Systematic Review

Treatment of Chondral Defects in the Athletes Knee

Joshua D. Harris, M.D., Robert H. Brophy, M.D., Robert A. Siston, Ph.D., and
David C. Flanigan, M.D.
Purpose: To determine which surgical technique(s) has improved outcomes and enables athletes to return
to their preinjury level of sports and which patient and defect factors signicantly affect outcomes after
cartilage repair or restoration. Methods: We conducted a search of multiple medical databases, evaluating
studies of articular cartilage repair in athletes. Results: We identied 11 studies for inclusion (658
subjects). Only 1 randomized clinical trial was identied. All other studies were prospective cohorts,
case-control studies, or case series reporting results after either microfracture or autologous chondrocyte
implantation (ACI) or osteoarticular transplantation (OATS). Eight different clinical outcomes measures
were used. Better clinical outcomes were observed after ACI and OATS versus microfracture. Results
after microfracture tended to deteriorate with time. The overall rate of return to preinjury level of sports
was 66%. The timing of return to the preinjury level of sports was fastest after OATS and slowest after
ACI. Defect size of less than 2 cm
2
, preoperative duration of symptoms of less than 18 months, no prior
surgical treatment, younger patient age, and higher preinjury and postsurgical level of sports all correlated
with improved outcomes after cartilage repair, especially ACI. Results after microfracture were worse
with larger defects. The rate of return to sports was generally lower after microfracture versus ACI or
OATS, and if a patient was able to return to sports, performance was diminished as well. Conclusions:
Management of chondral defects in the athlete is complex and multifactorial. There is little high-level
evidence to support one procedure over another, although good short-term and midterm outcomes with a
fair rate of return to preinjury level of sports can be achieved with cartilage repair and restoration in the
athlete. Level of Evidence: Level IV, systematic review.
A
lthough the natural history of focal articular car-
tilage injury in the knee is not completely under-
stood, it is known that articular cartilage has little
inherent capacity for healing.
1
Athletes place a high
demand on the knee and are at risk for the develop-
ment of early osteoarthritis.
2-6
The incidence and
prevalence of focal chondral defects in an athletic
population, as well as the proportion of defects that
are symptomatic, are unknown. Nevertheless, both
acute traumatic injury and chronic repetitive damage
to the articular cartilage are increasingly recognized in
the athletic population.
7,8
Because of the increased
stress placed on an athletes knee joint, a biome-
chanically durable tissue is desired after cartilage
repair or restoration. Ideally, surgical treatment of
symptomatic defects would allow this challenging
group of patients to return to sports at their pre-
symptom level. The purpose of this study was to
review the literature systematically to determine
which articular cartilage surgery techniques im-
prove clinical outcomes and enable athletes to re-
turn to their preinjury level of sports and which
patient and defect factors signicantly affect out-
comes after cartilage repair or restoration.
From the Department of Orthopaedics, Division of Sports Medicine
Cartilage Repair Center, The Ohio State University Medical Center
(J.D.H., D.C.F.) and Department of Mechanical Engineering, The
Ohio State University (R.A.S.), Columbus, Ohio; and the Depart-
ment of Orthopaedic Surgery, Washington University School of Med-
icine (R.H.B.), St. Louis, Missouri, U.S.A.
D.C.F. is on the speakers bureau for Genzyme. The authors
report no conict of interest.
Received December 8, 2009; accepted December 28, 2009.
Address correspondence and reprint requests to David C. Flani-
gan, M.D., The Ohio State University Sports Medicine Center,
2050 Kenny Rd, Ste 3100, Columbus, OH 43221-3502, U.S.A.
E-mail: david.anigan@osumc.edu
2010 by the Arthroscopy Association of North America
0749-8063/9721/$36.00
doi:10.1016/j.arthro.2009.12.030
841 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 26, No 6 (June), 2010: pp 841-852
METHODS
A systematic review of the literature was per-
formed, including Level I to IV studies based on
criteria established by the Oxford Centre for Evi-
dence-Based Medicine.
9
A search was performed by
use of the following databases: MEDLINE, Em-
base, CINAHL (Cumulative Index to Nursing and
Allied Health Literature), PubMed, SPORTDiscus,
and Cochrane Collaboration of Systematic Reviews.
The search was performed on October 23, 2009, and
repeated on October 24, 2009, to ensure accuracy.
Search key words included the following: knee,
articular cartilage, chondral, defect, lesion, athlete,
sport(s), treatment, debridement, lavage, chondro-
plasty, microfracture, autologous chondrocyte implan-
tation (ACI), osteochondral autograft, allograft, mosa-
icplasty, and osteoarticular transplantation (OATS).
All studies identied were independently reviewed by
all 4 authors and checked for potentially inclusive
references. In the event of disagreement over whether
an article should be included, the corresponding au-
thor made the nal determination. The heterogeneity
of identied studies precluded performance of a meta-
analysis, with specic attention paid to different ath-
letic populations, different inclusion criteria, different
assessments and classications of defects, different
treatments and techniques, and different outcome
measures; thus a systematic review was performed.
Inclusion criteria included the following:
English language
Human subjects
Between years 1981-2009
Randomized controlled trials, prospective cohort
studies, case-control studies, and case series
Results of studies describing the treatment of par-
tial- and full-thickness chondral defects in the knee
joints of athletes
Results of studies with a minimum follow-up of 12
months
Exclusion criteria included the following:
Non-English language
Basic science or animal studies
Expert opinion, Level V evidence studies
Surgical technique articles
Results of studies in nonathlete populations
Results of studies with less than 12 months of
follow-up
Different studies including identical subject popu-
lations, unless evaluating different data parameters
Results of studies on articular cartilage repair or
restoration in joints other than the knee
Results of studies on articular cartilage repair or
restoration in osteoarthritis
Initial search of all databases used yielded 6,532
citations. Figure 1 shows the application of the inclu-
sion and exclusion criteria. Limitation to the knee
joint yielded 1,530 citations. Further limitation to ath-
lete or sport(s) yielded 18 citations. Two studies were
excluded because they reported on results of subjects
with osteoarthritis.
10,11
One study was written in
Spanish and excluded.
12
Two studies were review
articles with expert opinion and were excluded.
8,13
FIGURE 1. Systematic review search algorithm. After application
of all inclusion and exclusion criteria, 11 studies were identied for
review.
842 J. D. HARRIS ET AL.
One study included treatment of talar defects and was
excluded.
14
Two studies reported the same data on the
same subject population,
15,16
and the more recent ar-
ticle was retained for analysis.
16
Two studies reported
on the same subject population; however, they evalu-
ated different parameters and were both included.
17,18
After application of all inclusion and exclusion crite-
ria, 11 studies were identied for this report.
16-26
RESULTS
After our initial database search and independent
manual review of all potential articles and references,
11 studies were identied for inclusion. Table 1 lists
study demographics, including patient-specic and
defect-specic data. One randomized prospective clin-
ical trial (Level I evidence) was identied, comparing
autologous osteochondral transplantation and micro-
fracture.
16
Only 1 prospective cohort study (Level II
evidence) was found, evaluating the effect of sports
and physical training after ACI.
22
Two case-control
studies (Level III evidence) were identied, evaluat-
ing the results after microfracture in active National
Basketball Association (NBA) players.
17,18
The re-
maining 7 studies included were case series (Level IV
evidence).
19-21,23-26
Surgical Interventions
There were a total of 730 subjects within the 11
articles. There were 658 subjects who underwent 1 of
3 surgical techniques and 72 control subjects (2 stud-
ies). Eight articles reported on 447 subjects who un-
derwent microfracture.
16-18,21,23-26
Three articles re-
ported on 183 subjects who underwent ACI.
19,20,22
One article reported on 28 subjects who underwent
OATS.
16
There were no studies using osteochondral
allograft in this patient group.
Microfracture was the most common intervention,
being performed in 8 of the 11 studies (73%) included
(Table 1).
16-18,21,23-26
The surgical technique was uni-
form in all studies, using the Steadman technique
27
:
By use of surgical awls of variable angles directed
perpendicular to the subchondral bone, 2- to 4-mm-
deep holes were placed 3 to 4 mm apart so as not to
coalesce and damage the general shape of the sub-
chondral architecture. After microfracture, rehabilita-
tion was similar among studies, including protected
weight bearing and continuous passive motion or as-
sisted range of motion for weight-bearing femoral
condylar defects and protected weight-bearing range
of motion for patellofemoral defects.
ACI was performed in 3 of the 11 studies (27%)
included.
19,20,22
All 3 studies used an arthrotomy for
cell implantation under a periosteal cover. One study
used chondrocytes cultured by Metreon Bioproducts
(CellGenix Technologie Transfer, Freiburg, Germany
[ARTROcell and CartiGro])
22
and the other two by Gen-
zyme Biosurgery (Cambridge, MA [Carticel]).
19,20
The
sandwich technique was used as necessary for os-
teochondral defects deeper than 1 cm, by use of either
iliac crest or proximal tibia bone graft with 2 perios-
teal patches placed supercial to the bone graft and
supercial to the implanted cells (6 of 45 [13%] and
14 of 23 [61%], respectively).
19,20
One study per-
formed mosaic-type autologous osteochondral plug
transfer using 5.5-mm plugs with 4.3 plugs per knee
(range, 3 to 6) placed in an all-arthroscopic manner.
16
Concomitant procedures were performed in 212
patients (32% overall) (Table 2). Anterior cruciate
ligament reconstruction was the most common con-
current procedure, being performed as an isolated
procedure in 14% of patients. Partial meniscectomy
and meniscal repair were performed in 11% and 2% of
patients, respectively. All other concurrent procedures
were performed in fewer than 1% of patients. Only 5
of the articles included in this review were studies of
isolated cartilage repair or restoration.
16-18,22,26
Outcomes Measures and Primary Outcomes
Outcomes measures included International Knee Doc-
umentation Committee (IKDC) score,
24
International
Cartilage Repair Society (ICRS) score,
16,22
modied
Hospital for Special Surgery score,
16
modied Cincin-
nati score,
22,26
Brittberg rating scale,
19,21
Tegner activity
score,
19-21,24
Lysholm score,
20,23,24
Marx activity score,
21
ability to return to preinjury level of sports, timing of
return to preinjury level of sports, and performance on
return to sports. The Knee Injury and Osteoarthritis Out-
come Score (KOOS) was not used in any article. Table
3 shows outcomes data per study.
The rate of return to preinjury level of sports was
66% overall (Fig 2A). After microfracture, OATS,
and ACI, the rate of return to preinjury level of sports
was 59% (range, 25% to 100%), 93%, and 78%
(range, 27% to 100%), respectively. Timing of return
to preinjury level of sports was fastest with OATS,
followed by microfracture and ACI (Fig 2B). Profes-
sional athletes (NBA and National Football League
[NFL]) had a 74% rate of return to professional sports
(range, 67% to 79%) at a mean of 7.8 months (range,
6.3 to 10 months).
17,18,23
The only randomized prospective trial comparing 2
843 CHONDRAL DEFECT TREATMENT IN ATHLETES KNEE
TABLE 1. Study Demographics, Including Patient-Specic and Defect-Specic Data
Study
Level of
Evidence Journal Intervention
Total No. of
Subjects
Mean
Age (yr)
Duration of
Symptoms
Preoperatively (mo)
Mean No. of
Previous
Operations
Defect Size
(cm
2
)
Defect
Depth*

(%)
Defect
Etiology (%)
Gudas et al.
16
I KSSTA (2006) OATS

v MFx

57 24.3 21.3 NR 2.8 III (56)*

IV (44)*
Trauma
Kreuz et al.
22
II AJSM (2007) ACI (periosteum)

118 35 6-12 NR 6.5 (3-16) IIIa or IIIb (100)* Sports trauma

Cerynik et al.
18
III KSSTA (2009) MFx

24 (24 controls) 29 NR NR NR NR Preseason (25)

Season (66)
Playoffs (9)
Namdari et al.
17
III AJSM (2009) MFx

24 (48 controls) 28.6 NR 10/24 (42%) 1 NR NR Basketball injury

Riyami and
Rolf
26
IV JOSR (2009) MFx

24 NR 0.5 (0.4-0.6) NR 1.97 (0.63-2.75) II/III or IV (100)* Acute knee trauma

with effusion
Mithoefer et al.
21
IV AJSM (2006) MFx

32 38 28 (0.5-372) 0.9 (0-7) 4.92 (0.24-20) III (84)*

IV (16)*
Trauma (59)
No trauma (41)
Mithfer et al.
20
IV AJSM (2005) ACI (periosteum)

20 15.9 21 (3-60) 2.5 (1-6) 6.4 (2.4-14) IV (39)

OCD (61)
Trauma (39)
OCD (61)
Mithfer et al.
19
IV AJSM (2005) ACI (periosteum)

45 26 26 2 (0-13) 5.7 IV (87)

OCD (13)
Acute or chronic
soccer injury
Gobbi et al.
24
IV KSSTA (2005) MFx

53 38 NR NR 4 (3-6) IV (100)

Sports trauma (58.5)

No trauma (37.5)
PF malalignment (4)
Steadman et al.
23
IV JKS (2003) MFx

25 29 NR NR 3.8 (1.5-10) IV (100)

NFL injuries
Blevins et al.
25
IV Orthopedics
(1998)
MFx

236 32 44.5 (0.2-195) NR 2.2 IV (100)

Sport injuries
Abbreviations: KSSTA, Knee Surgery, Sports Traumatology, and Arthroscopy; MFx, microfracture; NR, not reported; AJSM, American Journal of Sports Medicine; JOSR, Journal of Orthopaedic Surgery
and Research; OCD, osteochondritis dissecans; OATS, osteochondral autograft transfer; PF, patellofemoral; JKS, Journal of Knee Surgery.
*ICRS classication.
Outerbridge classication.
All-arthroscopic, 5.5-mm plugs; mean, 4.3 plugs per knee (range, 3 to 6).
Steadman technique.
27
ARTROcell and CartiGro (Metreon Bioproducts [CellGenix Technologie Transfer]).
Carticel (Genzyme Biosurgery).
8
4
4
J
.
D
.
H
A
R
R
I
S
E
T
A
L
.
techniques for the treatment of cartilage defects in
athletes showed superior results with OATS versus
microfracture.
16
In this study 57 high-performance
athletes were randomized between the 2 surgical tech-
niques and followed up annually for 3 years. Out-
comes (modied Hospital for Special Surgery and
ICRS scores) improved with OATS over time (86% at
1 year and 96% at 3 years), but microfracture results
worsened with time (76% at 1 year and 52% at 3
years). Furthermore, 93% of athletes were able to
return to their preinjury level of sports after OATS
versus 52% after microfracture. However, the rela-
tively small number of patients receiving each treat-
ment (28), as well as the possibility of bias caused by
this small sample size, could have affected the results
regarding both timing of return and rate of return to
preinjury level of sports.
Results of the microfracture technique in profes-
sional athletes are conicting. Steadman et al.
23
showed that microfracture was safe and effective and
improved symptoms, function, and activity level in
NFL players. Of 25 players, 19 (76%) were able to
return to professional football the season after micro-
fracture, and those who did return played 4.6 more
seasons and 56 more games on average. Performance
after return to play was not reported. In contrast, results
in other high-level professional athletes after microfrac-
ture showed that 21% to 33% of NBA players never
returned to play after microfracture.
17,18
Performance
was signicantly worse after microfracture, with fewer
points (P .01) and rebounds (P .05) per 40
minutes, fewer minutes per game (P .01), and
worse player efciency rating/power rating (P .01).
Athletes who underwent microfracture were 8.15
times less likely to return to the NBA after microfrac-
ture compared with control subjects (matched for age,
body mass index, position, and experience).
17
Al-
though 58% were able to play at least 1 more season
after surgery, 76% of those were on the injured list at
least once during the rst full postoperative season.
17
No studies reporting results after ACI exclusively in
professional athletes were found. However, good or
excellent results were reported at nal follow-up in
72% to 95% of competitive and recreational ath-
letes.
19,20
Signicant increases in modied Cincinnati
and ICRS scores were also reported after ACI.
22
More
competitive sports and more frequent play after ACI
were found to correlate with improved clinical scores
and increased rate of return to sports.
19,22
Kreuz et
al.
22
reported a 95% rate of return to sports at 18
months and a 100% rate of return at 36 months. Of
these athletes, the group with frequent sports (1 to 7
per week) had signicantly better ICRS and modied
Cincinnati scores than the group with little or no
sports (1 to 3 times per month). This correlation
between sports level and clinical score was observed
from 6 to 36 months. Although the overall rate of
return to sports was 33% in the study by Mithfer et
al.,
19
this rate was 83% in competitive athletes versus
16% in recreational athletes.
Patient Factors That Inuenced Outcomes
Patient age was found to be a signicant predictor
of outcomes and return to sports after cartilage repair.
The mean patient age in all studies ranged from 15.9
to 38 years. Mithoefer et al.
21
showed a signicantly
higher rate of return to high-impact sports after mi-
crofracture in athletes aged less than 40 years (P
.03). In a separate study of a group of soccer players,
Mithfer et al.
19
showed that athletes who success-
fully returned to their preinjury level of soccer after
ACI were signicantly younger (P .05), with 71%
of those aged less than 25 years returning versus 29%
of those aged greater than 25 years. Mithfer et al.
20
also evaluated a group of adolescent athletes with a
mean age of 15.9 years after ACI and showed good or
excellent results in 95% of athletes and return to
high-impact sports in 95% of athletes, with 60% re-
turning at a level equal to or higher than the preinjury
TABLE 2. Concomitant Procedures
Procedure
% of Subjects
With 1
Additional
Procedure
Total No. of
Additional
Procedures
Overall 32 212
ACL reconstruction 14 93
Partial meniscectomy 11 72
Meniscal repair 2 14
PCL reconstruction 0.8 5
TTO 0.8 5
Meniscal trephination 0.8 5
MCL repair 0.6 4
Lateral release 0.5 3
PLC reconstruction/repair 0.3 2
LCL repair 0.3 2
ACL reconstruction/HTO 0.3 2
ACL/partial meniscectomy 0.3 2
HTO 0.3 2
HTO/TTO 0.2 1
Abbreviations: ACL, anterior cruciate ligament; PCL, posterior
cruciate ligament; TTO, tibial tubercle osteotomy; MCL, medial
collateral ligament; PLC, posterolateral corner; LCL, lateral col-
lateral ligament; HTO, high tibial osteotomy.
845 CHONDRAL DEFECT TREATMENT IN ATHLETES KNEE
TABLE 3. Clinical, Imaging, Arthroscopic, and Return-to-Sport Outcomes
Study Clinical Outcomes Imaging Outcomes Arthroscopic Outcomes % RTS RTS Performance Duration of RTS Complications
Gudas et al.
16
Preoperative HSS: MFx,
77; OATS, 78
3-yr follow-up HSS: MFx,
81; OATS, 91 (P
.01)
Preoperative ICRS: MFx,
51; OATS, 51
3-yr follow-up ICRS:
MFx, 75; OATS, 89
(P .001)
No evidence of OA in
either group (on
radiographs)
MRI after OATS:
Complete osseous
incorporation (23/25)
Joint surface congruency
on MRI: OATS, 24/
25; MFx, 11/21
ICRS macroscopic
evaluation (12 mo):
Grade I/II: MFx, 9/20
(45%); OATS, 11/14
(79%)
93% OATS; 52% MFx 93% OATS and 52%
MFx RTS at preinjury
level
NR MFx, 9 failures;
OATS, 1 failure
Kreuz et al.
22
Frequent (1-7 times week)
sports group had
signicantly better
ICRS and Cincinnati
scores than little (1-3/
mo) or no sports groups
(P .001 and P
.001, respectively)
MRI showed periosteal
hypertrophy in 28% at
6 mo;
27% of those with
hypertrophy required
revision surgery;
defect ll correlated best
with ICRS score
NR All by 36 mo, 95% by
18 mo
95% and 100% at
preinjury level by 1.5
yr and 3 yr,
respectively
NR 4% revision for
periosteal
hypertrophy
Cerynik et al.
18
Power rating and minutes
per game decreased
(P .05 and P .01,
respectively) during 2
yr after surgery (v
controls)
21% (5/24) never returned
to play in NBA game
NR NR 79.2% by 30 wk Power rating, minutes per
game signicantly
decreased
70.8% played 2 NBA
seasons
NR
Namdari et al.
17
Compared with control
subjects, points and
minutes per game were
reduced (P .008 and
P .045, respectively)
during rst year after
surgery and cases
8.15 less likely to be
in NBA after index
year; 33% (8/24) never
returned to play in
NBA
NR NR 67% by 6.3 mo Points scored and
minutes played
signicantly reduced
58.3% played 1 NBA
season
NR
Riyami and
Rolf
26
Modied Cincinnati
scores improved from
preoperatively to 18 mo
Greatest modied
Cincinnati score
improvement was from
6 to 12 mo
MRI scores improved
from preoperatively to
18 mo; greatest
improvement from 6
to 12 mo
ICRS macroscopic
evaluation (6 mo):
Grade I/II, 8/10 (80%)
100% by 18 mo, 83%
by 6 mo
83% resumed full
training by 6.2 mo
NR None
Mithoefer et al.
21
66% good or excellent
results (Brittberg)
Activity scores (Marx and
Tegner) increased (P
.01 and P .01,
respectively) at nal
follow-up
NR NR 44% RTS (high impact,
pivoting)
57% RTS at preinjury
level
47% declined after
initial improvement
None
8
4
6
J
.
D
.
H
A
R
R
I
S
E
T
A
L
.
TABLE 3. Continued
Study Clinical Outcomes Imaging Outcomes Arthroscopic Outcomes % RTS RTS Performance Duration of RTS Complications
Mithfer et al.
20
95% reported good or
excellent results
(questionnaire)
Tegner score: 8.2 (before
injury) to 3.8
(preoperatively) to 7.8
(follow-up) (P .01)
Lysholm score: NR
(before injury) to 64
(preoperatively) to 87
(follow-up) (P .01)
NR NR 95% overall; 100% if
preoperative
symptoms 12 mo v
33% if preoperative
symptoms 12 mo
60% RTS at level equal
to or higher than
preinjury level
NR 15% graft
hypertrophy
Mithfer et al.
19
72% good or excellent
results (Brittberg) (93%
in single MFC lesions)
Tegner score: 3.6
(preoperatively) to 6.1
(follow-up) (P .001)
NR NR 33% overall; 83% v
16% for competitive
v recreational
80% of those RTS
returned at preinjury
level
87% of those RTS
maintained at last
follow-up
13% failure
Gobbi et al.
24
Lysholm score: 100
(before injury) to 57
(preoperatively) to 87
(follow-up)
Tegner score: 7 (before
injury) to 3.2
(preoperatively) to 6
(2 yr) to 5 (6 yr)
IKDC score: 3/53 (6%)
(preoperatively) to 37/
53 (70%) (follow-up)
Increased degenerative
changes in 30% of
patients on
radiographs, CT, and
MRI
10/53 underwent
second-look AKS,
showing rm to hard
brocartilage hybrid
tissue
NR NR NR 2 patients required
revision to ACI
Steadman et al.
23
Lysholm score: 52
(preoperatively) to 90
(follow-up)
Pain, swelling, ADL,
running, cutting,
squatting, strenuous
work, and strenuous
sports improved
NR NR 76% NR Those who RTS played
mean of 4.6 seasons
and 56 games
NR
Blevins et al.
25
Symptom improvement
(questionnaire) in both
professional and
recreational groups v
functional improvement
in professional group
only; pain scores
(questionnaire)
deteriorated in
recreational group at
3-4 yr after surgery
NR 80/178 (45%)
underwent second-
look AKS: 8% and
35% showed no
improvement
(exposed bone) in
groups A and B,
respectively
77% of professional
athletes RTS
71% of professional
athletes RTS at level
equal to or higher than
preinjury level
NR 86% of second-
look AKS were
because of new
injuries
Abbreviations: RTS, return to sports; HSS, Hospital for Special Surgery; MFx, microfracture; NR, not reported; OATS, osteochondral autograft transfer; OA, osteoarthritis; MFC, medial femoral condyle;
CT, computed tomography; AKS, arthroscopic knee surgery; ADL, activities of daily living.
8
4
7
C
H
O
N
D
R
A
L
D
E
F
E
C
T
T
R
E
A
T
M
E
N
T
I
N
A
T
H
L
E
T
E

S
K
N
E
E
level. In addition, Gudas et al.
16
showed signicantly
better clinical and functional outcomes in younger
athletes (aged 30 years) after both microfracture and
OATS (P .008).
Duration of symptoms before surgical intervention
was also an important predictor of outcome but varied
considerably between studies, with 1 study including
patients symptomatic for less than 2 to 3 weeks
26
and
2 studies including patients with symptoms for longer
than 15 years.
21,25
It appears that athletes with a
shorter preoperative period of time with symptoms
have improved outcomes after cartilage repair or res-
toration. In 3 separate studies, Mithoefer et al. showed
that athletes with symptoms for less than 12 months
had signicantly better outcomes and greater return to
sports after microfracture
21
(P .009) and ACI (P
.01 and P .05).
19,20
All athletes with symptoms for
less than 12 months returned after ACI versus only
33% of those with symptoms for greater than 12
months.
20
Soccer players with symptoms for less than
18 months before ACI had a 55% rate of return to
sports and a 66% rate of return if less than 12
months.
19
Before microfracture, symptoms for less
than 12 months correlated with a 67% rate of return to
high-impact sports versus 14% if greater than 12
months.
21
These ndings imply that early evaluation
and surgical treatment may provide better results and
a greater rate of return to sports. The reason for this
nding has yet to be determined.
The number of previous knee surgeries was also a
signicant predictor of outcome. In 1 study 86% of
athletes who successfully returned to high-impact
sports underwent microfracture as their index proce-
dure, whereas 67% of athletes who had undergone
prior surgical interventions did not return (P
.014).
21
After ACI in adolescent athletes, fewer prior
surgeries was correlated with return to preinjury ath-
letics (r 0.453, P .05).
20
Greater improvement in outcomes for high-level
competitive athletes versus recreational athletes has
been observed after both ACI and microfracture.
22,25
Whereas Blevins et al.
25
showed signicant symptom-
atic improvement in both professional and recreational
groups after microfracture, functional improvement
and increased return to sports were observed only in
the professional group.
Defect Characteristics Affecting Outcomes
The medial femoral condyle was the most common
location for all defects treated in all but 1 study
23
(Fig
3). Only 2 articles reported results (Brittberg knee
rating score) based on lesion location. In a case series
of 45 soccer players, Mithfer et al.
19
reported good or
excellent results after ACI in 72% of subjects overall
and in 93% of those players with isolated, single
medial femoral condyle defects. In a separate study
after microfracture, Mithoefer et al.
21
reported good or
FIGURE 2. (A) Overall rate of return to preinjury level of sports
(percentage of subjects who return to sports [RTS]) and return to
sports after microfracture (MFx), OATS, and ACI. (B) Timing (in
months) of return to preinjury level of sports (mean number of
subjects who return to sports [RTS]).
FIGURE 3. Distribution of defect location across studies (mean
percentage). MFC, medial femoral condyle; LFC, lateral femoral
condyle.
848 J. D. HARRIS ET AL.
excellent results in 66% of subjects overall and in 65%
of those athletes with isolated, single medial femoral
condylar lesions.
In all studies included in this review, the defects
treated were nearly full-thickness or full-thickness
chondral or osteochondral defects (ICRS grade III or
IV/Outerbridge grade III or IV). The mean defect size
in all studies ranged from 1.97 to 6.5 cm
2
. The mean
size of defects within all studies in which patients
were treated with microfracture ranged from 1.97 to
4.92 cm
2
.
16-18,21,23-26
The mean size of defects within
all studies in which patients were treated with ACI
was signicantly larger, ranging from 5.7 to 6.5
cm
2
.
19,20,22
Only 3 studies showed a difference in
clinical outcomes based on defect size. One compar-
ative study of microfracture and OATS reported sig-
nicantly worse clinical outcomes after microfracture
in defects greater than 2 cm
2
versus those defects less
than 2 cm
2
(P .04).
16
This association between
defect size was not shown in the OATS group; it
should be noted that this study had few patients, and
the results could be biased because of the sample size.
A case series of 32 athletes with minimum 2-year
follow-up after microfracture showed that athletes
with a defect smaller than 2 cm
2
had a signicantly
higher rate of return to high-impact sports (64%)
versus those who had defects larger than 2 cm
2
(22%)
(P .04).
21
A case series of 45 soccer players (mean
defect size in study, 5.7 cm
2
) with a mean of 41
months follow-up after ACI showed that defect size
was signicantly smaller (5.0 0.5 cm
2
) in soccer
players who had good or excellent results at nal
follow-up versus players who had fair or poor results
(8.5 1.9 cm
2
) (P .05).
19
Defect size in the latter
had no signicant effect on the rate of return to soccer.
Complications
Complications after ACI included periosteal hyper-
trophy
20,22
and traumatic graft delamination.
19
A pro-
spective cohort study of 118 patients with a minimum
of 36 months follow-up after ACI reported that 28%
of patients had hypertrophy on magnetic resonance
imaging (MRI) at 6 months postoperatively, but only
27% of those (n 5; 4.2% overall) were symptom-
atic, requiring revision surgery.
22
A case series of 20
adolescent athletes after ACI reported a 15% rate (n
3) of hypertrophy on MRI, successfully treated with
arthroscopic chondroplasty in all 3 cases.
20
A case
series of 45 soccer players reported failure in 13% of
patients (n 6).
19
Graft delamination was observed in
3 of these patients (50%) and was associated with
trauma. Patients with atraumatic failure (n 3) had
larger-than-average defects (9.0 3.9 cm
2
) and
longer duration of symptoms (69 16 months). Re-
vision ACI was performed in all 6 cases, with a 50%
rate of good or excellent clinical outcomes, but only 1
player (17%) was able to return to soccer.
Complications after microfracture were reported in
only 3 studies.
16,24,25
Persistent pain requiring revision
surgery was reported in 2 case series.
24,25
Of 178
patients in 1 case series, 80 (45%) were managed with
second-look arthroscopy
25
; 2 of 53 patients (4%) in
another case series had successful revision to ACI.
24
A
comparative study between microfracture and OATS
reported failure in 9 patients after microfracture, as
compared with 1 patient failure after OATS.
16
Of the
9 failures after microfracture, 8 were due to loosening
of brocartilage reparative tissue from the defect site,
requiring revision OATS, and the other 1 had arthro-
brosis, requiring debridement. The revision in the
OATS group was for a proud osteochondral plug,
which was scratching the opposite tibial cartilage.
Again, given the relatively few OATS patients in
comparison to the number of ACI and microfracture
patients, the results may be biased.
DISCUSSION
This systematic review has shown that athletes may
successfully return to competition after microfracture,
ACI, and OATS (Fig 2). Patient-specic factors, such
as younger age, shorter preoperative duration of
symptoms, no prior surgical interventions, and higher
preinjury and postsurgical level of sports, all corre-
lated with improved clinical outcomes and higher rate
of return to sports. Defect-specic factors, such as
smaller lesion size and isolated medial femoral con-
dyle lesion location, also correlated with successful
return to sports and better clinical results. Athletes are
a unique patient population that places a wide spec-
trum of demands on their knees. This review has also
shown that with a higher demand on the knee, as in
more competitive professional and amateur sports, the
ability to return to play and ability to perform on
return are affected positively after ACI
19,22
and neg-
atively after microfracture.
17,18
Little high-level evidence exists regarding the opti-
mal management of focal chondral injury in athletes.
Only 1 prospective randomized clinical trial was iden-
tied, and the small sample size of OATS patients
(28) relative to ACI (183) and microfracture (447)
may bias the results.
16
This study, however, showed
continued improvement over 1 to 3 years after OATS
849 CHONDRAL DEFECT TREATMENT IN ATHLETES KNEE
versus an initial improvement at 1 year and declining
outcomes and return to sports from 1 to 3 years after
microfracture.
The rst results on microfracture in high-level ath-
letes (NFL players) showed the procedure to be ef-
cacious with good outcomes and with a likely return to
sports the following season.
23
These latter results,
however, have not been reproduced in other groups.
Studies have shown good initial results after micro-
fracture at up to 18 to 24 months, but 4 studies in this
review have shown that outcomes deteriorate with
time, with worsening pain and decreased activity and
sport.
16,21,24,25
Defects larger than 2 cm
2
treated with
microfracture have worse clinical outcomes and a
lower rate of return to sports.
16,21
Even if able to return
to play after microfracture, performance is dimin-
ished.
17,18
Despite showing improvement in Lysholm,
Tegner, and IKDC scores at 2 years postoperatively,
Gobbi et al.
24
showed a decreased sports activity level
in 80% of patients at a mean nal follow-up of 6 years.
Furthermore, after an initial improvement through 3
years, Blevins et al.
25
reported deteriorating pain
scores at 3 to 4 years. Mithoefer et al.
21
reported
declining Tegner and Marx activity scores after a
signicant initial improvement in 47% of patients. In
a recent prospective cohort study comparing second-
generation ACI and microfracture, Kon et al.
28
showed a similar rate of return to preinjury level of
sports in both groups at 2 years. At 5 years, however,
this rate remained unchanged in the ACI group but
deteriorated in the microfracture group.
Further high-quality randomized studies are needed
to elucidate the effect of lesion size on outcome after
cartilage repair. In this review defects treated with
ACI were 2 to 3 times larger than those treated by
microfracture (mean, 2 to 5 cm
2
for microfracture v 5
to 7 cm
2
for ACI). Although smaller lesions may do
well after microfracture, ACI or OATS may be more
appropriate for larger lesions. Single defects located
on the medial femoral condyle tended to have better
outcomes.
19,21
Patient characteristics prognostic of outcome in-
cluded age, level of sports, preoperative duration of
symptoms, and number of previous surgeries before
the index cartilage repair. Patients with symptoms for
less than 12 to 18 months before microfracture or ACI
had better clinical outcomes and greater return to
sports.
19-21
Patients with fewer prior surgeries also had
better outcomes.
20,21
Patient age less than 25 to 40
years was found to correlate with better outcomes and
greater rate of return to high-impact sports.
16,19-21
More frequent and competitive sports correlated with
improved clinical scores and increased rate of return
to sports.
19,22,25
The optimal treatment approach in these patients,
especially high-level athletes, is unknown. A recent
study of articular cartilage injuries in NFL athletes
reported a mean of 8 such injuries per year over 14
years in the entire league.
29
Nearly half of these ath-
letes were treated surgically, with a signicantly
longer return to play (mean, 124 days) than those
treated nonsurgically (mean, 36 days). The NFL da-
tabase did not contain data on the actual surgical
treatment of these injuries, but a survey of NFL team
physicians within the study reported that microfrac-
ture was the most popular treatment approach (43%),
followed by debridement (31%) and nonoperative
management (13%). Smaller lesions (1 cm
2
) were
most likely to be managed with microfracture. Allo-
grafts and ACI were never selected as treatment op-
tions for these smaller lesions. Larger lesions (5
cm
2
) were still treated most often by microfracture.
Treatment choices in the tibiofemoral compartments
were similar to the overall data and to each other. For
the patellofemoral joint, however, the NFL team phy-
sicians were much more likely to select debridement
or nonsurgical management. None of the team physi-
cians would treat articular cartilage lesions in the
patellofemoral joint with allograft transplantation or
ACI. This study shows the lack of any consensus on
the treatment of articular cartilage defects in athletes.
Limitations
As a systematic review, this investigation is limited
by the quality of the studies on which it reports.
Inconsistent outcomes and heterogeneous outcome
measures are shown among most of the studies de-
scribed in this review. Outcome measures are but 1 of
the several methodologic deciencies that may ac-
count for the large variations in reported outcomes
among treatment modalities.
30
The Lysholm
31
and
Cincinnati
32
knee scores are validated outcomes mea-
sures after anterior cruciate ligament reconstruction.
The Lysholm, Tegner activity, and IKDC scores are
validated outcomes measures after meniscus re-
pair.
33,34
The only measure that has shown reliability
and validity in the evaluation of treatment of chondral
injury of the knee is the KOOS.
35
The KOOS has been
evaluated against the IKDC for evaluation of cartilage
repair patients, and although the IKDC was a higher-
scoring instrument overall in this comparative study, it
was a self-reported online questionnaire that lacked
validation and was not specic for assessing ath-
850 J. D. HARRIS ET AL.
letes.
36
No studies in this review used the KOOS as an
outcome measure, and only 1 study used the IKDC.
24
No studies were identied that reported outcomes
after osteochondral allograft treatment in athletes. The
lack of a control group in all studies on cartilage repair
or restoration precludes analysis of the real effect of
the treatment given. Although many case series do
show improved outcomes in athletes over time with
cartilage repair and restoration, only 1 comparative
study has been completed.
16
Although the isolated focal chondral defect lacks
innate healing and is thought to progress over time,
the natural history of an untreated defect is currently
unknown and the factors that determine symptomatology
are not well characterized. The long-term outcomes of
treatment are also unknown. Many studies have shown
the effect of defect size on treatment outcomes after
ACI, OATS, and microfracture.
16,19,21
Nevertheless, the
evidence to support a specic threshold size or intra-
articular location to dictate management is currently in-
conclusive, and treatment algorithms are multifactorial.
The theoretic advantage of cartilage restorative proce-
dures such as ACI and OATS is the production of
hyaline or hyaline-like cartilage with greater durability
than brocartilage produced with marrow-stimulation
procedures. Current data in athletic populations report
only midterm outcomes, and the true long-term effects
are currently unknown. Finally, as we have noted, the
number of OATS patients is small relative to ACI and
microfracture. This makes it difcult to draw a strong
conclusion on any advantage of OATS over ACI, al-
though its advantage over microfracture is suggestive.
CONCLUSIONS
Improved outcomes after cartilage repair or resto-
ration in athletes were observed in smaller defects in
younger patients with a shorter preoperative duration
of symptoms, without any prior surgical intervention,
and higher preinjury and postoperative levels of
sports. Results of microfracture appeared inferior in
rate of return and performance on return relative to
ACI or OATS, and clinical outcomes in this patient
population may deteriorate with time after microfrac-
ture. Although the literature suggests that ACI and
OATS may provide better results for athletes com-
pared with microfracture, only further randomized
prospective clinical trials will elucidate the optimal
surgical treatment for focal chondral injury in the
athletes knee.
REFERENCES
1. Buckwalter J. Articular cartilage: Injuries and potential for
healing. J Orthop Sports Phys Ther 1998;28:192-202.
2. Mandelbaum B, Browne J, Fu F, et al. Articular cartilage
lesions of the knee. Am J Sports Med 1998;26:853-861.
3. Piasecki D, Spindler K, Warren T, Andrish J, Parker R. Intra-
articular injuries associated with ACL tear: Findings at liga-
ment reconstruction in high school and recreational athletes.
Am J Sports Med 2003;31:601-605.
4. Arendt E, Dick R. Knee injury patterns among men and
women in collegiate basketball and soccerNCAA data and
review of literature. Am J Sports Med 1995;23:694-701.
5. Kujala U, Kettunen J, Paananen H. Knee osteoarthritis in
former runners, soccer players, weight lifters, and shooters.
Arthritis Rheum 1995;38:539-546.
6. Drawer S, Fuller C. Propensity for osteoarthritis and lower
limb joint pain in retired professional soccer players. Br J
Sports Med 2001;35:402-408.
7. Levy A, Lohnes J, Sculley S, LeCroy M, Garrett W. Chondral
delamination of the knee in soccer players. Am J Sports Med
1996;24:634-639.
8. Mithoefer K, McAdams T, Scopp J, Mandelbaum B. Emerging
options for treatment of articular cartilage injury in the athlete.
Clin Sports Med 2009;28:25-40.
9. Obremskey W, Pappas N, Attallah-Wasif E, Tornetta P, Bhan-
dari M. Levels of evidence in orthopaedic journals. J Bone
Joint Surg Am 2005;87:2632-2638.
10. Goldman R, Scuderi G, Kelly M. Arthroscopic treatment of the
degenerative knee in older athletes. Clin Sports Med 1997;16:
51-68.
11. Calvert G, Wright R. The use of arthroscopy in the athlete with
knee osteoarthritis. Clin Sports Med 2005;24:133-152.
12. Mina N, Cuellar E, Chapparro I. Treatment of chondral lesions
from knee injuries in the professional soccer players. Review
of 34 cases. Acta Ortop Mex 2007;21:267-273.
13. Micheli L, Curtis C, Shervin N. Articular cartilage repair in the
adolescent athlete: Is autologous chondrocyte implantation the
answer? Clin J Sports Med 2006;16:465-470.
14. Kish G, Modis L, Hangody L. Osteochondral mosaicplasty for
the treatment of focal chondral and osteochondral lesions of the
knee and talus in the athlete. Clin Sports Med 1999;18:45-67.
15. Gudas R, Kalesinkas R, Kimtys V, et al. A prospective ran-
domized clinical study of mosaic osteochondral autologous
transplantation versus microfracture for the treatment of os-
teochondral defects in the knee joint in young athletes. Ar-
throscopy 2005;21:1066-1075.
16. Gudas R, Stankevicius E, Monastyreckiene E, Pranys D, Kale-
sinkskas R. Osteochondral autologous transplantation versus
microfracture for the treatment of articular cartilage defects in
the knee joint in athletes. Knee Surg Sports Traumatol Ar-
throsc 2006;14:834-842.
17. Namdari S, Baldwin K, Anakwenze O, Park M, Huffman G,
Sennett B. Results and performance after microfracture in
National Basketball Association athletes. Am J Sports Med
2009;37:943-949.
18. Cerynik D, Lewullis G, Joves B, Palmer M, Tom J. Outcomes
of microfracture in professional basketball players. Knee Surg
Sports Traumatol Arthrosc 2009;17:1135-1139.
19. Mithfer K, Peterson L, Mandelbaum B, Minas T. Articular
cartilage repair in soccer players with autologous chondrocyte
transplantation. Am J Sports Med 2005;33:1639-1646.
20. Mithfer K, Minas T, Peterson L, Yeon H, Micheli L. Func-
tional outcome of knee articular cartilage repair in adolescent
athletes. Am J Sports Med 2005;33:1147-1153.
21. Mithoefer K, Williams RJ III, Warren RF, Wickiewicz TL, Marx
RG. High-impact athletics after knee articular cartilage repair: A
851 CHONDRAL DEFECT TREATMENT IN ATHLETES KNEE
prospective evaluation of the microfracture technique. Am J
Sports Med 2006;34:1413-1418.
22. Kreuz P, Steinwachs M, Erggelet C, et al. Importance of sports
in cartilage regeneration after autologous chondrocyte implan-
tation: A prospective study with a 3-year follow-up. Am J
Sports Med 2007;35:1261-1268.
23. Steadman J, Miller B, Karas S, Schlegel T, Briggs K, Hawkins
R. The microfracture technique in the treatment of full-thick-
ness chondral lesions of the knee in NFL players. J Knee Surg
2003;16:83-86.
24. Gobbi A, Nunag P, Malinowski K. Treatment of full thickness
chondral lesions of the knee with microfracture in a group of
athletes. Knee Surg Sports Traumatol Arthrosc 2005;13:213-
221.
25. Blevins F, Steadman J, Rodrigo J, Silliman J. Treatment of
articular cartilage defects in athletes: An analysis of functional
outcome and lesions appearance. Orthopedics 1998;21:761-
767.
26. Riyami M, Rolf C. Evaluation of microfracture of traumatic
chondral injuries to the knee in professional football and rugby
players. J Orthop Surg Res 2009;4:1-8.
27. Steadman J, Rodkey W, Rodrigo J. Microfracture: Surgical
technique and rehabilitation to treat chondral defects. Clin
Orthop Relat Res 2001:S362-S369 (Suppl).
28. Kon E, Gobbi A, Filardo G, Delcogliano A, Zaffagnini S,
Marcacci M. Arthroscopic second-generation autologous
chondrocyte implantation compared with microfracture for
chondral lesions of the knee. Am J Sports Med 2009;37:33-41.
29. Brophy R, Rodeo S, Barnes R, Powell J, Warren R. Knee
articular cartilage injuries in the National Football League. J
Knee Surg 2009;22:331-338.
30. Jakobsen R, Engebretsen L, Slauterbeck J. An analysis of the
quality of cartilage repair studies. J Bone Joint Surg Am
2005;87:2232-2239.
31. Lysholm J, Gillquist J. Evaluation of knee ligament surgery
results with special emphasis on use of a scoring scale. Am J
Sports Med 1982;10:150-154.
32. Barber-Westin S, Noyes F, McCloskey J. Rigorous statistical
reliability, validity, and responsiveness testing of the Cincin-
nati Knee Rating System in 350 subjects with uninjured,
injured, or ACL-reconstructed knees. Am J Sports Med 1999;
27:402-416.
33. Crawford K, Briggs K, Rodkey W, Steadman J. Reliability,
validity, and responsiveness of the IKDC score for meniscus
injuries for the knee. Arthroscopy 2007;23:839-844.
34. Briggs K, Kocher M, Rodkey W, Steadman J. Reliability,
validity, and responsiveness of the Lysholm Knee Score and
Tegner Activity Scale for patients with meniscal injury of the
knee. J Bone Joint Surg Am 2006;88:698-705.
35. Bekkers J, deWindt T, Raijmakers N, Dhert W, Saris D.
Validation of the Knee Injury and Osteoarthritis Outcome
Score (KOOS) for the treatment of focal cartilage lesions.
Osteoarthritis Cartilage 2009;17:1434-1439.
36. Hambly K, Griva K. IKDC or KOOS? Which measures dis-
ability symptoms and disabilities most important to postoper-
ative articular cartilage repair patients? Am J Sports Med
2008;36:1695-1704.
852 J. D. HARRIS ET AL.