The autologous osteochondral transplantation of the knee: clinical results, radiographic findings and histological aspects.

Abstract

The osteochondral transplantation (OCT) is a well accepted treatment option for focal cartilage lesions in the knee joint, whereas the fate of the transplanted cartilage is still unclear and the clinical outcome is variable. The purpose of this study was to evaluate the histological character of autologous transplanted cartilage and to correlate technical aspects and the patients' history with the clinical outcome.
The OCT was performed in 27 patients (median age of 32 (22-43) years) with a focal chondral lesion at the medial femoral condyle. We investigated the clinical outcome after a median follow-up of 13.5 (5-28) months using the Lysholm-score and the integration of the transplanted plugs using an MRI-scoring system. Biopsy specimens from representative patients (n = 8) were evaluated with histological staining and immunohistochemistry.
The median Lysholm-score was 80 (range 45-98). The wide range of the Lysholm-score in clinical outcome did not show significant differences in: follow-up, concomitant injuries, defect size or genesis. The MRI analysis revealed in all cases a regular osseous integration of the subchondral bone, but a failed chondral integration. The congruency of the plugs to the joint surface was often incorrect, however a correlation between the MRI-score and the clinical outcome could not be shown. Histology of the transplanted cartilage revealed small changes in immunohistochemistry after a relatively short-term follow-up, whereas the cartilage has still the typical hyaline character. Often, the surrounding cartilage consists of fibrous and granulation tissue.
The congruency of the joint surface can not be restored to the original status, particularly in larger defects with irregular shapes. However, we did not find any aspects which affected the function of the knee joint following OCT. It can be assumed that remaining lesions at the surrounding cartilage could maintain the inflammatory process and therefore maintain the pain and a low knee function. Further investigations are needed to specify the effects of the OCT on the transplanted cartilage and its influence on the later clinical outcome.

Full text

Page 1

ORIGINAL ARTICLE
Tim Rose Æ Steffen Craatz Æ Pierre Hepp
Conrad Raczynski Æ Joachim Weiss Æ Christoph Josten
Helmut Lill
The autologous osteochondral transplantation of the knee:
clinical results, radiographic findings and histological aspects
Received: 6 August 2004
? Springer-Verlag 2005
Abstract Introduction: The osteochondral transplanta-
tion (OCT) is a well accepted treatment option for focal
cartilage lesions in the knee joint, whereas the fate of the
transplanted cartilage is still unclear and the clinical
outcome is variable. The purpose of this study was to
evaluate the histological character of autologous trans-
planted cartilage and to correlate technical aspects and
the patients’ history with the clinical outcome. Material
and Methods: The OCT was performed in 27 patients
(median age of 32 (22–43) years) with a focal chondral
lesion at the medial femoral condyle. We investigated
the clinical outcome after a median follow-up of 13.5 (5–
28) months using the Lysholm-score and the integration
of the transplanted plugs using an MRI-scoring system.
Biopsy specimens from representative patients (n=8)
were evaluated with histological staining and immuno-
histochemistry. Results: The median Lysholm-score was
80 (range 45–98). The wide range of the Lysholm-score
in clinical outcome did not show significant differences
in: follow-up, concomitant injuries, defect size or gene-
sis. The MRI analysis revealed in all cases a regular
osseous integration of the subchondral bone, but a failed
chondral integration. The congruency of the plugs to the
joint surface was often incorrect, however a correlation
between the MRI-score and the clinical outcome could
not be shown. Histology of the transplanted cartilage
revealed small changes in immunohistochemistry after a
relatively short-term follow-up, whereas the cartilage
has still the typical hyaline character. Often, the sur-
rounding cartilage consists of fibrous and granulation
tissue. Conclusion: The congruency of the joint surface
can not be restored to the original status, particularly in
larger defects with irregular shapes. However, we did not
find any aspects which affected the function of the knee
joint following OCT. It can be assumed that remaining
lesions at the surrounding cartilage could maintain the
inflammatory process and therefore maintain the pain
and a low knee function. Further investigations are
needed to specify the effects of the OCT on the trans-
planted cartilage and its influence on the later clinical
outcome.
Keywords Cartilage defect Æ Osteochondral
transplantation Æ MRI Æ Histology Æ Clinical outcome
Introduction
The healing of articular cartilage defects is limited [21]
because of their avascularity and missed innervation.
Different treatment options have been used to stimulate
the regeneration of such defects, as subchondral drilling
[2], abrasion-chondroplasty [9], and microfractures [39].
However, these techniques result only in a fibrous sub-
stitute tissue covering the cartilage defect with poor
biomechanical characteristics [6]. The restoration of
hyaline cartilage using autologous chomdrocyte trans-
plantation results also mostly in hyaline-like cartilage or
fibrocartilage tissue [10–20].
The osteochondral transplantation (OCT) is an op-
tion to fill a chondral defect with hyaline cartilage with a
reliable osseous ingrowth. The first OCT was noted by
Wagner in 1964, whereas an autologous transplantation
was performed in three patients and a homologous
transplantation was performed in two patients [42].
Following these promising results, several surgeons
performed allogenic and autologic OCT with slightly
T. Rose (&) Æ P. Hepp Æ C. Raczynski Æ C. Josten Æ H. Lill
Department for Trauma and Reconstructive Surgery,
University of Leipzig, Leipzig, Germany
E-mail: t.rose@lrz.tum.de
Tel.: +49-89-28924471
Fax: +49-89-28924484
S. Craatz Æ J. Weiss
Institute of Anatomy, University of Leipzig,
Leipzig, Germany
T. Rose
Arzt fu ¨ r Chirurgie und Sportmedizin Oberarzt Abteilung fu ¨ r
Sportorthopa ¨ die, Technische Universita ¨ t Mu ¨ nchen Klinikum
Rechts der Isar, Connolly Strasse 32, 80809 Mu ¨ nchen, Germany
Arch Orthop Trauma Surg (2005)
DOI 10.1007/s00402-005-0010-8

Page 2

good results [7, 8, 18, 32, 35, 43]. At the beginning of
1990’s the technique of transplantation was improved
mainly by Hangody et al. [11] and Bobic [5], whereas
Matsusue et al. [31] were able to use this technique for
arthroscopic approach. Following middle and long term
studies had shown encouraging clinical results after
OCT [3, 15, 28]. Furthermore, experimental and clinical
experience have shown that the transplanted hyaline
cartilage has a good rate of survival [15].
The histological structure of the transplanted carti-
lage is an important marker for their biomechanic
characteristic and their survival. Hangody et al. [12] still
detected hyaline cartilage in a histological evaluation of
arthroscopic biopsies 1–5 years following OCT. Cadaver
studies mostly supported these findings [15, 38], but it
seems that the histological properties of the transplanted
cartilage are different from that of the normal articular
cartilage in animal model [30]. Additionally, a chondral
integration of the transplanted cartilage failed and a gap
between the transplanted cartilage and the adjacent
cartilage reached down to bone [15, 20]. However, the
characteristic of the histological structure of the grafted
cartilage in human has not been reported in detail.
Furthermore, the congruency of the transplanted plugs
and the osseous integration are propagated as an
important requirement for the clinical satisfaction of the
patient. But it is still unclear which circumstances of the
injured knee really affect the clinical outcome of the
patient following OCT.
The purpose of this study was to evaluate the histo-
logical character of autologous transplanted cartilage in
humans at different time points, and to correlate tech-
nical aspects and the patients’ history to the clinical
outcome. Our hypothesis was that a focal chondral de-
fect at the femoral condyle treated with OCT shows
hyaline cartilage during the follow-up. Furthermore we
assumed that the clinical outcome is be affected by the
size of the chondral lesion, by concomitant lesions,
genesis of the lesion, the age of the patient, and the
MRI-findings within the follow-up.
Material and methods
Study design
The study was approved by the ethics commission of the
University of Leipzig, Germany. The patients were in-
cluded regarding the inclusion and exclusion criteria,
after they were informed about the content of the study.
The inclusion criteria for the study were a osteochondral
lesion Grad IV of Outerbridge [34] in the weight bearing
area of the medial femoral condyle. Clinical symptoms
as swelling, pain, and locking of the knee joint and a
positive MRI-result regarding a chondral lesion were
used for a preoperative selection. Exclusion criteria were
a general osteoarthritis of the knee joint, remaining knee
joint instability, kissing osteochondral lesion, defects
larger than 4 cm2, and/or multiple defects. An axial
mal-alignment or a knee joint instability should be
corrected at the same time or previous to the OCT.
Patients
All study patients underwent an MRI previous to the
surgery and during the follow-up. At the same time
points the clinical examinations were evaluated with use
of the score described by Lysholm and Gillquist [29]. A
total of 27 patients (six female, 21 male) were included in
the study with a median age of 32 (22–43) years. The
median follow-up were 13.5 (5–28) months. The sizes of
the cartilage defect ranges from 1.0 to 3.0 cm2(median
1,2 cm2), whereas the defect was filled 16-times with one
cylinder, eight times with two osteochondral cylinders
and three times with three osteochondral cylinders. The
size of the cylinders was between 8 mm and 11 mm. The
transplantation was performed in 13 cases arthroscopi-
cally and in 14 cases with mini-arthrotomy. Only the
one-cylinder-technique was made in a arthroscopic
manner.
Additional therapeutic relevant lesions were found in
13 patients; we found anterior cruciate ligament (ACL)
rupture in two patients, meniscal lesion in six patients,
and the combination of a meniscal lesion and an ACL-
rupture in five patients. If the cruciate ligament was
turned, an ACL-reconstruction was performed at the
same time of OCT or in a previous procedure. In six
cases the meniscus was partially resected and in five
cases the meniscus was refixed using an inside-out
technique.
Eight patients underwent a re-arthroscopy for eval-
uation of the operative result between 5 and 21 months.
A biopsy, using a metal-needle with a diameter of
2.8 mm, for histological assessment was harvested from
the centre of the osteochondral plug and from the
transition zone between the transplanted cylinder and
the recipient site. The histological slides were evaluated
by a histologist who was blinded with regard to patient
allocation.
All chondral lesions were a grade IV lesion regarding
the Outerbridge-classification [34] with a total loss of
cartilage. The indication for the OCT were in four cases
an acute lesion (less than 6 weeks before transplanta-
tion) with trauma. A chronic lesion (n=23) was defined
as a traceable lesion which exists longer than 6 weeks,
whereas in 16 cases a trigger trauma was reproducible.
In seven cases an osteochondrosis dissecans was the
indication of the need for OCT. The median size of the
chondral lesions was 12 mm (range: 9–30). The OCT
was performed at the medial femoral condyle of the knee
joint over a 3-year period using one (n=16), two (n=8)
or three (n=3) graft cylinders. The harvesting site of the
cylinders was in case of an arthroscopic approach to the
femoral notch. The anterior lateral condyle and the
femoral notch were used for the donor cylinders in case
of an open approach. Concomitant intra-articular inju-
ries included nine injuries to the medial meniscus and
five injuries to the lateral meniscus (six refixed). An

Page 3

anterior instability of the knee joint was present in seven
cases, which required a reconstruction of the anterior
cruciate ligament.
Surgical technique
All operations were performed by the same surgeon,
who had extensive experience in the OCT-procedure
(open and arthroscopic), prior to the study period. The
arthroscopic assessment of the defect size led to the
decision regarding the number of the used osteochondral
cylinders. The shape of the recipient surface was fol-
lowed for harvesting the osteochondral graft to allow a
correct remodeling. The Osteochondral Autologous
Transplantation System (OATS, Arthrex, Karlsfeld,
Germany) was used for harvesting and implantation of
the osteochondral cylinders at the recipient and donor
site. The diameter of the recipient site was 1 mm smaller
compared with the donor site, so that a press-fit fixation
of the implanted osteochondral cylinder could be
achieved. Therefore an additional fixation was not nec-
essary. This process was repeated in succession for the
coverage of larger defects, which required two or three
osteochondral cylinders (Fig. 1). The medial arthrotomy
was used in all open procedures.
Rehabilitation
Full weight-bearing was allowed from the first day post-
surgery, if no refixation of the meniscus was applied.
Otherwise a weight bearing of 20 kg was allowed within
the first 6 weeks. In the phase immediately following
surgery (days 1–4), continuous passive motion (CPM)
was performed using a motorized CPM device to im-
prove the nutrition of the cartilage [22]. Analgesic
treatment was given, when necessary, in combination
with local cooling and anti-inflammatory medication.
From the fifth day onward, the CPM device was used to
obtain up to 90? in flexion and full extension of the knee.
No knee brace was provided. A more intensive training
therapy for specific muscle development was initiated in
the third week using muscle sequences and closed-chain
exercises. Beginning at week 7 post-surgery, coordina-
tion exercises were introduced to improve propriocep-
tion. Sports activities (e.g., running and open chain
exercise) were permitted after 12 weeks and contact
sport was permitted after 6 months of rehabilitation.
MRI analysis
The morphological evaluation after the OCT was per-
formed using an MRI as a noninvasive method.
Assessment of transplant vitality, osseous fixation, and
stability is possible [17]. Furthermore the congruence of
the transplant and the junction between transplant and
local cartilage can be assessed.
Table 1 describes the scoring system which was
developed for the evaluation, modified according to the
descriptions by Sanders et al. [36]. The MRI was re-
viewed by two examiners and consensus was reached for
all findings. The clinical outcome in all cases were con-
cealed from the examiners. Surface congruency was
graded as good, moderate, and poor. The protuberance
was measured as the distance from the surface of the
cylinder above the adjacent articular surface. The sub-
chondral edema of the osteochondral cylinder was
characterized as no edema, edema at the central part of
the transplant, or a general edema of the transplant.
Graft cartilage thickness was compared to the adjacent
cartilage thickness and was graded as similar (100%),
thinner (<100%), or less than the half (<50%) of the
adjacent articular cartilage thickness. No visible margin
between the transplanted bone and the adjacent bone is
determined as a good osseus integration without se-
paration. The osseus integration is incomplete when a
visible margin can be identified, and no bony integration
or graft loosening is present if fluid is around the cy-
linder.
Histological analysis
The osteochondral cylinders for the histological evalu-
ation were fixed directly after harvesting with BOUIN-
solution. The next step was to decalcify the bone using a
Chelaplex-solution following by dehydration and bed-
ding in paraffin. The paraffin-block was cut in 7 lm
slides using longitudinal sections. The histological
staining included: Hematoxylin-Eosin staining for the
Fig. 1 Example for a mosaicplasty using three osteochondral plugs
at the medial femoral condyle within the weight bearing zone. One
plug was harvested at the intercondylar notch

Page 4

overview and to evaluate the chondrocytes and Cross-
man-staining to evaluate the fibrous tissue, silver-GO-
MORI-staining to differentiate collagen type I and III,
and alcia blue staining to mark the matrix of the carti-
lage.
Immunohisochemistry was performed following the
peroxidase-anti-peroxidase-method
developed by Sternberger [40]. The endogen activity of
the tissue peroxides were blocked using 3% hydrogen
peroxide at the deparaffined slides before they were
prepared using a protein binding solution (0.25% Casein
in Phosphate-buffered saline with 0.015 molar sodium
acid (DAKO, Glostrup, Denmark). A monoclonal
mouseantibody—antihuman—was
0696; DPC Biermann, Bad Nauheim, Germany) to de-
tect aggrecan, a typical marker in the cartilage matrix.
Furthermore we used to evaluate:
(PAP method)
used(Lot.-Nr.:
Collagen type I primary antibody (mouse), Anti-
hCL(I), monoclonal antibody to
human, Clone No.: I-8H5—Lot.
Nr. IX 1 229-1 (DPC Biermann,
Bad Nauheim, Germany)
Collagen type IIprimary antibody (mouse), Anti-
hCL(II), monoclonal antibody to
human, Clone No.:II-4C11—Lot.
Nr. GU 1219-1 (DPC Biermann,
Bad Nauheim, Germany)
Collagene type III primary antibody (mouse), mono-
clonal Antihuman Collagen Type
III, Lot-Number:
Biomedicals, Shelton, USA)
2707B (ICN
Factor VIII primary antibody (rabbit) anti-hu-
man von Willebrandt factor (factor
VIII)polyclonal
(DAKO, Glostrup, Denmark) to
mark the blood vessels
Lot. Nr.105
Statistical analysis
Data are presented by descriptive statistics (i.e., mean,
standard deviation, median, range). The nonparametric
Mann-Whitney U test was used to assess variance in the
performance scores used to identify possible differences
between two groups. All significance tests were two-
tailed and statistical significance was set at 0.05.
Results
Clinical results
The median Lysholm-score was 80 (range 45–98). The
wide range of the Lysholm-score in clinical outcome did
not show significant differences in: follow-up, concomi-
tant injuries, defect size, or genesis (Fig. 2a–d). Post-
operative, three patients needed a revision because of an
haemarthros. Technical problems were once a broken
osteochondral cylinder, another time a chondral defect
at the graft cylinder, and at another time a cartilage
defect at the adjacent cartilage, detected during a re-
arthroscopy after 10 months.
MRI analysis
The MRI analysis revealed in all cases a regular
osseous integration of the subchondral bone. Only in
one case, a clear visible border remained 6 months
postoperative. A subchondral edema was detected
during the earlier follow-up (up to 1 year). However,
often the cartilage congruency was moderate or poor,
and/or the cartilage thickness reached less than 100%
of the donor cartilage. A protuberance of the cylinder
exist in 15 cases up to 2 mm. Only in three cases the
protuberance reached at least than 2 mm. A correla-
tion between the MRI-score and the clinical outcome
(Lysholm-Gillquistscore)
(Fig. 3a+b).
couldnot be shown
Results in re-arthroscopy
Macroscopic findings
At all time points between 6 and 21 months after
transplantation, a slight gap around the implanted
osteochondral plug was still visible. The cartilage
around the transplant had a soft consistency three times.
In one case the cartilage on the recipient site had a lesion
(diameter: 8 mm) starting from the transition zone to the
transplanted cylinder. The transplanted cartilage itself
was smooth and showed the typical biomechanical
character and color of hyaline cartilage. No loosing or
dislocation of the osteochondral plugs were seen at the
follow-up. The harvest site always was filled with fibrous
cartilage.
Table 1 MRI-scoring system.
Each MRI at the follow-up was
evaluated and the points were
distributed and added regarding
the findings by the examiner.
The lowest score (0) represents
the best result
Points
012
Surface congruency
Protuberance of the cylinder
Edema of the cylinder
Cartilage thickness
Osseous integration
Good
No
No
100%
Complete
Moderate
<2 mm
Center
<100%
Visible margin
Poor
2 mm and more
General
<50%
Surrounding fluid

Page 5

Histological findings
We did not find time-depending differences in the his-
tological evaluation during the follow-up. Overall, the
structure of the cartilage had shown hyaline cartilage in
HE staining with characteristic superficial tangential,
middle, and deep zones. The subchondral bone of the
graft was already united with the subchondral bone of
the host site. The cartilage surface has shown irregular
structure and was partially fissured (Fig. 4a) with sepa-
ration of cartilage fragments. The chondroid structure
dissolved at the apical and the margin of the transplant,
meanwhile the cell density decreased excepting the basal
zone of the cartilage, where the cell density increased
(Fig. 4b). Mostly the tidemark was preserved in a good
condition, but it was penetrated from single blood ves-
sels (Fig. 4c). These reactions of articular cartilage were
seen in the entire region of the implanted graft.
The surrounding cartilage of the transplanted plug
often consist of fibrous and granulation tissue (Fig. 5a)
and contain a high percentage of collagen type I
(Fig. 5b). The remodeling and degeneration process was
marked due to the ingrowth of the blood vessels starting
from the graft (Fig. 5c)
Immunohistochemistry
The special immunhistochemistry staining could detect
the characteristic collagen types for bone, hyaline car-
tilage or fibrous cartilage. The collagen type I was
mainly present in subchondral bone and the trans-
planted cartilage never contained collagen type I
(Fig. 6a), whereas the detection of collagen type II in the
osteochondral transplant was a typical marker for hya-
line cartilage (Fig. 6b). The increased occurrence of
collagen type III around the chondroids is also charac-
teristic of hyaline cartilage. The bone and the cartilage
within the gap showed no positive staining for collagen
type II or collagen type III.
The adjacent original cartilage around the trans-
plant has shown a focal clustering of proliferating
hypertrophic chondrocytes. The splitting surface was
marked with micro fissures and the tidemark was
crossed by blood vessels and showed mesenchymal
tissue. The staining was lighter in the surface area,
Fig. 2 a–d Median values of the Lysholm-score divided by: a
Follow-up; b Number of transplanted cylinders; c Concomitant
lesions; d Genesis. No significant differences were found between
these groups regarding the Lysholm-score