Single-injection coracoid infraclavicular brachial plexus block produces inconsistent anesthesia of the upper limb. In this study, we sought to determine the number of injections needed to provide a reasonably complete anesthesia of the upper limb with this approach. Seventy-five patients were randomly assigned to receive a coracoid block guided by nerve stimulator with 42 mL of 1.5% mepivacaine wish a single-injection (Group 1), dual-injection (Group 2), or triple-injection (Group 3) technique. No search for a specific motor response was performed in any group. Sensory and motor block was assessed 5 and 20 min after the end of the injection of local anesthetic. Significantly less complete anesthesia to pinprick in the distributions of the axillary, musculocutaneous, radial, ulnar, and medial cutaneous forearm nerves was found in Group 1 at 20 min. Significantly less complete paralysis for arm, wrist, and hand movements was found in Group 1 at 20 min. No significant difference was found between Groups 2 and 3. We conclude that dual and triple injection of local anesthetic guided by nerve stimulator increases the efficacy of coracoid block when compared with a single-injection technique.