www.elsevier.com/locate/knee

The Knee 12 (2005

Autologous chondrocyte implantation for treatment of focal chondral

defects of the knee—a clinical, arthroscopic, MRI and histologic

evaluation at 2 years

Ian Hendersona,*, Ramces Franciscoa, Barry Oakesb, Julie Cameronc

aOrthopaedic Research Department, St. Vincent’s and Mercy Private Hospital, 166 Gipps Street, East Melbourne, Victoria 3002, AustraliabDepartment of Anatomy and Cell Biology, Monash University, Clayton, Victoria, Australia

cMercy Radiology, East Melbourne, Victoria, Australia

Abstract

To determine the efficacy of autologous chondrocyte implantation (ACI) in treating focal chondral defects of the knee, we reviewed the 2-

year treatment outcome of ACI in 53 patients (72 lesions) through clinical evaluation, MRI, second-look arthroscopy and biopsies obtained.

Improvement in mean subjective score from preoperative (37.6) to 12 months (56.4) and 24 (60.1) months post-ACI were observed. Knee

function levels also improved [86% International Cartilage Repair Society (ICRS) III/IV to 66.6% I/II] from preoperative period to 24 months

postimplantation. Objective IKDC score of A or B were observed in 88% preoperatively. This decreased to 67.9% at 3 months before

improving to 92.5% at 12 months and 94.4% at 24 months post implantation. Transient deterioration in all these clinical scores was observed

at 3 months before progressive improvement became evident.

MRI studies demonstrated 75.3% with at least 50% defect fill, 46.3% with near normal signal, 68.1% with mild/no effusion and also

66.7% with mild/no underlying bone marrow oedema at 3 months. These values improved to 94.2%, 86.9%, 91.3% and 88.4%, respectively,

at 12 months. At 24 months, further improvements to 97%, 97%, 95.6% and 92.6%, respectively, were observed.

Second-look arthroscopy carried out in 22 knees (32 lesions) demonstrated all grafts to be normal/nearly normal based on the International

Cartilage Repair Society (ICRS) visual repair assessment while core biopsies from 20 lesions demonstrated 13 grafts to have hyaline/hyaline-

like tissue.

Improvement in clinical and MRI findings obtained from second-look arthroscopy and core biopsies evaluated indicate that, at 24 months

post-ACI, the resurfaced focal chondral defects of the knee remained intact and continued to function well.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Autologous chondrocyte implantation; Focal chondral defects; MRI

1. Introduction

It is well established that traditional methods of repair for

focal chondral defects of the knee such as microfracture,

subchondral drilling, abrasion chondroplasty and spongial-

ization result in fibrocartilaginous tissue with inferior

biomechanical properties when compared to normal articular

0968-0160/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.knee.2004.07.002

* Corresponding author. Orthopaedic Research Department, St. Vincent’s

and Mercy Private Hospital, 166 Gipps Street, East Melbourne, Victoria

3002, Australia. Tel.: +61 3 94158000; fax: +61 3 94158100.

E-mail address: ijphenderson@hotmail.com (I. Henderson).

cartilage [1–5]. Attempts to produce a better quality of repair

have led to the application of various techniques including

grafting of autologous tissues (synovial, perichondrial, or

mesenchymal), autologous osteochondral transplantation

and autologous chondrocyte implantation [6,7]. While

several studies can be found which advocate the use of each

of these methods, to date, only autologous osteochondral

transplantation and autologous chondrocyte implantation

(ACI) have found extensive application in clinical practice.

It is now known that with both techniques, a decrease in

the patients’ knee symptoms can be achieved. However,

problems associated with osteochondral transplants have

) 209–216

I. Henderson et al. / The Knee 12 (2005) 209–216210

been recognized. The most common pertain to the limited

available donor site and the resultant surface incongruity

when dealing with large lesions [8].

Available literature on ACI has already documented good

or excellent short-term clinical outcomes [9,10]. Long-term

outcomes have recently been reported by Peterson et al. [11]

demonstrating good or excellent clinical results in 91% of

58 patients treated in one series and 84% of 61 patients in

another [12]. The capacity of this technique to resurface

these focal chondral lesions with hyaline articular cartilage

affords the knee the protection needed to endure loads

passing through it. Methods employed to confirm the

quality of repair achieved mainly consist of invasive

(arthroscopic visual evaluation, indentometry and core

biopsy) techniques [6,13–16]. However, noninvasive modes

of evaluation, specifically that of MRI, have also been

validated [17–20]. This prospective investigation carried out

presents the clinical, arthroscopic, histologic and MRI

findings obtained 2 years after treatment of focal chondral/

osteochondral knee lesions with ACI.

2. Materials and methods

From November 2000 to November 2001, 72 consec-

utive focal chondral lesions (grade III or IV by modified

Outerbridge) [14] of the knee in 53 patients were treated

with ACI. To repair these defects, the Peterson periosteal

patch technique was used. In this method, an initial

arthroscopy was carried out to confirm the suitability of

the lesion for ACI and, when appropriate, cells were

harvested either from the margins of the lesion, the

intercondylar notch or both. At the same time, the size of

the lesion was documented to approximate the cell density

needed for implantation. The harvested cells were then

taken to the laboratory for in vitro proliferation. Three to 4

weeks later, the cells were implanted in the defect with a

medial or lateral parapatellar arthrotomy approach. To

contain the cells within the defect, an autologous periosteal

patch was used. The patch was secured with a 6.0 PDS

suture and sealed watertight with TisseelTM fibrin sealant. A

standardized postoperative rehabilitation protocol was then

commenced 24 h after implantation. This consisted of

passive knee range of motion exercises on a CPM machine,

static quadriceps and prone knee curl exercises. Variations

in the regimen depended on the location of the lesion. For

condylar lesions, the involved extremities were kept non-

weight bearing with crutches and extension splint for 6

weeks after which the patient was weaned off crutches and

fitted with an unloading brace to allow progressive weight

bearing. Then at 12 weeks post implantation, the brace was

discarded. For patellofemoral lesions, the knees were kept in

an extension splint and allowed to weight bear as tolerated.

Crutches were removed 3 weeks post-ACI or once

satisfactory static quadriceps strength was obtained. The

splint, on the other hand was maintained until 12 weeks

post-ACI. High impact activities were discouraged until 12

months post implantation.

2.1. Patient data

Fifty-three patients (72 lesions) with focal chondral

defects of the knee were included in this study. The mean

age of the patients was 41 years (range, 19–64 years). There

were 40 males and 13 females. Twenty-eight cases had the

left knee involved, 24 had the right knee affected and one

patient had both knees involved. Eighteen patients had

gradual onset of knee symptoms while 35 had associated

trauma to their knees. Forty-seven patients had previous

surgery to their knee prior to undergoing ACI. The

procedures included (37) knee arthroscopies (debdribe-

ment/synovectomy: 15; partial/total meniscectomy: 11;

osteo/chondroplasty: 9; and lateral retinacular release: 2);

(7) ACL reconstructions; (2) open meniscectomies; and (1)

extensor realignment.

2.2. Lesion data

Of the 54 knees involved, 37 had a single lesion, 16 had

double lesions and one patient had a triple lesion. The mean

size of the defect was 3.7 cm2 (range, 1–7.5 cm2). The

majority were condylar lesions which comprised 61.1% (44)

while patellofemoral lesions were evident in 38.9% (28).

The distribution of the defects was as follows: medial

femoral condyle (32), trochlea (25) lateral femoral condyle

(12) and patella (3).

2.3. Clinical and MRI assessment

Clinical evaluation was carried out using the IKDC

evaluation form endorsed by the International Cartilage

Repair Society (ICRS). An orthopaedic fellow examined the

patients preoperatively and at 3, 12 and 24 months

postimplantation to obtain the objective score. Subjective

knee scores and functional status were determined from the

questionnaires completed by the patients at the same

intervals.

Included as part of the evaluation was MRI scan at 3, 12

and 24 months postimplantation. The MRI unit (Signa LX,

General Electric Systems, Milwaukee, USA) had a dedicated

knee coil to optimize signal and the technical sequence by

which the images were documented followed a targeted

approach to produce high resolution. The scanned images

were then reviewed by a musculoskeletal radiology con-

sultant familiar with the procedure. To facilitate evaluation, a

scoring system with a scale of 1 to 4 was adopted [17]. In this

system, a score of 1 always indicated the best, while 4 the

worst result obtained. Four parameters were evaluated fill,

signal, effusion and bone marrow oedema. For assessing fill

of the grafted area, the score can be complete (1), greater than

50% (2), less than 50% (3) or no fill/full thickness defect (4).

Signal of the repaired area was ascribed as normal (1) if

Table 1

Objective, subjective and functional scores

Objective

IKDC

score

Pre-op

(%)

3

months

(%)

12

months

(%)

24

months

(%)

A (normal) 24 (48) 6 (11.3) 31 (58.5) 35 (64.8)

B (nearly

normal)

20 (40) 30 (56.6) 18 (34) 16 (29.6)

C

(abnormal)

5 (10) 12 (22.6) 4 (7.5) 3 (5.6)

D (severely

abnormal)

1 (2) 5 (9.4) 0 0

Pre-op

(mean)

3

months

(mean)

12

months

(mean)

24

months

(mean)

Subjective

knee

score

37.6 33.2 56.4 60.1

ICRS knee

functional

level

Pre-op

(%)

3

months

(%)

12

months

(%)

24

months

(%)

I (no 1 (2) – 3 (5.7) 10 (18.5)

I. Henderson et al. / The Knee 12 (2005) 209–216 211

identical to the adjacent articular surface, nearly normal (2)

when slight areas of hyperintensity exist, abnormal (3) when

with larger areas of hyperintensity, or absent (4). Effusion and

bone marrow oedema were both graded as absent (1), mild

(2), moderate (3) and severe (4). The overall MRI score was

also tabulated which corresponded to the worst score in the

four parameters assessed. Other significant features of the

repaired area, such as graft hypertrophy, fibrillations and

clefts were also noted.

2.4. Second-look arthroscopy and biopsy

At a mean duration of 13.4 months (range, 6.5–23.3

months) postimplantation, second-look arthroscopy was

carried out in 22 knees to asses the grafted area using the

ICRS visual cartilage repair assessment. In this visual

evaluation, the grafted area was assessed for depth of the

repair, integration to border zone and macroscopic appear-

ance. Each part of this subjective evaluation was ascribed a

maximum score of 4 points. The score was then tabulated to

derive the overall repair assessment grade. A repair was

classified as Grade I (12 points) when findings were normal,

grade II (8–11 points) nearly normal, grade III (4–7 points)

abnormal and grade IV (0–3 points) severely abnormal

morphologic appearance of the grafted area. Once eval-

uated, a 2-mm core biopsy was taken from the central and

marginal areas of the graft with the use of a Giebel needle

(Karl Storz, Germany) for histological evaluation. In this

series, biopsy samples from 20 lesions were available for

review. Specimens were prepared using ruthenium red/

osmium. Samples were then embedded in Epon-Araldite

after which 2-Am cut sections were prepared and Toluidine

blue was used for staining to facilitate microscopic

examination.

2.5. Statistical analysis

The Student’s t-test ( pb0.05) was used to analyze the data

gathered to determine any significant differences between

IKDC scores obtained at designated intervals post-ACI. The

same test was used for comparing subjective scores (pre-

operatively, 12 and 24 months) andMRI scores obtained at 3,

12 and 24 months post implantation. To establish the

presence of any linear correlation between IKDC and MRI

scores, the Pearson correlation coefficient (r2) was deter-

mined. All MRI parameters at different intervals were

assessed for correlation with IKDC scores obtained.

restriction)

II (mild

restriction)

6 (12) – 27 (58.5) 26 (48.1)

III (moderate

restriction)

32 (64) – 15 (28.3) 17 (31.5)

IV (severe

restriction)

11 (22) – 3 (7.5) 1 (1.9)

*Number of knees examined (n): n=50 for pre-op; n=53 for 3 and 12

months and n=54 for 24 months.

**Not scored.

3. Results

3.1. IKDC score

Objective knee scores obtained prior to surgery demon-

strated 88% (44) of the knees examined to be classified as A

or B. At 3 months post-ACI, the number of knees with A or

B classification significantly decreased ( pb0.001) to 67.9%

(36). However, continuous follow-up showed the decrease

to be only temporary as 12-month IKDC score revealed

92.5% (49) to have A or B classification which further

improved to 94.4% (51) at 24 months (Table 1). The

improvement in clinical score from three to 12 months was

significant ( pb0.001). However, further increase in scores

documented from 12 to 24 months was not statistically

significant (p=0.27). Comparison of the preoperative and 12

months score demonstrated no significant difference

( p=0.25) while preoperative and 24 months knee score

showed significant ( pb0.05) improvement. At the other end

of the spectrum, six knees were classified as C or D

preoperatively. Of these, only four remained as C and none

of the knees had D rating at 12 months. This improved

further as only three remained with C rating at 24 months

postimplantation. The same trend was also observed with

the subjective scores as gradual improvement was docu-

mented. A mean improvement of 22.5 points from

preoperative period to 24 months post-ACI was observed.

In terms of the location and number of lesions in each

knee examined, the objective IKDC, subjective and func-

tional scores were noted to have no significant differences

( pb0.05) (Table 2).

Table 2

Clinical outcome according to lesion site

Site of lesion Objective IKDC at 24 monthsa Functional status at 24 monthsb Subjective evaluation (mean)

A B C D I II III IV Pre-op 12 months 24 months

Patellofemoral lesion (n=15) 10 4 1 0 3 7 5 0 32.7 55.4 60.3

Condylar lesion (n=22) 12 8 2 0 6 7 8 1 31.4 42.6 56.8

Double/triple lesions (n=17) 13 4 0 0 1 12 4 0 43.6 62 61

Some patients failed to fill out some parts of the evaluation form.a (A) Normal, (B) nearly normal, (C) abnormal, (D) severely abnormal.b (I) Can do everything with the joint, (II) can do nearly everything with the joint, (III) restricted and many things can’t be done with the joint, (IV) very

restricted and can do almost nothing with the joint without severe pain and disability.

I. Henderson et al. / The Knee 12 (2005) 209–216212

3.2. Subjective knee score and knee functional level score

Subjective Knee Scores obtained preoperatively had a

mean of 37.6 points. The score decreased to 33.2 at 3

months before improving to 56.4 at 12 months and 60.1

at 24 months postimplantation. This reflected a signifi-

cant improvement ( pb0.001) between the preoperative

and the 12- and 24-month subjective scores. However,

further improvement in scores from 12 to 24 months was

no longer significant ( pN0.05).

Analysis of the Knee Functional Status of the patients

using the ICRS 4-level scale demonstrated 14% (7) to

have level I or II function preoperatively; 64% (32) with

level III while 22% (11) had level IV classification.

Reassessment at 12 months demonstrated significant

improvement ( pb0.001) to 64.2% (30) with level I or

II function, while 28.3% (15) remains at level III and

7.5% (3) with level IV functions. By 24 months post-

ACI, further improvement to 66.6% (36) with level I or

II was observed, while 31.5% (17) remained with level

III function. Only 1.9% (1) demonstrated level IV

functional status at this time. In terms of the correlation

between the overall MRI score and the IKDC and

subjective scores, no direct correlation was established

(r2=0.2) at 3, 12 and 24 months.

Table 3

MRI scores at 3, 12 and 24 months

MRI score Fill Signal

3 months 12 months 24 months 3 months 1

1 29 (42) 37 (53.6) 40 (58.8) 1 (1.4) 2

2 23 (33.3) 28 (40.6) 26 (38.2) 31 (44.9) 3

3 7 (10.1) 2 (2.9) 1 (1.5) 26 (37.7)

4 10 (14.5) 2 (2.9) 1 (1.5) 11 (15.9)

MRI scores Bone Marrow Oedema

3 months 12 months 24 mon

1 20 (29) 33 (47.8) 34 (50)

2 26 (37.7) 28 (40.6) 29 (42.6

3 21 (30.4) 6 (8.7) 4 (5.9)

4 2 (2.9) 2 (2.9) 1 (1.5)

n=69 for 3 and 12 months and n=68 for 24 months; not all patients returned for

3.3. MRI evaluation of cartilage repair

Three months post-ACI, a total of 69 lesions had MRI

scans available for review (Table 3). Based on the four

parameters used for MRI scoring, it was observed that

75.3% (52) of the grafted defects demonstrated N50% to

100% fill; 46.3% (32) had a normal or nearly normal signal;

68.1% (47) presented with mild or absent effusion and

66.7% (46) had mild or absent bone marrow oedema.

Overall MRI score was good in 31.9% (22) of the lesions

reviewed (Fig. 1). On reassessment of the grafted lesions 12

months postimplantation, significant improvement ( pb0.05)

in all MRI parameters was demonstrated as 94.2% (65) had

50% to 100% fill; 86.9% (60) had normal/nearly normal

signal; 91.3% (63) with mild or absent effusion; and 88.4%

(61) had mild or absent bone marrow oedema. Good or

excellent overall MRI score was also apparent in 68.1% (47)

of these lesions. At 24 months post-ACI, 97% (66) of the

lesions treated had N50% to 100% fill and demonstrated

normal/nearly normal signal; 95.6% (65) had mild or no

effusion; 92.6% (63) had mild or absent subchondral

oedema and 82.4% (56) had good or excellent overall

MRI score. The improvement in score from 3 to 12 months

was noted to be significant ( pb0.05). However, between 12

and 24 months, the further increase in scores was no longer

Effusion

2 months 24 months 3 months 12 months 24 months

1 (30.4) 26 (38.2) 18 (26.1) 33 (47.8) 34 (50)

9 (56.5) 40 (58.8) 29 (42) 30 (43.5) 31 (45.6)

7 (10.1) 1 (1.5) 20 (29) 6 (8.7) 3 (4.4)

2 (2.9) 1 (1.5) 2 (2.9) 0 0

Overall score

ths 3 months 12 months 24 months

0 8 (11.6) 8 (11.8)

) 22 (31.9) 39 (56.5) 48 (70.6)

35 (50.7) 19 (27.5) 10 (14.7)

12 (17.4) 3 (4.3) 2 (2.9)

their 2-year MRI.

Fig. 1. Medial femoral condyle lesion after ACI. (A) Sagittal MRI at 3

months post-ACI demonstrating good fill with heterogenous signal. (B)

Sagittal MRI at 12 months showing slight hypertrophy of graft (long

arrow), increased signal intensity and moderate underlying bone marrow

oedema (short arrows). (C) Sagittal MRI at 24 months demonstrating

minimal surface irregularity and slight graft overfill, decreased signal and

minimal subchondral bone marrow oedema.Fig. 2. Arthroscopic evaluation. (A) Arthroscopic visual assessment prior to

ACI demonstrating the chondral lesion at the area of the medial femoral

condyle (MFC). (B) Arthroscopic evaluation 10 months post-ACI showing

good fill and integration to adjacent articular cartilage with minor surface

irregularity.

I. Henderson et al. / The Knee 12 (2005) 209–216 213

statistically significant ( pN0.05). Although this steady

improvement in MRI scores paralleled the improvements

seen earlier with the objective IKDC and subjective scores,

no direct correlation (r2b0.3) was established with any of

the parameters tested.

3.4. Significant MRI findings

Distinct features noted on 3-month MRI examination

included fibrillation on graft surface (7), graft hypertrophy

(7), fluid undermining graft (4), displaced grafts (2) and

cleft in graft (3). Twelve-month studies still demonstrated

graft hypertrophy (2) and cleft in graft (2). At 24 months

post-ACI, deficit on graft margin (2) and subchondral cyst

(3) were observed. These findings at different time intervals

from ACI did not seem to have affected the clinical status of

the patients as most still presented with an IKDC score of A

or B. It also noted that earlier findings seemed to have

resolved on later MRI assessments.

3.5. Arthroscopic assessment of cartilage repair and core

biopsy

At the time of review, 32 lesions in 40.7% (22) of the 54

knees treated have undergone arthroscopic evaluation. Core

biopsy samples for histologic examination were taken from

27.8% (20) of the 72 lesions treated. Earlier knee arthros-

copies were done without taking core biopsies as appropriate

instrumentation (Gieble Needle) was not available. Arthro-

scopy was carried out at a mean duration of 13.4 months post-

ACI (Fig. 2). Twenty-one of the 22 knees that were

arthroscoped presented with persistent mechanical symptoms

such as catching, clicking and/or locking whilst one knee had

arthroscopy for graft assessment at the time of ACL

reconstruction. Visual scores obtained (mean, 10.2 points)

at second-look arthroscopy demonstrated 30 lesions to be

nearly normal (Grade II) while two lesions were normal in

macroscopic appearance (Grade I). Visual scores obtained

from second-look arthroscopy had no direct correlation with

I. Henderson et al. / The Knee 12 (2005) 209–216214

24-month overall MRI score (r2=0.2). Histologic examina-

tion of core biopsies taken from the grafts revealed eight to be

hyaline articular cartilage, five hyaline-like cartilage, four to

be fibrocartilage and another three mixed fibro-hyaline tissue.

The majority (6) of the hyaline articular tissue obtained was

from biopsies taken from the medial femoral condyle defects.

Two other sites which produced the hyaline articular tissue

were the lateral femoral condyle and the patella. In contrast,

hyaline like tissue was obtained from different sites (medial/

lateral femoral condyle and trochlea). Hyaline and hyaline-

like cartilage tissue obtained demonstrated the distinctive

organizational patterns of cells and extracellular matrix found

in the structure of normal articular cartilage. Microscopic

examination rendered these zones (superficial, middle, deep

and calcified) to be identified at times revealing the presence

of periosteal patch remnants on the most superficial layer.

Typical flattened appearance of chondrocytes on the super-

ficial layer becoming more rounded in the deeper layer was

observed. In the same manner, the distinct boundary marking

the transition from deep to calcified layer (tidemark) was also

Fig. 3. Core biopsy. (A) Central core biopsy showing the deep zone (DZ)

with good incorporation of graft to the subchondral layer. (B) Marginal core

biopsy demonstrating seamless integration between the adjacent articular

cartilage (AC) and hyaline-like cartilage (HLC).

noted. The increased cell density in the deeper zone of

hyaline-like tissue makes it distinct from hyaline articular

cartilage. Common findings in all tissue examined was the

excellent subchondral bone integration displayed and the

seamless integration between the neocartilage formed by the

graft and the adjacent host articular cartilage on marginal core

biopsies (Fig. 3). The core biopsies examined were taken

from grafted defects at the medial femoral condyle (11),

trochlea (6), lateral femoral condyle (2) and patella (1). All of

the fibrocartilage tissue were biopsied from the trochlea (4),

while the hyaline-like tissue were obtained from lesions

involving the medial femoral condyle (3), lateralfemoral

condyle (1) and the trochlea (1). Mixed fibro-hyaline tissues

were biopsied from the trochlear (1) and medial femoral

condyle (2) regions.

4. Discussion

This study was carried out with the objective of

determining the clinical outcome of ACI in the first 72

consecutive focal chondral lesions in 54 knees with a

minimum 2-year review. As with the series reviewed by

Erggelet et al. [21] and Peterson et al. [11], most of the lesions

treated in this study were located on the medial femoral

condyle. To evaluate the outcome for this group, the IKDC

2000 knee exam form from the ICRS Cartilage Injury

Evaluation Package was utilized.

Normal/near normal results were evident in 88% of the

knees examined preoperatively. This seemingly good pre-

operative status declined significantly at 3 months as only

67.9% maintained a normal/near normal score. However,

further reassessment at 1 year and then at 2 years after the

surgery reflected an improvement to 92.5% and 94.4%,

respectively. We now know from a previous study [17] that

the preoperative and 1-year objective score was not

significantly different and that benefits gained in the short-

term with this procedure is better demonstrated with the

subjective scores (86.8% of 53 knees demonstrated

improvement) and level of knee function (64.2% having

mild or no restriction). On the other hand, the objective knee

score at 2 years was significantly improved when compared

to the preoperative scores. It was also observed in this series

that of the six knees preoperatively evaluated as abnormal or

severely abnormal, four improved to normal or near normal

at 1 year. Subsequent evaluation carried out 2 years post

implantation demonstrated only one knee to have abnormal

findings. In this knee, the moderate effusion persisted while

passive motion and ligament stability remained normal. The

apparent trend therefore appears to be that at 3 months post-

ACI, transient deterioration in objective, subjective and

knee functional status is demonstrated before a steady,

progressive improvement is seen.

Regarding the location of the chondral defect in the knee

[patellofemoral, condylar or both) as a possible factor

influencing outcome, no significant difference ( pN0.05)]

I. Henderson et al. / The Knee 12 (2005) 209–216 215

was established with the clinical outcome at 2 years post-

ACI. A reliable, noninvasive means of evaluating the status

of the graft is always desirable. At present, this is achieved

with the use of MRI scans taken at designated intervals (3, 12

and 24 months) postoperatively. In a study conducted by

Burkart and Imhoff, [15] it was observed that MRI taken

between 3 and 6 months post-ACI still showed signal

irregularities with partial gadolinium uptake at the area of

repair. At 1 year from implantation, the reparative process at

the grafted areas seemed to have been completed as

gadolinium uptake was no longer demonstrated and the

borders of the graft were hardly visible. Wada et al. [20] on

the other hand observed the volume of the reparative tissue to

remain at a constant level with MRI studies done in five cases

treated with ACI. Moreover, the signal intensity of the grafted

areas approached that of normal articular cartilage with time.

In this study, the MRI parameters used for evaluation

included fill, signal, effusion and bone marrow oedema.

Assessment of graft fill was used to measure the amount of

growth of the reimplanted tissue. Signal observed from the

graft material was compared with adjacent articular cartilage

to assess its maturity. Presence of effusion was also noted and

subchondral marrow oedema observed was interpreted as a

possible sign of graft immaturity and inability to cushion the

underlying bone from excessive loads. Marked improvement

was evident in all these parameters evaluated between 3 and

12 months. Further improvement was still observed from 12

to 24 months, however this no longer proved to be statisti-

cally significant ( pN0.05). This progressive improvement

indicates that the graft at 2 years maintains its structural

integrity and continues to function well.

Even as good results were obtained from MRI studies,

second-look arthroscopy with biopsy remains the ideal

means by which the status of the graft repair is assessed. In

this series, 22 (40.7%) of the 54 knees had second-look

arthroscopy at a mean duration of 13.4 months post-

implantation. As the process of graft maturation takes

several months, even beyond a year to be completed, it

was not surprising to find the graft repair site to be soft to

probing when arthroscopies were performed within 1 year

post-ACI [22]. Later arthroscopic evaluation showed the

grafts to be firmer eventually attaining a consistency

comparable to the adjacent normal articular cartilage.

Further examination revealed most of the grafts were

already level with the surrounding cartilage with complete

integration to adjacent normal tissue. Minor fibrillations

visualized on the surface of the graft were arthroscopically

trimmed. All 22 knees evaluated were rated as normal/

nearly normal with a mean visual score of 10.2 points.

However, on MRI four of these nearly normal repairs were

rated as having fair outcome with a score of 3 at 24 months.

The less favorable results seen with MRI could mean that

MRI is more sensitive in picking up subtle irregularities in

the repair than a visual score would allow. This is under-

standable as MRI has the advantage of determining the

condition not just of the graft fill and signal but also of the

immediate area around the repair to detect the presence of

subchondral bone plate oedema and effusion.

Once visual assessment of the graft has been completed,

core biopsy samples for histologic analysis were obtained. In

a prospective study conducted by Briggs et al. [13] in 14

patients, core biopsy samples taken 1 year from ACI

revealed eight (57.1%) to have produced hyaline cartilage.

In this review, the 20 core biopsies obtained revealed 13

(65%) to be hyaline or hyaline-like, four (20%) samples to be

fibrocartilaginous tissue and three (15%) were mixed fibro-

hyaline cartilage. Of the 13 repairs with hyaline or hyaline-

like tissue three obtained abnormal MRI scores (3) at 24

months while only one of the three fibrocartilage repairs was

rated abnormal. This finding suggests the usefulness of MRI

in monitoring the progression of grafts implanted but

inaccuracy in determining the nature of the neocartilage.

ACI like any other surgical procedure is not free from

complications. Adverse postoperative events like adhesions,

arthrofibrosis, hypertrophic changes, phlebitis and graft

failure have been known to occur [1,11,22–24]. Erggelet

et al. in a series of 1051 patients treated with ACI reported

postoperative complication rate of 4.8%. So far, in this

series of the first 62 cases with a 2-year review, none have

experienced any of the complications mentioned.

5. Conclusion

In this 2-year review, it has been shown that ACI is an

effective means of managing focal chondral lesions of the

knee. The clinical, MRI, arthroscopic and biopsy findings

reflect the good outcome obtained with the procedure.

Clinically, although objective knee examination demon-

strated deterioration at 3 months, this regression proved to

be only temporary as continuous improvement was

observed at 12 and 24 months postimplantation. Although

the preoperative objective score was not significantly

different to scores obtained at 12 months, significant

improvement was observed when the 24 month scores

were compared with the preoperative knee scores. A

similar trend was also demonstrated with the MRI scores

where images reviewed demonstrated progressive improve-

ment from 3 to 24 months postimplantation. Second look

arthroscopy again demonstrated a satisfactory outcome as

all lesions were rated normal/nearly normal by ICRS visual

scores while biopsies obtained revealed production of

hyaline/hyaline-like tissue in 65% of the core biopsy

samples examined.

References

[1] Bentley G, Biant LC, Carrington RW, Akmal M, Goldberg A,

Williams AM, et al. A prospective, randomised comparison of

autologous chondrocyte implantation versus mosaicplasty for osteo-

chondral defects in the knee. J Bone Joint Surg Br 2003

(Mar.);85(2):223–30.

I. Henderson et al. / The Knee 12 (2005) 209–216216

[2] Brittberg M, Lindahl A, Nilsson A, et al. Treatment of full thickness

cartilage defects in the human knee with cultured autologous

chondrocytes. N Engl J Med 1994;331:889–95.

[3] Gillogly SD, Voight M, Blackburn T. Treatment of articular cartilage

defects of the knee with autologous chondrocyte implantation. J

Orthop Sports Phys Ther 1998;28:241–51.

[4] S.L. Sledge, Microfracture techniques in the treatment of osteochon-

dral injuries. Clin. Sports Med. Apr.; 20 (2) 365–77, Apr. (Abstract).

[5] Steadman JR, Rodkey WG, Rodrigo JJ. Microfracture: surgical

technique and rehabilitation to treat chondral defects. Clin Orthop

2001 (Oct.);391:S362 [Suppl.].

[6] Mainil-Varlet P, Aigner T, Brittberg M, Bullough P, Hollander A,

Hunziker E, et al. Histological assessment of cartilage repair: a report

by the histological endpoint committee of the ICRS. J Bone Joint Surg

Am 2003;85A(Suppl. 2):45–57.

[7] Minas T. Autologous chondrocyte implantation for focal chondral

defects of the knee. Clin Orthop 2001 (Oct.);391:S349 [Suppl].

[8] Hangody L, Fules P. Autologous osteochondral mosaicplasty for the

treatment of full-thickness defects of weight-bearing joints. Ten years

of experimental and clinical experience. J Bone Joint Surg Am

2003;85A(Suppl. 2):25–32.

[9] Brittberg M, Tallheden T, Sjogren-Jansson B, Lindahl A, Peterson L.

Autologous chondrocytes used for articular cartilage repair: an update.

Clin Orthop 2001;(391 Suppl.):S337–48.

[10] Peterson L, Minas T, Brittberg M, Nilsson A, et al. Two-to 9 year

outcome after autologous chondrocyte transplantation of the knee.

Clin Orthop 2000;374:212–34.

[11] Peterson L, Minas T, Brittberg M, Lindahl A. Treatment of

osteochondritis dissecans of the knee with autologous chondrocyte

transplantation: results at two to ten years. J Bone Joint Surg Am

2003;85A(Suppl. 2):17–24.

[12] Peterson L, Brittberg M, Kiviranta I, Akerlund EL, Lindahl A.

Autologous chondrocyte transplantation. Biomechanics and long-term

durability. Am J Sports Med 2002 (Jan.–Feb.);30(1):2–12.

[13] Briggs TW, Mahroof S, David LA, Flannelly J, Pringle J, Bayliss M.

Histological evaluation of chondral defects after autologous chon-

drocyte implantation of the knee. J Bone Joint Surg Br 2003

(Sep.);85(7):1077–83.

[14] Brittberg M, Winalski C. Evaluation of cartilage injuries and repair. J

Bone Joint Surg Am 2003;85A(Suppl. 2):58–69.

[15] Burkart A, Imhoff AB. Diagnostic imaging after autologous chon-

drocyte transplantation. Correlation of magnetic resonance tomog-

raphy, histological and arthroscopic findings. Orthopade 2000

(Feb.);29(2):135–44.

[16] Laasanen M, Toyras J, Vasara A, Hyttinen M, Saarakkala S, Hirvonen

J, et al. Mechano-acoustic diagnosis of cartilage degeneration and

repair. J Bone Joint Surg Am 2003;85A(Suppl. 2):78–84.

[17] Henderson IJP, Tuy B, Oakes B, Connell D, Hettwer W. Prospective

clinical study of autologous chondrocyte implantation (ACI) and

correlation with MRI at 3 and 12 months. J Bone Joint Surg Br 2003

(Sep.);85(7):1060–6.

[18] Poole R. What type of cartilage repair are we attempting to attain? J

Bone Joint Surg Am 2003;85A(Suppl. 2):40–4.

[19] Potter HG, Linklater JM, Allen AA, Hannafin JA, Haas SB. Magnetic

resonance imaging of articular cartilage in the knee. An evaluation

with use of fast-spin-echo imaging. J Bone Joint Surg Am

1998;80A:1276–84.

[20] Wada Y, Watanabe A, Yamashita T, Isobe T, Moriya H. Evaluation of

articular cartilage with 3D-SPGR MRI after autologous chondrocyte

implantation. J Orthop Sci 2003;8(4):514–7.

[21] Erggelet C, Browne JE, Fu F, Mandelbaum BR, Micheli LJ, Mosely

JB. Autologous chondrocyte transplantation for treatment of cartilage

defects of the knee joint. Clinical results. Zentralbl Chir

2000;125(6):516–22.

[22] Bahuaud J, Maitrot RC, Bouvet R, Kerdiles N, Tovagliaro F, Synave

J, et al. Implantation de Chondrocytes Autologues pour Lesions

Cartilagineuses du Sujet Jeune. Etude de 24 cas. Chiurgie

1998;123:568–71.

[23] Micheli LJ, Browne JE, Erggelet C, Fu F, Mandelbaum B, Moseley

JB, et al. Autologous chondrocyte implantation of the knee: multi-

center experience and minimum 3-year follow-up. Clin J Sport Med

2001 (Oct.);11(4):223–8.

[24] Muellner T, Knopp A, Ludvigsen T, Engebretsen L. Failed autologous

chondrocyte implantation. Complete atraumatic graft delamination

after two years. Am J Sports Med 2001 (Jul.–Aug.);29(4):516–9.