No non-invasive quantitative method is currently available for measuring the amount of regional skeletal muscle necrosis in humans. A threshold method using 99Tcm pyrophosphate (PPi) with single photon emission computed tomography (SPECT) has been reported to quantify muscle necrosis successfully in a canine model. This method has been modified to measure muscle necrosis in the lower extremities of human patients. Modifications include adding background subtraction and using the contralateral unaffected lower extremity as background region. A region of interest was also manually drawn around the affected lower extremity muscle to exclude the tibia. In order to identify the group of patients with no significant necrosis and with diffuse hyperaemia, nine control subjects were studied. The variance of the counts within the pixels of the lower extremities was 132 (S.D. = 27). A variance of greater than 159 was considered as being significant for necrosis. Thirteen patients were studied. Six patients had significant muscle necrosis (462 ml, S.D. = 280 ml). All except one patient (who had a small amount of muscle necrosis) developed foot drop. Seven patients had no significant necrosis. All patients had normal ankle dorsiflexion. It is possible to quantify regional muscle necrosis using PPi with SPECT. Volume of necrosis determined by this method can also predict clinical outcome.