Cartilage oligomeric matrix protein and hyaluronan levels in synovial fluid from horses with osteoarthritis of the tarsometatarsal joint compared to a control population

REASONS FOR PERFORMING STUDY: Quantification of cartilage oligomeric matrix protein (COMP) levels within synovial fluid from the tarsometatarsal joint has not previously been reported and an effective synovial fluid marker would allow monitoring of disease progression and treatment. OBJECTIVES: To quantify levels of COMP and hyaluronan (HA) in synovial fluid from the tarsometatarsal joint, identify differences in levels from horses with osteoarthritis (OA) of the tarsometatarsal joint compared to a control population and to correlate levels with radiographic changes in horses with OA. METHODS: Synovial fluid was collected from the tarsometatarsal joint of 25 horses without hindlimb lameness (controls) and 25 lame horses, subjected to analgesia of the joint. COMP concentrations were measured using a homologous inhibition ELISA. Immunoblots of synovial fluid from 3 lame horses and 3 controls were performed to identify fragmentation of COMP. Hyaluronan (HA) concentration in synovial fluid was determined using a competition ELISA. Radiographs of the lame horses with OA were scored and correlated with levels of COMP and HA. RESULTS: Concentrations of COMP in OA of the tarsometatarsal joint were significantly lower than in the control samples. An additional fragment band of COMP (approximately 30 kDa) was identified on the immunoblots of the horses with OA and this fragment was not identified in controls. No significant difference was identified in the HA or HA:COMP ratio between lame and control horses. There was no correlation between levels of synovial fluid COMP and HA, and radiographic changes. CONCLUSIONS AND POTENTIAL RELEVANCE: Lowered levels of COMP in synovial fluid of tarsometatarsal joints correlates with the presence of osteoarthritis. However, a single value cannot be used to stage the disease process. Levels of HA may not be a useful marker for this disease. Decreased, rather than increased COMP levels, may reflect significant loss of cartilage in established osteoarthritis. A specific assay for the COMP fragment generated with osteoarthritis may allow the earlier detection of clinical cases.     

Prevalence and relevance of antibodies to type-I and -II collagen in synovial fluid of dogs with cranial cruciate ligament damage

OBJECTIVE: To measure and compare synovial fluid antibody titers to type-I and -II collagen in stifle joints with instability caused by complete or partial cranial cruciate ligament (CCL) rupture and joints with osteoarthrosis secondary to other pathologic changes in dogs. ANIMALS: 82 dogs with diseased stifle joints. PROCEDURE: Synovial fluid samples were collected from 7 dogs with clinically normal stifles (control group) and 82 dogs with diseased joints (50 stifle joints with complete rupture of the CCL, 20 with partial damage of the CCL, and 12 joints with radiographic signs of osteoarthritis secondary to other arthropathies). Synovial fluid samples were tested for autoantibodies to type-I and -II collagen by an ELISA. RESULTS: In dogs with complete and partial CCL rupture, synovial fluid antibody titers to type-I and -II collagen were significantly increased, compared with control dogs. Forty-eight percent (24/50) of samples from dogs with complete CCL rupture and 35% (7/20) of samples from dogs with partial CCL rupture had antibody titers to type-I collagen that were greater than the mean plus 2 standard deviations of the control group titers. Synovial fluid antibody titers to type-II collagen were high in 40% of the dogs with partial or (8/20) complete (20/50) CCL rupture. Dogs with osteoarthrosis secondary to other pathologic changes had significantly increased synovial fluid antibodies to type-I and -II collagen, compared with control dogs. CONCLUSION: Increases in autoantibodies to collagen in synovial fluid are not specific for the type of joint disorder. It is unlikely that the anticollagen antibodies play an active role in the initiation of weakening of the CCL.     

Evaluation of avocado and soybean unsaponifiable extracts for treatment of horses with experimentally induced osteoarthritis

OBJECTIVE: To evaluate the use of a combination of avocado and soybean unsaponifiable (ASU) extracts for the treatment of experimentally induced osteoarthritis in horses. ANIMALS: 16 horses. PROCEDURES: Osteoarthritis was induced via osteochondral fragmentation in 1 middle carpal joint of each horse; the other joint underwent a sham operation. Horses were randomly allocated to receive oral treatment with ASU extracts (1:2 [avocado-to-soybean] ratio mixed in 6 mL of molasses; n = 8) or molasses (6 mL) alone (placebo treatment; 8) once daily from days 0 to 70. Lameness, response to joint flexion, synovial effusion, gross and histologic joint assessments, and serum and synovial fluid biochemical data were compared between treatment groups to identify effects of treatment. RESULTS: Osteochondral fragmentation induced significant increases in various variables indicative of joint pain and disease. Treatment with ASU extracts did not have an effect on signs of pain or lameness; however, there was a significant reduction in severity of articular cartilage erosion and synovial hemorrhage (assessed grossly) and significant increase in articular cartilage glycosaminoglycan synthesis, compared with placebo-treated horses. CONCLUSIONS AND CLINICAL RELEVANCE: Although treatment with ASU extracts did not decrease clinical signs of pain in horses with experimentally induced osteoarthritis, there did appear to be a disease-modifying effect of treatment, compared with findings in placebo-treated horses. These objective data support the use of ASU extracts as a disease-modifying treatment for management of osteoarthritis in horses.     

Glycosaminoglycans in horses with osteoarthritis

Horse articular cartilage glycosaminoglycans (GAGs) were measured in synovial fluids from 48 joints affected with osteoarthritis (OA), 22 normal joints, four joints with osteochondritis, three joints with traumatic arthritis and seven joints infected with bacteria. Serum and urine from individual horses were also examined for the presence of GAGs. High levels of GAGs were found in synovial fluids (SF) from horses with OA. In each case, the level was higher in the synovial fluid than in the serum or urine from the same horse. Horses with OA showed high GAG levels in SF, serum and urine compared to horses with normal and infected joints. High levels were also found in horses with osteochondritis and traumatic arthritis. Levels of synovial fluid GAG reflect cartilage destruction in arthritis and may be useful for monitoring disease progression in the equine species.     

Canine distemper viral antigens and antibodies in dogs with rheumatoid arthritis

Dogs with canine rheumatoid arthritis had significantly elevated levels of antibodies to canine distemper virus. This increase was particularly seen in the synovial fluids, compared with paired sera, and was not found in dogs with infective arthropathies, osteoarthritis or in osteoarthritis secondary to rupture of the cranial cruciate ligament. Analysis of the immune complexes precipitated from synovial fluids showed immunoglobulins in all types of arthropathy. Western blotting analyses showed reactivity with anti-distemper antisera in immune complexes from dogs with rheumatoid arthritis, but not in immune complexes from dogs with other joint diseases. These results suggest that there are increased immune responses to distemper in canine arthritis and that these may be due to the presence of this paramyxovirus in affected joints. The implications for the role of a possible infectious agent in rheumatoid arthritis in the dog are considerable.     

Concentrations of serum amyloid A in serum and synovial fluid from healthy horses and horses with joint disease

OBJECTIVE: To determine serum amyloid A (SAA) concentrations in serum and synovial fluid from healthy horses and horses with joint disease and assess the effect of repeated arthrocentesis on SAA concentrations in synovial fluid. Animals-10 healthy horses and 21 horses with various types of joint disease. PROCEDURES: Serum and synovial fluid samples were obtained from each horse. In 5 of the 10 healthy horses, arthrocentesis was repeated 9 times. Concentrations of SAA were determined via immunoturbidometry. RESULTS: Serum and synovial fluid SAA concentrations were less than the assay detection limit in healthy horses and did not change in response to repeated arthrocentesis. Synovial fluid SAA concentrations were significantly higher in horses with suspected bacterial joint contamination or infectious arthritis, or tenovaginitis than in healthy controls, and serum concentrations were significantly higher in horses with infectious conditions than in the other groups. Neither serum nor synovial fluid SAA concentrations in horses with low-inflammation joint conditions differed significantly from those in healthy controls. Concentrations of SAA and total protein in synovial fluid were significantly correlated. CONCLUSIONS AND CLINICAL RELEVANCE: Synovial fluid SAA concentration was a good marker of infectious arthritis and tenovaginitis and appeared to reflect changes in inflammatory activity. The advantages of use of SAA as a marker include the ease and speed of measurement and the fact that concentrations in synovial fluid were not influenced by repeated arthrocentesis in healthy horses. Further study of the SAA response in osteoarthritic joints to assess its usefulness in diagnosis and monitoring of osteoarthritis is warranted.     

Assessment of the effects of age and joint disease on hydroxyproline and glycosaminoglycan concentrations in synovial fluid from the metacarpophalangeal joint of horses

OBJECTIVE: To assess the effects of age and joint disease on hydroxyproline and glycosaminoglycan (GAG) concentrations in synovial fluid from the metacarpophalangeal joint of horses and evaluate the association of those concentrations with severity of osteoarthritis and general matrix metalloproteinase (MMP) activity. SAMPLE POPULATION: Synovial fluid was collected from the metacarpophalangeal joints of foals at birth (n = 10), 5-month-old foals (10), 11-month-old foals (5), and adult horses (73). PROCEDURE: Hydroxyproline and GAG concentrations were determined in synovial fluid samples. The severity of osteoarthritis in adult joints was quantified by use of a cartilage degeneration index (CDI) and assessment of general MMP-activity via a fluorogenic assay. RESULTS: Hydroxyproline and GAG concentrations in synovial fluid were highest in neonates and decreased with age. Concentrations reached a plateau in adults by 4 years and remained constant in healthy joints. In synovial fluid from osteoarthritic joints, hydroxyproline and GAG concentrations were not increased, compared with unaffected joints, but hydroxyproline were significantly correlated with the CDI and general MMP activity. There was no significant correlation between GAG concentration and CDI value or MMP activity. CONCLUSIONS AND CLINICAL RELEVANCE: Changes in hydroxyproline concentration in synovial fluid appeared to indicate damage to collagen of the articular cartilage. In joints with osteoarthritis, the lack of high GAG concentration in synovial fluid and the absence of a significant correlation between GAG concentration and CDI values or MMP activity may severely limit the usefulness of this marker for monitoring equine joint disease.     

Serum level of cartilage oligomeric matrix protein (COMP) in equine osteoarthritis

This study was designed to assay and compare cartilage oligomeric matrix protein (COMP) in horse sera, in samples from normal and joint diseased horses, and to investigate the relationships between COMP in sera and synovial fluids (SF) with keratan sulphate (KS) data. Sera from 38 horses free of any joint pathology (controls) and from horses with aseptic joint disease (AJD horses, n = 40) were assayed for COMP and KS concentrations. Of the 78 horses in the study, 53 were also assayed for COMP and KS concentrations in SF. COMP and KS were measured by inhibition ELISA, using monoclonal antibodies 12C4 and 5D4, respectively. The COMP concentration in sera from AJD horses (mean +/- s.d. 10.7 +/- 7.4 microg/ml) was significantly (P<0.02) lower than in control sera (14.8 +/- 7.8 microg/ml). The joint disease sera also had significantly lower (P<0.01) KS levels (180.5 +/- 61.8 ng/ml) than controls (237.1 +/- 116.1 ng/ml). A significant correlation (r = 0.52, n = 53, P<0.001) was seen between serum and SF in COMP levels; no such relationship was seen in KS levels. It is possible that serum COMP concentration could be a more specific marker of equine joint disease than any other described to date.     

Metalloproteinases and tumor necrosis factor-alpha activities in synovial fluids of horses: correlation with articular cartilage alterations

Early detection of osteoarthritis in horses represents a challenge for equine practitioners. Several biological markers have been implicated in the pathological processes involved in articular cartilage destruction. To further document cartilage matrix proteases production, synovial fluid was collected from 14 horses (90 joints) before they were subjected to euthanasia. Growth macroscopic examination of the joints gave information on cartilage alterations. Samples were analyzed for matrix metalloproteinase (MMPs) activities by gelatin zymography and tumor necrosis factor alpha (TNF-alpha) cytotoxicity using L929 cells. Significant increase of MMP-9 monomer and dimer were found in synovial fluids of joints with severe cartilage alterations. On the contrary, the activity of TNF-alpha was not correlated to the degree of joint damage. The levels of MMP-9 monomer and dimer in the synovial fluid could reflect cartilage alteration in arthritis in the horse.     

Evaluation of anticollagen type I antibody titers in synovial fluid of both stifle joints and the left shoulder joint of dogs with unilateral cranial cruciate disease

OBJECTIVE: To evaluate anticollagen type I antibodies in synovial fluid of the affected stifle joint, the contralateral stifle joint, and the left shoulder joint of dogs with unilateral cranial cruciate ligament (CrCL) rupture during an extended period of 12 to 18 months. ANIMALS: 13 client-owned dogs with CrCL rupture and 2 sham-operated dogs. PROCEDURES: All dogs were examined and arthrocentesis of all 3 joints was performed every 6 months after surgery. Synovial fluid samples were tested for anticollagen type I antibodies by use of an ELISA. RESULTS: Dogs with partial CrCL rupture had higher antibody titers than dogs with complete rupture. Six of 13 dogs ruptured the contralateral CrCL during the study, whereby higher antibody titers were found for the stifle joints than for the shoulder joint. Seronegative dogs or dogs with extremely low antibody titers and 2 dogs with high antibody titers did not sustain a CrCL rupture in the contralateral stifle joint. CONCLUSIONS AND CLINICAL RELEVANCE: In most dogs that had a CrCL rupture of the contralateral stifle joint, a distinct antibody titer gradient toward the stifle joints was detected, suggesting that there was a local inflammatory process in these joints. However, only a small number of sham-operated dogs were used to calculate the cutoff values used to determine the anticollagen type I antibody titers in these patients. Synovial fluid antibodies against collagen type I alone do not initiate CrCL rupture because not all dogs with high antibody titers sustained a CrCL rupture in the contralateral stifle joint.     

Measurement of glycosaminoglycans and keratan sulphate in canine arthropathies

Glycosaminoglycans (gag) and keratan sulphate (ks) were measured in sera and synovial fluids from dogs with either osteoarthritis (oa) or rupture of the cranial cruciate ligament (ccl) and normal dogs. The dogs with oa had higher synovial fluid gag levels (P<0·002) and serum KS (P<0·03) compared to the normal dogs. No significant differences in serum gag were found in either group. In both oa and rupture of the ccl, gag levels were increased in the synovial fluid from the affected joint compared with the clinically normal (inactive) contralateral joint. Neither gag nor ks measurements correlated with serum and synovial fluid antibodies to collagen type II, synovial fluid white cell count or age of dog. It is unlikely that the measurement of these cartilage breakdown products is of value for diagnostic or prognostic use in canine arthropathies.     

Synovial Fluid Studies in Navicular Disease

The purpose of this study was to investigate biochemical changes in synovial fluid in navicular disease, and to establish if synovial fluid from the distal interphalangeal joint (DIP) could be used diagnostically to assess alterations in the synovial fluid of the navicular bursa. Cartilage oligomeric matrix protein (COMP), total glycosaminoglycans (GAG), hyaluronan (HA), metalloproteinases 2 and -9 (MMP-2 and MMP-9) and total protein (TP) levels were determined in synovial fluids obtained from 18 navicular bursae and 35 DIP -joints from animals suffering from navicular disease, and the same synovial structures in 16 joints of horses with no evidence of abnormalities involving the foot. To avoid dilution effects, GAG/COMP, HA/COMP, MMP-2/ COMP and MMP-9/COMP ratios were also calculated for different synovial cavities. There was a good correlation, for COMP, GAG, HA, MMP-2 and TP levels, between synovial fluid from the navicular bursa and fluid from the DIP -joint in healthy animals. However, in animals with navicular disease, only COMP levels showed no difference between the navicular bursal fluid and the DIP-joint fluid concentration. Thus, enabling the use of COMP to standardise other biochemical concentration measurements from the synovial joint fluids. In horses with navicular disease, there was a significantly lower absolute concentration of GAG, and a significantly lower GAG/COMP ratio, in the synovial fluid of the navicular bursa and the DIP-joint compared to synovial fluid from the same joints from healthy horses. In contrast, the absolute HA concentration and HA/ COMP, MMP-2/COMP and MMP-9/COMP ratios were higher in synovial fluid from the DIP-joint of horses with navicular disease, and MMP-2 and MMP-9 relative activity levels and MMP-2/COMP and MMP-9/ COMP ratios were increased in fluid from navicular bursae in horses with navicular disease when compared to a control group.     

Substance P in the synovial membrane and fluid of the equine middle carpal joint.

This preliminary study was designed to determine whether the neurotransmitter substance P was present in the middle carpal synovial membrane of the normal horse and whether the neuropeptide could be identified in the synovial fluid of normal horses and those with joint diseases. Immunocytochemistry on middle carpal synovial membrane biopsies from fresh cadavers was used to demonstrate substance P-containing neural elements. Substance P was most abundant in the subintimal portion of the membrane, with occasional filaments coursing via synovial fronds to the intimal portion. Radioimmunoassay techniques were used on acidified acetonitrile-preserved synovial fluid samples to measure substance P concentrations. Fluid from 9 joints of 5 normal horses and 6 joints of 4 horses with joint diseases were analysed. Disease conditions included acute and chronic osteoarthritis and osteochondrosis. Synovia from normal horses contained a mean concentration of substance P significantly less than that of horses with joint diseases (P less than 0.05). Elevated concentrations of neurotransmitters in diseased joints suggests a potential contribution to the pathophysiology of joint disorders in horses.     

Recombinant equine growth hormone administration: effects on synovial fluid biomarkers and cartilage metabolism in horses

REASONS FOR PERFORMING STUDY: Recombinant equine growth hormone (reGH) has recently been evaluated for effects on body condition and wound healing. It has the potential to influence articular cartilage via stimulation of IGF-1. OBJECTIVES: To investigate effects of administration on synovial joint metabolism. METHODS: Six mature horses were given 20 microg/kg bwt reGH daily for 8 weeks by i.m. injection. Three control horses were injected with sterile water. Serum and synovial fluid samples were collected at 6, 8, 11 and 16 weeks for GH and IGF-1 assays. Articular cartilage harvested at week 16 was evaluated by Western analysis using monoclonal antibodies BC-13, BC-4, 8-A-4 and CH-3. RESULTS: Concentrations of IGF-1 in serum and synovial fluid were significantly elevated (P < 0.05) at 6 and 8 weeks in the reGH group. Glycosaminoglycan concentrations in synovial fluid were significantly less than controls at these time points, suggesting that reGH may modulate proteoglycan metabolism in articular cartilage. In the reGH group, there were not any alterations in synovial fluid content of 3B3(-) epitope or aggrecan metabolite, or in aggrecan or link protein catabolites retained within cartilage, that might be expected with development of osteoarthritis. CONCLUSIONS: Intramuscular administration of reGH may be a more efficient means of delivery of IGF-1 to joints for cartilage resurfacing initiatives. POTENTIAL RELEVANCE: We found no alterations in cartilage metabolism indicative of development of osteoarthritis.     

Plasma and synovial fluid endothelin-1 and nitric oxide concentrations in horses with and without joint disease

OBJECTIVE: To compare plasma and synovial fluid endothelin-1 (ET-1) and nitric oxide (NO) concentrations in clinically normal horses and horses with joint disease. ANIMALS: 36 horses with joint disease, and 15 horses without joint disease. PROCEDURE: Horses with joint disease were assigned to 1 of the 3 groups (ie, synovitis, degenerative joint disease [DJD], or joint sepsis groups) on the basis of findings on clinical and radiographic examination and synovial fluid analysis. Endothelin-1 and NO concentrations were measured in plasma from blood samples, collected from the jugular vein and ipsilateral cephalic or saphenous vein of the limb with an affected or unaffected joint, as well as in synovial fluid samples obtained via arthrocentesis from the involved joint. RESULTS: Plasma ET-1 concentrations between affected and unaffected groups were not significantly different. Median concentration and concentration range of ET-1 in synovial fluid obtained from the joint sepsis group (35.830 pg/mL, 7926 to 86.614 pg/mL; n = 7) were significantly greater than values from the synovitis (17.531 pg/mL, 0.01 to 46.908 pg/mL; 18), DJD (22.858 pg/mL, 0.01 to 49.990 pg/mL; 10), and unaffected (10.547 pg/mL, 0.01 to 35.927 pg/mL; 10) groups. Plasma and synovial fluid NO concentrations between affected and unaffected groups were not significantly different. CONCLUSIONS AND CLINICAL RELEVANCE: Endothelin-1 is locally synthesized in the joints of horses with various types of joint disease. Synovial fluid concentrations of ET-1 varied among horses with joint disease, with concentrations significantly higher in the synovial fluid of horses with joint sepsis. These results indicate that ET-1 may play a role in the pathophysiologic mechanism of joint disease in horses.     

Measurement of equine myeloperoxidase (MPO) activity in synovial fluid by a modified MPO assay and evaluation of joint diseases - an initial case study

The aim of this study was to develop a specific myeloperoxidase (MPO) activity assay in the synovial fluid of horses and investigate whether MPO activity is increased in different forms of joint diseases. Synovial fluid samples were taken from affected joints from horses with osteoarthritis, chronic non-septic arthritis and septic arthritis, and from healthy control horses. MPO activity was measured using a specific modified o-dianisidine-assay containing 4-aminobenzoic acid hydrazide as a potent and specific inhibitor of the MPO. This assay is characterized by high reproducibility. The results reveal only a slight elevation of MPO activity in the synovial fluid of horses with osteoarthritis and chronic non-septic arthritis. However, in the cases of septic arthritis a significant increase in MPO activity was found when compared to the controls. In conclusion the first field study suggests that synovial fluid MPO may be used as a marker for septic arthritis in horses.     

Inhibition of interleukin-1 activity by equine synovial fluid.

The presence, in equine synovial fluid, of inhibitors of interleukin-1 (IL-1) activity has been investigated by means of an assay involving IL-1-mediated production of PGE2 by synovial cells. Inhibitors of IL-1 alpha and IL-1 beta were identified in normal synovial fluid and synovial fluid from two horses with early joint disease. Inhibitors of IL-1 alpha were also present in synovial fluid from two horses with long-standing joint disease. However, IL-1 beta inhibitory activity was not present in fluid from the horses with more chronic joint disease. The effect appeared to be specific for IL-1, and not a direct action on PGE2 production, as synovial fluid had no effect on lipopolysaccharide-mediated PGE2 production. It is suggested that the inhibitory activity may be involved physiologically in the control of IL-1 activity in the joint, and the loss of IL-1 inhibition may be at least as important biologically as increased production of IL-1.     

1H NMR investigation of normal and osteo-arthritic synovial fluid in the horse.

Proton magnetic resonance spectroscopy (1H NMR) has been successfully used in the study of many biological fluids. The data presented here report on the metabolic profiles of normal equine synovial fluids compared with osteoarthritic (OA) fluids. Twenty-five OA synovial fluid samples and eight normal ones were collected from the forelimb fetlock joint in 22 horses, aged between five and 24 years. 1H NMR spectroscopy was carried out with a Bruker Avance DRX 500 equiped with a cryo-magnet working at 11 Tesla, and 'Mestre-C 4.9.9.6' software was used to analyze the spectra. The study assessed the increase of lactate, alanine, acetate, N-acetylglucosamine, pyruvate, citrate, creatine/creatinine, glycerol, HDL choline, and a-glucose in OA synovial fluid. The variations observed in samples from horses with OA compared to those in the control group, and similar data found in other studies, confirm that this technique may be useful in the study of joint metabolism. Its practical application may be in the evaluation of the treatment of OA in athletic horses.     

Pharmacokinetics of amoxycillin in normal horses and horses with experimental arthritis

The serum and synovial pharmacokinetics of amoxycillin (AMX) were studied after i.v. administration at a dosage of 40 mg/kg to normal horses and horses with induced aseptic carpal arthritis. The best estimates of serum and synovial pharmacokinetic parameters were calculated by mono or bivariable non-linear regression analysis. A biexponential equation was used to describe the concentration vs. time profiles in both normal and arthritic horses. There were no serum kinetic differences between normal and arthritic horses. There were, however, major synovial kinetic changes between these groups. The rate of penetration from serum to synovial fluid was larger in arthritic animals, indicating better penetration in this case. On the other hand, the rate of disappearance from synovial fluid was larger in normal horses, indicating more persistence of the drug in the diseased joint. Synovial AMX availability increased from 21% in normal horses to 79% in arthritic horses. These findings support the use of AMX for the treatment of infectious synovial joint disease produced by susceptible organisms in horses.     

Differences in the concentration of various synovial fluid constituents between the distal interphalangeal joint, the metacarpophalangeal joint and the navicular bursa in normal horses

As a prerequisite for the identification of navicular disease markers, the concentrations of cartilage oligomeric matrix protein (COMP), total glycosaminoglycans (GAG), hyaluronan, metalloproteinases (MMP) 2 and 9 and total protein were measured in synovial fluid samples obtained from the distal interphalangeal joint (DIP), the metacarpophalangeal joint (MCP) and the navicular bursa of 24 horses. Mean GAG, COMP and total protein levels were significantly higher in the DIP joint and in the navicular bursa compared to the MCP joint. Hyaluronan content was lower. MMP -2 activity was present in all fluids measured and had similar levels in different joints. MMP -9 was present in 42 per cent of MCP joint samples and 58 per cent of DIP joint samples and of navicular bursal samples. In relation to the constituents measured, the composition of navicular bursal fluid was similar to the articular synovial fluids, in particular that obtained from the DIP joint. Correlation between the constituents of DIP joint fluid and navicular bursal fluid obtained from the same legs was statistically significant for all the parameters measured.