Osteoarthritis in horses: intramuscular polysulfated glycosaminoglycan vs placebo — MAYBE (weak evidence)

In horses with osteoarthritis, does intramuscular polysulfated glycosaminoglycan vs placebo improve lameness?

Shelf #804 · gate-verified and approved by the developer

MAYBE (weak evidence): The intramuscular (IM) polysulfated glycosaminoglycan (PSGAG) vs placebo evidence in horses comes entirely from small experimental joint-disease models, not from naturally occurring osteoarthritis, and the two randomised records point in different directions on different outcomes. In Verde 2010 (PMID: 20646197; 12 healthy horses with monoiodoacetate-induced carpitis allowed to develop for 7 days, then randomly divided), horses receiving PSGAG 500 mg IM every 4 days for seven treatments showed less joint circumference enlargement and less lameness and greater maximum carpal flexion than horses receiving IM normal saline on the same schedule, assessed on days 0, 7, 14, 21, 28 and 35 by area-under-the-curve analysis — but the record's abstract reports no effect sizes, no scores and no P values, so the size and statistical security of the lameness benefit cannot be verified from the record. In Kawcak 2011 (PMID: 21627523; 24 healthy 2- to 3-year-old horses with an osteochondral fragment created in one middle carpal joint, randomly allocated, 8 per group), IM PSGAG every 4 days for 28 days had no significant effect on any subchondral bone variable and no effect on serum or synovial fluid biomarkers, against a placebo group that received a sham shock-wave probe rather than placebo injections — and that record reports no lameness outcome at all. Design class (b): randomisation is established by both carded records for this shelf's own route and comparison (IM PSGAG vs a control), but blinding is not established — neither record states masked lameness or outcome assessment in its abstract, and in Kawcak 2011 the control was a sham procedure for a different treatment arm, not a matched IM placebo. The stamp is MAYBE (weak evidence), not POSSIBLE, because the only record measuring lameness is a 12-horse chemically induced carpitis study whose abstract gives direction without numbers, the population is an acute induced model rather than clinical osteoarthritis, and the second randomised record is null on every outcome it measured. One-home cross-points: intra-articular PSGAG in experimentally induced equine osteoarthritis (Frisbie 2009, PMID: 19231952) homes on shelf #198 (intra-articular triamcinolone vs hyaluronic acid) and is not re-carded here — the intra-articular route is a different intervention; IM PSGAG in dogs homes on shelf #187 and no canine record is imported; intra-articular triamcinolone combinations home on shelves #802/#805. Tier C is not engaged: no mortality or survival claim is made. Drafted Nov 10, 2026 (VetMed build, topic #804); figures from the abstracts, fetched and verified at drafting.

Verde 2010: in 12 horses with chemically induced carpitis, IM polysulfated glycosaminoglycan (500 mg every 4 days × 7) gave less lameness, less joint enlargement and greater carpal flexion than IM saline — with no effect sizes or P values in the record

Randomised controlled experimental-model study (record states horses were randomly divided; publication types: Journal Article, Randomized Controlled Trial) — blinding not stated in the record

Who: Horses: 12 healthy horses subjected to the monoiodoacetate (MIA) carpitis model, allowed to develop for 7 days, then randomly divided into two groups. Group A (control) received an intramuscular injection of normal saline every 4 days for a total of seven injections; group B received 500 mg of a polysulfated glycosaminoglycan preparation (recorded in the record as SYNTEX CSY36) intramuscularly every 4 days for seven treatments. Efficacy was evaluated by three clinical outcomes — lameness score, carpal circumference and maximum carpal flexion — measured on days −8 (before carpitis induction), 0 (before drug treatment), 7, 14, 21, 28 and 35; areas under the curve of each clinical outcome over time were used as the statistical variables. The record does not state that assessors were masked, and the abstract does not give the number of horses per group.

Result: There was less joint circumference enlargement and lameness and greater carpal flexion in PSGAG-treated horses compared with controls, and the authors state the compound demonstrated efficacy in the treatment of chemically induced carpitis in horses. The record's abstract reports these directions of effect only: it gives no lameness scores, no circumference or flexion values, no effect sizes and no P values, so neither the magnitude nor the statistical significance of the lameness difference can be stated from the record.

Watch out: This is an acute chemically induced carpitis in 12 healthy horses, not naturally occurring osteoarthritis — the shelf's target condition — so transport to clinical OA is indirect. The abstract's complete absence of numbers is the central limitation: 'less lameness' is the authors' summary of an area-under-the-curve analysis whose scores and significance are not reported in the record, and the per-group n is not stated (12 horses in two groups). Lameness scoring is subjective and no masking of assessors is stated. The tested preparation is identified in the record by a product code (SYNTEX CSY36); results are stated here for polysulfated glycosaminoglycan as a class only as far as this one preparation was tested.

Verde C, Ferrante M, Simpson MI, Babusci M, Broglia G, Landoni MF. Efficacy of intramuscular polysulfated glycosaminoglycan in a controlled study of equine carpitis. J Vet Pharmacol Ther. 2010;33(4):357-62. PMID: 20646197. Registry ID: none in the record. MeSH: Animals · Arthritis/drug therapy/veterinary* · Carpal Joints/drug effects*/pathology · Disease Models, Animal · Glycosaminoglycans/administration & dosage/standards/therapeutic use* · Horse Diseases/drug therapy*/pathology · Horses · Injections, Intramuscular/veterinary · Lameness, Animal/drug therapy* · Treatment Outcome

Kawcak 2011: in 24 horses with fragment-induced carpal osteoarthritis, IM polysulfated glycosaminoglycan every 4 days for 28 days had no effect on subchondral bone or on serum/synovial biomarkers vs the placebo group — and the record reports no lameness outcome

Randomised controlled experimental-model study (record states horses were randomly allocated; publication types: Journal Article, Randomized Controlled Trial) — the placebo arm is a sham-procedure control, not a matched IM placebo, and masking is not stated in the record

Who: Horses: 24 healthy 2- to 3-year-old horses with experimentally induced osteoarthritis — an osteochondral fragment was created on the distal aspect of the radial carpal bone in one middle carpal joint of each horse. Horses were randomly allocated to local extracorporeal shock wave therapy (ESWT; days 14 and 28; n = 8), polysulfated glycosaminoglycan treatment (PSGAG; IM, every 4 days for 28 days; n = 8), or a sham ESWT probe (placebo; days 14 and 28; n = 8). Serum biomarkers were measured every 7 days and synovial fluid biomarkers every 14 days; bone density was measured by computed tomography on days 0 and 70, and microdamage and bone formation variables were compared at day 70. Note the placebo was a sham shock-wave procedure, not a placebo intramuscular injection.

Result: There was no significant effect of ESWT or PSGAG treatment on any subchondral bone variable. Serum osteocalcin concentration was significantly greater in ESWT-treated horses than in placebo-treated horses, and serum concentration of the C-terminal telopeptide of type I collagen was significantly higher in ESWT-treated horses than in placebo- and PSGAG-treated horses; synovial fluid CS846 epitope concentrations were significantly higher in joints treated with ESWT. The authors conclude that treatment of osteoarthritis with ESWT had no effect on subchondral bone but induced increases in serum biomarkers indicative of bone remodeling, and that treatment with PSGAG had no effect on subchondral bone or biomarkers. No lameness scores or clinical-sign outcomes are reported in the record's abstract for any group.

Watch out: This card is carded for its randomised IM-PSGAG-vs-control comparison in induced equine osteoarthritis, but its outcomes are bone and biomarker variables only — it neither confirms nor refutes a lameness effect, and it must not be quoted as a lameness trial. The control group received a sham shock-wave probe on days 14 and 28 rather than repeated placebo IM injections, so the PSGAG contrast is not a fully matched placebo comparison (no injection ritual in controls, and masking of assessors is not stated). Only 8 horses per group, a fragment-induced model in young healthy horses, and a 70-day horizon. Its null biomarker/bone results bound any disease-modifying claim for IM PSGAG; they do not contradict Verde 2010's symptomatic lameness claim, which rests on a different model and unreported numbers.

Kawcak CE, Frisbie DD, McIlwraith CW. Effects of extracorporeal shock wave therapy and polysulfated glycosaminoglycan treatment on subchondral bone, serum biomarkers, and synovial fluid biomarkers in horses with induced osteoarthritis. Am J Vet Res. 2011;72(6):772-9. PMID: 21627523. Registry ID: none in the record. MeSH: Animals · Biomarkers/metabolism · Carpus, Animal/injuries/pathology · Cartilage, Articular/pathology · Collagen Type II/metabolism · Epitopes/blood · Female · Glycosaminoglycans/administration & dosage/therapeutic use* · High-Energy Shock Waves/therapeutic use* · Horse Diseases/blood/pathology/therapy* · Horses · Lameness, Animal/blood/classification/therapy* · Male · Osteoarthritis/blood/pathology/therapy/veterinary* · Synovial Fluid/chemistry

Synthesis: IM PSGAG has one small randomised positive lameness result in induced carpitis (numbers not reported) and one randomised null on bone/biomarkers in fragment-induced OA — MAYBE (weak evidence)

Synthesis

Who: Synthesis of the two cards above (Verde 2010, 12 horses, MIA carpitis model; Kawcak 2011, 24 horses, osteochondral-fragment OA model) and of the near-miss records documented in the result, fetched at drafting and deliberately not carded as claim-carriers.

Result: (1) Lameness: only Verde 2010 measured it — less lameness, less joint enlargement and greater flexion with IM PSGAG 500 mg every 4 days × 7 vs IM saline in MIA carpitis — but the record reports no scores, effect sizes or P values, so the benefit is directional and unquantified. (2) Structure/biomarkers: Kawcak 2011 found no effect of the same IM regimen class (every 4 days for 28 days) on subchondral bone variables or serum/synovial biomarkers in fragment-induced OA. (3) Near-misses not carded: Trotter 1989 (PMID: 2469533) gave the same IM regimen (500 mg every 4 days × 7 vs IM saline, 8 horses) after created cartilage defects/MIA injury and found no effect on healing of articular cartilage lesions and only minimal chondroprotective effect (slightly greater matrix glycosaminoglycan staining in MIA joints) — cartilage outcomes, no lameness comparison, consistent with the weak stamp. Kawcak 2011's shock-wave arm and the intra-articular PSGAG literature (e.g., PMID: 8236786, intra-articular PSGAG after osteochondral defects in ponies; PMID: 2929085, a 4-case intra-articular series) are different routes/interventions and are not imported. Practitioner-survey evidence (PMID: 8899019) records perceived efficacy only and carries no claim here. (4) Design class (b): randomisation established by both carded records; blinding not established by either. No randomised IM-PSGAG trial in naturally occurring equine osteoarthritis with a lameness outcome was found in the searches run at drafting — the clinical-OA lameness question remains untested in the fetched records.

Watch out: One-home check at drafting: none of PMIDs 20646197, 21627523 or 2469533 appears in any other pack on disk. Boundaries honoured: shelf #187 is IM PSGAG in DOGS (canine records never imported); shelf #198 homes the intra-articular PSGAG record (PMID: 19231952) alongside IA triamcinolone vs hyaluronan; shelves #802 and #805 cover IA triamcinolone + hyaluronan and IA autologous conditioned serum respectively — different routes and interventions, cross-pointed here only. Do not upgrade Verde 2010's 'less lameness' into a quantified benefit, and do not present Kawcak 2011 as evidence about lameness in either direction — it measured none.

Synthesis of Verde C et al. PMID: 20646197; Kawcak CE et al. PMID: 21627523; Trotter GW et al. PMID: 2469533; PMID: 8236786; PMID: 2929085; PMID: 8899019. Registry IDs as on the individual cards (none in the records).

Back to the library index · VetMed home

VetMed is an educational resource, not veterinary advice.