We have investigated the effects of stimulation frequency and pulse duration on fatigue and energy metabolism in rat gastrocnemius muscle during a single bout of neuromuscular electrical stimulation (NMES). Electrical pulses were delivered at 100 Hz (1-ms pulse duration) and 20 Hz (5-ms pulse duration) for the high (HF) and low (LF) frequency protocols, respectively. As a standardization procedure, the averaged stimulation intensity, the averaged total charge, the initial peak torque, the duty cycle, the contraction duration and the torque-time integral were similar in both protocols. Fatigue was assessed using two testing trains delivered at a frequency of 100 Hz and 20 Hz before and after each protocol. Metabolic changes were investigated in vivo using 31P-magnetic resonance spectroscopy (31P-MRS) and in vitro in freeze-clamped muscles. Both LF and HF NMES protocols induced the same decrease in testing trains and metabolic changes. We conclude that, under carefully controlled and comparable conditions, the use of low stimulation frequency and long pulse duration do not minimize the occurrence of muscle fatigue or affect the corresponding stimulation-induced metabolic changes so that this combination of stimulation parameters would not be adequate in the context of rehabilitation.