Observational study of the clinical value of the canine osteoarthritis staging tool

The aim of this multi-centred, cohort, prospective, observational study was to evaluate the clinical application of the canine osteoarthritis staging tool (COAST), in dogs affected by osteoarthritis (OA). In all dogs, a COAST stage was obtained based on the Liverpool Osteoarthritis in Dogs (LOAD) score and a complete orthopaedic and radiographic examination. The severity of OA (COAST stage) was scored as 1 (preclinical), 2 (mild), 3 (moderate), and 4 (severe). These scores were compared with the overall subjective opinion of experienced orthopaedic surgeons who examined the dogs (Clinical Opinion). Data were analysed with descriptive statistic, multiple regression analysis, chi-square and Bland-Altman tests (P < 0.05). In total, 362 evaluations were performed in 202 dogs. Clinical Opinion had a greater proportion of stage 1 and 2 cases compared to COAST (P < 0.0001). The proportion of stage 4 cases was higher in COAST compared to Clinical Opinion (P < 0.0001). The proportions of Stage 3 cases were similar in the two evaluation systems. COAST had a strong correlation (r = 0.79; P < 0.01) with Clinical Opinion. Overall, the two evaluation systems exhibited strong agreement (mean bias 0.51). Stages 1 and 2 had weaker agreement (mean bias 1.04 and 0.75, respectively), than stage 3 and stage 4 (mean bias 0.46 and 0.0, respectively).