Studies suggest that acupuncture is more effective when induced before the induction of general anesthesia than afterwards. We tested the hypothesis that electro-acupuncture initiated 30 min before the induction reduces anesthetic requirement more than acupuncture initiated after the induction. Seven volunteers were each anesthetized with desflurane on 3 study days. Needles were inserted percutaneously at four acupuncture points thought to produce analgesia in the upper abdominal area and provide generalized sedative and analgesic effects: Zusanli (St36), Sanyinjiao (Sp6), Liangqiu (Sp34), and Hegu (LI4). Needles were stimulated at 2 Hz and 10 Hz, with frequencies alternating at 2-s intervals. On Preinduction day, electro-acupuncture was started 30 min before the induction of anesthesia and maintained throughout the study. On At-induction day, needles were positioned before the induction of anesthesia, but electro-acupuncture stimulation was not initiated until after the induction. On Control day, electrodes were positioned near the acupoints, but needles were not inserted. Noxious electrical stimulation was administered via 25-gauge needles on the upper abdomen (70 mA; 100 Hz; 10 s). The desflurane concentration was increased 0.5% when movement occurred and decreased 0.5% when it did not. These up-and-down sequences continued until volunteers crossed from movement to no movement four times. The P(50) of logistic regression identified desflurane requirement. Desflurane requirement was similar on the Control (mean +/- sd; 5.2% +/- 0.6%), Preinduction (5.0% +/- 0.8%), and At-induction (4.7% +/- 0.3%; P = 0.125) days. This type of acupuncture is thus unlikely to facilitate general anesthesia or decrease the requirement for anesthetic drugs.