It was reported recently that oral administration of prednisone, 0.3 mg/kg/day, decreased osteophyte size and cartilage ulceration in dogs with osteoarthritis (OA) produced by "stab" transection of the anterior cruciate ligament. Since this dose would be equivalent to 17 mg/day for a 70-kg human, our study was performed to determine whether a lower, clinically more realistic, dose of prednisone also had ameliorating effects on OA in the canine model. Following arthrotomy and transection of the anterior cruciate ligament 10 adult mongrel dogs were each given prednisone, 0.1 mg/kg/day, begun either 0, 2, or 4 weeks after arthrotomy and continued until sacrifice 12 weeks after surgery; 7 others underwent the identical surgical procedure but received no prednisone and served as OA controls. No differences were noted between dogs treated with prednisone and the controls with respect to size or extent of osteophytes, severity of synovial inflammation, morphologic changes of OA in articular cartilage, or in vitro synthesis of glycosaminoglycans. Increases in cartilage thickness in the OA knee were similar in the 2 groups, as were increases in H2O content and uronic acid concentration. Regardless of the interval between arthrotomy and start of treatment, the low dose of prednisone used here did not reduce the severity of OA in this model.