Synthetic osteochondral replacement of the femoral articular surface

Abstract

We have studied damage to the tibial articular surface after replacement of the femoral surface in dogs. We inserted pairs of implants made of alumina, titanium and polyvinyl alcohol (PVA) hydrogel on titanium fibre mesh into the femoral condyles. The two hard materials caused marked pathological changes in the articular cartilage and menisci, but the hydrogel composite replacement caused minimal damage. The composite osteochondral device became rapidly attached to host bone by ingrowth into the supporting mesh. We discuss the clinical implications of the possible use of this material in articular resurfacing and joint replacement.

Full text

VOL. 79-B, N
O
. 6, NOVEMBER 1997 1003
M. Oka, MD, Professor
Y.-S. Chang, MD
J. Toguchida, MD, PhD
H.-O. Gu, PhD
Department of Artificial Locomotive Systems, Research Centre for Bio-
medical Engineering
T. Nakamura, MD
K. Ushio, MD
Department of Orthopaedic Surgery
Kyoto University, 53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-01,
Japan.
Correspondence should be sent to Professor M. Oka.
©1997 British Editorial Society of Bone and Joint Surgery
0301-620X/97/67606 $2.00
SYNTHETIC OSTEOCHONDRAL REPLACEMENT OF THE
FEMORAL ARTICULAR SURFACE
MASANORI OKA, YONG-SHUN CHANG, TAKASHI NAKAMURA, KAZUYASU USHIO,
JUNYA TOGUCHIDA, HAI-OU GU
From Kyoto University, Japan
We have studied damage to the tibial articular surface
after replacement of the femoral surface in dogs. We
inserted pairs of implants made of alumina, titanium
and polyvinyl alcohol (PVA) hydrogel on titanium fibre
mesh into the femoral condyles.
The two hard materials caused marked pathological
changes in the articular cartilage and menisci, but the
hydrogel composite replacement caused minimal
damage. The composite osteochondral device became
rapidly attached to host bone by ingrowth into the
supporting mesh.
We discuss the clinical implications of the possible use
of this material in articular resurfacing and joint
replacement.
J Bone Joint Surg [Br] 1997;79-B:1003-7.
Received 17 January 1997; Accepted after revision 9 May 1997
Hemiarthroplasty using a metal prosthesis articulating with
articular cartilage has been widely used for the hip,
1,2
shoulder
3
and other joints
4,5
since the 1950s. The overall
clinical results are usually satisfactory, but pathological
changes in the opposing articular cartilage have often been
observed.
6-9
Erosion of the acetabular articular cartilage,
with associated pain, is a recognised late complication of
hemiarthroplasty.
10,11
Furthermore, the use of prosthetic replacements usually
requires the removal of healthy cancellous bone which has
an important shock-absorbing role.
12
It is desirable therefore
to develop new materials with mechanical properties which
are more similar to those of normal articular cartilage.
We have developed an artificial replacement for articular
cartilage which preserves healthy subchondral cancellous
bone. This substance is polyvinyl alcohol (PVA) hydrogel
and we have improved its mechanical properties using a
new synthetic process.
13-15
In previous studies we characterised the mechanical
properties of this material in relation to its use as an
artificial articular cartilage including its lubricating and
shock-absorbing functions. A series of in vivo tests showed
that it had good biocompatibility.
16
In addition, histological
findings in synovial membranes and articular cartilage two
months after the introduction of small particles (50 to
300 m) of PVA hydrogel and ultra-high-molecular-weight
polyethylene (UHMWPE) into contralateral knees of the
same rats showed that the PVA hydrogel particles caused
much less inflammation. They were not resorbed because
of their bioinert properties.
17
A major problem, however, was the attachment of the
material to the bone. To overcome this, we infiltrated PVA
solution into the pores of one half of a block of titanium
fibre mesh to obtain a composite material. The shear force
between the PVA hydrogel and the titanium fibre mesh was
approximately 2.2 MPa.
17
Our aim in this study was to determine the changes
which occurred in menisci and articular cartilage in contact
with various implants. We implanted alumina, titanium and
our composite ‘osteochondral’ device (COD) into the load-
bearing regions of the femoral condyles of dogs and exam-
ined the time-dependent changes in the tibial articular
cartilage. We also determined the ingrowth of new bone
into the pores of the titanium fibre mesh and the remodel-
ling of the bone around the implants.
MATERIALS AND METHODS
Rectangular test pieces (10  6  5 mm) were manufac-
tured from pure titanium fibre mesh with a porosity of 60%
and a mean pore size of 170 m. The composite prostheses
were made by infiltrating PVA (molecular weight 220 000)
solution into the pores of the distal half of the titanium fibre
mesh, and binding it by gelling the PVA (Fig. 1). The joint
surfaces were shaped using the mirror-finished PVA hydro-
gel which we have developed. Implants of the same shape,
finish and size were manufactured from alumina and titani-

um. All the implants were produced by the Kyocera Cor-
poration, Japan. The artificial CODs were sterilised using
ethylene oxide gas and the titanium and alumina implants
in an autoclave.
Thirty-two of each of the three types of implant (total 96)
were inserted in pairs into the stifle joints of 24 mature adult
mongrel dogs weighing approximately 15 to 20 kg. Each
dog received two types of implant, one pair in each hind
leg. Under general anaesthesia with intravenous ketamine
HC1 combined with intramuscular atropine sulphate, both
hind legs of each dog were prepared and draped. A medial
parapatellar approach was used to expose the femoral con-
dyles and holes for the implants were made using a broach
and a dental burr. After an undersize trial, the test prostheses
were tapped into place. Routine anteroposterior and lateral
radiographs were taken immediately after surgery to ensure
that the implants were properly positioned (Fig. 2). No
postoperative external fixation was used. Two identical
implants were inserted into the medial and lateral femoral
condyles of one joint, and a different pair into the opposite
hind leg at the same operation. This ensured that the animal
would be obliged to stand on the operated legs.
The animals were killed in two batches, 12 at eight
weeks and 12 at 24 weeks using intravenous injections of
10% KC1. Immediately after death all the femoral and
tibial joint surfaces, including the menisci, were removed.
Colour macrophotographs of samples of the tibial articular
cartilage were taken and graded using improved evaluation
criteria (Table I) according to the classification of Cook,
Thomas and Kester.
18
These data were grouped by implant
type and analysed using the Kruskal-Wallis non-parametric
analysis of variance.
Each tibial plateau and its menisci, including the sub-
chondral bone, was fixed with 10% neutral buffered forma-
lin solution. The samples of tibial articular cartilage,
including the subchondral bone, were decalcified using the
Plank-Rychlo method.
19
These, together with the menisci,
were then dehydrated in a graded alcohol series and embed-
ded in paraffin. Thin sections (5 m) of each specimen
were cut and stained with haematoxylin and eosin.
20
Thio-
nin was used to stain for proteoglycans.
To study new bone ingrowth into the titanium fibre mesh
and bone remodelling around the implanted materials, each
femoral condyle was fixed in 10% neutral buffered forma-
lin solution. The samples were dehydrated in a graded
alcohol series and embedded in polyester resin. Thin unde-
calcified sections of each femoral condyle and its implant
were cut (EXAKT Cutting Machine, model BS-3000A,
Norderstedt, Germany) and ground to a thickness of 100 to
150 m using a bench-top polishing wheel (EXAKT
Microgrinding Machine, Norderstedt, Germany) with grad-
ed levels of waterproof abrasive paper (Rikagaku Co Ltd,
Sankyo, Japan). Each section was then subjected to Giemsa
surface staining, contact microradiographic examination
and studied by light microscopy.
RESULTS
Within one week of implantation all the animals were able
to bear weight on their operated legs and were allowed to
move freely outside their cages at two weeks after the
operation. No animal died and there were no intra- or
postoperative complications. All the knees had a nearly
normal range of movement and all the alumina, titanium
1004 M. OKA, Y.-S. CHANG, T. NAKAMURA,
ET AL
THE JOURNAL OF BONE AND JOINT SURGERY
Fig. 1
Photograph of titanium fibre-mesh implant with mirror-finished PVA
hydrogel joint surface.
Fig. 2a Fig. 2b
Anteroposterior (a) and lateral (b) radiographs of dog knees showing the
position of two metallic implants within the femur.
Table I. Improved criteria of evaluation for gross appearance
Grade
0 Normal cartilage
2 Minimal wear, no surface irregularities, slight colour change
4 Obvious cartilage thinning, limited regions of fibrillation, obvious
colour change
6 Areas of eburnation, severe fibrillation, obvious colour change
8 Partial loss of cartilage, no exposed subchondral bone
10 Partial exposure of subchondral bone, pannus formation

and artificial COD surfaces remained intact. No effusion or
inflammation was found and all of the test specimens were
firmly fixed to the underlying bone.
Macroscopic examination of 24 joints at eight weeks and
24 at 24 weeks showed considerable changes in the menisci
and tibial articular cartilage in those with alumina and
titanium implants. Ulceration and cartilage loss were appar-
ent in all specimens studied at eight weeks after implanta-
tion, and in some dogs the subchondral bone was exposed
(Fig. 3). The surfaces of the menisci which had articulated
with femoral condyles implanted with the two rigid materi-
als were no longer glossy and bright. In joints with a COD,
the menisci and tibial articular cartilages were still intact,
even in the 24-week specimens (Fig. 4). Table II sum-
marises the gross appearances of the tibial articular cartil-
age according to the stated evaluation criteria. There were
no significant differences between the two rigid materials,
but there were significant differences (p < 0.01) between
the findings for the rigid implants and the COD at both
eight and 24 weeks. At 24 weeks, we noted very slight
colour changes in the cartilage in joints with a COD. There
was much more severe cartilage damage and the formation
of pannus on the surface of the cartilage in tibial specimens
articulating with either of the rigid materials.
Thionin staining showed partial articular cartilage loss
and exposed subchondral bone in the tibial plateau at eight
weeks postoperatively in seven joints with three alumina
and four titanium implants, respectively (Fig. 5). By con-
trast, in those with a COD, the tibial articular cartilage
showed normal metachromatic staining, an intact surface,
normal cells and a bright matrix (Fig. 6).
There was abundant new bone ingrowth into the pores
of the titanium fibre mesh at eight weeks postoperatively
(Fig. 7), and at 24 weeks lamellar bone remodelling had
advanced further. Higher magnification showed mature
bone ingrowth in the deep pore spaces. At 24 weeks, no
demarcated or radiolucent zones were observed around the
material and there was no interposing fibrous tissue sur-
1005SYNTHETIC OSTEOCHONDRAL REPLACEMENT OF THE FEMORAL ARTICULAR SURFACE
VOL. 79-B, N
O
. 6, NOVEMBER 1997
Fig. 3 Fig. 4
Figure 3 – Photograph showing damage to the articular cartilage at eight weeks postoperatively in a knee with an alumina implant. Figure 4 – Photograph
of articular surface in a knee with a PVA hydrogel implant showing that it was still intact at 24 weeks postoperatively.
Fig. 5
Photomicrograph eight weeks after implantation showing considerable
erosion of the articular cartilage and exposure of subchondral bone in a
joint with alumina implants (thionin 70).
Table II. Changes of tibial articular cartilage of the three different biomaterials according to the criteria of evaluation (see
Table I)
8 weeks (n = 16) 24 weeks (n = 16)
Mean Median Maximum Minimum Mean Median Maximum Minimum
Alumina 8.38 8 10 8 9.38 10 10 8
Titanium 8.50 8 10 8 9.50 10 10 8
PVA hydrogel 0.25 0 2 0 0.25 0 2 0

1006 M. OKA, Y.-S. CHANG, T. NAKAMURA,
ET AL
THE JOURNAL OF BONE AND JOINT SURGERY
Fig. 6 Fig. 7
Fig. 8
Figure 6 – Photomicrograph of the tibial articular surface in
a knee with a COD at 24 weeks after implantation. There
are no pathological changes (thionin 140). Figure 7 –
Photomicrograph of a COD at eight weeks after implanta-
tion. The dark, blunt area at the top of the titanium fibre
mesh shows intact artificial cartilage PVA hydrogel (Giem-
sa surface staining 8). Figure 8 – Photomicrograph of a
COD at 24 weeks after implantation. The gap between the
PVA hydrogel and the surrounding tissues is seen on both
sides of the composite material (Giemsa surface staining
8).

rounding the titanium fibre mesh. Mature interstitial bone
was seen in the pores of the titanium fibre mesh. The
polished alumina and titanium implants were surrounded
by a thin fibrous tissue layer at 24 weeks.
Figure 8 shows a typical gap between the surrounding
normal articular cartilage and the PVA hydrogel. These
gaps were not filled by fibrous repair tissues, and they
persisted at 24 weeks after operation. The surface of the
PVA hydrogel showed no signs of wear at 24 weeks after
the operation.
DISCUSSION
Various methods of repairing articular surfaces have been
proposed using autografts, allografts, xenografts, periosteal
autografts and cultured chondrocytes, all of which may be
designated as ‘biological resurfacing’. Remarkable progress
has recently been made in this field, particularly with the
development of tissue engineering.
21,22
In addition to these
biological methods of resurfacing, different materials such
as polymers, hydrogels
23
and porous hydroxyapatite have
been used. All these artificial materials suffer from the
disadvantage of insufficient mechanical strength, which
leads to degradation of the restored surface after only two
or three months. A second problem is that the implant
cannot be firmly attached to adjacent bone and is thus
readily displaced. The scope of such repairs is therefore
limited to small areas; larger areas require a new bio-
material which has mechanical properties more similar to
those of articular cartilage. The properties required for
artificial articular cartilage include good lubrication, suffi-
cient shock-absorbing ability, good biocompatibility, high
wear resistance and the ability to be attached to the under-
lying bone.
Our findings have shown that our composite material has
abundant new bone ingrowth into the pores of the titanium
fibre mesh and can be firmly attached to the underlying
bone. In regard to the response of the tibial articular
surface, the two hard materials caused marked pathological
changes but the COD produced minimal changes.
Cook et al
18
investigated the low modulus and surface
chemistry and energy characteristics of low-temperature
isotropic (LTI) pyrolytic carbon. They reported that con-
sistently greater glycosaminoglycan loss and bony changes
occurred in cartilage articulating with metallic than with
LTI carbon implants.
There may also be a difference in the stability of the
implants. We did not perform formal pull-out tests of the
implants from bone, but the two hard materials seemed to
be fixed firmly to the bone. We believe that the more
favourable surface properties and interface mechanics are
relevant to the wear and degeneration of articular cartilage
replacement surfaces. Although a gap between the COD
and surrounding cartilage was found 24 weeks after opera-
tion, this composite material was firmly fixed to sub-
chondral bone.
Since the most important problem of firm attachment of
the material to the underlying bone has now been solved,
clinical application of the COD becomes possible. The
main potential application is in partial surface replacement
of the femoral head after aseptic necrosis. Others could
include articular resurfacing and the replacement of inter-
vertebral discs.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this
article.
REFERENCES
1. Thompson FR. Two and a half years’ experience with a vitallium
intramedullary hip prosthesis. J Bone Joint Surg [Am] 1954;36-A:
489-502.
2. Moore AT. The self-locking metal hip prosthesis. J Bone Joint Surg
[Am] 1957;39-A:811-27.
3. Neer CS II. Articular replacement for the humeral head. J Bone Joint
Surg [Am] 1955;37-A:215-28.
4. MacIntosh DL. Hemiarthroplasty of the knee using a space occupying
prosthesis for painful varus and valgus deformities. J Bone Joint Surg
[Am] 1958;40-A:1431.
5. Levitt RL. A long-term evaluation of patellar prostheses. Clin Orthop
1973;97:153-7.
6. D’Arcy J, Devas M. Treatment of fractures of the femoral neck by
replacement with the Thompson prosthesis. J Bone Joint Surg [Br]
1976;58-B:279-86.
7. Meyer S. Prosthetic replacement in hip fractures: a comparison
between the Moore and Christiansen endoprostheses. Clin Orthop
1981;160:57-62.
8. Leyshon RL, Matthews JP. Acetabular erosion and the Monk “hard
top” hip prosthesis. J Bone Joint Surg [Br] 1984;66-B:172-4.
9. Kaltsas DS, Klugman DJ. Acetabular erosion: a comparison between
the Austin Moore and the Monk hard top prostheses. Injury 1986;
17:230-6.
10. Shepherd MM. Assessment of function after arthroplasty of the hip.
J Bone Joint Surg [Br] 1954;36-B:354-63.
11. McDonough JM, Brandfass WT, Stinchfield FE. Idiopathic osteo-
arthritis of the hip. J Bone Joint Surg [Am] 1967;49-A:625-31.
12. Radin E, Paul O. Response of joints to impact loading: in vitro wear.
Arthritis Rheum 1971;14:356-62.
13. Oka M, Chang YS, Nakamura T, et al. Bone remodelling around
implanted materials. In: Hirasawa Y, Sledge CB, Woo S.L.-Y., eds.
Clinical biomechanics and related research. Springer-Verlag, Tokyo,
1994:124-37.
14. Oka M, Ikeuchi K, Tsutsumi S, et al. Biomechanics of natural and
artificial joints. In: Niwa S, Perren SM, Hattori T, eds. Biomechanics
in orthopedics. Springer, Berlin etc: 1992:282-98.
15. Oka M, Noguchi T, Kumar P. Development of an artificial articular
cartilage. Clin Mater 1990;6L361-81.
16. Noguchi T, Yamamuro T, Oka M, et al. Poly(vinyl alcohol) hydrogel
as an artificial articular cartilage: evaluation of biocompatibility.
J Appl Biomater 1991;2:101-7.
17. Oka M. Mise au point d’un cartilage articulaire artifiel. In: Passuti N,
ed. Biomateriaux de substitution de L OS et du cartilage. 57 Cahier’s
d’enseignement de la SOFCOT Expans Scientif Française, Paris,
1996:62-74.
18. Cook SD, Thomas KA, Kester MA. Wear characteristics of the
canine acetabulum against different femoral prostheses. J Bone Joint
Surg [Br] 1989;71-B:189-97.
19. Plank J, Rychlo A. Eine Schnellentkalkungsmethode. Z Path Anat
1952;89:252-4.
20. Bulstra SK, Drukker J, Juijer R, Buurman WA, Van Der Linden
AJ. Thionin staining of paraffin and plastic embedded sections of
cartilage. Biotech Histochem 1993;68:20-8.
21. Langer R, Vacanti JP. Tissue engineering. Science 1993;260:920-6.
22. Vacanti CA, Kim W, Schloo B, Upton J, Vacanti JP. Joint resurfac-
ing with cartilage grown in situ from cell-polymer structures. Am J
Sports Med 1994;22:485-8.
23. Murray DG, Dow JA, Fairchild E. Surface interaction between
hydrophilic gels and articular cartilge. J Surg Res 1978;24:537-43.
1007SYNTHETIC OSTEOCHONDRAL REPLACEMENT OF THE FEMORAL ARTICULAR SURFACE
VOL. 79-B, N
O
. 6, NOVEMBER 1997