 weeks.
Continued passive motion is usually applied Since its conception 20 years ago, satisfac-
intensively until 90 degrees of flexion is tory clinical and radiographic (magnetic res-
96 attained, to avoid joint adherence and favor onance imaging [MRI]) outcomes have been

Stannard_9781604068580_Ch08.indd 96 1/30/13 2:46 PM

reported consistently at medium- to long- These studies demonstrated that patients
term follow-up for the ACI procedure.310 report good and excellent clinical and func-
After the preliminary promising results,2 tional outcomes after ACI at long-term
the indication of this treatment has been follow-up. With regard to imaging, encourag-
broadened, and good patient-reported out- ing data have also been documented.21 Even
comes have been reported also for more chal- though intralesional osteophytes, subchon-
lenging lesions. Browne et al8 documented dral cysts, and bone marrow edema were
good results in large defects and in patients common, the defect area was restored in
who previously failed prior cartilage repair. most patients, with the quality of the repair
Minas et al4 treated patients affected by tissue being similar to the surrounding nor-
early osteoarthritis successfully. Rosenberger mal cartilage, thus confirming the good long-
et al17 analyzed ACI treatment in older term outcome offered by this procedure.
patients and found outcomes comparable to However, these good results have to be
those reported in the literature for younger weighed against several limitations related
patients if all articular comorbidities were to the complexity and morbidity of the sur-
recognized and treated concomitantly. Farr18 gical procedure, which requires a large open
and Pascual-Garrido et al19 showed that even surgical approach and thus entails a high
more complex patellofemoral lesions can risk of joint stiffness and arthrofibrosis. The
successfully be treated as long as corrective periosteal patch is believed to have biological
osteotomies are being performed to unload properties, but it also requires a second inci-
the repair tissue. The ACI procedure has also sion and causes a high rate of hypertrophy
been modified to expand the treatment to with a high reoperation rate.9,2224
deep osteochondral lesions. This sandwich A much lower complication rate has been
technique procedure shows good results for shown by several authors using a type I/III col-
the treatment of osteochondritis dissecans lagen membrane in place of periosteum.25,26
(OCD) at medium-term follow-up.5 A further improvement of the procedure has
Recently, Peterson et al7 investigated the been introduced with the development of a

8 ACI and MACI

20-year outcomes of the ACI procedure with new technology: characterized chondrocyte
periosteum in 224 patients. The subjective implantation (CCI), which aims to improve
scores documented a significant improve- the results of articular regeneration through
ment compared with the preoperative val- the use of a selected cell population (approxi-
ues. Seventy-four percent of the patients mately 5% deselection of inferior whole cul-
reported that they were better or stable, ture populations of chondrocytes).27,28
and 92% were satisfied with the operation However, despite the substantial devel-
and would undergo the ACI again. Further opment undergone by the procedure since
analysis was performed to determine factors its introduction, some problems remain
that could influence the final outcome. The unsolved. One of those factors is the concern
authors found that the age at the time of the about the maintenance of the chondrocytic
operation and the size of the lesion did not phenotype during the prolonged mono-
correlate with the results, and interestingly layer culture, which is a critical factor. In
the presence of meniscal injuries before ACI fact, it is known that chondrocytes in two-
or history of bone marrow procedures before dimensional cell cultures alter their pheno-
the implantation did not affect the outcome type and dedifferentiate to fibroblast cells
in this series either. This is in contrast to a that no longer possess the capacity to pro-
report by Minas et al, who showed a three duce type II collagen or proteoglycans,6 and
times higher rate of failures for defects that it is still unclear whether transplanted cells
had prior treatments affecting the subchon- re-express their phenotype after transplanta-
dral bone.20 Better results have been obtained tion. Another important concern is whether
in cases of isolated femoral condyle lesions chondrocytes will be homogeneously distrib-
and OCD, whereas patients with multiple uted in the three-dimensional (3D) space of
lesions undergo a progressive decline, with the defect when used in liquid cell suspen-
the bipolar lesions having an inferior out- sion. Even with meticulous technique, there
come at 20 years. is the risk of chondrocyte leakage. 97

Stannard_9781604068580_Ch08.indd 97 1/30/13 2:46 PM

MACI a decline in clinical outcome at 5 years.
Interestingly, even though the data showed a
MACI has been developed with the attempt small decline, clinical and histological scores
to address most of the concerns related to were still good, suggesting that autologous
the cell culture and the surgical technique chondrocytes seeded on a hyaluronan-based
observed with the first-generation approach. scaffold can be considered a viable treatment
Tissue engineering technology allows for the option even for these lesions.
use of cell-loaded scaffolds. These are 3D A nonrandomized prospective cohort
structures developed to provide a support study documented good and excellent results
for cell adhesion, proliferation and produc- over time in the MACI group. Interestingly,
tion of matrix, and in the end the formation over the same time frame, the microfractures
of a cartilage-like tissue.29 Maintenance of a control group showed a decline in patient-
chondrocyte differentiated phenotype, lost reported outcomes.31 These results were
during liquid culture, has been documented confirmed in a study on highly demand-
with the use of 3D scaffolds.30 In addition, the ing patients, soccer players; despite similar
II Science and Techniques for Cartilage Repair

MACI technique can be done arthroscopically success in returning to competitive sport,

or through minimal exposure, thus avoiding microfractures allowed a faster recovery but
the large open approach necessary for the presented a clinical deterioration over time,
traditional ACI procedure.13,14 whereas arthroscopic second-generation ACI
Several different second-generation tissue- offered more durable clinical results.38 A clini-
engineered products have been introduced. cal improvement was found using MACI even
Most of the currently available products are in older patients.16 Patients over 40 years old
either collagen- or hyaluronan-based matri- were treated with chondrocytes seeded on a
ces. While there are many studies reporting collagen-based membrane or a hyaluronic-
good short-term success, long-term evalua- based scaffold. Despite the higher failure rate
tions of these procedures are not available to in this population and inferior results with
date.11,3134 respect to those previously found for younger
Behrens et al32 treated localized cartilage patients, a significant improvement was doc-
defects using a cell-seeded collagen matrix. umented. Results were consistent comparing
They obtained good clinical outcome in 8 out the two treatment groups, with the only dif-
of 11 patients at 5 years after transplantation. ference of a faster recovery in the group in
Ebert et al33 used a different chondrocyte- which an arthroscopic approach was used.
seeded collagen membrane and found clini- More recently, stable results were reported
cal and functional improvements and a high up to 7 years of follow-up,39 and a good out-
patient satisfaction rate. They also reported come was documented, using a modified
good MRI outcomes in 41 patients at 5 years technique with an associated bone grafting,
follow-up. Kreuz et al34 used a bioresorbable also for the treatment of OCD, at 6 years of
two-component gel-polymer scaffold for the follow-up.40
treatment of mild degenerative and focal oste- Many other scaffolds have been recently
oarthritic defects of the knee, showing that proposed in preclinical studies; only a few
the good clinical outcome was stable over the have been introduced in clinical practice,
course of a period of 4 years. Ferruzzi et al35 though, presenting promising results at
analyzed the results obtained with a 3D hya- shorter follow-up.41,42
luronic acid scaffold, reporting  50 patients
affected by OCD and traumatic lesions. They
reported good and excellent clinical out- Discussion
comes at a minimum 5 years follow-up. This
study also reported MRI findings suggest- Despite thousands of treated patients and
ing that the cartilage repair tissue was well many published studies suggesting good
integrated in 93% of the patients. Nehrer et and durable clinical results of this regenera-
al36 confirmed good results over time with tive surgical treatment approach, there is no
this scaffold. Gobbi et al37 treated more chal- agreement to date about the effective superi-
98 lenging patellofemoral defects and reported ority of one of these techniques over another,

Stannard_9781604068580_Ch08.indd 98 1/30/13 2:46 PM

and both indications and results are still preliminary results obtained. The modern
being discussed controversially.10 regenerative procedures can replace the
Studies comparing ACI with mosaicplasty injured articular surface with a hyaline-
are inconclusive; whereas the outcome is like tissue, but the properties of the healthy
similar or even worse in small to medium- cartilage tissue are still unmatched by any
size lesions,43,44 it seems that the regenera- available substitute. Additionally the cur-
tive (ACI) approach may offer better results rent techniques are still cell culture-based;
for bigger lesions.45 The comparison between thus, they still pose the problem of cost-
ACI and bone marrow stimulation proce- effectiveness and two-step surgery.
dures also reported contradictory findings. Even though different approaches have
Fu et al observed that patients who received been studied to avoid the problems related to
ACI obtained higher levels of knee function the ex vivo chondrocyte culture and expansion
and greater relief from pain and swelling at in a scaffold, cell-free implants5153 sufficiently
3 years compared with debridement.46 intelligent to introduce the appropriate cues
Visna et al47 evaluated an original method to induce orderly and durable tissue regen-
of chondrograft preparation based on cul- eration are under investigation in numerous
tivated autologous chondrocytes in a 3D animal studies and are gaining increasing
carrierfibrin glue, showing a better short- interest in the clinical practice. These new and
term outcome when compared with abrasive emerging technologies represent a new field
techniques, and Basad et al48 compared MACI of regenerative treatments that will be excit-
with microfractures, reporting MACI superi- ing to follow in the future.
ority at 2 years.
By contrast, Knutsen et al49,50 showed that
there was no difference in clinical outcomes References
between ACI and microfractures at up to
1. Buckwalter JA, Mankin HJ. Articular cartilage:
5 years follow-up, and no differences were tissue design and chondrocyte-matrix interac-
detected in the macroscopic or histological tions. Instr Course Lect 1998;47:477486
2. Brittberg M, Lindahl A, Nilsson A, Ohlsson C,

8 ACI and MACI

results at 2 years and radiographic findings
Isaksson O, Peterson L. Treatment of deep carti-
at 5 years follow-up. However, they also lage defects in the knee with autologous chon-
demonstrated that none of the failures pre- drocyte transplantation. N Engl J Med 1994;
sented with high-quality repair cartilage, and 331(14):889895
3. Vasiliadis HS, Lindahl A, Georgoulis AD, Peterson L.
that patients with ACI tended to have a more Malalignment and cartilage lesions in the patel-
hyaline-like appearance of the repair tissue lofemoral joint treated with autologous chondro-
in the biopsies that were taken. The studies cyte implantation. Knee Surg Sports Traumatol
Arthrosc 2011;19(3):452457
of Saris et al27,28 confirmed that the repair tis-
4. Minas T, Gomoll AH, Solhpour S, Rosenberger R,
sue quality influences risk of failure and out- Probst C, Bryant T. Autologous chondrocyte implan-
come over time. In fact, whereas the superior tation for joint preservation in patients with
early osteoarthritis. Clin Orthop Relat Res 2010;
histomorphometric and histological score 468(1):147157
observed by Saris et al in the CCI group did 5. Peterson L, Minas T, Brittberg M, Lindahl A. Treat-
not correlate with the short-term outcome, ment of osteochondritis dissecans of the knee
the quality of the repair tissue significantly with autologous chondrocyte transplantation:
results at two to ten years. J Bone Joint Surg Am
influenced the later follow-up; at 3 years, 2003;85-A(Suppl 2):1724
CCI offered a further improvement with bet- 6. Peterson L, Brittberg M, Kiviranta I, Akerlund EL,
ter clinical results compared with microfrac- Lindahl A. Autologous chondrocyte transplanta-
tion. Biomechanics and long-term durability. Am
tures, whose results reached a plateau after J Sports Med 2002;30(1):212
18 months. 7. Peterson L, Vasiliadis HS, Brittberg M, Lindahl A.
These most recent findings suggest the Autologous chondrocyte implantation: a long-
term follow-up. Am J Sports Med 2010;38(6):
higher healing potential offered by regenera- 11171124
tive procedures, but the role of many vari- 8. Browne JE, Anderson AF, Arciero R, et al. Clinical
ables that may influence the final outcome outcome of autologous chondrocyte implantation
still need to be clarified to optimize results, at 5 years in US subjects. Clin Orthop Relat Res
and further systematic long-term evalua- 9. Micheli LJ, Browne JE, Erggelet C, et al. Autologous
tion is necessary to confirm the promising chondrocyte implantation of the knee: multicenter 99

Stannard_9781604068580_Ch08.indd 99 1/30/13 2:46 PM

experience and minimum 3-year follow-up. Clin J of focal chondral defects of the kneea clinical,
Sport Med 2001;11(4):223228 arthroscopic, MRI and histologic evaluation at
10. Vasiliadis HS, Wasiak J. Autologous chondrocyte 2 years. Knee 2005;12(3):209216
implantation for full thickness articular cartilage 25. Gomoll AH, Probst C, Farr J, Cole BJ, Minas T. Use of
defects of the knee. Cochrane Database Syst Rev a type I/III bilayer collagen membrane decreases
2010;(10):CD003323 reoperation rates for symptomatic hypertrophy
11. Kon E, Delcogliano M, Filardo G, Montaperto C, after autologous chondrocyte implantation. Am J
Marcacci M. Second generation issues in cartilage Sports Med 2009;37(1, Suppl 1):20S23S
repair. Sports Med Arthrosc 2008;16(4):221229 26. Gooding CR, Bartlett W, Bentley G, Skinner JA,
12. Kon E, Verdonk P, Condello V, et al. Matrix-assisted Carrington R, Flanagan A. A prospective, ran-
autologous chondrocyte transplantation for the domised study comparing two techniques of
repair of cartilage defects of the knee: systematic autologous chondrocyte implantation for osteo-
clinical data review and study quality analysis. Am chondral defects in the knee: Periosteum cov-
J Sports Med 2009;37(1, Suppl 1):156S166S ered versus type I/III collagen covered. Knee
13. Marcacci M, Zaffagnini S, Kon E, Visani A,
Iacono F, Loreti I. Arthroscopic autologous chon- 27. Saris DB, Vanlauwe J, Victor J, et al. Character-
drocyte transplantation: technical note. Knee Surg ized chondrocyte implantation results in better
Sports Traumatol Arthrosc 2002;10(3):154159 structural repair when treating symptomatic car-
tilage defects of the knee in a randomized con-
14. Erggelet C, Sittinger M, Lahm A. The arthroscopic trolled trial versus microfracture. Am J Sports Med
II Science and Techniques for Cartilage Repair

implantation of autologous chondrocytes for the 2008;36(2):235246

treatment of full-thickness cartilage defects of the
knee joint. Arthroscopy 2003;19(1):108110 28. Saris DB, Vanlauwe J, Victor J, et al; TIG/
ACT/01/2000&EXT Study Group. Treatment of
15. Della Villa S, Kon E, Filardo G, et al. Does intensive symptomatic cartilage defects of the knee: char-
rehabilitation permit early return to sport without acterized chondrocyte implantation results in bet-
compromising the clinical outcome after arthro- ter clinical outcome at 36 months in a randomized
scopic autologous chondrocyte implantation in trial compared to microfracture. Am J Sports Med
highly competitive athletes? Am J Sports Med 2009;37(1, Suppl 1):10S19S
29. Freed LE, Marquis JC, Nohria A, Emmanual J,
16. Kon E, Filardo G, Condello V, et al. Second- Mikos AG, Langer R. Neocartilage formation in
generation autologous chondrocyte implantation: vitro and in vivo using cells cultured on synthetic
results in patients older than 40 years. Am J Sports biodegradable polymers. J Biomed Mater Res
Med 2011;39(8):16681675 1993;27(1):1123
17. Rosenberger RE, Gomoll AH, Bryant T, Minas 30. Grigolo B, Lisignoli G, Piacentini A, et al. Evidence
T. Repair of large chondral defects of the knee for redifferentiation of human chondrocytes grown
with autologous chondrocyte implantation in on a hyaluronan-based biomaterial (HYAff 11):
patients 45 years or older. Am J Sports Med 2008; molecular, immunohistochemical and ultrastruc-
36(12):23362344 tural analysis. Biomaterials 2002;23(4):11871195
18. Farr J. Autologous chondrocyte implantation 31. Kon E, Gobbi A, Filardo G, Delcogliano M,
improves patellofemoral cartilage treatment out- Zaffagnini S, Marcacci M. Arthroscopic second-
comes. Clin Orthop Relat Res 2007;463:187194 generation autologous chondrocyte implantation
19. Pascual-Garrido C, Slabaugh MA, LHeureux DR, compared with microfracture for chondral lesions
Friel NA, Cole BJ. Recommendations and treatment of the knee: prospective nonrandomized study at
outcomes for patellofemoral articular cartilage 5 years. Am J Sports Med 2009;37(1):3341
defects with autologous chondrocyte implantation: 32. Behrens P, Bitter T, Kurz B, Russlies M.
prospective evaluation at average 4-year follow- Matrix-associated autologous chondrocyte
up. Am J Sports Med 2009;37(1, Suppl 1):33S41S transplantation/implantation (MACT/MACI)
20. Minas T, Gomoll AH, Rosenberger R, Royce RO, 5-year follow-up. Knee 2006;13(3):194202
Bryant T. Increased failure rate of autologous 33. Ebert JR, Robertson WB, Woodhouse J, et al. Clini-
chondrocyte implantation after previous treat- cal and magnetic resonance imaging-based out-
ment with marrow stimulation techniques. Am J comes to 5 years after matrix-induced autologous
Sports Med 2009;37(5):902908 chondrocyte implantation to address articular
21. Vasiliadis HS, Danielson B, Ljungberg M, cartilage defects in the knee. Am J Sports Med
McKeon B, Lindahl A, Peterson L. Autologous 2011;39(4):753763
chondrocyte implantation in cartilage lesions of 34. Kreuz PC, Mller S, Ossendorf C, Kaps C, Erggelet C.
the knee: long-term evaluation with magnetic Treatment of focal degenerative cartilage defects
resonance imaging and delayed gadolinium- with polymer-based autologous chondrocyte
enhanced magnetic resonance imaging technique. grafts: four-year clinical results. Arthritis Res Ther
Am J Sports Med 2010;38(5):943949 2009;11(2):R33
22. Peterson L, Minas T, Brittberg M, Nilsson A, 35. Ferruzzi A, Buda R, Faldini C, et al. Autologous
Sjgren-Jansson E, Lindahl A. Two- to 9-year out- chondrocyte implantation in the knee joint: open
come after autologous chondrocyte transplanta- compared with arthroscopic technique. Compari-
tion of the knee. Clin Orthop Relat Res 2000;(374): son at a minimum follow-up of five years. J Bone
212234 Joint Surg Am 2008;90(4, Suppl 4):90101
23. Minas T. Autologous chondrocyte implantation 36. Nehrer S, Dorotka R, Domayer S, Stelzeneder D,
for focal chondral defects of the knee. Clin Orthop Kotz R. Treatment of full-thickness chondral defects
Relat Res 2001;(391, Suppl):S349S361 with hyalograft C in the knee: a prospective clini-
24. Henderson I, Francisco R, Oakes B, Cameron J. cal case series with 2 to 7 years follow-up. Am J
Autologous chondrocyte implantation for treatment Sports Med 2009;37(Suppl 1):81S87S

Stannard_9781604068580_Ch08.indd 100 1/30/13 2:46 PM

37. Gobbi A, Kon E, Berruto M, et al. Patellofemoral for osteochondral defects in the knee. J Bone Joint
full-thickness chondral defects treated with Surg Br 2003;85(2):223230
second-generation autologous chondrocyte 46. Fu FH, Zurakowski D, Browne JE, et al. Autolo-
implantation: results at 5 years follow-up. Am J gous chondrocyte implantation versus debride-
Sports Med 2009;37(6):10831092 ment for treatment of full-thickness chondral
38. Kon E, Filardo G, Berruto M, et al. Articular carti- defects of the knee: an observational cohort
lage treatment in high-level male soccer players: study with 3-year follow-up. Am J Sports Med
a prospective comparative study of arthroscopic 2005;33(11):16581666
second-generation autologous chondrocyte 47. Visna P, Pasa L, Cizmr I, Hart R, Hoch J. Treatment
implantation versus microfracture. Am J Sports of deep cartilage defects of the knee using autolo-
Med 2011;39(12):25492557 gous chondrograft transplantation and by abra-
39. Filardo G, Kon E, Di Martino A, Iacono F, Marcacci M. sive techniquesa randomized controlled study.
Arthroscopic second-generation autologous chon- Acta Chir Belg 2004;104(6):709714
drocyte implantation: a prospective 7-year follow- 48. Basad E, Ishaque B, Bachmann G, Strz H,
up study. Am J Sports Med 2011;39(10):21532160 Steinmeyer J. Matrix-induced autologous chon-
40. Filardo G, Kon E, Berruto M, et al. Arthroscopic drocyte implantation versus microfracture in the
second generation autologous chondrocytes treatment of cartilage defects of the knee: a 2-year
implantation associated to bone grafting for the randomised study. Knee Surg Sports Traumatol
treatment of knee osteochondritis dissecans: Arthrosc 2010;18(4):519527
results at 6 years. Knee 2011; November 22 (Epub 49. Knutsen G, Engebretsen L, Ludvigsen TC, et al.
ahead of print) Autologous chondrocyte implantation compared
41. Selmi TA, Verdonk P, Chambat P, et al. Autologous with microfracture in the knee. A randomized
chondrocyte implantation in a novel alginate- trial. J Bone Joint Surg Am 2004;86-A(3):455464
agarose hydrogel: outcome at two years. J Bone 50. Knutsen G, Drogset JO, Engebretsen L, et al. A
Joint Surg Br 2008;90(5):597604 randomized trial comparing autologous chon-
42. Nehrer S, Chiari C, Domayer S, Barkay H, Yayon A. drocyte implantation with microfracture. Find-
Results of chondrocyte implantation with a fibrin- ings at five years. J Bone Joint Surg Am 2007;
hyaluronan matrix: a preliminary study. Clin 89(10):21052112
Orthop Relat Res 2008;466(8):18491855 51. Gille J, Schuseil E, Wimmer J, Gellissen J, Schulz AP,
43. Dozin B, Malpeli M, Cancedda R, et al. Com- Behrens P. Mid-term results of autologous matrix-
parative evaluation of autologous chon- induced chondrogenesis for treatment of focal
drocyte implantation and mosaicplasty: a cartilage defects in the knee. Knee Surg Sports
multicentered randomized clinical trial. Clin J Traumatol Arthrosc 2010;18(11):14561464
Sport Med 2005;15(4):220226 52. Bedi A, Foo LF, Williams RJ, et al. The maturation of
44. Horas U, Pelinkovic D, Herr G, Aigner T, Schnettler R. synthetic scaffolds for osteochondral donor sites
Autologous chondrocyte implantation and osteo- of the knee: an MRI and T2-mapping analysis. Car-

8 ACI and MACI

chondral cylinder transplantation in cartilage tilage 2010;1(1):2028
repair of the knee joint. A prospective, comparative 53. Kon E, Delcogliano M, Filardo G, Busacca M,
trial. J Bone Joint Surg Am 2003;85-A(2):185192 Di Martino A, Marcacci M. Novel nano-composite
45. Bentley G, Biant LC, Carrington RW, et al. A pro- multilayered biomaterial for osteochondral regen-
spective, randomised comparison of autologous eration: a pilot clinical trial. Am J Sports Med
chondrocyte implantation versus mosaicplasty 2011;39(6):11801190


Stannard_9781604068580_Ch08.indd 101 1/30/13 2:46 PM

Particulated Articular Cartilage:
CAIS and DeNovo NT
Jack Farr, Brian J. Cole, Seth Sherman, and Vasili Karas

Articular cartilage lesions are a common stimulation is easy to perform but may also be
cause of knee symptoms.1,2 The ultimate limited with regard to the extent of durable
goal of surgical intervention is to restore hyaline-like cartilage formation, lesion size,
the patients comfort and function while the and long-term sustained clinical gains.15,16
secondary goal is to prevent or delay osteo- ACI was the first cultured chondrocyte-based
arthritis.35 As with other tissues, articular therapy, but it has variable long-term benefits
cartilage form follows function, and recent when compared with microfracture and is
studies suggest that improved clinical results technically tedious. Today, it is indicated for
correlate with better cartilage restoration second-line treatment, especially for those
constructs.6 Current surgical treatment options patients with larger chondral defects.8,17,18
for symptomatic cartilage lesions include Similar to ACI, other cultured chondrocyte
debridement/lavage, marrow stimulation, techniques (e.g., ChondroCelect; DePuy/
osteochondral autograft implantation, fresh Mitek, Raynham, MA) have promising mid-
osteochondral allograft implantation, and term results.6,16 Second- and third-generation
autologous chondrocyte implantation (ACI).712 cultured chondrocyte techniques culture the
More recently, minced cartilage autograft chondrocytes on a matrix, which improves the
(Cartilage Autograft Implantation System technical aspects, yet the results are similar to
[CAIS]; DePuy/Mitek, Raynham, MA) and par- first generation and still require two-staged
ticulated juvenile cartilage allograft (DeNovo surgical procedures for harvest and implan-
Natural Tissue [NT]; ISTO, St. Louis, MO) tation. Until recently, allograft treatment
options have been reported.3,13 options have been limited to osteochondral
One repair strategy is to use a bioactive com- grafts, as graft incorporation to host tissue was
ponent (i.e., cells or growth factors) that drives possible only at the bone level. The biologic
the biological process and a matrix (biomate- requirement of transplant bone remodeling/
rial that serves as a carrier or scaffold) that incorporation to host bone at the basilar bone
provides architectural support and facilitates layer remains a challenge, and availability is
the integration of the repaired tissue with the limited.1,2 In light of these limitations, ongoing
contiguous tissue.3 Current treatment options research continues to search for cartilage res-
have unique advantages and disadvantages. toration techniques that form durable tissue,
Autograft osteochondral plugs provide a living are technically easier for the surgeon to per-
osteochondral unit but are limited to smaller form, and are less disruptive to patients lives
lesions ranging from 1 to 2.5 cm2.14,15 Marrow during the recovery phase.

Stannard_9781604068580_Ch09.indd 102 1/30/13 2:51 PM

The concept that cartilage could be trans- pivotal study of the technique, which began
planted without its underlying bony compo- recruiting patients in 2010 and will enroll
nent and heal would have been considered over 300 patients for a randomized prospec-
heretical even a few years ago by most carti- tive comparison of CAIS to MFX (microfrac-
lage surgeons. However, the potential safety ture) (i.e., CAIS is not available for general use
and efficacy of both CAIS and DeNovo NT in the United States; use is limited to study
are challenging this paradigm. As in many patients).
aspects of science, the key to advances is In another laboratory, Yao noted these
seeing what has been there all along. While early preclinical reports and decided to
the phenomenon of hyaline cartilage repair evaluate similar studies using particulated
using particulated articular cartilage is rela- juvenile cartilage allograft (DeNovo NT; dis-

9 Particulated Articular Cartilage: CAIS and DeNovo NT

tively new to the English language literature, tributed by Zimmer, Warsaw, IN) in place of
a thorough literature review reveals a pub- autograft.1,2 This alternative approach was
lished report by Albrecht et al in the German based on two factors: (1) allograft allows
literature dating back to 1983.19 Their work conceptually no limit to the amount of har-
showed that cartilage autograft implanta- vested tissue and (2) juvenile cartilage has
tion without bone can lead to cartilage defect the potential of more robust cellular activity
healing if the cartilage is cut into small pieces. than older cartilage tissue.13,2225 Yao demon-
Most scientists in the English-speaking world strated that new extracellular matrix can be
were unaware of this article until recently, formed from juvenile cartilage cubes in an
after US-based scientists noted the produc- explant culture study.1 In addition, he dem-
tion of new chondrocyte and matrix forma- onstrated that particulated juvenile articular
tion adjacent to minced cartilage fragments. cartilage xenografts healed chondral defects
Researchers Lu and Binette began a series of on the trochlea of the horse knee joint.1
experiments to investigate these findings. Given the momentum from these positive
What followed was a rapid progression from results, the FDA now considers DeNovo NT
in vitro experiments to the mouse, goat, and as a minimally manipulated human tissue
finally the horse model.20,21 All these studies allograft, regulated as a 361 HCT/P product
together demonstrated that autograft carti- similar to fresh osteochondral allograft and
lage, when mechanically minced into cubes of bone-tendon-bone allograft. It is available for
1 to 2 mm, could affect cartilage repair.19,21 In use in clinical applications without an inves-
essence, chondrocytes in the cartilage pieces tigational device exemption (IDE) study and,
could escape from the extracellular matrix, to date, over 2,200 patients have received this
migrate, multiply, and form a new hyaline- product.2 During this same market release,
like cartilage tissue matrix that would inte- the sponsor supported a prospective study of
grate with the surrounding host tissue. In 25 patients in a multicenter study with pre-
addition, unlike cultured chondrocytes that liminary results reported that complement a
take on a spindle-shaped morphology during case report in the literature.2,26
culture, the chondrocytes from the minced
cartilage retained the standard chondrocyte
spheroid shape.21 Indications/Contraindications
These preclinical data were compelling
enough for the Food and Drug Administration The indications for CAIS (DePuy/Mitek) and
(FDA) to approve a proof of concept and safety DeNovo NT (ISTO) are evolving. In general,
pilot study of what the sponsor referred to as they mirror the selection criteria for other
CAIS.3 The clinical outcomes are now pub- cell-based cartilage procedures. On the basis
lished at 2 years, and an extension follow-up of limited clinical trials, these products are
study is complete to 4 years postoperative indicated for treatment of symptomatic artic-
with publication to follow. Based on a paral- ular cartilage defects in patients from age 18
lel European study, the technique received to 55. Prior to treatment, same-day arthro-
a CE (Conformit Europenne) mark and is scopic evaluation should confirm a cartilage
available through a limited release in Europe. lesion that is at least International Cartilage
In the United States, the FDA has approved a Repair Society (ICRS) grade 3 or higher. After 103

Stannard_9781604068580_Ch09.indd 103 1/30/13 2:51 PM

peripheral cartilage debridement, lesion achieved using epinephrine-soaked sponges
size should range from 1 to 5 cm2. As with and/or punctuated amounts of fibrin glue. An
the treatment of all cartilage defects, careful arthroscopic ruler is used to measure width,
attention must be paid to meniscal status and length, and depth of the prepared lesion.
to restoring or maintaining knee alignment Subsequently, a template sizes the area of the
and stability. Potential contraindications to lesion. Sterile paper or foil is used to make a
CAIS (DePuy/Mitek) or DeNovo NT (ISTO) template of the cartilage defect and used to
include bipolar lesion ICRS grade 2, signifi- cut the minced cartilage/scaffold construct to
cant underlying subchondral bony edema, or the appropriate size. The trimmed CAIS scaf-
osteochondritis dissecans lesion with 6 mm fold (DePuy/Mitek) implant is transferred to
subchondral bone loss, as the two last scenar- the defect with the cartilage fragments fac-
ios may require an osteochondral allograft or ing the subchondral bone and affixed with
alternative techniques. two or more biodegradable staple anchors
(prototype, Advanced Technologies and
Regenerative Medicine), which consist of
II Science and Techniques for Cartilage Repair

Surgical Technique PDO straps and tip (Advanced Technologies

and Regenerative Medicine).
DeNovo NT
Standard arthroscopic portals are established
and the lesion(s) are evaluated to confirm After confirmatory arthroscopy, a limited
size, location, and appropriateness for treat- medial or lateral arthrotomy is performed
ment. If CAIS (DePuy/Mitek) is indicated, to fully visualize the lesion(s) as shown in
hyaline cartilage is then harvested arthro- Fig. 9.1a The defect is outlined with a scalpel
scopically from a low load-bearing surface to create a shoulder (vertical peripheral wall)
(i.e., lateral wall of the intercondylar notch of normal or nearly normal host articular car-
or trochlear margin with an amount similar tilage. The cartilage within the outlined area is
to that harvested for ACI, roughly 200 mg) removed carefully with a curet to the vertical
using a unique device that minces the car- wall of the host cartilage shoulder and the base
tilage into 1- to 2-mm pieces. After harvest, of the defect (Fig. 9.1b). The base is cleared
the device (DePuy/Mitek, Raynham, MA) uni- of all cartilage tissue, including the calcified
formly disperses the minced cartilage onto layer, without entering into the subchondral
a biodegradable scaffold. (The CAIS scaffold bone. No marrow stimulation procedure is
implant consists of an absorbable copoly- performed. Hemostasis, without a tourniquet,
mer foam of 35% polycaprolactone and 65% is achieved with epinephrine-soaked cot-
polyglycolic acid, reinforced with a polydiox- tonoids and fibrin glue. After measuring the
anone [PDO mesh] [Advanced Technologies defect dimensions and recording the visual
and Regenerative Medicine, Raynham, MA].) findings with photographs, a thin aluminum
The polymer foam is designed to keep the sterile foil is pressed into the defect to cre-
tissue fragments in place and serves as a ate a three-dimensional mold, as a complete
three-dimensional scaffold for cartilage replica of the defect (Fig. 9.1c). Once formed,
matrix generation. The reinforcing PDO mesh the foil mold is removed from the defect and
enables the foam to have adequate mechani- placed on the back table of the operating
cal strength during implant handling. The room. Using the measured defect dimensions,
fragments are then secured to this scaffold the defect surface area was calculated. One
using a commercially available fibrin seal- package of DeNovo NT graft (ISTO) is used for
ant (Tisseel, Baxter, IL). A mini-arthrotomy each 2.5-cm2 defect. Larger defects require
is performed, and the defect is identified and proportionally more packages of DeNovo NT
prepared similar to the technique used for graft (ISTO).
ACI, whereby vertical lesion walls are cre- The DeNovo NT graft (ISTO), in a specially
ated and the damaged cartilage is removed formulated nutrient preservation medium, is
to the level of the subchondral bone using a shipped in an aseptic temperature-controlled
104 ring curet. If bleeding is noted, hemostasis is packaging (Fig. 9.1d). The medium is aspirated

Stannard_9781604068580_Ch09.indd 104 1/30/13 2:51 PM

9 Particulated Articular Cartilage: CAIS and DeNovo NT
Fig. 9.1 Surgical technique for DeNovo NT. (a) The are allowed to cure (g). At this time the construct
defect is identified, and a curet is used to clear the is lifted in one piece (h). Fibrin glue is added to the
base of the defect (b). (c) A thin aluminum sterile base of the defect and the cartilage construct is then
foil is pressed into the defect to create a three- placed in the defect and sealed with fibrin glue and
dimensional mold. The DeNovo NT is shipped in a allowed to cure for 10 minutes. The cartilage con-
temperature-controlled package (d). The nutrient- struct is recessed relative to the surrounding cartilage
preserving medium (e, f) is aspirated. The cells are (i). (Reprinted with permission from Farr J, Yao JQ.
then transferred to the foil mold approximately 1 mm Chondral defect repair with particulated juvenile
apart (g). Fibrin glue is then added and the cells cartilage allograft. Cartilage 2011;2(4):346353.)

(Fig. 9.1e) and the particulated cartilage the Zimmer/ISTO technique, some surgeons
pieces are transferred to the foil mold and dis- are directly applying the particulated carti-
tributed 1 to 2 mm apart (potentially less lage into the defect and gluing it in situ.26 It is
separation depending upon the ratio between imperative that the fibrin gluecartilage tis-
the implanted tissue volume and the surface sue construct is thinner (average 1 mm) than
area of the defect) (Fig. 9.1f). Fibrin glue is the surrounding cartilage shoulders (aver-
then added to the cartilage pieces until the age 2 to 3 mm), to minimize the potential for
foil mold is filled to within 1 mm of its full shear or direct compressive load.
depth (Fig. 9.1g). The glue is allowed to cure
(typically 3 to 10 minutes). At that point, the
Rehabilitation Protocol
fibrin glue/cartilage tissue construct is gently
separated and then lifted from the foil in one In general, the rehabilitation program focuses
piece (Fig. 9.1h). Fresh fibrin glue is applied initially on protection of the cartilage repair
at the base of the patients cartilage lesion process and then progresses toward con-
and the fibrin glue/particulated cartilage con- trolled loading, increased range of motion,
struct is pressed into the defect and the glue and progressive muscle strengthening.3
allowed to cure (Fig. 9.1i). As an alternative to Patients receive a different rehabilitation 105

Stannard_9781604068580_Ch09.indd 105 1/30/13 2:51 PM

protocol depending on whether they had a indicated an overall improvement in both
lesion in the patellofemoral compartment or groups, and no differences in the number of
the tibiofemoral compartment. Immediately adverse effects were noted in comparisons
after surgery, all patients receive a hinged between the CAIS (DePuy/Mitek) and MFX
knee brace locked in extension. Patients with groups. The IKDC score of the CAIS (DePuy/
a lesion on the femoral condyle are made Mitek) group was significantly higher com-
nonweightbearing for the first 2 weeks and pared with the MFX group at both 12 and
are advanced to partial weight bearing with 24 months. Select subdomains () in the
an unlocked brace from weeks 2 through 6. KOOS instrument were significantly differ-
Patients with a trochlear lesion are allowed ent at 12 and 18 months, and all subdomains
to bear weight as tolerated immediately with (Symptoms and Stiffness, Pain, Activities of
the brace locked in extension. Regardless of Daily Living, Sports and Recreation, Knee-
lesion location, the brace is removed each day Related Quality of Life) were significantly
for continuous passive motion during the first increased at 24 months in CAIS (DePuy/Mitek)
4 weeks, which is progressively increased (as versus MFX. These significant improvements
II Science and Techniques for Cartilage Repair

tolerated) during the subsequent 3 weeks. were maintained at 24 months in both IKDC
Muscle strength is maintained using isomet- and KOOS.
ric quadriceps sets, straight-leg raises, and Qualitative analysis of the imaging data did
isometric contraction of the hamstrings, hip not note differences between the two groups
abductors, and hip adductors. When toler- in fill of the graft bed, tissue integration,
ated, patients use a stationary bike with- or presence of subchondral cysts. Patients
out resistance to maintain passive range of treated with MFX had a significantly higher
motion. Patients return to low-load activity incidence of intralesional osteophyte forma-
levels at week 6 to 8 and progress in activity tion (54 and 70% of total number of lesions
as strength and comfort permitted. treated) at 6 and 12 months when compared
with CAIS (DePuy/Mitek) (8 and 25% of total
number of lesions treated).
Clinical Results

There is only one study in the clinical lit- To date, there are only two clinical studies on
erature reporting outcomes of single-stage the use of DeNovo NT (ISTO) for symptomatic
CAIS (DePuy/Mitek) for symptomatic knee cartilage lesions in the knee that are reported
cartilage defects.3 The goal of this FDA- in the literature.2,26
approved study was to establish the safety The first is a case report on the use of partic-
of CAIS (DePuy/Mitek) and to test whether ulated juvenile cartilage tissue for a sympto-
CAIS (DePuy/Mitek) improves quality of life matic full-thickness patella cartilage defect.26
by using standardized outcomes assessment At 2-year follow-up, the patient experienced
tools. A total of 29 patients was randomized substantial clinical improvement in both
with the intent to treat with either MFX or pain and function when evaluated with both
CAIS. Patients were followed at predeter- the IKDC subjective evaluation and the KOOS
mined time points for 2 years using several outcome measures. MRI at final follow-up
standardized outcomes assessment tools demonstrated fill of the defect with repair
(Short Form-36 [SF-36], International Knee tissue, and nearly complete resolution of
Documentation Committee [IKDC], Knee preoperative bony edema. Figure 9.2 shows
Injury and Osteoarthritis Outcome Score a preoperative MRI from a patient implanted
[KOOS]). Magnetic resonance imaging (MRI) with DeNovo NT (ISTO) and at 21 months
was performed at baseline, 3 weeks, and 6, postoperatively (Fig. 9.3).
12, and 24 months. The second is an early interim report
Lesion size and ICRS grade were similar of patients who are a part of an ongoing
in both groups. General outcome measures multicenter, prospective, single-arm study
106 (e.g., physical component score of the SF-36) of 25 subjects.2 This study is designed to

Stannard_9781604068580_Ch09.indd 106 1/30/13 2:51 PM

9 Particulated Articular Cartilage: CAIS and DeNovo NT
Fig. 9.3 Postoperative magnetic resonance imag-
Fig. 9.2 Preoperative magnetic resonance imag- ing at 21 months reveals near-resolution of the
ing demonstrating a full-thickness chondral defect bone edema and repair tissue within the previ-
of the patella with underlying bone edema and ous defect site. (Reprinted with permission from
early subchondral cyst formation (indicated by the Bonner KF, Daner WD, Yao JQ. 2-year postoperative
arrow). (Reprinted with permission from Bonner KF, evaluation of a patient with a symptomatic full-
Daner WD, Yao JQ. 2-year postoperative evaluation of thickness patellar cartilage defect repaired with
a patient with a symptomatic full-thickness patellar particulated juvenile cartilage tissue. J Knee Surg
cartilage defect repaired with particulated juvenile 2010;23:109113.)
cartilage tissue. J Knee Surg 2010;23:109113.)

evaluate clinical outcomes such as IKDC, MRI taken preoperatively at 12 months

KOOS, and visual analog scale (VAS) scores, and at 24 months is shown in Fig. 9.4. This
as well as extent and quality of repair demonstrates good defect fill at 24 months
with MRI and optional biopsies. To date, postoperative.
25 patients with one or two chondral lesions
on the femoral condyle or trochlea have been
enrolled at three study sites. Four patients Discussion
have completed 24 months of follow-up
and their outcomes have been recently CAIS (DePuy/Mitek) and DeNovo NT (ISTO)
reported in Cartilage.2 Detailed results of all are somewhat similar surgical procedures
25 patients will be reported once they have that involve the single-stage implantation
all reached the 2-year postoperative follow- of minced articular cartilage using either
up milestone. autograft or juvenile allograft, respectively.
Of the four patients with 2-year follow- Several in vitro and in vivo models demon-
up, three had nontraumatic cartilage lesions strate the unique ability of both particulated
and one had a traumatic cartilage injury. The autograft and juvenile allograft chondro-
average age was 43 5.4 years and body cytes to escape from the extracellular matrix,
mass index was 27 5.8 lb/in2. The aver- migrate, and form new hyaline-like cartilage
age lesion size was 2.71 1.2 cm2. Two tissue that integrates with the surround-
patients had isolated trochlear lesions, one ing host cartilage.1921 In short-term clinical
had an isolated condylar lesion, and one studies, both procedures appear to be safe,
had focal lesions of both the femoral con- feasible, and effective, with improvements in
dyle and the trochlea. KOOS, IKDC, and VAS subjective patient scores and with MRI evi-
scores demonstrate clear improvements in dence of good defect fill.2,3,26
all scores across the 24-month follow-up There are several potential advantages to
period. Most of these improvements, espe- these techniques. Both CAIS (DePuy/Mitek)
cially in KOOS and VAS, were achieved at the and DeNovo NT (ISTO) do not require the
12-month mark and maintained throughout violation of the subchondral bone, as is nec-
the study period. A representative patient essary for marrow stimulation procedures. 107

Stannard_9781604068580_Ch09.indd 107 1/30/13 2:51 PM

II Science and Techniques for Cartilage Repair

Fig. 9.4 Note that the sagittal plane is not lesion was chosen. (Reprinted with permission
identical from one image to the next secondary from Farr J, Yao JQ. Chondral defect repair with
to slight positional differences of the knee in the particulated juvenile cartilage allograft. Cartilage
MRI. The image plane that best demonstrated the 2011;2(4):346353.)

These truly represent a burn no bridge strategy of cartilagecartilage healing in the

procedure, unlike reports that prior marrow defect bed. This may help to avoid problems
stimulation may compromise subsequent of bony healing as seen in failed osteochon-
revision surgeries.27 Similarly, these pro- dral allograft procedures, including lack of
cedures avoid the need to surgically create bone incorporation, necrosis, and avascu-
an osteochondral defect, as is necessary for lar necrosislike collapse. Other potential
osteochondral allograft transplantation. CAIS advantages include (1) the use of fibrin fixa-
108 (DePuy/Mitek) and DeNovo NT (ISTO) use a tion, which eliminates problems relating to

Stannard_9781604068580_Ch09.indd 108 1/30/13 2:51 PM

flap hypertrophy, as seen with other tech- risks comparable to those of MFX. In that
niques28; (2) CAIS (DePuy/Mitek) and DeNovo study, CAIS (DePuy/Mitek) patients had
NT (ISTO) are single-stage procedures unlike consistent and progressive improvement
techniques such as ACI; (3) DeNovo NT (ISTO) during the second year after surgery, when
lacks any autogenous donor site morbid- compared with the MFX group.3 Similarly,
ity; (4) the autograft tissue portion of CAIS the preliminary results of DeNovo NT (ISTO)
(DePuy/Mitek) is obviously without charge compare favorably with 2-year postoperative
(as compared with cultured cells or allograft). KOOS pain scores for ACI and microfracture,
A disadvantage specific to CAIS (DePuy/ and with IKDC subjective scores for ACI.8,16,31
Mitek) is the potential for donor site morbid- Comparison of MRI results from CAIS (DePuy/
ity at cell harvest. This risk is minimal and Mitek) and MFX patients suggests a differ-

9 Particulated Articular Cartilage: CAIS and DeNovo NT

is, in theory, similar to the risk involved in ence in the biologic repair process. MRI from
ACI harvest. Potential disadvantages specific DeNovo NT (ISTO) patients also demonstrates
to DeNovo NT (ISTO) include the theoretical good lesion fill at early follow-up.2,8,10,26
risk of disease transmission and/or immu- Future study will require sophisticated imag-
nological rejection, which is inherent to any ing or second-look biopsies to determine
allograft procedure. The risk of disease trans- whether the quality and quantity of hyaline-
mission is extremely low in allograft proce- like fill correlates with subjective and objec-
dures due to stringent donor requirements tive clinical outcomes.
by the FDA and standard allograft screening
tests to ensure tissue safety.2 Thirty years of
cumulative knowledge has similarly shown Conclusion
that immune rejection is an extremely rare
phenomenon with osteochondral allograft CAIS (DePuy/Mitek) and DeNovo NT (ISTO)
transplantation. No immune responses have appear to be promising new treatment
been reported to the cartilage component of options for the young patient with a symp-
osteochondral allografts.2 In addition, articu- tomatic focal chondral defect in the knee.
lar cartilage has been shown to be immune Further study is needed before there
privileged, partly due to a lack of vascular- are evidence-based recommendations.
ity and the dense extracellular matrix of the Prospective randomized controlled studies
tissue.29 will certainly help to refine the indications
Other clinicians have tried to treat chondral and contraindications for both CAIS (DePuy/
defects of the knee with particulated chon- Mitek) and DeNovo NT (ISTO).
dral or osteochondral tissue from mature
donors. In particular, Stone et al implanted a
paste of autologous osteochondral tissue into References
defects concomitantly treated with micro-
fracture.30 While good clinical results were 1. Chondral Defect Repair with Particulated
Juvenile Cartilage Allograft. Updated 2010; avail-
reported, several animal studies have shown able at: http://www.zimmer.com/web/enUS/
that a combination bone and cartilage paste pdf/DeNovo_Chondral_Defect_Repair_White_
Paper_05_ 2010.pdf
forms both bone and cartilage, whereas car-
2. Farr J, Yao JQ. Chondral defect repair with
tilage pieces alone formed cartilage.19,21,30 We particulated juvenile cartilage allograft. Cartilage
believe that CAIS (DePuy/Mitek) and DeNovo 2011;2(4):346353
NT (ISTO) potentially may improve upon the 3. Cole BJ, Farr J, Winalski CS, et al. Outcomes after
a single-stage procedure for cell-based cartilage
paste-grafting concept by using a homogene- repair: a prospective clinical safety trial with
ous cartilage-only approach and by avoiding 2-year follow-up. Am J Sports Med 2011;39(6):
concomitant microfracture. 11701179
Despite the obvious limitations of short- 4. Hjelle K, Solheim E, Strand T, Muri R, Brittberg M.
Articular cartilage defects in 1,000 knee arthros-
term outcomes, the results of both CAIS copies. Arthroscopy 2002;18(7):730734
(DePuy/Mitek) and DeNovo NT (ISTO) com- 5. Maletius W, Messner K. The effect of partial
pare favorably to other procedures for simi- meniscectomy on the long-term prognosis of
knees with localized, severe chondral damage. A
lar cartilage lesions. Cole et al demonstrated twelve- to fifteen-year followup. Am J Sports Med
that CAIS (DePuy/Mitek) is safe to use, with 1996;24(3):258262 109

Stannard_9781604068580_Ch09.indd 109 1/30/13 2:51 PM

6. Saris DBF, Vanlauwe J, Victor J, et al; TIG/ACT/ with microfracture in the knee. A randomized
01/2000&EXT Study Group. Treatment of symp- trial. J Bone Joint Surg Am 2004;86-A(3):455464
tomatic cartilage defects of the knee: character- 19. Albrecht F, Roessner A, Zimmermann E. Closure of
ized chondrocyte implantation results in better osteochondral lesions using chondral fragments
clinical outcome at 36 months in a randomized and fibrin adhesive. Arch Orthop Trauma Surg
trial compared to microfracture. Am J Sports Med 1983;101(3):213217
2009;37(Suppl 1):10S19S
20. Frisbie DD, Lu Y, Kawcak CE, DiCarlo EF, Binette F,
7. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, McIlwraith CW. In vivo evaluation of autologous
Isaksson O, Peterson L. Treatment of deep carti- cartilage fragment-loaded scaffolds implanted
lage defects in the knee with autologous chondro- into equine articular defects and compared with
cyte transplantation. N Engl J Med 1994;331(14): autologous chondrocyte implantation. Am J Sports
889895 Med 2009;37(Suppl 1):71S80S
8. Zaslav K, Cole B, Brewster R, et al; STAR Study Prin- 21. Lu Y, Dhanaraj S, Wang Z, et al. Minced cartilage
cipal Investigators. A prospective study of autolo- without cell culture serves as an effective intraop-
gous chondrocyte implantation in patients with erative cell source for cartilage repair. J Orthop Res
failed prior treatment for articular cartilage defect 2006;24(6):12611270
of the knee: results of the Study of the Treatment 22. Cheung HS, Cottrell WH, Stephenson K, Nimni ME.
of Articular Repair (STAR) clinical trial. Am J Sports In vitro collagen biosynthesis in healing and nor-
Med 2009;37(1):4255 mal rabbit articular cartilage. J Bone Joint Surg Am
II Science and Techniques for Cartilage Repair

9. Gross AE, Shasha N, Aubin P. Long-term followup 1978;60(8):10761081

of the use of fresh osteochondral allografts for 23. Adkisson HD IV, Martin JA, Amendola RL, et al. The
posttraumatic knee defects. Clin Orthop Relat Res potential of human allogeneic juvenile chondro-
2005;(435):7987 cytes for restoration of articular cartilage. Am J
10. Hangody L, Kish G, Krpti Z, Szerb I, Udvarhelyi I. Sports Med 2010;38(7):13241333
Arthroscopic autogenous osteochondral mosaic- 24. Stockwell RA. The interrelationship of cell density
plasty for the treatment of femoral condylar artic- and cartilage thickness in mammalian articular
ular defects. A preliminary report. Knee Surg cartilage. J Anat 1971;109(Pt 3):411421
Sports Traumatol Arthrosc 1997;5(4):262267
25. Namba RS, Meuli M, Sullivan KM, Le AX, Adzick NS.
11. Moseley JB, OMalley K, Petersen NJ, et al. A con- Spontaneous repair of superficial defects in artic-
trolled trial of arthroscopic surgery for osteoarthri- ular cartilage in a fetal lamb model. J Bone Joint
tis of the knee. N Engl J Med 2002;347(2):8188 Surg Am 1998;80(1):410
12. Steadman JR, Rodkey WG, Singleton SB, Briggs KK. 26. Bonner KF, Daner W, Yao JQ. 2-year postoperative
Microfracture technique for full-thickness chon- evaluation of a patient with a symptomatic full-
dral defects: technique and clinical results. Oper thickness patellar cartilage defect repaired with
Tech Orthop 1997;7:300304 particulated juvenile cartilage tissue. J Knee Surg
13. Farr J. DeNovo NT natural graft tissue. Poster pre- 2010;23(2):109114
sented at: 8th World Congress of the International 27. Minas T, Gomoll AH, Rosenberger R, Royce RO,
Cartilage Repair Society 2009; Miami, FL Bryant T. Increased failure rate of autologous
14. Asik M, Ciftci F, Sen C, Erdil M, Atalar A. The chondrocyte implantation after previous treat-
microfracture technique for the treatment of ment with marrow stimulation techniques. Am J
full-thickness articular cartilage lesions of the Sports Med 2009;37(5):902908
knee: midterm results. Arthroscopy 2008;24(11): 28. Gomoll AH, Probst C, Farr J, Cole BJ, Minas T. Use of
12141220 a type I/III bilayer collagen membrane decreases
15. Cole BJ, Pascual-Garrido C, Grumet RC. Surgical reoperation rates for symptomatic hypertrophy
management of articular cartilage defects in the after autologous chondrocyte implantation. Am J
knee. J Bone Joint Surg Am 2009;91(7):17781790 Sports Med 2009;37(Suppl 1):20S23S
16. Saris DBF, Vanlauwe J, Victor J, et al. Character- 29. Gibson T, Brian Davis W, Curran R. The long-term
ized chondrocyte implantation results in better survival of cartilage homografts in man. Br J Plast
structural repair when treating symptomatic car- Surg 1959;11:177187
tilage defects of the knee in a randomized con- 30. Stone KR, Walgenbach AW, Freyer A, Turek TJ,
trolled trial versus microfracture. Am J Sports Med Speer DP. Articular cartilage paste grafting to full-
2008;36(2):235246 thickness articular cartilage knee joint lesions: a
17. Knutsen G, Drogset JO, Engebretsen L, et al. A 2- to 12-year follow-up. Arthroscopy 2006;22(3):
randomized trial comparing autologous chon- 291299
drocyte implantation with microfracture. Find- 31. Crawford DC, Heveran CM, Cannon WD Jr, Foo LF,
ings at five years. J Bone Joint Surg Am 2007;89 Potter HG. An autologous cartilage tissue implant
(10):21052112 NeoCart for treatment of grade III chondral injury
18. Knutsen G, Engebretsen L, Ludvigsen TC, et al. to the distal femur: prospective clinical safety trial
Autologous chondrocyte implantation compared at 2 years. Am J Sports Med 2009;37(7):13341343


Stannard_9781604068580_Ch09.indd 110 1/30/13 2:51 PM

Nontraditional Modification
to Articular Cartilage
Vasili Karas, Neil Ghodadra, Ellen Kroin, and Brian J. Cole

Biomechanical imbalance, trauma, and age- pharmacologic treatments for degenerative

related degeneration often lead to chondral arthritis aim to reduce inflammation and
lesions, which may lead to overt osteoar- decrease associated pain. Topical treatments
thritis over time. Such cartilage pathology is include nonsteroidal anti-inflammatories
frequently accompanied by persistent pain such as diclofenac gels that isolate the patho-
and loss of normal joint function. As a result, logic joint, localizing treatment and decreas-
patients who suffer from biologically active ing the possibility of systemic side effects.
articular cartilage lesions are often unable Traditional injectables such as cortisone
to function in high-level activities and may injections and viscosupplementation have
exhibit compromised activities of daily living. been found to decrease pain for short and
The limited potential for self-regeneration of medium time periods. Corticosteroids have
hyaline cartilage has led to the emergence of been shown to provide a 30 to 50% decrease in
new technologies to solve this difficult clini- pain that is most evident in the first 4 weeks
cal problem. In the event that the chondral after treatment.3 Viscosupplementation with
lesion remains superficial to the subchondral various formulations and molecular weights
bone, repair relies on the proliferation of sur- of hyaluronic acid has been shown to impart
rounding cells and cells within the synovium similar but longer-lasting results.4
as lesions are not exposed to the cellular and It is our purpose to discuss the nontradi-
protein components of circulating blood. tional and innovative nonsurgical treatments
Lesions that include the subchondral bone for articular cartilage pathology. Weight loss,
and expose the marrow cavity rely on com- physical therapy, oral anti-inflammatories,
ponents therein for regeneration and repair. and corticosteroids are, at present, the
Cartilage synthesized without exogenous standard of care for conservative treatment
intervention usually resembles type I fibrous modalities for arthritis. The use of biologic
cartilage, with inferior biomechanical prop- injectables such as growth factors, platelet-
erties when compared with native, hyaline rich plasma (PRP), autologous conditioned
cartilage replete with type II collagen.1 serum (ACS), and stem cell therapy is cur-
Treatment of arthritis and chondral lesions rently under investigation and will be the
includes alleviation of pain and return of func- present focus. Although the clinical evidence
tion through pharmacologic intervention and/ supporting the use of these modalities is
or attempts at cartilage reparative, restora- sparse, their potential is clear, as is the need
tive, and reconstructive options.2 Systemic for their continued development.

Stannard_9781604068580_Ch10.indd 111 1/30/13 2:55 PM

Growth Factors and The clinical success of BMP-2 in orthopedics
has spurred basic science research investi-
Cytokines gating its potential effect on cartilage regen-
eration. In multiple studies, it has been shown
Osteoarthritis is largely a cytokine-driven
in vitro to partially reverse dedifferentiated
disease process. The synovial membrane,
chondrocytes found in osteoarthritic models.24
cartilage, and subchondral bone are all
In addition, BMP-2 stimulates the synthesis
potential factors in cartilage degeneration as
and turnover of extracellular matrix, and spe-
each is capable of producing large amounts
cifically that of proteoglycans and type II col-
of cytokines. A thorough understanding of
lagen. Augmentation of a microfracture model
the clinically relevant interactions between
with BMP-2 has also been reported in a rab-
cytokines, mediators, growth factors, and
bit model. Although surgical intervention is
mechanisms of action in this local envi-
beyond the present scope, it is valuable to note
ronment is needed to ameliorate cartilage
that BMP-2 may guide differentiating cells
degeneration caused by the catabolic milieu
to produce more hyaline-like cartilage.2527
II Science and Techniques for Cartilage Repair

present in osteoarthritis. Accompanying the

Although the effects of BMP-2 on chondro-
increased interest in nontraditional treat-
cyte metabolism seem promising, synovial
ment methods for articular cartilage disease
thickening, fibrosis, and, in some cases, osteo-
is an increased interest in the use of cytokines
phytes have been shown to develop after mul-
as the basis for biological treatments such as
tiple injections.28 In addition, a recent animal
PRP and ACS.
study suggests temporal limitations to the use
Growth factors are commonly defined as
of BMP-2.29 Although the efficacy of BMP-2
biologically active polypeptides that contrib-
seems promising, further studies are needed
ute to the regulation of growth and homeo-
to develop the most efficacious dosing, timing,
stasis of tissues throughout life.5,6 The use of
and route of administration.
growth factors such as transforming growth
BMP-7/OP-1 is the most investigated
factor (TGF), fibroblast growth factor (FGF),
member of the TGF- superfamily for its
and bone morphogenic (BMP) to influence
potential to regenerate articular cartilage.
cell differentiation and anabolism is a possi-
Not only does BMP-7 increase ECM syn-
ble solution in the context of osteoarthritis.79
thesis, it decreases the activity of catabolic
Recent basic science studies have shown an
cytokines such as interleukin (IL-1), IL-6,
increasingly important role for growth factors
IL-8, matrix matalloproteinase-1 (MMP-1),
in cartilage regeneration and have become
and MMP-7.30 BMP-7 expression has been
the basis for the potential clinical benefits of
shown to decrease with age. Although
modification of articular cartilage.9,10
decreased BMP-7 expression is a factor in
TGF-1 has been shown, in vitro, to stim-
cartilage breakdown, BMP-7 continues to
ulate the synthesis of extracellular matrix
have autocrine effects for both anabolism
within cartilage, induce synovial prolifera-
and catabolism.3134 Finally, although basic
tion, and increase mesenchymal stem cell
science studies suggest a beneficial effect
(MSC) proliferation.1114 Positive effects of
from the administration of BMP-7, recent
TGF-1 have also been documented in carti-
basic science and clinical literature has not
lage defects within rabbit models.1519 Despite
shown a trend between endogenous levels of
the positive effects of TGF-1, safety concerns,
BMP-7 and higher symptomatic pain relief in
specifically the presence of osteophytes and
patients with osteoarthritis.35 The efficacy of
synovial fibrosis in murine and lapine stud-
BMP-7 seems to be clear; however, the need
ies, have limited extensive human testing.14,20
to develop the proper dosing, timing, and
Albeit on a smaller scale, compared with
route of administration remains uncertain.
TGF-1, TGF-3 has been shown to stimulate
Insulin-like growth factor-I (IGF-1) has
extracellular matrix (ECM) formation in ani-
been investigated within the context of car-
mal models2123 without these adverse effects.
tilage metabolism in both native and patho-
BMP-2 is a close structural relative to both
logic states.30,3639 IGF-1 has been shown to
TGF-1 and TGF-3 and has been studied
increase the anabolic response and decrease
112 extensively in fracture care and spine surgery.
catabolism.40 In contrast to evidence found in

Stannard_9781604068580_Ch10.indd 112 1/30/13 2:55 PM

BMP-7, IGF-1 shows a decreased responsive- spheres, exposed to chromium sulfate, and
ness in aging and osteoarthritic cartilage.41,42 placed into a centrifuge to separate into the
Although IGF-1 may not be a viable option plasma with platelets.48 ACS is believed to
alone, it may offer a synergistic effect in con- be effective through its increased concen-
junction with other growth factors.36 Further trations of cytokines and growth factors.
studies are necessary to determine the opti- Multiple studies have shown that the expres-
mal combination of growth factors. sion of IL-4, IL-10, IL-1Ra (receptor antago-
Recent evidence suggests that platelet- nist), fibroblastic growth factor-1, hepatocyte
derived growth factor (PDGF) has a possible growth factor, and TGF-1 are increased in
place in cartilage repair based on its role in human ACS. While there is an increase in
wound healing and stimulation of ECM pro- these anti-inflammatory agents, there is no
liferation in bone growth.4346 Multiple ani- increase in proinflammatory cytokines-like

10 Nontraditional Modification to Articular Cartilage

mal studies have shown that PDGF has an IL-1 or TNF-.47
excellent safety profile when used in isola- In particular, IL-1Ra expression has been
tion. PDGF has had an increasingly promi- shown to increase as much as 140-fold in
nent role in research and media as in vivo ACS. IL-1Ra is a competitive receptor antago-
use of PDGF remains largely within the con- nist of IL-1, a proinflammatory cytokine that
text of PRP. PRP has been used successfully triggers the destruction of hyaline cartilage
in various clinical situations and has drawn and its matrix.48 Thus IL-1Ra may play a role
national attention as it has shown promising in the clinical improvement of osteoarthri-
results for tendon healing. tis patients injected with ACS. IL-1 has also
been identified as being the major mediator
of cartilage loss in osteoarthritis. Currently,
Blood-Derived Products it is not clear if all biologically active IL-1
receptors need to be blocked to have a sig-
Although growth factors show promise, they nificant impact on treating conditions such
must be carefully synthesized and stored as osteoarthritis; however, it is known that
and are thus very expensive to produce. As other anti-inflammatory cytokines that are
evidenced above, they may also have a syn- expressed in ACS also affect IL-1 receptor
ergistic effect and would thus require varied signaling.48 In gene therapy studies, it was
concentrations of multiple growth factors, found that IL-1Ra decreases synovial effu-
a practice that is not sanctioned by the U.S. sion, gross articular cartilage erosion, and
Food and Drug Administration. Thus, there synovial membrane vascularity as compared
has been a recent resurgence in interest in the with placebo-treated joints.47
use of the bodys own combination of growth To induce the de novo production of
factors and cytokines using autologous blood IL-1Ra, aspirated venous blood is incubated
as a medium from which to extract growth with borosilicate glass spheres in a syringe.
factor and cytokine-containing components The anti-inflammatory cytokines, which are
such as platelets. produced by peripheral blood leukocytes,
accumulate and are recovered within the
serum. The cytokine concentrations do vary
Autologous Conditioned Serum between individual samples, and their syner-
gistic action contributes to the effects.48 After
Autologous conditioned serum (ACS) was centrifugation, ACS can be injected into the
developed in the mid-1990s and marketed osteoarthritic area in a series of six intraartic-
under the name Orthokine (Arthrex, Inc., ular injections twice a week for 3 weeks.47,52
Naples, FL). It has been reported not only to
be beneficial in the treatment of osteoarthri-
tis, but also to be beneficial in rheumatoid Platelet Rich Plasma
arthritis, spinal disorders, and muscle inju-
ries in humans.4751 To prepare an ACS injec- The contemporary definition of PRP is a
tion, human whole blood from the patient is sample of plasma with a two-fold or more
incubated with medical-grade glass beads or increase in platelet concentration or greater 113

Stannard_9781604068580_Ch10.indd 113 1/30/13 2:55 PM

than 1.1 106 platelets/L.53 Presently, sev- Stem Cell Therapy
eral different manufacturers have developed
systems for PRP preparation for augmenta- Stem cell therapy serves as another possible
tion or as primary orthopedic treatments.54 It method of treatment of articular cartilage
is important to understand that preparations defects. Not only do MSCs have the ability to
differ across manufacturers in final platelet self-renew but they possess the potential to
count, presence of leukocytes and number of differentiate into other specialized cells when
centrifugations for preparation. placed in appropriate culture conditions.62
The concept of PRP as a possible treatment For the purpose of treating cartilage defects,
for osteoarthritis derives from the platelets MSCs need to be differentiated toward chon-
role in wound healing55 as platelets contain drogenic lineage of cells and more specifically
many of the cytokines and growth factors toward the formation of hyaline type II car-
delineated above. In addition, platelets con- tilage. Aside from MSC differentiation prop-
tain approximately another 1,500 proteins, erties, they also have a trophic activity and
some of which modulate the inflammatory secrete bioactive factors that have a protec-
II Science and Techniques for Cartilage Repair

response inherent in degenerative joint dis- tive immunoregulatory effect on the local
ease as well as the attraction of fibroblasts tissue environment. Their anti-inflammatory
and stem cells to the site of injury.56,57 and differentiation properties make MSCs
The use of PRP and its reported clinical suc- good contenders for a possible tissue repair
cess in treating various tendon pathologies modality in osteoarthritis.63
throughout the body has led to increased Synovial membranederived, bone
interest in its potential role in cartilage repair. marrowderived MSCs (BMSCs), and adi-
The use of PRP as treatment for articular car- pose-derived stem cells (ASCs) from adult
tilage repair is new, and thus there are sparse tissues have the potential to form a hyaline-
data on the clinical outcomes of its use. In like cartilage matrix, with the latter being
the laboratory, injected PRP has been shown a more abundant and minimally morbid
to increase production of chondrocytes and source (Hildner).62 For example, recent stud-
MSCs, leading to increased proliferation and ies suggest that the infrapatellar fat pad of
synthesis of ECM, collagen II, and proteogly- adult knees is a good source of cells that can
cans.58,59 In animal models, damaged cartilage be induced to differentiate into chondrocytes
treated with PRP also demonstrated higher that synthesize cartilage matrix molecules.63
degree of degeneration when compared with BMSCs and ASCs require a different growth
control. In a recent trial of hyaluronic acid factor treatment to differentiate into the
(HA) versus PRP for the treatment of osteo- sought-after material. Aggrecan upregulation
arthritis, Kon et al60 compared the two treat- in ASCs is seen when treated with BMP-6,
ment modalities over a 6-month time period while in BMSCs, TGF-3 is needed instead. In
to evaluate patient-reported outcomes. The addition, several studies have concluded that
study concluded that three weekly injections BMSCs are more easily differentiated toward
of autologous PRP, when compared with a the chondrogenic lineage than ASCs.62,63
series of three HA injections, showed more Park et al showed that MSCs from both
and longer efficacy in mitigating the symp- bone marrow and periosteum formed hyaline
toms of osteoarthritis. It also concluded that cartilaginous tissue when transplanted into
younger, more active patients with presum- cartilage defects in rats. This study also dem-
ably a lesser degree of cartilage degeneration onstrated that MSCs derived from bone mar-
improve to a higher degree with PRP injec- row were superior to adipose-derived MSCs
tions as compared with HA. These results in forming hyaline cartilage in vivo.64 Bone
are the most promising to date. However, a marrow, synovium, adipose tissue, and mus-
randomized controlled trial with more objec- cle of adult rabbits have also been studied to
tive outcomes is needed to shed more defini- compare their in vivo chondrogenic potential.
tive light on PRP as a treatment for cartilage Results have shown that the potential of syn-
degeneration.60 A recent study reported a ovial and bone marrow MSCs to repair carti-
decrease in pain and an increase in function lage defects is higher than those from skeletal
114 by the time patients reached 24 months.61 muscle and adipose tissue, and they produced

Stannard_9781604068580_Ch10.indd 114 1/30/13 2:55 PM

more cartilage matrix than the other cells in under investigation with renewed vigor to
the cartilage defects. More specifically, the provide additional solutions to articular car-
MSCs taken from the synovial tissue had the tilage repair. Recent basic science and clini-
greatest proliferation potential.65 cal research have initiated a paradigm shift
Wakitani et al performed a clinical study in our understanding of the role of cytokines,
using BMSCs resuspended in a collagen type growth factors, and stem cells in poten-
I gel and transplanted with an autologous tial cartilage repair. Although results have
periosteal flap. This cell-containing scaf- been promising in animal studies, exten-
fold was placed into osteoarthritic cartilage sive human clinical studies are necessary to
defects in the patients medial femoral con- ascertain the benefit of the use of growth
dyles. This was compared with patients who factors or blood-derived products to repair
were transplanted with a cell-free scaffold articular cartilage defects.

10 Nontraditional Modification to Articular Cartilage

in a similar defect.66 Results showed that the
cell-treated groups clinical scores were not References
significantly different 64 months after trans-
plantation compared with the control group. 1. Nehrer S, Breinan HA, Ramappa A, et al. Matrix
In this situation, longer observation might collagen type and pore size influence behaviour
of seeded canine chondrocytes. Biomaterials
be required and/or MSC transplantation may 1997;18(11):769776
not be as effective in an osteoarthritic knee 2. McNickle AG, Provencher MT, Cole BJ. Overview of
environment.66 However, both the arthro- existing cartilage repair technology. Sports Med
Arthrosc 2008;16(4):196201
scopic and histological scores were better in
3. Bellamy N, Campbell J, Gee T, Robinson V, Bourne R,
the group treated with the MSC transplant.66 Wells G. Efficacy of intra-articular corticosteroid
In addition, three case reports from the same treatment in knee osteoarthritis. Cochrane Review
group did report that the clinical symptoms 2006;64:494494
4. Gigante A, Callegari L. The role of intra-articular
in the patients with the MSC transplant had hyaluronan (Sinovial) in the treatment of osteoar-
improved.65 thritis. Rheumatol Int 2011;31(4):427444
In another clinical study, human bone mar- 5. Quintana L, zur Nieden NI, Semino CE. Morphoge-
row MSCs were used to treat a 20 30-mm netic and regulatory mechanisms during devel-
opmental chondrogenesis: new paradigms for
full-thickness cartilage medial femoral con- cartilage tissue engineering. Tissue Eng Part B Rev
dyle defect in a 31-year-old male athlete. 2009;15(1):2941
Bone marrow was aspirated from the patient 6. Gaissmaier C, Koh JL, Weise K. Growth and differ-
entiation factors for cartilage healing and repair.
4 weeks before surgery, and the cells were Injury 2008;39(Suppl 1):S88S96
expanded in culture and then covered with 7. Hill DJ, Logan A. Peptide growth factors and their
an autologous periosteal flap once trans- interactions during chondrogenesis. Prog Growth
Factor Res 1992;4(1):4568
ferred to the defect. In a 7-month postsur-
8. Hoffmann A, Gross G. BMP signaling pathways in
gical evaluation, the defect was covered cartilage and bone formation. Crit Rev Eukaryot
with a hyaline-like type of cartilage tissue. Gene Expr 2001;11(13):2345
This smooth tissue also stained positively 9. Plaas A, Velasco J, Gorski DJ, et al. The relation-
ship between fibrogenic TGF1 signaling in the
with Safranin-O. Clinically, the symptoms joint and cartilage degradation in post-injury
resolved significantly and the patient was osteoarthritis. Osteoarthritis Cartilage 2011;19
able to return to full physical activity level (9):10811090
with no pain or complications.65,67 10. Puetzer JL, Petitte JN, Loboa EG. Comparative
review of growth factors for induction of three-
dimensional in vitro chondrogenesis in human
mesenchymal stem cells isolated from bone mar-
Conclusion row and adipose tissue. Tissue Eng Part B Rev
11. Dickhut A, Dexheimer V, Martin K, Lauinger R,
Since articular cartilage defects have lim- Heisel C, Richter W. Chondrogenesis of human
ited intrinsic regenerative properties, there mesenchymal stem cells by local transforming
growth factor-beta delivery in a biphasic resorba-
is an interest in providing nontraditional ble carrier. Tissue Eng Part A 2010;16(2):453464
modifications to injured articular cartilage 12. Fan H, Hu Y, Li X, et al. Ectopic cartilage formation
in patients. To explore methods to repair induced by mesenchymal stem cells on porous
gelatin-chondroitin-hyaluronate scaffold contain-
articular cartilage, transplantation of various ing microspheres loaded with TGF-beta1. Int J
progenitor cells other than chondrocytes is Artif Organs 2006;29(6):602611 115

Stannard_9781604068580_Ch10.indd 115 1/30/13 2:55 PM

13. Reem T, Kaminer-Israeli Y, Ruvinov E, Cohen S. 29. Krawczak DA, Westendorf JJ, Carlson CS, Lewis JL.
Chondrogenesis of hMSC in affinity-bound TGF- Influence of bone morphogenetic protein-2 on the
beta scaffolds. Biomaterials 2012;33(3):751761 extracellular matrix, material properties, and gene
14. Blaney Davidson EN, van der Kraan PM, van den expression of long-term articular chondrocyte
Berg WB. TGF-beta and osteoarthritis. Osteoar- cultures: loss of chondrocyte stability. Tissue Eng
thritis Cartilage 2007;15(6):597604 Part A 2009;15(6):12471255
15. Wang W, Li B, Yang J, et al. The restoration of 30. Elshaier AM, Hakimiyan AA, Rappoport L,
full-thickness cartilage defects with BMSCs and Rueger DC, Chubinskaya S. Effect of interleukin-
TGF-beta 1 loaded PLGA/fibrin gel constructs. Bio- 1beta on osteogenic protein 1-induced signaling
materials 2010;31(34):89648973 in adult human articular chondrocytes. Arthritis
Rheum 2009;60(1):143154
16. Olivos-Meza A, Fitzsimmons JS, Casper ME, et al.
Pretreatment of periosteum with TGF-beta1 in 31. Chubinskaya S, Hurtig M, Rueger DC. OP-1/BMP-7
situ enhances the quality of osteochondral tis- in cartilage repair. Int Orthop 2007;31(6):773781
sue regenerated from transplanted periosteal 32. Gavenis K, Schneider U, Wallich R, Mueller-
grafts in adult rabbits. Osteoarthritis Cartilage Rath R, Schmidt-Rohlfing B, Andereya S. Effects
2010;18(9):11831191 of low concentrated BMP-7 administered by co-
17. Guo X, Park H, Young S et al. Repair of osteochondral cultivation or plasmid transfection on human
defects with biodegradable hydrogel composites osteoarthritic chondrocytes. Int J Artif Organs
encapsulating marrow mesenchymal stem cells in 2010;33(6):339347
II Science and Techniques for Cartilage Repair

a rabbit model. Acta Biomater 2010;6(1):3947 33. Hunter DJ, Pike MC, Jonas BL, Kissin E, Krop J,
18. Fan H, Liu H, Zhu R, et al. Comparison of chondral McAlindon T. Phase 1 safety and tolerability study
defects repair with in vitro and in vivo differen- of BMP-7 in symptomatic knee osteoarthritis.
tiated mesenchymal stem cells. Cell Transplant BMC Musculoskelet Disord 2010;11:232
2007;16(8):823832 34. Hunter DJ, Pike MC, Jonas BL, Kissin E, Krop J,
19. Guo X, Zheng Q, Yang S, et al. Repair of full- McAlindon T. Phase 1 safety and tolerability study
thickness articular cartilage defects by cultured of BMP-7 in symptomatic knee osteoarthritis.
mesenchymal stem cells transfected with the BMC Musculoskelet Disord 2010;11:232
transforming growth factor beta1 gene. Biomed 35. Schmal H, Niemeyer P, Zwingmann J, Stoffel F,
Mater 2006;1(4):206215 Sdkamp NP, Mehlhorn AT. Association between
20. Bakker AC, van de Loo FA, van Beuningen HM, expression of the bone morphogenetic proteins
et al. Overexpression of active TGF-beta-1 in the 2 and 7 in the repair of circumscribed cartilage
murine knee joint: evidence for synovial-layer- lesions with clinical outcome. BMC Musculoskelet
dependent chondro-osteophyte formation. Osteo- Disord 2010;11:170
arthritis Cartilage 2001;9(2):128136 36. Chubinskaya S, Hakimiyan A, Pacione C, et al. Syn-
21. Tang QO, Shakib K, Heliotis M, et al. TGF-beta3: A ergistic effect of IGF-1 and OP-1 on matrix forma-
potential biological therapy for enhancing chondro- tion by normal and OA chondrocytes cultured in
genesis. Expert Opin Biol Ther 2009;9(6):689701 alginate beads. Osteoarthritis Cartilage 2007;15
22. Fan H, Tao H, Wu Y, Hu Y, Yan Y, Luo Z. TGF-3
immobilized PLGA-gelatin/chondroitin sulfate/ 37. Busschers E, Holt JP, Richardson DW. Effects of
hyaluronic acid hybrid scaffold for cartilage regen- glucocorticoids and interleukin-1 beta on expres-
eration. J Biomed Mater Res A 2010;95(4):982992 sion and activity of aggrecanases in equine chon-
drocytes. Am J Vet Res 2010;71(2):176185
23. Lee CH, Cook JL, Mendelson A, Moioli EK, Yao H,
Mao JJ. Regeneration of the articular surface of 38. Allen KD, Adams SB Jr, Mata BA, et al. Gait and
the rabbit synovial joint by cell homing: a proof of behavior in an IL1-mediated model of rat knee
concept study. Lancet 2010;376(9739):440448 arthritis and effects of an IL1 antagonist. J Orthop
Res 2011;29(5):694703
24. Gouttenoire J, Valcourt U, Ronzire MC, Aubert-
Foucher E, Mallein- Gerin F, Herbage D. Modulation 39. Ley C, Svala E, Nilton A, et al. Effects of high mobil-
of collagen synthesis in normal and osteoarthritic ity group box protein-1, interleukin-1, and
cartilage. Biorheology 2004;41(34):535542 interleukin-6 on cartilage matrix metabolism in
three-dimensional equine chondrocyte cultures.
25. Claus S, Aubert-Foucher E, Demoor M, et al.
Connect Tissue Res 2011;52(4):290300
Chronic exposure of bone morphogenetic pro-
tein-2 favors chondrogenic expression in human 40. Davies LC, Blain EJ, Gilbert SJ, Caterson B, Duance VC.
articular chondrocytes amplified in monolayer The potential of IGF-1 and TGFbeta1 for promoting
cultures. J Cell Biochem 2010;111(6):16421651 adult articular cartilage repair: an in vitro study.
Tissue Eng Part A 2008;14(7):12511261
26. Lee CH, Cook JL, Mendelson A, Moioli EK, Yao H,
Mao JJ. Regeneration of the articular surface of 41. Kapoor M, Martel-Pelletier J, Lajeunesse D, Pel-
the rabbit synovial joint by cell homing: a proof of letier JP, Fahmi H. Role of proinflammatory
concept study. Lancet 2010;376(9739):440448 cytokines in the pathophysiology of osteoarthritis.
Nat Rev Rheumatol 2011;7(1):3342
27. Toh WS, Liu H, Heng BC, Rufaihah AJ, Ye CP, Cao T.
Combined effects of TGFbeta1 and BMP2 in serum- 42. Fortier LA, Miller BJ. Signaling through the small
free chondrogenic differentiation of mesenchymal G-protein Cdc42 is involved in insulin-like growth
stem cells induced hyaline-like cartilage forma- factor-I resistance in aging articular chondrocytes.
tion. Growth Factors 2005;23(4):313321 J Orthop Res 2006;24(8):17651772
28. Postlethwaite AE, Raghow R, Stricklin G, Ballou L, 43. Eli N, Oragui E, Khan W. Advances in meniscal tis-
Sampath TK. Osteogenic protein-1, a bone morpho- sue engineering. Ortop Traumatol Rehabil 2011;
genic protein member of the TGF- superfamily, 13(4):319326
shares chemotactic but not fibrogenic properties 44. Brandl A, Angele P, Roll C, Prantl L, Kujat R,
with TGF-. J Cell Physiol 1994;161(3):562570 Kinner B. Influence of the growth factors PDGF-BB,

Stannard_9781604068580_Ch10.indd 116 1/30/13 2:55 PM

TGF-beta1 and bFGF on the replicative aging of 57. Senzel L, Gnatenko DV, Bahou WF. The platelet pro-
human articular chondrocytes during in vitro teome. Curr Opin Hematol 2009;16(5):329333
expansion. J Orthop Res 2010;28(3):354360 58. Mishra A, Tummala P, King A, et al. Buffered
45. Gomez-Camarillo MA, Almonte-Becerril M, platelet-rich plasma enhances mesenchymal
Vasquez Tort M, Tapia-Ramirez J, Kouri Flores JB. stem cell proliferation and chondrogenic differ-
Chondrocyte proliferation in a new culture sys- entiation. Tissue Eng Part C Methods 2009;15(3):
tem. Cell Prolif 2009;42(2):207218 431435
46. Mishima Y, Lotz M. Chemotaxis of human articu- 59. Akeda K, An HS, Okuma M, et al. Platelet-rich
lar chondrocytes and mesenchymal stem cells. J plasma stimulates porcine articular chondrocyte
Orthop Res 2008;26(10):14071412 proliferation and matrix biosynthesis. Osteoar-
47. Frisbie DD, Kawcak CE, Werpy NM, Park RD, thritis Cartilage 2006;14(12):12721280
McIlwraith CW. Clinical, biochemical, and his- 60. Kon E, Mandelbaum B, Buda R, et al. Platelet-rich
tologic effects of intra-articular administration plasma intra-articular injection versus hyalu-
of autologous conditioned serum in horses with ronic acid viscosupplementation as treatments for
experimentally induced osteoarthritis. Am J Vet cartilage pathology: from early degeneration to oste-
Res 2007;68(3):290296

10 Nontraditional Modification to Articular Cartilage

oarthritis. Arthroscopy 2011;27(11):14901501
48. Wehling P, Moser C, Frisbie D, et al. Autologous 61. Filardo G, Kon E, Buda R, et al. Platelet-rich plasma
conditioned serum in the treatment of ortho- intra-articular knee injections for the treatment of
pedic diseases: the orthokine therapy. BioDrugs degenerative cartilage lesions and osteoarthritis.
2007;21(5):323332 Knee Surg Sports Traumatol Arthrosc 2011;19(4):
49. Milano G, Deriu L, Sanna Passino E, et al. The effect 528535
of autologous conditioned plasma on the treat- 62. Hildner F, Albrecht C, Gabriel C, Redl H, van
ment of focal chondral defects of the knee. An Griensven M. State of the art and future per-
experimental study. Int J Immunopathol Pharma- spectives of articular cartilage regeneration: a
col 2011;24(1, Suppl 2):117124 focus on adipose-derived stem cells and platelet-
50. Fortier LA, Barker JU, Strauss EJ, McCarrel TM, derived products. J Tissue Eng Regen Med 2011;
Cole BJ. The role of growth factors in cartilage repair. 5(4):e36e51
Clin Orthop Relat Res 2011;469(10):27062715 63. Toghraie FS, Chenari N, Gholipour MA, et al. Treat-
51. Georg R, Maria C, Gisela A, Bianca C. Autolo- ment of osteoarthritis with infrapatellar fat pad
gous conditioned plasma as therapy of tendon derived mesenchymal stem cells in Rabbit. Knee
and ligament lesions in seven horses. J Vet Sci 2011;18(2):7175
2010;11(2):173175 64. Park J, Gelse K, Frank S, von der Mark K,
52. Baltzer AW, Moser C, Jansen SA, Krauspe R. Aigner T, Schneider H. Transgene-activated mes-
Autologous conditioned serum (Orthokine) is an enchymal cells for articular cartilage repair: a com-
effective treatment for knee osteoarthritis. Osteo- parison of primary bone marrow-, perichondrium/
arthritis Cartilage 2009;17(2):152160 periosteum- and fat-derived cells. J Gene Med
53. Miller Y, Bachowski G, Benjamin R, Eklund D, 2006;8(1):112125
Hibbard A, Lightfoot T. Practice Guidelines for 65. Koga H, Muneta T, Nagase T, et al. Comparison
Blood Transfusion: A Compilation from Recent of mesenchymal tissues-derived stem cells for in
Peer-Reviewed Literature. Washington, DC: vivo chondrogenesis: suitable conditions for cell
American Red Cross; 2007 therapy of cartilage defects in rabbit. Cell Tissue
54. Hall MP, Band PA, Meislin RJ, Jazrawi LM, Res 2008;333(2):207215
Cardone DA. Platelet-rich plasma: current con- 66. Wakitani S, Goto T, Young RG, Mansour JM,
cepts and application in sports medicine. J Am Goldberg VM, Caplan AI. Repair of large full-
Acad Orthop Surg 2009;17(10):602608 thickness articular cartilage defects with allograft
55. Nurden AT, Nurden P, Sanchez M, Andia I, Anitua articular chondrocytes embedded in a collagen
E. Platelets and wound healing. Front Biosci gel. Tissue Eng 1998;4(4):429444
2008;13:35323548 67. Kuroda R, Ishida K, Matsumoto T, et al. Treatment
56. Qureshi AH, Chaoji V, Maiguel D, et al. Proteomic of a full-thickness articular cartilage defect in the
and phospho-proteomic profile of human plate- femoral condyle of an athlete with autologous
lets in basal, resting state: insights into integrin bone-marrow stromal cells. Osteoarthritis Carti-
signaling. PLoS ONE 2009;4(10):e7627 lage 2007;15(2):226231


Stannard_9781604068580_Ch10.indd 117 1/30/13 2:55 PM

Osteochondral Autograft
Brett McCoy and Anthony Miniaci

Healthy articular cartilage transmits load to and difficulty matching the native contour of
subchondral bone while minimizing friction the condyle. Allograft transplantation has also
between articulating surfaces.1 Articular car- been performed, but availability can be an
tilage has minimal inherent healing potential, issue and it carries a theoretical risk of immu-
and the natural history of untreated lesions is nologic rejection and infection.24 The use of
progressive degenerative changes and dete- multiple smaller autologous osteochondral
rioration in functional outcomes scores.26 grafts emerged as an option to minimize
The treatment of patients with full-thickness donor morbidity and more accurately match
chondral lesions remains a difficult task for the native contour without the inherent risks
physicians. of an allograft.
Multiple treatment options exist to address The mosaicplasty technique was initially
full-thickness lesions. Early procedures such tested in canine and equine models with
as abrasion chondroplasty and microfrac- promising results.25 Further studies in the
ture targeted bone marrow stimulation to goat model demonstrated successful incor-
elicit a fibrocartilaginous healing response. poration of the graft with 86% chondrocyte
These procedures had promising short-term viability at a 6-month follow-up.26 These his-
results, but long-term outcomes have been tological results were reaffirmed in a clini-
less predictable.7,8 Osteochondral autograft cal study with longer follow-up.27 Clinical
transplantation (OAT/mosaicplasty) is a tech- use began in 1992, and long-term studies
nique that addresses these lesions with the have demonstrated successful results.2831
goal of preserving hyaline cartilage. Initial When compared with autologous chondro-
treatments predominantly addressed post- cyte implantation (ACI), mosaicplasty offers
traumatic tibiofemoral and talar pathol- the benefit of a single-stage procedure with
ogy but have been subsequently described lower cost and a shorter duration for graft
for multiple etiologies in varying anatomic adaptation and remodeling.32,33
Reports of osteochondral grafts date back to
the early 20th century.22 In 1985, Yamashita Diagnosis
et al described the transplantation of autolo-
gous osteochondral grafts for the treatment The etiology of chondral lesions includes
of large lesions.23 This technique had notable both traumatic injury and repetitive micro-
limitations, including donor site morbidity trauma. Patients will frequently present with

Stannard_9781604068580_Ch11.indd 118 1/30/13 1:58 PM

11 Osteochondral Autograft Transplantation/Mosaicplasty
Fig. 11.1 Magnetic resonance imaging demonstrating chondral injury. Sagittal fat-saturated spoiled gradient-
recalled-echo image (a), sagittal fast-spin echo intermediate-weighted image (b), and coronal image (c).

pain, swelling, and mechanical symptoms. Further advances in MRI, such as isotropic
Concomitant pathology such as meniscal or resolution reconstruction, may allow for
ligamentous injury may be the predominant improved preoperative assessment of chon-
factor in the initial symptomatology. The dral lesions, but, despite the sophistication of
chondral defect may evoke a more insidious current MRI techniques, articular lesions can
clinical picture. Thus, a high suspicion for be accurately defined only at the time of ini-
chondral lesions should be maintained dur- tial arthroscopy. It is important to counsel the
ing clinical and imaging evaluation. It is also patient about the possibility of mosaicplasty
important to note that full-thickness chon- (via open or arthroscopic means) before the
dral defects are common in athletes, and surgery.
many of these are asymptomatic.34
The physical examination should include
observation of the patients gait and over- Indications
all limb alignment. The assessment should
include evaluation for an effusion, patellar Mosaicplasty is indicated for symptomatic
maltracking, crepitance, and tenderness over focal, unipolar, full-thickness lesions (chon-
the affected area. Plain radiographs should dral and osteochondral) of the knee, includ-
include anteroposterior, Rosenberg, lateral, ing patients with OCD lesions in situ or with
and patellar views.35 These films should be the fragment missing (Fig. 11.2). The knee
scrutinized for evidence of degenerative
changes, osteochondritis dissecans (OCD),
or loose bodies. If concern for malalignment
exists, long-standing views can be obtained.
Bone scan (technetium-99 isotope) and
computed tomography (CT) (with or with-
out arthrography) have limited utility in
diagnosing chondral defects. Magnetic reso-
nance imaging (MRI) remains the preferred
advanced imaging modality.36 The most
sensitive sequence is the T1-weighted fat-
suppressed three-dimensional spoiled echo
gradient images.37 This technique utilizes
the high spatial resolution of T1-weighted Fig. 11.2 Magnetic resonance imaging pre- (a) and
images and optimizes the signal-to-noise postoperatively (b) demonstrating treatment of a full-
ratio via gradient echo techniques (Fig. 11.1). thickness lesion with mosaicplasty. 119

Stannard_9781604068580_Ch11.indd 119 1/30/13 1:58 PM

Table 11.1 Indications and contraindications for which protects the graft and functions as a
osteochondral autologous transplantation concomitant realignment procedure (if clini-
cally indicated).
Indications Contraindications
If performed arthoscopically, a post or
Full-thickness lesion Previous total
padded leg holder can be utilized per sur-
between 1 and 4 cm2 meniscectomy geon preference. The perpendicularity of
portal placement should be assessed with
Symptomatic patient Noncompliant
patient an 18-gauge spinal needle before formal
establishment. The contralateral leg can be
Contact-bearing surface Advanced age
positioned as desired, but it should undergo
Acceptable alignment Malalignment sterile prep for larger lesions as it may be
Stable joint Unstable joint needed as a site to obtain additional grafts.
Arthroscopic portals should be established
Osteochondritis dissecans
Fragment in situ in a vertical direction to allow incorporation
Fragment missing into an arthrotomy, if necessary. For arthro-
II Science and Techniques for Cartilage Repair

scopic procedures, the anteromedial and

anterolateral portals should be established
1 cm off the patellar tendon and will yield
should be stable and normally aligned. The three to four 4.5-mm grafts. Accessory portals
lesions should be greater than 1 cm2 and less can be established proximally to obtain a total
than 4 to 5 cm2 due to limitations of donor of 9 to 12 plugs depending on the size of the
availability.38 The defect should extend femur. If more graft is necessary the con-
 10 mm into the subchondral bone. Larger tralateral knee is an appropriate donor site.
lesions may be amenable to treatment with After identification of an appropriate-size
mosaicplasty in conjunction with an alterna- defect, the recipient site should be prepared.
tive technique such as microfracture or ACI, Any loose tissue should be excised and the
although limited clinical data exist at this rim should be debrided to a clean, stable
time (Table 11.1).39,40 margin using various tools (arthroscopic
resector, curet, or scalpel blade). The edges
should be oriented at 90 degrees. After the
Technical Considerations stable edges are obtained, a rasp or burr can
be applied to the base of the lesion to expose
Positioning/Preparation subchondral bone. This will allow fibrocarti-
lage ingrowth between the plugs placed. The
Patient positioning depends on surgeon graft chisel can then be placed over the lesion
preference and the location of the lesion. In to accurately determine the location of the
general, the patient should be supine and the plugs and the number required. The chisel
limb positioned to accommodate 120 degrees can gently score the recipient sites as a refer-
of flexion to ensure perpendicular access to ence for plug placement.
more posterior lesions. The decision for an
open versus arthroscopic procedure should Donor Harvest
be dictated by the location of the lesion and
the surgeon experience. Several cadaveric The ideal donor site is easily accessible and
studies demonstrate similar graft suitability provides appropriate functional tissue qual-
in open and arthroscopic procedures.41,42 ity with minimal morbidity. Traditionally,
Open procedures can be accomplished via the sites include the medial and lateral mar-
a vertical mini-arthrotomy (anterolateral or gins of the femoral trochlea and the inter-
medial parapatellar) for femoral lesions. For condylar notch (Fig. 11.3). One study noted
tibial or patellar lesions, a standard medial lower contact pressures in the medial troch-
parapatellar arthrotomy enhances visualiza- lea (when compared with lateral) and rec-
tion. Patellar lesions can also be addressed ommended this as the ideal site for harvest.43
with a lateral parapatellar arthrotomy in The intercondylar notch has several notable
120 conjunction with a tibial tubercle osteotomy, shortcomings, including thinner cartilage and

Stannard_9781604068580_Ch11.indd 120 1/30/13 1:58 PM

negatively impact chondrocyte viability.49
The grafts are harvested with double-edged
tubular cutting chisels that will allow for
accuracy in both length and diameter. If the
base of the graft is asymmetric, it can be
modified to create a flat surface and thus has
a more consistent length measurement. After
harvest, the grafts should be placed in saline-
soaked gauze and the donor sites can be filled
Fig. 11.3 Locations for graft harvest (red circles) and to potentially minimize hemarthroses. In a

11 Osteochondral Autograft Transplantation/Mosaicplasty

recipient sites (green circles). canine study, compressed collagen demon-
strated the best fibrocartilaginous fill during
histologic evaluation of the donor sites.50
a concave contour that will not match recipi-
ent sites on the femoral condyles but may
adequately address central trochlear defects. Graft Insertion
Cadaveric CT studies utilizing topographic
mapping noted that the medial and lateral The different systems for mosaicplasty
patellar groove were a better topographic require a varying amount of insertional force
match than the intercondylar notch for and some degree of toggling during graft
lesions of the weight-bearing aspect of the removal.51 Clinicians should remain aware
medial and lateral femoral condyle.44,45 Grafts of the principles of an ideal system, which
harvested from the intercondylar notch were preserves the maximal amount of viable tis-
also less perpendicular.41 The posterior con- sue with minimal tissue trauma. The grafts
dyle has also been suggested as a potential should be placed gently as excess forces have
donor site, but cadaveric data found unsuit- been demonstrated to negatively impact
able grafts based on the angle of harvest and the chondrocyte viability.52,53 If the recipient
should not be considered as a routine harvest hole is shorter than the graft, excess force
site.46 In our experience, the lateral condyle will be required to achieve congruency; thus,
is the most accessible area for graft harvest. the recommendation is equal length.54,55 The
After preparation of the donor site, mul- stability of the press fit plug is dependent
tiple systems exist for graft harvest and on several factors. In a porcine model, grafts
include both reusable and disposable types. were found to be more stable with larger
The diameter of the harvested plugs varies. diameters and shorter dilation length, and
Donor site morbidity is a concern with larger single grafts were superior to multiple grafts.
plugs ( 6 mm). Animal studies with larger No difference was noted between grafts
plugs have demonstrated the formation of aligned in a row versus a circular pattern.56
cavitary lesions and sclerotic walled cysts The grafts are anticipated to expand 0.1 to
that can result in collapse adjacent to the 0.2 mm after harvest. Thus, a conical dilator
donor site, which can result in osteoarthritic is used to help prepare the tunnel to mini-
changes.47 Smaller plugs minimize donor site mize the stresses required to insert the graft.
morbidity and result in fibrocartilaginous When the dilator is placed in the next recipi-
fill of the defects.48 The difficulty with small ent hole it will compress the bone adjacent to
plugs ( 3 mm) pertains to fragility and dif- the previously placed graft.
ficulty handling the graft. Manipulation can Congruency of the transplanted graft with
also be problematic and an increased risk the adjacent native articular cartilage is a
of fragmentation during insertion has been crucial technical aspect of the procedure
reported. The authors suggest that a 4.5-mm- (Fig. 11.4). Huang et al57 demonstrated a
diameter plug is an ideal graft with mini- limited tolerance for incongruity in a sheep
mal donor morbidity, reasonable ease of model, noting that all grafts countersunk
handling, and less concern for fragmentation.  2 mm had cartilage necrosis or over-
Grafts should be harvested manually growth. In a cadaveric study, grafts that were
as power trephination has been shown to 1 mm proud experienced a 21% increase in 121

Stannard_9781604068580_Ch11.indd 121 1/30/13 1:58 PM

peak contact pressure.58 In the setting of tis-
sue loss, graft congruency can be more dif-
ficult. For example, if a lesion has 5-mm
depth of tissue loss and the donor plug has a
length of 20 mm, then drilling to 15 mm will
achieve the ideal congruency. In other words,
the graft may remain proud of the recipient
drill hole in the setting of tissue loss to obtain
congruency with adjacent cartilage.
The reproduction of joint congruency
requires accurately positioning the plugs
to match the native contour of the articu-
lar surface. These grafts will be predomi-
nantly placed in convex locations. Starting
at the periphery of the lesion and working
II Science and Techniques for Cartilage Repair

toward the center helps to avoid a flat graft

(Figs. 11.4 and 11.5). A flat graft increases
the risk of fibrocartilaginous overgrowth,

Fig. 11.5 A depiction of a common pattern for

order of insertion for an osteochondral lesion (a) of
Fig. 11.4 Improper graft placement falls to restore the medial femoral condyle. Placement is peripheral
the contour (a) or the curvature (b). Proper graft (b) followed by central (c).
placement (c) with restoration of both the contour
and curvature due to slight obliquity.

Stannard_9781604068580_Ch11.indd 122 1/30/13 1:58 PM

the lesion. The lesion should be probed to
assess stability and debrided if the plugs do
not adequately stabilize it. The plug from the
recipient site can be placed in the donor site
if dilation is not performed.59
Postoperative cyst formation deep to the
grafts has several theoretical causes. They
include trapped or communicating syno-
vial fluid, graft necrosis, and increased graft
motion. Adequate planning can help elimi-

11 Osteochondral Autograft Transplantation/Mosaicplasty

nate some of these potential risks; for exam-
Fig. 11.6 Magnetic resonance imaging demonstrat- ple, avoiding power during graft harvesting
ing graft convergence (white arrows) because perpen-
dicular placement was not obtained. T1 (a), T2 (b).
and placement reduces the risk of thermal
necrosis. A press fit graft will eliminate
motion and synovial communication. When
which supplants the beneficial component sized properly the graft will abut the bottom
of hyaline cartilage preservation. Grafts are of the recipient hole and have good contact
generally 15 to 20 mm in depth, but the cen- along the peripheral margin (Fig. 11.9).
tral grafts may be longer than peripherally
placed grafts. The grafts should be placed in
a perpendicular or slightly oblique fashion Postoperative Course
with an attempt to avoid graft convergence
(Fig. 11.6). Weight bearing before graft incorporation can
In an OCD lesion where the fragment is be detrimental to the patients outcome. Toe-
missing, the procedure will be similar to touch weight bearing is advocated for the first
posttraumatic defects (Fig. 11.7). If the frag- 6 weeks. Range of motion of the knee can be
ment is intact, mosaicplasty can be utilized beneficial during this time period. We recom-
to confer stability to the lesion and allow for mend a brace that allows motion from 0 to
vascular inflow and the theoretical benefit 90 degrees. The patient should be encouraged
of improved healing. A Kirschner wire can to perform isometric quadriceps exercises,
be used to stabilize the graft during the pro- calf pumps, and straight-leg raises. The use
cedure. Alternatively, a screw can be placed of continuous passive motion after surgery
to lag an unstable fragment while plugs are has ample basic science support, but a sys-
placed peripherally. The screw can then be tematic review notes a lack of well-conducted
removed and replaced with a plug (Fig. 11.8). clinical trials and thus is not currently recom-
The central plug should be of adequate mended.60 Gradual weight bearing is insti-
length to reach the cancellous bone deep to tuted at 6 weeks if radiographs are acceptable.
Return to athletics requires minimal range-of-
motion deficit and quadriceps strength com-
parable to the contralateral extremity.
MRI with newer matrix assessment tech-
niques can be a useful tool for evaluating
healing but requires an experienced inter-
preter. The appearance will vary on the
basis of the technique utilized and the time
interval from intervention.36,61 Some studies
suggest that persistent edema on MRI is com-
mon after osteochondral grafting with mini-
mal relationship to clinical outcome.62,63 In a
recent study, correlation was noted between
Fig. 11.7 Magnetic resonance imaging pre- (a) and long-term clinical outcome scores and MRI
postoperative (b) after mosaicplasty. findings.31

Stannard_9781604068580_Ch11.indd 123 1/30/13 1:58 PM

II Science and Techniques for Cartilage Repair

Fig. 11.8 Osteochondritis dissecans in situ (unstable) (a) fixed first with lag screw (b) followed by peripheral
grafts (c) and then replacement of the lag screw with a graft (d).

The clinical outcome data regarding mosaic-
plasty are predominantly drawn from pro-
spective cohorts, case-control studies, and
case series. Multiple studies demonstrate
good results after mosaicplasty during short-
and midterm follow-up.6467 Solheim et al
noted a decrease in objective outcome scores
between the 1-year and 5- to 9-year follow-
up.28 However, 88% of the patients stated that
they would undergo the procedure again.
Marcacci et al noted a decrease in sports
activities during a 2- to 7-year follow-up
Fig. 11.9 Intraoperative image of mosaicplasty. after mosaicplasty (Table 11.2).30

Stannard_9781604068580_Ch11.indd 124 1/30/13 1:58 PM

Table 11.2 A summary of clinical studies evaluating mosaicplasty

Study Type Number Age (y) follow-up Complications Return to sport Outcomes

Solheim Prospective 69 33 7y Two hemarthroses N/A Lysholm 48 preop, 81 at 12 mo,

et al28 One DVT 68 after 12 mo
Three superficial wound issues VAS 62 preop, 24 at 12 mo,
One septic arthritis 32 after 12 mo
Marcacci Prospective 30 29.3 7 y No serious complications 2 y: ICRS 76% good or excellent

Stannard_9781604068580_Ch11.indd 125
et al30 Same level 73% IKDC 72 (35 preop)
Decreased level 13% Tegner 5.6 (2.9 preop)
7 y:
Same level 23%
Decreased level 47%
Gudas Prospective 60 24.3 37.1 mo Two superficial infections Same level HSS and ICRS good or excellent
et al38 randomized 93% mosaicplasty 96% mosaicplasty
52% microfracture 52% microfracture
Muller Retrospective 15 2752 42 mo Three reoperations (two persis- 92% IKDC 58.3
et al64 tent pain) (one septic arthritis) Lysholm 80.9
Subjective 7.2 (4.7 preop)
Lahav Retrospective 16 N/A 10 mo N/A N/A KOOS 80.6
et al65 IKDC 68.2
Subjective 8 (3.1 preop)
Oztrk Retrospective 19 33.1 32.4 mo None N/A Lysholm 87.5 (45.8 preop)
et al66 85% good or excellent results
Marcacci Prospective 37 29.5 2448 mo N/A 73% same level ICRS 78% good or excellent results
et al67 14% lower level
Hangody Case series 354 24.3 9.6 y 5% donor site morbidity 63% same level Good to excellent results
et al68 8% hemorrhage 27% lower level 92% talar
Two cases of septic arthritis 91% femoral
86% tibial
74% patellofemoral
Gudas Prospective 50 14.3 4.2 y One superficial infection Same level: ICRS good or excellent
et al75 randomized 1y 91% mosaicplasty
84% mosaicplasty 56% microfracture
33% microfracture
81% mosaicplasty
14% microfracture
Bentley Prospective 100 31.3 19 mo One DVT Good to excellent functional results
et al77 randomized One arthrofibrosis 88% ACI
One superficial infection 69% mosaicplasty

Abbreviations: y, year; DVT, deep venous thrombosis; preop, preoperative; mo, month; ICRS, International Cartilage Repair Society; IKDC, International Knee
Documentation Society; HSS, Hospital for Special Surgery; KOOS, Knee and Osteoarthritis Outcome Score; ACI, autologous chondrocyte implantation.

1/30/13 1:58 PM
In one of the larger series with long-term systematic review of chondral defects in the
follow-up, Hangody and Fles reported athletes knee, ACI and OAT had better clinical
on the clinical outcome of 831 patients outcomes than microfracture.73 Microfracture
who underwent mosaicplasty for small was noted to have a deterioration of results
to medium-sized lesions.48 Good to excel- over time and was less effective for larger
lent results were noted in 94% of talar, 92% lesions.74 Gudas et al demonstrated in a rand-
of femoral condylar, 87% of tibial, and 79% omized control trial that OAT returned 92% of
of patellofemoral/trochlear lesions, respec- the young athletes to an equivalent preinjury
tively. Subsequent follow-up at a mean of level versus only 52% with microfracture.75 In
9.6 years demonstrated a slight deterioration the treatment of OCD lesions, a prospective
in clinical results but good to excellent results randomized trial also demonstrated that the
in 92% of talar, 91% of femoral condylar, 86% results of microfracture deteriorate over time
of tibial, and 74% of patellofemoral lesions.68 when compared with mosaicplasty.75
Donor site morbidity assessed with the The results comparing ACI with mosaic-
Bandi score was noted in 3% of the patients. plasty are varied. A recent Cochrane review
II Science and Techniques for Cartilage Repair

Paul et al noted that when harvesting grafts identified three studies that directly compared
from asymptomatic knees for talar lesions ACI with mosaicplasty.76 In a prospective ran-
increased, body mass index was a risk factor domized trial, Bentley et al noted that ACI was
for higher morbidity at the donor site.69 This superior to mosaicplasty in clinical outcome
morbidity was identified by changes in the scoring and second-look arthroscopy, but this
Lysholm and WOMAC scores. Age, number was limited to post hoc subgroup analysis of
of grafts, and diameter of the grafts were not medial condylar lesions.77 They noted that
statistically significant. Morbidity from graft mosaicplasty failed in all five cases for patellar
harvest can be difficult to discern in the set- lesions. In a separate multicenter randomized
ting of the intervention; however, it should control trial, ACI and mosaicplasty were found
be suspected if there is persistent pain or to be clinically equivalent.33 In one study, the
mechanical symptoms that do not correlate timing of return to athletics was quickest with
with the treated lesion. OAT and slowest with ACI.73
In addition to clinical outcome meas- The conflicting data regarding patient
ures, direct visualization and histology outcomes reaffirm the belief that more
data also demonstrate good results after high-quality studies are needed to compare
mosaicplasty. Hangody et al noted congru- the various treatment options for chondral
ent gliding surfaces and survival of hyaline defects. Long-term data will also further
cartilage during second-look arthroscopy in delineate the role each treatment has on pre-
81 of 98 patients.25 Barber and Chow found venting degenerative changes.
viable chondrocytes and graft survival in all
patients at 1-year follow-up during arthro-
scopic biopsy in a small series.27 A similar Future Directions
study noted excellent congruency and sur-
vival of hyaline cartilage with fibrocartilage Technical advances have helped mosaic-
fill between the plugs.70 During total knee plasty evolve as a procedure. In a recent
arthroplasty, Huntley et al harvested 4.5-mm study, computer-assisted surgery helped
plugs and noted with laser scanning micros- improve the accuracy and precision of har-
copy that, despite good survival centrally, vest and insertion angles.78 Mosaicplasty
one-third of the lateral margin underwent may also have a role in conjunction with uni-
cell death.71 The clinical implication of this compartmental arthroplasty or resurfacing
study has not been realized. procedures in an effort to preserve as much
In a recent systematic review, Benthien et native anatomy as possible.79
al note that level I and II evidence is needed Biologic interventions continue to pose
to determine the appropriate manner to treat great potential benefit to cartilaginous
cartilage defects in the knee.72 Multiple treat- lesions. These interventions may augment
ment options exist with promising results; mosaicplasty procedures. In animal studies
126 thus, comparative studies are useful. In a evaluating substances such as hepatocyte

Stannard_9781604068580_Ch11.indd 126 1/30/13 1:58 PM

growth factor, bone morphogenic protein osteochondrosis of the second metatarsal head: a
case report. J Med Case Reports 2011;5:308
(BMP-2), and hyaluronate sodium have dem-
10. Krych AJ, Lorich DG, Kelly BT. Treatment of focal
onstrated promise as adjuncts for graft heal- osteochondral defects of the acetabulum with
ing and chondrocyte survival.8082 There osteochondral allograft transplantation. Orthope-
remains a great deal of uncertainty regarding dics 2011;34(7):e307e311
11. Nam D, Shindle MK, Buly RL, Kelly BT, Lorich DG.
the long-term prognosis of joints affected by Traumatic osteochondral injury of the femoral
articular cartilage injuries. Analysis of carti- head treated by mosaicplasty: a report of two
lage adjacent to the site of intervention has cases. HSS J 2010;6(2):228234
been performed in a laboratory setting and 12. Park TS, Kim TS, Cho JH. Arthroscopic osteochon-
dral autograft transfer in the treatment of an osteo-
may help in discerning the overall milieu of

11 Osteochondral Autograft Transplantation/Mosaicplasty

chondral defect of the humeral head: report of one
the joint, and thus the role interventions may case. J Shoulder Elbow Surg 2006;15(6):e31e36
play in the long-term health of the knee.83 13. Ozyurekoglu T. Multiple osteochondral autograft
transfer to the proximal interphalangeal joint:
case report. J Hand Surg Am 2010;35(6):931935
14. Kircher J, Patzer T, Magosch P, Lichtenberg S,
Summary Habermeyer P. Osteochondral autologous trans-
plantation for the treatment of full-thickness
cartilage defects of the shoulder: results at nine
Osteochondral autograft transplantation years. J Bone Joint Surg Br 2009;91(4):499503
has a place in the algorithm for the treat- 15. Fansa AM, Murawski CD, Imhauser CW,
Nguyen JT, Kennedy JG. Autologous osteochondral
ment of chondral injuries in the knee. There transplantation of the talus partially restores con-
is increasing clinical evidence that it may be tact mechanics of the ankle joint. Am J Sports Med
superior to microfracture in a young patient 2011;39(11):24572465
16. Liu W, Liu F, Zhao W, Kim JM, Wang Z, Vrahas MS.
with a focal chondral lesion. Osteochondral autograft transplantation for acute
osteochondral fractures associated with an ankle
fracture. Foot Ankle Int 2011;32(4):437442
References 17. Iwasaki N, Kato H, Ishikawa J, Saitoh S, Minami A.
Autologous osteochondral mosaicplasty for capitel-
lar osteochondritis dissecans in teenaged patients.
1. Hasler EM, Herzog W, Wu JZ, Mller W, Wyss U. Am J Sports Med 2006;34(8):12331239
Articular cartilage biomechanics: theoretical mod- 18. Tsuda E, Ishibashi Y, Yamamoto Y, Maeda S,
els, material properties, and biosynthetic response. Kimura Y, Sato H. Osteochondral autograft trans-
Crit Rev Biomed Eng 1999;27(6):415488 plantation for advanced stage Freiberg disease in
2. Piasecki DP, Spindler KP, Warren TA, Andrish JT, adolescent athletes: a report of 3 cases and sur-
Parker RD. Intraarticular injuries associated gical procedures. Am J Sports Med 2011;39(11):
with anterior cruciate ligament tear: findings at 24702475
ligament reconstruction in high school and rec- 19. Taketomi S, Hiraoka H, Nakagawa T, et al. Osteo-
reational athletes. An analysis of sex-based differ- chondral autograft for medial femoral condyle
ences. Am J Sports Med 2003;31(4):601605 chondral lesions in a patient with multiple epi-
3. Mandelbaum BR, Browne JE, Fu F, et al. Articular physeal dysplasia: long-term result. J Orthop Sci
cartilage lesions of the knee. Am J Sports Med 2011; May 11 (Epub ahead of print)
1998;26(6):853861 20. Lee YK, Lee M, Lee JM. Osteochondral autograft
4. Buckwalter JA, Mankin HJ. Articular cartilage: transplantation for osteochondritis dissecans of
degeneration and osteoarthritis, repair, regenera- the scaphoid: case report. J Hand Surg Am 2011;
tion, and transplantation. Instr Course Lect 1998; 36(5):820823
47:487504 21. Gobbi A, Francisco RA, Lubowitz JH, Allegra F,
5. Kujala UM, Kettunen J, Paananen H, et al. Knee Canata G. Osteochondral lesions of the talus:
osteoarthritis in former runners, soccer play- randomized controlled trial comparing chondro-
ers, weight lifters, and shooters. Arthritis Rheum plasty, microfracture, and osteochondral auto-
1995;38(4):539546 graft transplantation. Arthroscopy 2006;22(10):
6. Drawer S, Fuller CW. Propensity for osteoarthritis 10851092
and lower limb joint pain in retired professional 22. Lexer E. Substitution of whole or half joints from
soccer players. Br J Sports Med 2001;35(6):402408 freshly amputated extremities by free plastic
7. Mithoefer K, Williams RJ III, Warren RF, et al. The operation. Surg Gynecol Obstet 1908;6:601607
microfracture technique for the treatment of artic- 23. Yamashita F, Sakakida K, Suzu F, Takai S. The
ular cartilage lesions in the knee. A prospective transplantation of an autogeneic osteochondral
cohort study. J Bone Joint Surg Am 2005;87(9): fragment for osteochondritis dissecans of the
19111920 knee. Clin Orthop Relat Res 1985;(201):4350
8. Steadman JR, Briggs KK, Rodrigo JJ, Kocher MS, Gill 24. Bugbee WD, Convery FR. Osteochondral allograft
TJ, Rodkey WG. Outcomes of microfracture for trau- transplantation. Clin Sports Med 1999;18(1):6775
matic chondral defects of the knee: average 11-year 25. Hangody L, Feczk P, Bartha L, Bod G, Kish G.
follow-up. Arthroscopy 2003;19(5):477484 Mosaicplasty for the treatment of articular defects
9. Nagura I, Fujioka H, Kokubu T, Kurosaka M. Autol- of the knee and ankle. Clin Orthop Relat Res
ogous osteochondral plug transplantation for 2001;(391, Suppl):S328S336

Stannard_9781604068580_Ch11.indd 127 1/30/13 1:58 PM

26. Lane JG, Massie JB, Ball ST, et al. Follow-up of osteo- 42. Keeling JJ, Gwinn DE, McGuigan FX. A comparison
chondral plug transfers in a goat model: a 6-month of open versus arthroscopic harvesting of osteo-
study. Am J Sports Med 2004;32(6):14401450 chondral autografts. Knee 2009;16(6):458462
27. Barber FA, Chow JC. Arthroscopic osteochondral 43. Garretson RB III, Katolik LI, Verma N, Beck PR, Bach
transplantation: Histologic results. Arthroscopy BR, Cole BJ. Contact pressure at osteochondral
2001;17(8):832835 donor sites in the patellofemoral joint. Am J Sports
28. Solheim E, Hegna J, Oyen J, Austgulen OK, Med 2004;32(4):967974
Harlem T, Strand T. Osteochondral autografting 44. Bartz RL, Kamaric E, Noble PC, Lintner D,
(mosaicplasty) in articular cartilage defects in the Bocell J. Topographic matching of selected donor
knee: results at 5 to 9 years. Knee 2010;17(1): and recipient sites for osteochondral autografting
8487 of the articular surface of the femoral condyles.
29. Hangody L, Vsrhelyi G, Hangody LR, et al. Autolo- Am J Sports Med 2001;29(2):207212
gous osteochondral graftingtechnique and long- 45. Ahmad CS, Cohen ZA, Levine WN, Ateshian GA,
term results. Injury 2008;39(Suppl 1):S32S39 Mow VC. Biomechanical and topographic consid-
30. Marcacci M, Kon E, Delcogliano M, Filardo G, erations for autologous osteochondral grafting in
Busacca M, Zaffagnini S. Arthroscopic autolo- the knee. Am J Sports Med 2001;29(2):201206
gous osteochondral grafting for cartilage defects 46. Thaunat M, Beaufils P. The posterior femoral con-
of the knee: prospective study results at a mini- dyles: a potential donor site for mosaic-like osteo-
mum 7-year follow-up. Am J Sports Med 2007; chondral autograft? Arthroscopy 2010;26(11):
II Science and Techniques for Cartilage Repair

35(12):20142021 15031509
31. Tetta C, Busacca M, Moio A, et al. Knee osteo- 47. Hurtig MB, et al. THe effect of graft size and
chondral autologous transplantation: long-term number on the outcome of mosaic arthroplasty
MR findings and clinical correlations. Eur J Radiol resurfacing: an experimental model in sheep. Pro-
2010;76(1):117123 ceedings of the Arthroscopy Association of North
32. Derrett S, Stokes EA, James M, Bartlett W, Bentley America. 1999;18:1617
G. Cost and health status analysis after autologous 48. Hangody L, Fles P. Autologous osteochondral
chondrocyte implantation and mosaicplasty: a mosaicplasty for the treatment of full-thickness
retrospective comparison. Int J Technol Assess defects of weight-bearing joints: ten years of
Health Care 2005;21(3):359367 experimental and clinical experience. J Bone Joint
33. Dozin B, Malpeli M, Cancedda R, et al. Com- Surg Am 2003;85-A(Suppl 2):2532
parative evaluation of autologous chondrocyte 49. Evans PJ, Miniaci A, Hurtig MB. Manual punch
implantation and mosaicplasty: a multicentered versus power harvesting of osteochondral grafts.
randomized clinical trial. Clin J Sport Med 2005; Arthroscopy 2004;20(3):306310
50. Feczk P, Hangody L, Varga J, et al. Experimental
34. Flanigan DC, Harris JD, Trinh TQ, Siston RA, Bro- results of donor site filling for autologous osteo-
phy RH. Prevalence of chondral defects in athletes chondral mosaicplasty. Arthroscopy 2003;19(7):
knees: a systematic review. Med Sci Sports Exerc 755761
51. Barber FA, Herbert MA, McGarry JE, Barber CA.
35. Rosenberg TD, Paulos LE, Parker RD, Coward DB, Insertion force of articular cartilage transplanta-
Scott SM. The forty-five-degree posteroanterior tion systems. J Knee Surg 2008;21(3):200204
flexion weight-bearing radiograph of the knee. J
Bone Joint Surg Am 1988;70(10):14791483 52. Pylawka TK, Wimmer M, Cole BJ, Virdi AS, Wil-
liams JM. Impaction affects cell viability in osteo-
36. Hayter C, Potter H. Magnetic resonance imaging
chondral tissues during transplantation. J Knee
of cartilage repair techniques. J Knee Surg 2011;
Surg 2007;20(2):105110
53. Patil S, Butcher W, DLima DD, Steklov N, Bugbee
37. Chung CB, Frank LR, Resnick D. Cartilage imag-
WD, Hoenecke HR. Effect of osteochondral graft
ing techniques: current clinical applications and
insertion forces on chondrocyte viability. Am J
state of the art imaging. Clin Orthop Relat Res
Sports Med 2008;36(9):17261732
2001;(391, Suppl):S370S378
38. Gudas R, Kalesinskas RJ, Kimtys V, et al. A pro- 54. Kords G, Szab JS, Hangody L. The effect of drill-
spective randomized clinical study of mosaic hole length on the primary stability of osteo-
osteochondral autologous transplantation versus chondral grafts in mosaicplasty. Orthopedics
microfracture for the treatment of osteochon- 2005;28(4):401404
dral defects in the knee joint in young athletes. 55. Kock NB, Van Susante JL, Buma P, Van Kampen A,
Arthroscopy 2005;21(9):10661075 Verdonschot N. Press-fit stability of an osteo-
39. Lane JG, Healey RM, Chen AC, Sah RL, Amiel D. chondral autograft: Influence of different plug
Can osteochondral grafting be augmented with length and perfect depth alignment. Acta Orthop
microfracture in an extended-size lesion of 2006;77(3):422428
articular cartilage? Am J Sports Med 2010;38(7): 56. Kords G, Szab JS, Hangody L. Primary stabil-
13161323 ity of osteochondral grafts used in mosaicplasty.
40. Gudas R, Simonaityt R, Cekanauskas E, Arthroscopy 2006;22(4):414421
Mickeviius T. Concomitant autologous chondro- 57. Huang FS, Simonian PT, Norman AG, Clark JM.
cyte implantation with osteochondral grafting for Effects of small incongruities in a sheep model
treatment of a massive osteochondral defect in of osteochondral autografting. Am J Sports Med
the bilateral knees of a child. Medicina (Kaunas) 2004;32:18421848
2011;47(3):170173 58. Harris JD, Solak KK, Siston RA, Litsky A, Richards J,
41. Diduch DR, Chhabra A, Blessey P, Miller MD. Oste- Flanigan DC. Contact pressure comparison of proud
ochondral autograft plug transfer: achieving per- osteochondral autograft plugs versus proud syn-
128 pendicularity. J Knee Surg 2003;16(1):1720 thetic plugs. Orthopedics 2011;34(2):97

Stannard_9781604068580_Ch11.indd 128 1/30/13 1:58 PM

59. Nakagawa Y, Matsusue Y, Nakamura T. A novel 72. Benthien JP, Schwaninger M, Behrens P. We do not
surgical procedure for osteochondritis dissecans have evidence based methods for the treatment
of the lateral femoral condyle: Exchanging osteo- of cartilage defects in the knee. Knee Surg Sports
chondral plugs taken from donor and recipient Traumatol Arthrosc 2011;19(4):543552
sites. Arthroscopy 2002;18(1):E5 73. Harris JD, Brophy RH, Siston RA, Flanigan DC.
60. Fazalare JA, Griesser MJ, Siston RA, Flanigan DC. Treatment of chondral defects in the athletes
The use of continuous passive motion following knee. Arthroscopy 2010;26 (6):841852
knee cartilage defect surgery: a systematic review. 74. Magnussen RA, Dunn WR, Carey JL, Spindler KP.
Orthopedics 2010;33(12):878 Treatment of focal articular cartilage defects in the
61. Choi YS, Potter HG, Chun TJ. MR imaging of carti- knee: a systematic review. Clin Orthop Relat Res
lage repair in the knee and ankle. Radiographics 2008;466(4):952962
2008;28(4):10431059 75. Gudas R, Simonaityte R, Cekanauskas E, Tamosinas

11 Osteochondral Autograft Transplantation/Mosaicplasty

62. Kokkinakis M, Kafchitsas K, Rajeev A, Mortier J. Is R. A prospective, randomized clinical study of
MRI useful in the early follow-up after autologous osteochondral autologous transplantation versus
osteochondral transplantation? Acta Orthop Belg microfracture for the treatment of osteochondri-
2008;74(5):636642 tis dissecans in the knee joint in children. J Pediatr
63. Nemec SF, Marlovits S, Trattnig S. Persistent bone Orthop 2009;29(7):741748
marrow edema after osteochondral autograft 76. Wasiak J, Clar C, Villanueva E. Autologous cartilage
transplantation in the knee joint. Eur J Radiol implantation for full thickness articular cartilage
2009;71(1):159163 defects of the knee. Cochrane Database Syst Rev
64. Muller S, Breederveld RS, Tuinebreijer WE. Results 2006;3(3):CD003323
of osteochondral autologous transplantation in 77. Bentley G, Biant LC, Carrington RW, et al. A pro-
the knee. Open Orthop J 2010;4:111114 spective, randomised comparison of autologous
65. Lahav A, Burks RT, Greis PE, Chapman AW, Ford chondrocyte implantation versus mosaicplasty
GM, Fink BP. Clinical outcomes following osteo- for osteochondral defects in the knee. J Bone Joint
chondral autologous transplantation (OATS). J Surg Br 2003;85(2):223230
Knee Surg 2006;19(3):169173 78. Koulalis D, Di Benedetto P, Citak M, OLoughlin P,
66. Oztrk A, Ozdemir MR, Ozkan Y. Osteochondral Pearle AD, Kendoff DO. Comparative study of
autografting (mosaicplasty) in grade IV cartilage navigated versus freehand osteochondral graft
defects in the knee joint: 2- to 7-year results. Int transplantation of the knee. Am J Sports Med
Orthop 2006;30(3):200204 2009;37(4):803807
67. Marcacci M, Kon E, Zaffagnini S, et al. Multiple 79. Unnithan A, Jimulia T, Mohammed R, Learmonth
osteochondral arthroscopic grafting (mosaic- DJ. Unique combination of patellofemoral joint
plasty) for cartilage defects of the knee: prospec- arthroplasty with Osteochondral Autograft Trans-
tive study results at 2-year follow-up. Arthroscopy fer System (OATS)a case series of six knees in
2005;21(4):462470 five patients. Knee 2008;15(3):187190
68. Hangody L, Dobos J, Bal E, Pnics G, Hangody 80. Siebert CH, Schneider U, Sopka S, Wahner T,
LR, Berkes I. Clinical experiences with autologous Miltner O, Niedhart C. Ingrowth of osteochon-
osteochondral mosaicplasty in an athletic popula- dral grafts under the influence of growth factors:
tion: a 17-year prospective multicenter study. Am 6-month results of an animal study. Arch Orthop
J Sports Med 2010;38(6):11251133 Trauma Surg 2006;126(4):247252
69. Paul J, Sagstetter A, Kriner M, Imhoff AB, 81. Tibesku CO, Daniilidis K, Szuwart T, Jahn UR,
Spang J, Hinterwimmer S. Donor-site morbid- Schlegel PM, Fuchs-Winkelmann S. Influence of
ity after osteochondral autologous transplanta- hepatocyte growth factor on autologous osteo-
tion for lesions of the talus. J Bone Joint Surg Am chondral transplants in an animal model. Arch
2009;91(7):16831688 Orthop Trauma Surg 2011;131(8):11451151
70. Rdulescu RA, Cirstoiu CF, Bdil AE. Arthro- 82. Tytherleigh-Strong G, Hurtig M, Miniaci A. Intra-
scopical and histological study of cartilagi- articular hyaluronan following autogenous
nous lesions treated by mosaicplasty. J Med Life osteochondral grafting of the knee. Arthroscopy
2010;3(4):407411 2005;21(8):9991005
71. Huntley JS, Bush PG, McBirnie JM, Simpson AH, 83. Strauss EJ, Goodrich LR, Chen CT, Hidaka C,
Hall AC. Chondrocyte death associated with Nixon AJ. Biochemical and biomechanical proper-
human femoral osteochondral harvest as per- ties of lesion and adjacent articular cartilage after
formed for mosaicplasty. J Bone Joint Surg Am chondral defect repair in an equine model. Am J
2005;87(2):351360 Sports Med 2005;33(11):16471653


Stannard_9781604068580_Ch11.indd 129 1/30/13 1:58 PM

Improved Preservation of Fresh
Osteochondral Allografts for
Clinical Use
Aaron M. Stoker, Joeseph T. Garrity, Clark T. Hung,
James P. Stannard, and James L. Cook

Fresh osteochondral allografts (OCAs) have tissue biochemical and biomechanical prop-
been used clinically to treat cartilage defects erties. However, concerns regarding disease
in the knee for over 30 years.1 The major transmission stimulated the implementation
advantages of OCAs over other currently of a mandatory disease and contamination
available biologic treatment options for testing period of 14 days once tissue banks in
cartilage defects of the knee include implan- the United States made OCAs commercially
tation of hyaline cartilage and bone in a graft available.
that is site- and size-matched with tissue With the 14-day testing period require-
architecture and material properties that can ment, tissue banks had to develop protocols
withstand the loads normally transmitted to for storing OCAs in a way that would main-
the joint. Fresh OCA grafting has a reported tain sterility and chondrocyte viability until
5- to 10-year functional survival rate of 75 they could be delivered and implanted.
to 85% for treatment of focal defects of the Initially, OCAs were stored in lactated
femoral condyle (FC).24 Additionally, the lon- Ringer solution (LRS) at 4C based on stand-
gevity of the fresh OCA tissue after implan- ard protocol.1 However, chondrocyte viability
tation has been documented to be as long as in OCAs stored in LRS at 4C rapidly declines
25 years,3 indicating that this procedure can to 60% of at-harvest levels by day 7,7,8 and
provide a long-term solution for treatment of 20% by day 147 after harvest. This essen-
osteochondral defects. The factor most con- tially renders the grafts unusable for clinical
sistently reported to influence the long-term patients. To maintain chondrocyte viability
success of OCAs is the viability of chondro- in