Evaluation of polysulfated or sodium hyaluronan administered intra-articularly for treatment of horses with experimentally induced

David D. Frisbie, DVM, PhD; Chris E. Kawcak, DVM, PhD; C. Wayne McIlwraith, BVSc, PhD; Natasha M. Werpy, DVM

Objective—To assess clinical, biochemical, and histologic effects of polysulfated glycosaminoglycan (PSGAG) or sodium hyaluronan administered intra-articularly in treatment of horses with experimentally induced osteoarthritis. Animals—24 horses. Procedures—Osteoarthritis was induced arthroscopically in 1 middle carpal joint of all horses. Eight horses received hyaluronan (20 mg) and amikacin (125 mg) intra-articularly on study days 14, 21, and 28. Eight horses received PSGAG (250 mg) and amikacin (125 mg) intra-articularly on study days 14, 21, and 28. Eight control horses received 2 mL of saline (0.9% NaCl) solution and amikacin (125 mg) intra-articularly on study days 14, 21, and 28. Clinical, radiographic, analysis, gross, histologic, histochemical, and biochemi- cal findings were evaluated. Results—No adverse treatment-related events were detected. Induced osteoarthritis caused a substantial change in lameness, response to flexion, joint effusion, and radio- graphic findings, and of these, synovial fluid effusion was reduced with PSGAG, compared with control horses. No changes in clinical signs were seen with PSGAG or hyaluronan, compared with control horses. Histologically, the degree of synovial membrane vascularity and subintimal fibrosis was significantly reduced with PSGAG treatment, compared with controls. Histologically, significantly less fibrillation was seen with hyaluronan treatment, compared with controls. Conclusions and Clinical Relevance—Results indicated that PSGAG and hyaluronan had beneficial disease-modifying effects and are viable therapeutic options for osteoarthritis in horses. (Am J Vet Res 2009;70:203–209)

oint disease, specifically osteoarthritis, is one of Abbreviations Jthe most prevalent and debilitating diseases affect- PSGAG Polysulfated glycosaminoglycan ing horses and has a notable economic impact on the PGE Prostaglandin E equine industry.1–4 Various medications have been 2 2 evaluated or used for treatment of horses with osteo- arthritis, including NSAIDs, corticosteroids, PSGAG, and hyaluronan.5 Although the latter 2 medications of cartilage damage and clinical cases of osteoarthritis, are used extensively in clinical practice, definitive PSGAG had substantial beneficial effects.13–15 Likewise, data from randomized placebo-controlled studies in beneficial8,14,16–18 and nonbeneficial19–21 results have been horses are limited.6 reported for equine hyaluronan in vitro and in vivo. Beneficial effects of PSGAG have been detected in An experimental model of osteoarthritis has been vitro, although not all effects have been realized when used in horses for more than 10 years for assessment evaluated in vivo. Specifically, some7–9 but not all10,11 of the pathophysiologic process as well as evaluation equine in vitro studies have detected upregulation of of the efficacy of therapeutic substances in a controlled glycosaminoglycan and collagen synthesis as well as a environment.17,22-25,a The purpose of the masked con- decrease in inflammatory mediators, including PGE2. trolled study reported here was to evaluate the effects of Todhunter et al12 failed to detect beneficial effects in PSGAG and hyaluronan on clinical signs and disease- vivo on cartilage healing, although in a chemical model associated variables, compared with effects of control treatments, in experimentally induced osteoarthritis Received April 28, 2008. by evaluation of clinical (joint lameness, range of mo- Accepted June 9, 2008. tion, response to flexion, and synovial effusion), radio- From the Gail Holmes Equine Orthopaedic Research Center, College graphic, gross, biochemical, immunologic, and histo- of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO 80523-1678. logic outcome measures. Our hypothesis was that the Supported by Luitpold Pharmacies Incorporated. outcome of horses treated with PSGAG and hyaluronan Address correspondence to Dr. Frisbie. would be more favorable than that of control horses.

AJVR, Vol 70, No. 2, February 2009 203

08-04-0134r.indd 203 1/22/2009 12:56:54 PM Materials and Methods Assessment of clinical outcomes—Animal care personnel assessed horses daily for comfort, move- Experimental design and induction of osteoar- ment, and respiratory character. Clinical examinations thritis—The Colorado State University Animal Care of both forelimbs were performed weekly from day 0 and Use Committee approved the study protocol for (baseline) throughout the study period. Lameness was this experiment, which included the use of 24 healthy graded on a standardized scale of 0 to 5.27 All other clin- 2- to 5-year-old horses. Prior to inclusion in the study, ical outcomes were graded on a scale of 0 to 4 (0 repre- horses underwent a lameness examination, and body sented normal, and 4 represented severe change). As an condition, radiographs of the carpal joints, range of indication of joint pain, carpal flexion was performed motion (flexion) of the carpal joints, and evidence of after lameness grading. As an indication of increased joint effusion were assessed to ensure all variables were volume of synovial in the middle carpal within reference limits. On day 7 (7 days after induc- joint, the extent of the effusion was graded after carpal tion of osteoarthritis), horses were evaluated for degree flexion. All clinical outcome variables were assessed by of lameness. The lameness scores were used to rank the a board-certified large animal surgeon who focuses on horses from most to least lame and this list was used to equine lameness. blindly assign horses to 1 of the 3 treatment groups: hy- Radiographic evaluation of both carpi was aluronan (n = 8), PSGAG (8), or control (8). All evalu- performed prior to inclusion in the study (day –7), fol- ators were unaware of treatment assignment. 26 lowing the induction of osteoarthritis (day 14), and at As described, on day 0, following anesthesia and termination of the study (day 70). A board-certified routine preparation for surgery, each horse underwent radiologist assessed images. The radiographic images bilateral arthroscopic surgery of the middle carpal were evaluated for bony proliferation at the joint cap- joints to ensure that there were no preexisting abnor- sule attachment, subchondral bone lysis, and osteo- malities. During this procedure, an osteochondral frag- phyte formation on a scale of 0 to 4 (0 represented nor- ment was created in 1 randomly selected middle carpal mal, and 4 represented severe change). joint. The fragment was generated by use of an 8-mm curved osteotome directed perpendicular to the articu- Synovial fluid—Beginning on day 0 until the lar cartilage surface of the radial carpal bone at the level end of the study (day 70), synovial fluid was asepti- of the medial synovial plica. The fragment was allowed cally aspirated once per week from both middle car- to remain adhered to the joint capsule proximally. A pal joints of each horse. Synovial fluid (2 to 4 mL) motorized arthroburr was used to debride the exposed was directly aspirated from the joints by use of a subchondral bone between the fragment and parent 20-gauge needle and syringe. Samples were placed bone. A 15-mm-wide defect bed for the 8-mm-wide in tubes containing EDTA for routine synovial fluid fragment was created and the debris was not actively analysis (total protein concentration, cytologic eval- flushed from the joint, thereby inducing osteoarthritis. uation, and total WBC count) or stored at –80oC for This joint was designated as the osteoarthritis-affected biochemical protein analysis. joint; the sham-operated joint was used as the control The conventional analysis of synovial fluid in- joint. The arthroscopic portals were closed with 2-0 cluded assessment of total protein concentration, WBC nylon suture in a simple interrupted pattern, and cya- count, and differential count. Total protein and WBC noacrylateb was applied to the incision. The forelimbs concentrations were determined via refractometry and were bandaged, and the horses were allowed to recover use of an automated cell counter, respectively. Smears from anesthesia and surgery. All study horses received of synovial fluid were examined microscopically to de- c 1 dose of ceftiofur (2.2 mg/kg) IM prior to surgery and termine the differential WBC count. d 2 g of phenylbutazone orally once daily for 5 days af- Two biomarker protein assays were performed on ter surgery. Bandages were changed every 3 to 5 days synovial fluid. A modified 1,9-dimethyl-methylene blue and maintained until suture removal at 10 days after dye–binding assay was used to determine glycosamino- surgery. glycan concentration.28 Synovial fluid concentration

Exercise—Horses were housed in stalls (3.65 X of PGE2 was assessed by use of a commercially avail- 3.65 m each). Beginning on day 15, horses were exer- able, competitive, high-sensitivity enzyme immunoas- cised on a high-speed treadmill 5 days each week until say kit as directed by the manufacturer’s instructions.h the end of the study. Each day, the horses were trot- This included extraction by use of C2 columnsi as sug- ted (16 to 19 km/h) for 2 minutes, galloped (approx 32 gested by the manufacturer. Samples were processed in km/h) for 2 minutes, and trotted again (16 to 19 km/h) duplicate, and results were expressed in picograms per for 2 minutes to simulate the strenuous exercise of race milliliter. training. Gross pathologic observations of joints—At the Treatment groups—Treatment began on day 14. end of the study, all horses were euthanatized by ad- The 8 horses treated with hyaluronane were adminis- ministration of pentobarbital sodium. For each horse, tered 20 mg and 125 mg of amikacinf intra-articularly a necropsy examination was performed during which on study days 14, 21, and 28. The 8 horses treated with both middle carpal joints were specifically examined PSGAGg were administered 250 mg and 125 mg of ami- for degree and location of articular cartilage fibrillation kacin intra-articularly on study days 14, 21, and 28. The or erosion. A subjective grade (scale of 0 to 4) was as- 8 placebo horses received 2 mL of saline (0.9% NaCl) signed for partial- and full-thickness cartilage erosion as solution and 125 mg of amikacin intra-articularly on well as synovial membrane hemorrhage; for each of the study days 14, 21, and 28. 3 variables, grade 0 represented no pathologic change

204 AJVR, Vol 70, No. 2, February 2009

08-04-0134r.indd 204 1/22/2009 12:56:54 PM and 4 represented a severe change. A total erosion score Without knowledge of treatment assignments, ar- was assigned on the basis of overall joint health, also ticular cartilage sections stained with safranin O, fast with a scale of 0 to 4. green were evaluated for intensity of staining in the tangential, intermediate, radiate territorial, and radiate Histologic examinations—At necropsy, samples of interterritorial zones of the third carpal, fourth carpal, synovial membrane and joint capsule were harvested and radial carpal bones.26 Numeric values ranging from from the region dorsal to the osteochondral fragment 0 to 4 were assigned to each variable (0 indicated no and placed in neutral-buffered 10% formalin for H&E stain uptake, and 4 indicated normal stain uptake), and staining. Five-micron sections of the tissue samples a cumulative score for each articular cartilage sample were prepared. An evaluator who was unaware of treat- was calculated by summation of the zonal scores. ment assignments assessed the sections of synovial membrane and joint capsule for cellular infiltration, sy- Articular cartilage matrix evaluation—To estimate novial intimal hyperplasia, subintimal , subinti- articular cartilage proteoglycan content, the total artic- mal fibrosis, and subintimal vascularity.26 Each variable ular cartilage glycosaminoglycan content was measured was graded on a scale of 0 to 4 (0 represented no abnor- by use of a 1,9-dimethyl-methylene blue technique.28 mal change, and 4 represented the most severe change). Articular cartilage pieces were obtained from the area A cumulative pathology score was also calculated for directly adjacent to the osteochondral fragment and a synovial membrane samples. remote site within each joint. Each piece was stored at Articular cartilage pieces (5 mm2) were obtained –80oC prior to further processing and analysis. Samples from each joint (Figure 1). Samples were stored in were processed in duplicate, and results were reported neutral-buffered 10% formalin for 7 days and then pro- in micrograms of glycosaminoglycan per milliliter (car- cessed routinely for histologic examination by an eval- tilage was digested prior to analysis at a ratio of 10 mg uator who was unaware of treatment assignment. The of wet weight/mL of papain digest). 5-µm sections were stained with both H&E and safra- For analysis of cartilage matrix metabolism, articu- nin O fast green. Collection locations were chosen to lar cartilage samples were aseptically collected from the represent an area directly adjacent to the osteochondral weight-bearing surface that was remote from the osteo- fragment, a portion of the opposing articulating surface chondral fragment within each joint, and incorporation (third carpal bone), and a remote location (fourth car- of sulfur 35 radiolabeled SO4 was measured by use of pal bone). reported methods.26 Samples were processed in dupli- The H&E-stained sections were evaluated for cate, and the results were reported as counts per minute articular cartilage fibrillation, chondrocyte necrosis, per milligram of dry weight. chondrone formation (chondrocyte division within a lacuna), and focal cell loss.26 Numeric values ranging Statistical analysis—Data were evaluated by use of from 0 to 4 were assigned to each variable (0 repre- an ANOVA framework with a software programj with sented no abnormal change, and 4 represented the the horse as a random variable. The ANOVA tables were most severe change). A cumulative pathology score used to determine significant (P < 0.05) main effects (modified Mankin score) for each articular cartilage and interactions between main effect variables. When sample was also determined. The analysis of the his- individual comparisons were made, least square means tologic outcome parameters also considered location was used and P < 0.05 was considered significant. Val- of sample collection. ues are reported as mean ± SEM.

Figure 2—Plot of the improvement in lameness score in control horses and horses with induced osteoarthritis treated with hyal- Figure 1—Illustration of cartilage collection sites (a) in the mid- uronan or PSGAG. A positive score represents improvement in dle carpal joint of horses. Articular cartilage was collected for lameness following treatment. Different letters indicate signifi- histologic evaluation from the intermediate carpal bone (CI). cant (P < 0.05) differences among groups. PCB = Placebo. HA CR = Radial carpal bone. CU = Ulnar carpal bone. C2 = Second = Hyaluronan. No OA = No osteoarthritis (control). OA = Osteo- carpal bone. C3 = Third carpal bone. C4 = Fourth carpal bone. arthritis.

AJVR, Vol 70, No. 2, February 2009 205

08-04-0134r.indd 205 1/22/2009 12:56:55 PM Results Joint effusion—All horses had a significant P ( < 0.001) increase in synovial effusion score in the osteo- Musculoskeletal variables—All horses had a sig- arthritis-affected (2.42 ± 0.13) joints, compared with nificant P( < 0.001) increase in lameness score in the that of control joints (1.13 ± 0.13) for day 14. Change osteoarthritis-affected (mean ± SEM, 2.25 ± 0.13) limb, in joint effusion was calculated by use of day 14 as the compared with that in the control limb (0.38 ± 0.13) postosteoarthritis pretreatment baseline. There was a for days 7 and 14. Change in lameness was calculated significant P ( < 0.001) improvement in joint effusion by use of day 14 (the last pretreatment evaluation) as for osteoarthritis-affected joints in horses treated with the postosteoarthritis pretreatment baseline (a positive PSGAG, compared witheffusion in control or hyaluro- change score indicates improvement). There was no nan-treated horses (Figure 3). significant improvement in lameness score with respect to treatment (Figure 2). Radiographic evaluation—A significant increase in radiographic joint changes was detected for each Flexion—All horses had a significant P( < 0.001) radiographic outcome variable after surgery. Total ra- increase in flexion score in the osteoarthritis-affected diographic scores before treatment for control (0.40 ± (1.80 ± 0.11) limb, compared with that in the control 0.35) versus osteoarthritis-affected (4.08 ± 0.26) joints limb (0.29 ± 0.11) for days 7 and 14. Change in flexion were significantly P( < 0.001) different. No significant was calculated by use of day 14 as the postosteoarthritis treatment effects were detected. pretreatment baseline. No significant treatment effects were observed. Synovial fluid—Results of routine synovial fluid analysis indicated that, as expected, the total protein concentration increased significantly (P < 0.001) with induction of osteoarthritis (2.37 ± 0.08 g/dL) through- out the study period, compared with that in control joints (2.04 ± 0.08 g/dL). Synovial fluid WBC count was significantly P( < 0.001) increased in osteoarthritic versus control joints (437 ± 30 cells/dL vs 308 ± 30 cells/dL, respectively). There were no significant treat- ment effects in synovial total protein concentration or WBC counts. The glycosaminoglycan concentration in synovial fluid was significantly (P < 0.001) increased (3.85 ± 0.05 ln µg of glycosaminoglycan/mL) versus control (3.36 ± 0.05 ln µg of glycosaminoglycan/mL) joints. There were no significant treatment effects.

Synovial fluid PGE2 concentrations were signifi- cantly (P < 0.001) increased with induction of osteoar- thritis (5.19 ± 0.10 ln pg/mL), compared with control (4.22 ± 0.10 ln pg/mL) joints. There were no significant Figure 3—Plot of the improvement in synovial fluid effusion score treatment effects. in the same horses as in Figure 2. A positive score represents im- provement or less effusion following treatment. See Figure 2 for Gross pathologic observations of joints—At remainder of key. necropsy, hemorrhage score within the synovial

Figure 4—Plot of synovial membrane vascularity scores for H&E- Figure 5—Plot of the synovial membrane subintimal fibrosis stained sections in the same horses as in Figure 2. See Figure 2 scores for H&E-stained sections in the same horses as in Figure for remainder of key. 2. See Figure 2 for remainder of key.

206 AJVR, Vol 70, No. 2, February 2009

08-04-0134r.indd 206 1/22/2009 12:56:55 PM membrane was significantly (P < 0.001) increased in ment effects were seen for any of the gross pathologic osteoarthritis-affected (1.83 ± 0.15) versus control observations. ± (0.63 0.15) joints. Similarly, the articular cartilage Synovial membrane—Induction of osteoarthritis < total erosion score was significantly (P 0.001) in- did not result in significant effects on degree of synovi- ± creased in osteoarthritis-affected (1.67 0.16) ver- al membrane cellular infiltration, subintimal edema, or ± sus control (0.42 0.16) joints. No significant treat- intimal hyperplasia, and no significant treatment effects were observed. For the effect of treatment on synovial membrane vascularity, the P value (0.085) approached significance. Specifically, when individual comparisons with control values were made, the P value for joints treated with hyaluronan was 0.061 and for joints treat- ed with PSGAG was 0.019 (Figure 4). Similarly, for sy- novial membrane fibrosis, theP value (0.057) for treat- ment effects approached significance. Specifically, when individual comparisons with control values were made, the P value for joints treated with hyaluronan was 0.077 and for joints treated with PSGAG was 0.022 (Figures 5 and 6). Articular cartilage—Histologic evaluation of the ar- ticular cartilage via H&E staining revealed a significant P(

Figure 6—Photomicrographs of sections of synovial membranes representing typical scores for vascularity and subintimal fibrosis in carpal joints of the same horses as in Figure 2. H&E stain; bar = Figure 8—Photomicrographs of sections of cartilage fibrillation in 200 µm. carpal joints of horses. H&E stain; bar = 200 µm.

Figure 7—Plot of fibrillation scores in H&E-stained sections of Figure 9—Plot of glycosaminoglycan synthesis by articular cartilage articular cartilage from third carpal bones (C3) in the same horses from third carpal bones in the same horses as in Figure 2. dpm = as in Figure 2. See Figure 2 for remainder of key. Disintegrations per minute. See Figure 2 for remainder of key.

AJVR, Vol 70, No. 2, February 2009 207

08-04-0134r.indd 207 1/22/2009 12:56:57 PM < 0.001) increase in the modified Mankin score for os- volvement. Other variables such as exercise protocol and teoarthritis-affected joints (3.12 ± 0.32), compared with bilateral fragments versus unilateral fragments also could the score for control joints (1.04 ± 0.32), when all loca- explain differences. The variability of the clinical metacar- tions were considered. There were no significant treat- pophalanygeal (fetlock) joint cases reported by Gaustad et ment effects observed on the basis of the total modified al14 makes comparison between results of that study and Mankin score; however, fibrillation was significantlyP ( = those of the present study difficult. It is interesting that 0.018) less with treatment by location from which carti- IV administration of hyaluronan did reduce lameness in a lage samples were obtained, compared with fibrillation in previous study17 that used a similar technique. However, at osteoarthritis-affected joints in control horses. Specifically, that time, the technique induced more acute inflammatory when individual comparisons were made, treatment with changes in the synovial membrane and more acute syno- hyaluronan in osteoarthritis-affected joints significantly vitis with less articular cartilage change. Specifically, great- (P = 0.007) improved fibrillation score and theP value for er improvements in synovial membrane histologic scores treatment with PSGAG (0.083) approached significance, and synovial fluid total protein concentrations including

compared with the value for control horses (Figures 7 and PGE2 were detected, compared with improvements in the 8). present study. In the present study, although beneficial im- provements were detected in disease-modifying variables, Articular cartilage stained with safranin O, fast no changes in clinical signs were detected. green —Evaluation of articular cartilage for safranin Significant improvement in synovial fluid effusion O fast green staining score revealed a significant (P = scores was seen with PSGAG, suggesting a beneficial 0.012) decrease in staining of osteoarthritis-affected effect on joint soft tissues and joint inflammation that joints (4.89 ± 0.47), compared with staining of con- was superior to treatment with hyaluronan or saline trol joints (6.33 ± 0.46) for the cumulative score on the solution. This was confirmed by improvements in sy- third carpal bone. No other comparisons were signifi- novial membrane histologic variables and was similar cantly different, including treatment effects. to previous observations in synovial fluid effusion af- Articular cartilage matrix evaluation—No sig- ter chemically induced synovitis.13 It was interesting nificant difference was detected regarding cartilage gly- that no change in synovial fluid PGE2 concentration cosaminoglycan content with respect to induction of was observed, especially because that would be ex- disease or treatment group. The glycosaminoglycan pected with a decrease in inflammation and that has synthesis was significantly (P = 0.013) higher in car- been seen with PSGAG treatment in vitro.8 This was tilage from osteoarthritis-affected joints (4,236 ± 262 not a unique finding26 and may suggest that a reduction disintigrations/min per µg of glycosaminoglycan), com- in PGE2 concentration is not necessary for decreased pared with that in control joints (3,379 ± 262 disintegra- synovial effusion and improvement in synovial mem- tions/min per µg of glycosaminoglycan). A significant P( = brane variables. When administered systemically, use of 0.021) difference was also detected in glycosaminoglycan hyaluronan induced a significant reduction in synovial synthesis on the basis of treatment group in osteoarthri- fluid PGE2 concentration as well as a more profound tis-affected joints, compared with that in control joints. effect on synovial membrane variables. Although not Specifically, osteoarthritis-affected joints treated with PS- significant, scores for vascularity and subintimal fibro- GAG had lower glycosaminoglycan synthesis, compared sis in the synovial membrane in treated veresus control with control joints or hyaluronan-treated osteoarthritis-af- horses in this study suggested improvement but, com- fected joints (Figure 9). pared with previous results,17 were of lesser magnitude. Conversely, the reduction in articular cartilage fibrilla- Discussion tion was greater in the present study and suggested a more potent effect of locally administered hyaluronan In the present study, experimentally induced osteo- to improve cartilage health in a diseased joint. arthritis effectively resulted in clinical, gross, histologic, A significant reduction in glycosaminoglycan syn- and biochemical changes indicative of osteoarthritis. The thesis was observed in osteoarthritis-affected joints lesions are thought to represent both acute synovitis and treated with PSGAG, compared with those treated with traditional osteoarthritis that includes full-thickness car- hyaluronan or saline solution. This finding suggested tilage erosion. During this study, no adverse events were a beneficial effect of PSGAG. The authors have also re- recorded with any of the treatments and a mild degree of ported an increase in glycosaminoglycan synthesis in lameness was induced (typically grade 2 on a scale pub- association with early osteoarthritis changes.29 lished by the American College of Equine Practitioners). Clinically, reports suggest that hyaluronan may work There was no significant improvement in clinical better in acute synovitis and PSGAG in osteoarthritis, but lameness in any treatment group. This was in contrast to these findings were not supported by the present study.6 results of a study by Auer et al,18 in which harsher car- Both medications yielded similar improvements in various pal osteochondral fragmentation but a similar hyaluronan outcome variables, suggesting good efficacy especially in product was used, and a study of clinical cases by Gaus- cases of synovitis. Hyaluronan appeared to have a greater tad et al14 with a similar PSGAG and different hyaluro- effect on decreasing the degree of articular cartilage fibril- nan. In the study by Auer et al, intra-articular treatment lation, compared with PSGAG, and the magnitude of pos- with hyaluronan improved force plate values in treated itive effects on the synovial membrane was greater with limbs. One explanation may be in the difference between PSGAG, compared with hyaluronan. Both PSGAG and use of arthroscopy versus arthrotomy to create articular hyaluronan had disease-modifying osteoarthritis drug ac- fragments, which may affect the degree of soft tissue in- tions in the present study.

208 AJVR, Vol 70, No. 2, February 2009

08-04-0134r.indd 208 1/22/2009 12:56:58 PM a. Frisbie DD, Kawcak CE, McIlwraith CW. Evaluation of extracor- gated glycosaminoglycan on chemical and physical defects in poreal shock wave therapy for osteoarthritis (abstr), in Proceed- equine articular cartilage. Am J Vet Res 1987;48:1407–1414. ings. 50th Annu Conv Am Assoc Equine Pract 2004;261–263. 14. Gaustad G, Dolvik NI, Larsen S. Comparison of intra-articu- b. Super Glue (cyanoacrylate adhesive), Pacer Technology, Rancho lar of 2 ml of 0.9% NaCl solution with rest alone Cucamonga, Calif. for treatment of horses with traumatic arthritis. Am J Vet Res c. Naxcel, Pfizer Inc, New York, NY. 1999;60:1117–1121. d. Equi-Phar, Schering-Plough Animal Health Corp, Union, NJ. 15. Kristiansen KK, Kold SE. Multivariable analysis of factors influ- e. Hyvisc, Anika Therapeutics Inc, Woburn, Mass. encing outcome of 2 treatment protocols in 128 cases of horses f. Amikacin, Sicor Pharmaceuticals Inc, Irvine, Calif. responding positively to intra-articular analgesia of the distal g. Adequan IA, Luitpold Pharmaceuticals Inc, Animal Health Divi- interphalangeal joint. Equine Vet J 2007;39:150–156. sion, Shirley, NY. 16. Santangelo KS, Johnson AL, Ruppert AS, et al. Effects of hyal-

h. PGE2 kit, Assay Designs, Ann Arbor, Mich. uronan treatment on lipopolysaccharide-challenged fibroblast- i. Amprep Mini-columns ethyl C2 columns, GE Healthcare, Pisca- like synovial cells. Arthritis Res Ther 2007;9:R1. taway, NJ. 17. Kawcak CE, Frisbie DD, Trotter GW, et al. Effects of intrave- j. The GLIMMIX Procedure, SAS Institute Inc, Cary, NC. nous administration of sodium hyaluronate on carpal joints in exercising horses after arthroscopic surgery and osteochondral References fragmentmentation. Am J Vet Res 1997;58:1132–1140. 18. Auer JA, Fackelman GE, Gingerich DA, et al. Effect of hyal- 1. NAHMS. Part 1: baseline reference of 1998 equine health and uronic acid in naturally occurring and experimentally induced management. In: National animal health monitoring system. Fort osteoarthritis. Am J Vet Res 1980;41:568–574. Collins, Colo: USDA, APHIS, Veterinary Services, Centers for 19. Lynch TM, Caron JP, Arnoczky SP, et al. Influence of exogenous Epidemiology and Animal Health, 1998. hyaluronan on synthesis of hyaluronan and collagenase by 2. NAHMS. Lameness and laminitis in US Horses. In: National ani- equine synoviocytes. Am J Vet Res 1998;59:888–892. mal health monitoring system. Fort Collins, Colo: USDA, APHIS, 20. Barr ARS, Duance VC, Wotton SF, et al. Influence of intra-articu- Veterinary Services, Centers for Epidemiology and Animal lar sodium hyaluronate and polysulphated Health, 2000. on the biochemical composition of equine articular surface re- 3. Rossdale PD, Hopes R, Digby NJW, et al. Epidemiological pair tissue. Equine Vet J 1994;26:40–42. study of wastage among racehorses 1982 and 1983. Vet Rec 21. Hilbert BJ, Rowley G, Antonas KN, et al. Changes in the sy- 1985;116:66–69. novia after the intra-articular injection of sodium hyaluronate 4. Todhunter R, Lust G. Pathophysiology of synovitis: clinical into normal horse joints and after arthrotomy and experimental signs and examination in horses. Compend Contin Educ Pract Vet cartilage damage. Aust Vet J 1985;62:182–184. 1990;12:980–992. 22. Foland JW, McIlwraith CW, Trotter GW, et al. Effect of beta- 5. McIlwraith CW, Trotter GW. Joint disease in the horse. Philadel- methasone and exercise on equine carpal joints with osteochon- phia: WB Saunders, 1996. dral fragments. Vet Surg 1994;23:369–376. 6. Caron JP, Kaneene JB, Miller R. Results of a survey of equine 23. Frisbie DD, Kawcak CE, Baxter GM, et al. The effects of 6-alpha practitioners on the use and perceived efficacy of polysul- methylprednisolone acetate on an equine osteochondral frag- fated glycosaminoglycan. J Am Vet Med Assoc 1996;209:1564– ment exercise model. Am J Vet Res 1998;59:1619–1628. 1568. 24. Frisbie DD, Kawcak CE, Trotter GW, et al. The effects of triam- 7. Glade MJ. Polysulfated glycosaminoglycan accelerates net cinolone acetate on an in vivo equine osteochondral fragment synthesis of collagen and glycosaminoglycans by arthrit- exercise model. Equine Vet J 1997;29:349–359. ic equine cartilage tissues and chondrocytes. Am J Vet Res 25. Frisbie DD, Kawcak CE, Werpy NM, et al. Clinical, biochemi- 1990;51:779–785. cal, and histologic effects of intra-articular administration of 8. Frean SP, Lees P. Effects of polysulfated glycosaminoglycan and autologous conditioned serum in horses with experimentally hyaluronan on prostaglandin E2 production by cultured equine induced osteoarthritis. Am J Vet Res 2007;68:290–296. synoviocytes. Am J Vet Res 2000;61:499–505. 26. Frisbie DD, Ghivizzani SC, Robbins PD, et al. Treatment of 9. Clegg PD, Jones MD, Carter SD. The effect of drugs commonly experimental equine osteoarthritis by in vivo delivery of the used in the treatment of equine articular disorders on the acitiv- equine interleukin-1 receptor antagonist gene. Gene Ther ity of equine matrix metalloproteinases-2 and 9. J Vet Pharmacol 2002;9:12–20. Ther 1998;21:406–413. 27. Anonymous. Definition and classification of lameness. In:Guide 10. Caron JP, Toppin DS, Block JA. Effect of polysulfated glycosami- for veterinary service and judging of equestrian events. Lexington, noglycan on osteoarthritic equine articular cartilage in explant Ky: American Association of Equine Practitioners, 1991;19. culture. Am J Vet Res 1993;54:1116–1121. 28. Farndale RW, Buttle DJ, Barrett AJ. Improved quantitation and 11. Caron JP, Eberhart SW, Nachreiner R. Influence of polysulfated discrimination of sulphated glycosaminoglycans by use of di- glycosaminoglycan on equine articular cartilage in explant cul- methylmethylene blue. Biochim Biophys Acta 1986;883:173– ture. Am J Vet Res 1991;52:1622–1625. 177. 12. Todhunter R, Freeman K, Yeager A, et al. Effects of exercise and 29. Frisbie DD, Ray CS, Ionescu M, et al. Measurement of synovial polysulfated glycosaminoglycan on the development of osteo- fluid and serum concentrations of the 846 epitope of chondroi- arthritis in equine carpal joints with osteochondral defects. Vet tin sulfate and of carboxy propeptides of type II procollagen for Surg 1993;22:330–342. diagnosis of osteochondral fragmentation in horses. Am J Vet Res 13. Yovich JV, Trotter GW, McIlwraith CW, et al. Effects of polysul- 1999;60:306–309.

AJVR, Vol 70, No. 2, February 2009 209

08-04-0134r.indd 209 1/22/2009 12:56:58 PM