To define the skeletal muscle abnormalities in patients undergoing exercise deconditioning and evaluate the metabolic effect of propionyl-L-carnitine (PLC), muscle biopsies were obtained from 28 patients with effort angina and 31 control subjects. Coronary artery disease patients received either placebo (n = 12), PLC (1.5 g i.v. followed by infusion of 1 mg/kg/min for 30 min, n = 10), or L-carnitine (1 g i.v. followed by infusion of 0.65 mg/kg/min for 30 min, n = 6) for 2 days. Exercise deconditioned patients treated with placebo showed normal muscle content of total carnitine and glycogen, and decrease in percentage of type 1 fibers (P < 0.01) and in the activity of citrate synthase (P < 0.05), succinate dehydrogenase (P < 0.05), and cytochrome oxidase (P < 0.05), as compared to controls. Both PLC and L-carnitine did not modify muscle fiber composition or enzyme activities, but significantly increased muscle levels of total carnitine by 42% and 31%, respectively (P < 0.05). Moreover, PLC significantly increased glycogen muscle content (P < 0.01), while the equimolar dose of L-carnitine did not. This effect, probably due to the anaplerotic activity of the propionic group of PLC, suggests that this drug may be effective in improving energy metabolism of muscles with impaired oxidative capacity.