The effect of dietary starch and fat content on serum creatine kinase (CK) activity and substrate availability was evaluated in 4 mares of Quarter Horse-related breeds with polysaccharide storage myopathy (PSSM). Four isocaloric diets ranging in digestible energy (DE) from 21.2% (diet A), 14.8% (B), 8.4% (C), to 3.9% (D) for starch, and 7.2% DE (diet A), 9.9% (B), to 12.7% DE (diet C and D) for fat were fed for 6-week periods (4 weeks with exercise) using a 4×4 Latin square design. Postprandial glucose and insulin responses were measured, and 4 hours postexercise, serum CK activity, glucose, insulin, free fatty acids (FFA), and β-hydroxybutyrate (β-HBA) were analyzed. Glycogen, glucose-6-phosphate, citrate synthase, 3-hydroxy-acyl-CoA dehydrogenase, lactate dehydrogenase as well as abnormal polysaccharide and lipid content were measured in middle gluteal muscle samples. Postprandial insulin and glucose response was higher for diet A versus D. Log CK activity was higher with diets A, B, and C versus D. Daily insulin was higher and FFA lower on diet A versus B, C, and D, whereas glucose varied only slightly with diet. Muscle oxidative capacity and lipid stores were low in PSSM horses and muscle glycogen and abnormal polysaccharide content high on both diets A and D. Individual variation occurred in the response of PSSM horses to diets differing in starch and fat content. However, for those horses with clinical manifestations of PSSM, a diet with <5% DE starch and >12% DE fat can reduce exertional rhabdomyolysis, potentially by increasing availability of FFA for muscle metabolism.