For sciatic nerve blockade, no study has defined the optimal volume of local anesthetic required to block the nerve. The current, prospective, randomized investigation was designed to find a minimum volume of 1.5% mepivacaine required to block the sciatic nerve using the subgluteal and posterior popliteal approaches. A total of 56 patients undergoing foot surgery were randomly assigned to receive sciatic nerve block by means of a posterior subgluteal (group subgluteal, n = 28) or a posterior popliteal (group popliteal, n = 28) approaches. All blocks were performed with the use a nerve stimulator (stimulating frequency, 2 Hz, intensity 1.5-0.5 mA) and a perineural stimulating catheter. In all patients, plantar flexion of the foot was elicited at <0.5 mA, to maintain consistency among groups. The volume of local anesthetic used in each patient was based on the modified Dixon's up-and-down method. Complete anesthesia was defined as complete loss of pinprick sensation in the sciatic nerve distribution with concomitant inability to perform plantar or dorsal flexion of the foot 20 min after injection. The mean volume of local anesthetic required to block the sciatic nerve was 12 +/- 3 mL in the subgluteal group and 20 +/- 3 mL in the popliteal group (P < 0.05). The ED95 for adequate block of the sciatic nerve was 17 mL in the subgluteal group and 30 mL in the popliteal group. The authors conclude that a larger volume of local anesthetic is necessary to block the sciatic nerve at a more distal site (popliteal approach) as compared with a more proximal level (subgluteal approach).