This prospective, randomized, blinded study tested the hypothesis that ultrasound guidance can shorten the onset time of axillary brachial plexus block as compared with nerve stimulation guidance when using a multiple injection technique. Sixty American Society of Anesthesiology physical status I-III patients receiving axillary brachial plexus block with 20 ml ropivacaine, 0.75%, using a multiple injection technique, were randomly allocated to receive either nerve stimulation (group NS, n = 30), or ultrasound guidance (group US, n = 30) for nerve location. A blinded observer recorded the onset of sensory and motor blocks, the need for general anesthesia (failed block) or greater than 100 microg fentanyl (insufficient block) to complete surgery, procedure-related pain, success rate, and patient satisfaction. The median (range) number of needle passes was 4 (3-8) in group US and 8 (5-13) in group NS (P = 0.002). The onset of sensory block was shorter in group US (14 +/- 6 min) than in group NS (18 +/- 6 min) (P = 0.01), whereas no differences were observed in onset of motor block (24 +/- 8 min in group US and 25 +/- 8 min in group NS; P = 0.33) and readiness to surgery (26 +/- 8 min in group US and 28 +/- 9 min in group NS; P = 0.48). No failed block was reported in either group. Insufficient block was observed in 1 patient (3%) of group US and 2 patients (6%) of group NS (P = 0.61). Procedure-related pain was reported in 6 patients (20%) of group US and 14 patients (48%) of group NS (P = 0.028); patient acceptance was similarly good in the two groups. Multiple injection axillary block with ultrasound guidance provided similar success rates and comparable incidence of complication as compared with nerve stimulation guidance.