The transarterial approach to brachial plexus block is a well-established method of producing anesthesia of the upper extremity. However, it is associated with a failure rate of 20% to 30%. Failure may be secondary to the common use of a relatively long needle, which can penetrate the posterior wall of the sheath and result in inadvertent injection of the local anesthetic into the surrounding tissue. The purpose of this investigation was to compare success rates following transarterial brachial plexus block with a standard 22-gauge, 1 1/2-in, B bevel needle or a 26-gauge, 1/2-in needle. We enrolled 98 subjects scheduled for elective surgery at or below the elbow and randomized them into 2 groups. The control group received a transarterial axillary block with a standard 22-gauge, 1 1/2-in, B bevel needle, and the experimental group received a transarterial axillary block with a 26-gauge, 1/2-in needle. Success was defined as no discomfort at the time of incision. Success rates were compared using a chi 2 test, and a P value of less than .05 was considered significant. The overall success rate was significantly higher with the 26-gauge, 1/2-in needle (42/48 [88%]) than with the 22-gauge, 1 1/2-in needle (39/49 [69%]; P = .035).