Osteoarthritis and Cartilage (2000) 8, 1 8 2000 OsteoArthritis Research Society International 1063 4584/00/010001+08 $35.00/0 Article No. joca.1999.0263, available online at http://www.idealibrary.com on Joint distraction in treatment of osteoarthritis (II): effects on cartilage in a canine model A. A. van Valburg, P. M. van Roermund*, A. C. A. Marijnissen, M. J. G. Wenting, A. J. Verbout*, F. P. J. G. Lafeber and J. W. J. Bijlsma From the Department of Rheumatology & Clinical Immunology and the *Department of Orthopaedics, University Medical Centre of Utrecht, PO Box 85500, 3508 GA, The Netherlands Summary Objective: From a clinical point of view, joint distraction as a treatment for osteoarthritis (OA) of hip and ankle has been demonstrated to be very promising. Pain, joint mobility and functional ability, the most important factors for a patient with severe OA, all improved. Although radiographic joint space enlargement in a significant number of patients suggested cartilage repair, actual cartilage repair remains difficult to evaluate. Therefore the present study was initiated to evaluate the actual effects of joint distraction on cartilage. Methods: For this purpose a canine model for OA, anterior cruciate ligament transection (ACLT) was used. Sixteen weeks after ACLT articulating Ilizarov joint distraction of the knee was carried out. Absence of mechanical contact between articular surfaces and presence of intra-articular intermittent fluid pressure, characteristics of Ilizarov joint distraction, were confirmed. Twenty-five weeks after ACLT joint tissue of the dogs was analyzed. Results: Biochemical analysis showed that after joint distraction the abnormal cartilage proteoglycan (PG) metabolism, characteristic for OA, had changed to a level found in control joints. Moreover, a mild degree of inflammation, present after ACLT, was reduced upon joint distraction. PG-content and histological cartilage degeneration had not (yet) improved within the time of treatment. Discussion: Results suggest that the promising clinical results of Ilizarov joint distraction in patients with OA are accompanied by changes in cartilage metabolism. A change in proteoglycan turnover, indicating normalization of overall chondrocyte function, might in the long term, with normal joint use, lead to actual repair of cartilage. 2000 OsteoArthritis Research Society International Key words: Osteoarthritis, Distraction, Cartilage, Canine. Introduction As yet there is no effective proven remedy for osteoarthritis (OA), a degenerative joint disease characterized by destruction of articular cartilage. 1,2 In end stages of OA the only remaining options are an arthrodesis or placement of an endoprosthesis. 3 Joint distraction is an alternative approach, experimentally used in treatment of joint disorders such as ankylosis and OA. It aims to temporarily release mechanical stress on articular cartilage by preventing articular surfaces to contact during movement. It has been described that short term joint distraction by use of an external fixation frame can result in prolonged significant relieve of pain and significant restoration of joint function. 3 6 In patients with severe ankle OA it was found that the need for an arthrodesis could be delayed for at least two years by Ilizarov joint distraction. 3,6 A remarkable finding was the persistent radiographic joint space widening after joint distraction which could be due to actual repair of cartilage. Received 23 March 1998; accepted 2 May 1999. Address correspondence and reprint requests to: F. Lafeber, Dept. Rheumatology & Clinical Immunology, F02.127, University Medical Centre of Utrecht, PO Box 85500, 3508GA UTRECHT, The Netherlands. Supported by a grant from the Nationaal Reumafonds. The positive clinical effects of joint distraction have not yet been explained. Aldegheri et al. suggested an important role of fluid flow during joint movement under distraction. 4 We found that during ankle distraction with an Ilizarov external ring fixator, joint loading resulted in intra-articular intermittent fluid pressure. 3 Such an intermittent hydrostatic pressure, applied in vitro, resulted in stimulation of cartilage matrix formation specifically in OA cartilage. 7,8 We therefore hypothesise that, due to the absence of mechanical stress in combination with intra-articular intermittent fluid pressure/flow, Ilizarov joint distraction may result in actual repair of articular cartilage. Prospective clinical studies on the effect of joint distraction on clinical status and on cartilage repair in OA, judged by radiographic joint space width, are ongoing. Unfortunately, actual repair of articular cartilage in the clinical situation is difficult to study. Histological evaluation of cartilage biopsies, or evaluation of cartilage by X-ray, MRI or arthroscopy, all have their restrictions (not representative or too little material, no quantitative measures). 9,10 For this reason in the present study an animal model was used to study the effects of joint distraction on the early changes in articular cartilage and on secondary inflammation in OA. Biochemical and histological parameters of cartilage and secondary inflammation characteristic for OA were evaluated. 1
2 A. A. van Valburg et al.: Joint distraction in OA; a canine model Material and methods EXPERIMENTAL DESIGN Thirteen beagle dogs, aged 15.5±2.7 months, weighing 13.1±0.9 kg, were obtained from the animal laboratory of the Utrecht University. In all dogs, experimental osteoarthritis (OA) was induced by anterior cruciate ligament transection (ACLT) in one knee joint (t=0) as described below. Eleven weeks after ACLT dogs were randomly divided into three groups. One group (N=5) was treated with articulating joint distraction as described below (OA treated). A second group of three dogs was treated with non-articulating joint distraction (blocked hinges). And five other dogs received no treatment after ACLT (OA controls). During the entire study period dogs were housed in groups of three dogs per pen and were fed a standard commercial diet and water ad libitum. Dogs were exercised in a group twice a day for half an hour on a patio. Twenty-five weeks after ACLT all dogs were killed and cartilage and synovium of both knee joints was analysed according to standard procedures as described below. ACLT was performed in the right knee joint; the other knee joint was left intact and served as a control. The dogs were halothane endotracheal anaesthetized, and a small incision of less than 2 cm was made close to the ligamentum patellae to view the cruciate ligament. The cruciate ligament was lifted and transected, with care taken not to damage other joint structures. The completeness of transection was confirmed by a positive anterior drawer sign. The incision was sutured subcutaneously and cutaneously. Dogs received analgesics (buprenorphin 0.01 mg/kg) and antibiotics (amoxyllin 400 mg/kg) during the first three days after surgery. Exercise was started two days after surgery. Biochemical changes occurring 10 to 22 weeks after ACLT in this model are similar to changes observed in early human OA. 11 JOINT DISTRACTION Joint distraction was carried out in two phases; first, external fixation of femur and tibia (11 weeks after ACLT), followed by actual distraction after 5 weeks (16 weeks after ACLT). This approach was used to let the dogs become accustomed to the fixation frame before actual distraction was carried out. Application of the Ilizarov external fixation frame (Fig. 1A) was carried out under general anaesthesia (halothane). Dogs received analgesics (buprenorphin 0.01 mg/kg) and antibiotics (amoxyllin 400 mg/kg) during the first three days after surgery. Exercise was continued between three and five days after surgery. For fixation of the femur, rings as used for clinical treatment of OA (see part I of this study 6 ) were adapted to beagle anatomy. In eight dogs external fixation of femur and tibia were connected via hinges medial and lateral of the joint. By use of the screw threaded connecting rods, distraction of femur and tibia was carried out. Distraction was checked every two to three weeks by radiographic comparison with the contralateral control joint in identical position (Fig. 1B). Joint distraction appeared to be persistent throughout the entire period of treatment. In five of these dogs care was taken that frictionless motion of the knee joint remained possible during distraction. In the other three dogs the joint motion was blocked in a neutral position by blocking the hinges. Treatment was continued for a period of 8 weeks (8.4±0.5 weeks). The aim of the treatment was to achieve the absence of mechanical stresses on the articular cartilage in combination with intra-articular intermittent fluid pressure, as observed during clinical treatment. 3 Since dogs did not load their treated joint significantly, changes in intra-articular fluid pressure were generated by motion of the joint. These changes in intra-articular fluid pressure were measured by means of a pressure transducer connected to an intraarticularly positioned canula. During measurement dogs were only sedated with atropine (0.5 g) and methadon (10 mg) preserving muscular tension. Registration using a data-acquisition program (MKR version 4.2; Biomed. Eng. University Medical Centre of Utrecht) revealed that flexion and extension of the canine knee joint, under distraction without joint loading, resulted in pressure levels from 2.9±2.2 kpa to 11.8±4.7 kpa. During daily exercise dogs had a frequent motion of the treated knee. In the dogs with non-articulating distraction (blocked hinges) no motion was possible and no intra-articular pressure changes could be measured. SYNOVIAL TISSUE ANALYSIS For the five control dogs and the five dogs with articulating distraction from every joint three samples of synovial tissue were obtained from the supra-patellar pouch of the knee. These samples were fixed in 4% phosphate buffered formalin (ph 7.0). Samples were embedded in paraffin-wax after standard processing of the tissue in an automatic tissue processing apparatus. Deparaffined sections of synovium were stained with haematoxylin eosin. Specimens were analysed for degree of inflammation (separately in random order and independently by two observers who were not aware of the source of synovium) using the slightly modified 12 criteria described by Goldenberg and Cohen. 13 In short: Severity of synovitis was graded from 0 10 by adding the scores of three histological criteria, synovial lining cell hyperplasia (0 2); villous hyperplasia (0 3) and cellular infiltration by mononuclear and polymorphonuclear leucocytes (0 5). In assessing the overall grade, the three specimens taken from each knee were considered as one unit. Analysis of cartilage From all dogs cartilage was obtained from femoral condyles and tibial plateau according to standardized procedures. 14 Slices of articular cartilage were cut as thick as possible with care taken to exclude the underlying bone, and were kept in phosphate-buffered saline. Within 1 hour after dissection slices were cut into square pieces (3 10 mg) used for histological and biochemical analysis. From the control dogs and the animals with articulating distraction histological gradation of cartilage degeneration was carried out by light microscopy according to the slightly modified criteria of Mankin on safranin-o fast-green iron haematoxylin stained histological sections. 14,15 This gradation includes, in addition to evaluation of cartilage structure and cells, five steps for Safranin-O staining of proteoglycans. Specimens were graded in random order and independently by two observers who were not aware of the source of the cartilage. Six specimens taken from femoral condyles and tibial plateau each were considered as one unit. For biochemical analysis cartilage explants were weighed aseptically and incubated in 96 wells tissue culture plates according to standard procedures. The culture
Osteoarthritis and Cartilage Vol. 8 No. 1 3 (A) (B) Fig. 1. (A) Illustration of joint distraction applied to the knee joint of the dog. Half pins are drilled into the femur and connected with half oval rings around the upper limb. Kirschner wires are drilled through the tibia and connected to rings around the lower limb. Connection of both fixations is carried out by screw threaded rods connected with hinges, used for distraction and mobility during distraction. (B) Representative radiograph showing radiographic joint space widening indicative for the absence of mechanical stress during distraction. medium was Dulbecco s Modified Eagle s medium 12 with 10% heat inactivated pooled beagle serum. For both femoral and tibial cartilage proteoglycan (PG) content, PG synthesis and PG release as well as the DNA content were determined for six explants per animal. The average value of the six explants was considered representative of that parameter. DNA-content As a measure of cellularity of the cartilage its DNA content was determined. Cartilage samples were digested (ex vivo) in papain (2 h; 65 C) as described previously. 14 In part of the papain digest DNA was stained by use of the fluorescent dye Hoechst 33258 as has been described by Young-Jo et al. 16 Calf thymus DNA (Sigma D-4764) was used as a reference. The sample DNA content is expressed as mg DNA normalized to the wet weight of cartilage tissue (mg/g). PG synthesis rate As a measure of PG synthesis the rate of sulphate incorporation was determined ex-vivo during a 4-h period after 1 h of pre-incubation. Procedures were used as described previously. 14 In short: Na 2 35 SO 4 was added to
4 A. A. van Valburg et al.: Joint distraction in OA; a canine model each 200-μl culture medium. After labelling, the samples were digested with papain. Glycosaminoglycans (GAG) were precipitated upon addition of cetylpyridinium chloride (CPC). The 35 SO 2 4 labelled GAG in the precipitate were measured by liquid scintillation analysis. Total sulphate incorporation rate, calculated using the specific activity of the medium, was normalized to the wet weight of the explants and is expressed as nmoles of sulphate incorporated per hour per gram wet weight of the cartilage (nmol/h.g). PG release For determination of the release of PGs the cartilage samples were incubated for a subsequent 3-day period (day 5 7). The total amount of GAG in the medium was determined by precipitation and staining with Alcian blue dye solution. Blue staining was quantified photometrically with chondroitin sulfate (Sigma C4384) used as a reference. The total amount of released GAG (mg GAG per g wet weight released in three days) was normalized to the initial GAG-content of the tissue (percentage release). PG content As a measure of PG content of the cartilage samples the amount of GAG was determined as described previously. 12 Papain digest of cartilage explants was diluted to the appropriate GAG concentration where after the GAG were precipitated and stained with Alcian Blue as described above. GAG content is expressed in mg GAG normalized to the wet weight of the cartilage explants (mg/g). STATISTICAL ANALYSIS A power calculation for proteoglycan synthesis, proteoglycan release en proteoglycan content based on previous pilots indicates group sizes of 2, 16 and 6, respectively, to be sufficient for detecting the differences found between control and OA knees in the OA model at α=0.05 and 1-β=0.80. For all parameters mean values for 5 (c.q. 3) dogs±sem are presented. For comparison of experimental and contralateral control joints the Wilcoxon test for correlated data was used. For comparison of OA controls and OA treated dogs we used the Mann Whitney-U test. P values <0.05 were considered statistically significant. Results EXPERIMENTAL MODEL ACLT had a significant influence on gait pattern of the dogs. Dogs showed joint motion of the treated knee, but no significant loading of the treated joints was observed. This motion, however, guaranteed intermittent intra-articular fluid pressures in the absence of mechanical stress on the joint surfaces (see methods section). The dogs with blocked hinges (non-articulating distraction) did not show movement, nor loading of the knee joint. Within one week after surgery all dogs showed, apart from the above mentioned restrictions, normal behaviour. At the end of the distraction period quadriceps muscles showed a moderate degree of atrophy in the case of articulated distraction and significant atrophy in the case of fixed distraction. No significant complications were observed, no (pen) infections developed and all dogs were active throughout the entire experiment. SYNOVIAL TISSUE No effusions were observed in the dogs. Histological gradation of inflammatory features in the synovial tissue revealed a mild degree of inflammation in the OA control group 25 weeks after ACLT (score 4.8±0.4) compared to contralateral control joints (score 1.6±0.8; P<0.03). After joint distraction inflammation in the OA treated group was graded 3.1±0.91 which was significantly lower than in the OA control group (P<0.008) and not statistically different from the contralateral control joint (score 2.1±0.7). In the case of non-articulating distraction synovial tissue was graded 3.0±0.0 and 3.5±0.4 for control and treated knee, not statistically different from each other. Synovial inflammation in the experimental joints after non-articulating distraction was graded statistically significantly lower than in untreated experimental joints (P 0.02). There were no statistical differences in degree of inflammation between contralateral control joints of the three groups. The results for the variables as listed in materials and methods and detailed statistics are given in Table I. CARTILAGE Histological analysis of the articular cartilage revealed only a minimal change in cartilage in the OA untreated joints (Mankin grade femur 1.0±0.1; tibia 1.4±0.8) compared to the contralateral control joints (femur 0.2±0.2; tibia 0.5±0.2). Effects of joint distraction on these histological changes were insignificant (OA treated joints; Mankin grade femur 1.8±0.1, tibia 1.9±0.2). Differences between experimental and contralateral control joints (femur 0.2±0.2; tibia 0.4±0.2) in the treated dogs also were not statistically significant. In the case of non-articulated distraction grades were 0.1±0.1 and 2.3±0.5 (femur/tibia) for control and treated knees, respectively. In the OA treated joints the Safranin-O staining, representing PG content, was reduced compared to the cartilage obtained from untreated or contralateral control joints an effect even more pronounced in the joints treated with non-articulating (blocked hinges) distraction. (See also biochemically determined proteoglycan content in Table II.) Cartilage cellularity, as measured by DNA content, had not changed by the induction nor by the subsequent treatment of the OA; DNA content was similar compared to the contralateral control joints and did not differ between groups (Table II). A significant increase in PG synthesis, percentage PG release and a decrease in PG content, all characteristic for OA, was observed in the experimental joints of the OA control dogs compared to the contralateral control joints (Table II). Proteoglycan synthesis rate had increased with 87% in femoral cartilage and 110% in tibial cartilage 25 weeks after ACLT. The release of PG, expressed as percentage PG release, had increased 38% and 27% for femoral and tibial cartilage, respectively after induction of OA. PG content was decreased in femoral and tibial cartilage in the experimental joints of the OA control dogs (18 and 17% respectively). After treatment of the OA with articulating joint distraction it appeared that with respect to PG synthesis and PG release these characteristic changes had
Osteoarthritis and Cartilage Vol. 8 No. 1 5
6 A. A. van Valburg et al.: Joint distraction in OA; a canine model
Osteoarthritis and Cartilage Vol. 8 No. 1 7 altered towards the values of PG metabolism in contralateral control joints. Thus after articulating distraction, with respect to PG synthesis and PG release of the cartilage, no statistically significant differences were found anymore between treated and contralateral control joints. In contrast, after non-articulating distraction PG synthesis was strongly reduced and PG release was considerably increased compared to the contralateral control joint, both differences statistically significant (Table II). Values of PG metabolism in contralateral control joints in the OA control group and the OA treated groups were not statistically different. Surprisingly, after articulating distraction, a stronger decrease in PG content was observed than in the untreated OA joints. In the case of non-articulating distraction this decrease was even stronger (P 0.01) resulting in an almost complete depletion of PG (Table II). PG content had remained unchanged in the contralateral control joints of the osteoarthritic control and osteoarthritic treated groups. Discussion Joint distraction is a new experimental approach in treatment of osteoarthritis. 3 6 Clinical results of this experimental treatment are very promising and clinical benefit appears to persist over a prolonged period of time. Interestingly, clinical benefit of the distraction in the case of human ankle osteoarthritis becomes evident only after a significant period of time (months) after treatment, gradually increasing over time (years). We have hypothesized that clinical improvement depends on actual repair of cartilage induced by the absence of mechanical stress in combination with intermittent fluid pressure/flow during the joint distraction. 3 Since in humans evaluation of joint tissue after treatment is very difficult, we studied the effects of Ilizarov joint distraction as treatment for OA in an animal model. We used the anterior cruciate ligament transection model in the dog which is a generally accepted model for OA. 17 20 Application of joint distraction in the dog has been described previously by others. 21 The conditions as a result of the Ilizarov joint distraction in the human situation, viz. the absence of mechanical stress in the presence of intra-articular intermittent fluid pressure (0 13 kpa), were assessed during distraction in this experimental model. An increased joint space appeared to persist over the entire period of treatment. Levels of intermittent hydrostatic fluid pressure measured intra-articularly during distraction, induced by joint movement, varied from 3 to 12 kpa. These levels applied to human articular cartilage in vitro (0 13 kpa), have been reported to be beneficial for OA cartilage. 7,8 Thus the used strategy in this animal model provides a suitable model which mimics the human situation. We previously found that anterior cruciate ligament transection in the dog results within weeks in a minor degree of secondary inflammation in the synovial tissue and in biochemical cartilage changes comparable to those observed in human early OA. 11 Features of OA observed 25 weeks after ACLT in the present study are comparable with those observed 10 weeks after anterior cruciate ligament transection. 11 The use of contralateral knees as control joints in the present study has been reported to be legitimate 19 and is justified by the observation that cartilage biochemistry in these contralateral control joints is not different between the OA control group and OA treated group. The changed loading of the contralateral knees during joint distraction apparently does not affect the cartilage biochemistry significantly. Articulating joint distraction in the osteoarthritic knee joints results in a decrease of secondary inflammation of the synovial tissue. This effect of intermittent intra articular fluid pressure in the absence of mechanical stresses has been confirmed in vitro using cocultures of human OA synovial fluid mononuclear cells with human articular cartilage. 8 Moreover, joint distraction after induction of OA results in a change of cartilage PG metabolism towards values of the contra-lateral control knees. The observed increased in PG synthetic activity in OA cartilage, accompanied by an increased PG release has been suggested to depend on a functional change of the chondrocyte, as suggested a dedifferentiation of the chondrocyte. 22 28 Apparently, as shown in the present study, the characteristics of chondrocytes in osteoarthritic cartilage can be counteracted by articulating joint distraction. This change in chondrocyte metabolism could be beneficial for the cartilage, representing a first step to actual repair as was suggested by the clinical studies. Unfortunately, in the presently used experimental set-up cartilage repair was not found. PG content, significantly decreased in OA cartilage, showed an even stronger decrease after joint distraction. This seems in contrast with the reported beneficial clinical results in humans. However, the beneficial clinical results took a long time to emerge, whereas in the animal model cartilage metabolic changes were measured immediately after the distraction. The adverse effects of joint distraction found in the animal model may be due to atrophic changes. The impaired joint loading and probably also limited joint movement during distraction may have caused these effects. It is known that immobilization of joints or absence of joint loading results in atrophy, and in the long term even in OA. 29,30 The fact that in non-articulating distraction cartilage integrity was even worse supports this idea. It could be that in humans during treatment initially a similar change in cartilage integrity occurs. The change in chondrocyte metabolism, assumed to be beneficial, may need a successive follow-up period with normal joint use to actually result in repair of the tissue. In the human treatment it indeed required months to a year before significant clinical benefit was achieved; clinical improvement was experienced gradually after treatment. 6 In the presently used animal model such a follow-up is impossible since anterior cruciate ligament transection is a permanent trigger for OA. Furthermore, we have to keep in mind that the present animal model concerns very early OA in relatively young animals when compared to the severe OA in the older individuals in our clinical study. The importance of joint movement and therewith intra articular intermittent fluid pressure is illustrated by the observation that in the animals with nonarticulating distraction a practically complete inhibition of PG metabolism and depletion of PG was observed. In conclusion, Ilizarov joint distraction, being a combination of the absence of mechanical stress in the presence of intermittent fluid pressure, induces reduction of inflammation and changes in PG turnover. These findings suggest that the promising clinical results of Ilizarov joint distraction in patients might be accompanied by actual changes in cartilage. Acknowledgments The authors would like to thank Helma Avezaath and Aline van der Grift for their help in exercising the dogs. We thank Hanneke Hanssen- van Roy for technical assistance.
8 A. A. van Valburg et al.: Joint distraction in OA; a canine model References 1. Katz WA. General approach to the management of rheumatic diseases. In: Katz WA, Ed. Diagnoses and Management of Rheumatic Diseases. Philadelphia: Lippincott Co 1988:777 84. 2. Buckwalter JA, Lohmander S. Operative treatment of osteoarthrosis: current practice and future development. J Bone Joint Surg Am 1994;76:1405 18. 3. Van Valburg AA, Van Roermund PM, Lammens J, Van Melkebeek J, Verbout AJ, Lafeber FPJG, Bijlsma JWJ. Can Ilizarov joint distraction delay the need for an arthrodesis of the ankle? A preliminary report. J Bone Joint Surg Br 1995;77:720 5. 4. Aldegheri R, Griampaolo T, Saleh M. Articulated distraction of the hip. Conservative surgery for arthritis in young patients. Clin Orthopaedics and Rel Res 1994;301:94 101. 5. Morrey BF. Post-traumatic contracture of the elbow. Operative treatment, including distraction arthroplasty. J Bone Joint Surg Am 1990;72:601 18. 6. Van Valburg AA, Van Roermund PM, Marijnissen ACA, Van Melkebeek J, Lammens J, Verbout AJ, Lafeber FPJG, Bijlsma JWJ. Joint distraction in treatment of osteoarthritis (I): A two years follow-up of the ankle. Osteoarthritis Cart 1999;7:474 9. 7. Lafeber FPJG, Veldhuijzen JP, Van Roy JL, Huber- Bruning O, Bijlsma JWJ. Intermittent hydrostatic compressive force stimulates exclusively the proteoglycan synthesis of osteoarthritic human cartilage. Br J Rheumatol 1992;31:437 42. 8. Van Valburg AA, Van Roy JLAM, Lafeber FPJG, Bijlsma JWJ. Beneficial effect of intermittent fluid pressure, of a low but physiological magnitude, on cartilage and inflammation in osteoarthritis. J Rheumatol 1998;25:515 20. 9. Brandt KD, Fife RS, Braunstein EM, Katz B. Radiographic grading of the severity of knee osteoarthritis: relation of the Kellgren and Lawrence grade to a grade based on joint space narrowing, and correlation with arthroscopic evidence of articular cartilage degeneration. Arthritis Rheum 1991;34:1381 6. 10. Martel W, Adler RS, Chan K, Niklason L, Helvie MA, Jonsson K. Overview: New methods in imaging osteoarthritis. J Rheumatol 1991;18:32 7. 11. Marijnissen ACA, Van Valburg AA, Van Roermund PM, Van Roy JLAM, Verbout AJ, Lafeber FPJG, Bijlsma JWJ. A new canine model for osteoarthritis without a permanent trigger; a comparison with ACLT and human OA. Transactions of the 44th Annual Meeting of the Orthopaedic Research Society 1998;23:462. 12. Pelletier J-P, Martel-Pelletier J, Ghandur-Mnaymneh L, Howell DS, Woessner FJ. Role of synovial membrane inflammation in cartilage matrix breakdown in the Pond-Nuki model of osteoarthritis. Arthritis Rheum 1985;28:554 61. 13. Goldenberg DL, Cohen AS. Synovial membrane histopathology in the differential diagnosis of rheumatoid arthritis, gout, pseudo-gout, systemic lupus erythematosus, infectious arthritis and degenerative joint disease. Medicine (Baltimore) 1978;57:239 52. 14. Lafeber FPJG, Van der Kraan PM, Van Roy JLAM, Huber-Bruning O, Bijlsma JWJ. Articular cartilage explant culture; an appropriate in vitro system to compare osteoarthritic and normal human cartilage. Conn Tissue Res 1993;29:287 99. 15. Mankin HJ, Johns ME, Lippiello L. Biochemical and metabolic abnormalities in articular cartilage from osteoarthritic human hips. J Bone Joint Surg Am 1981;63:131 40. 16. Young-Jo RLY, Doong JYH, Grodzinsky AJL. Fluorometric assay of DNA in cartilage explants using Hoechst 33258. Anal Biochem 1988;74:168 74. 17. Pond MJ, Nuki G. Experimentally induced osteoarthritis in the dog. Ann Rheum Dis 1973;32:387 8. 18. Sandy JD, Adams ME, Billingham MEJ, Plaas A, Muir H. In vivo and in vitro stimulation of chondrocyte biosynthetic activity in early experimental osteoarthritis. Arthritis Rheum 1984;27:388 97. 19. Manicourt DH, Pita JC. Progressive depletion of hyaluronic acid in early experimental osteoarthritis in dogs. Arthritis Rheum 1988;31:538 44. 20. McDevitt CA, Muir H. Biochemical changes in the cartilage of the knee in experimental and natural osteoarthritis in the dog. J Bone Joint Surg Br 1976;58:94 101. 21. Judet R, Judet T. Arthrolyse et arthroplastie sous distracteur articulaire. Rev Chir Orthop 1978;64: 353 6. 22. Venn G, Billingham MEJ, Hardingham TE. Increased proteoglycan synthesis in cartilage in experimental canine osteoarthritis does not reflect permanent change in chondrocyte phenotype. Arthritis Rheum 1995;38:525 32. 23. Adams ME, Brandt KD. Hyperthropic repair of canine articular cartilage in osteoarthritis after anterior cruciate ligament transection. J Rheumatol 1991;18: 428 35. 24. Lafeber FPJG, Van der Kraan PM, Van Roy JLAM, Vitters EL, Huber Bruning O, Vanden Berg WB, Bijlsma JWJ. Local changes in proteoglycan synthesis during culture are different for normal and osteoarthritic cartilage. Am J Pathol 1992;140: 1421 9. 25. Carney SL, Billingham MEJ, Muir H, Sandy JD. Structure of newly synthesized ( 35 S)-Proteoglycans and ( 35 S)-Proteoglycan turnover products of cartilage explant cultures from dogs with experimental osteoarthritis. J Orthop Res 1985;3:140 7. 26. Pidd JG. Ultrastructural changes in the femoral condylar cartilage of mature American foxhounds following transection of the anterior cruciate ligament. J Rheumatol 1988;15:663 9. 27. Maroudas A, Venn M. Chemical composition and swelling of normal and osteoarthritic femoral head cartilage. Ann Rheum Dis 1977;36:399 406. 28. Walker GD, Fischer M, Gannon J, Thompson RC, Oegema TR. Expression of type X collagen in osteoarthritis. J Orthop Res 1995;13:4 12. 29. Palmoski M, Perricone E, Brandt KD. Development and reversal of a proteoglycan aggregation defect in normal canine knee cartilage after immobilization. Arthr Rheum 1979;22:508 17. 30. Langeskiöld A, Mechelsson J, Videman T. Osteoarthritis of the knee in the rabbit produced by immobilization. Acta Orthop Scand 1979;50:1 14.