Study Objective: To determine whether treatment with ondansetron, a new antiemetic drug, affects nondepolarizing neuromuscular blockade. Design: Randomized, double-blind, prospective study. Setting: Operating room at a university medical center. Patients: 30 ASA physical status I and II patients scheduled for elective surgery. Interventions: After the induction of anesthesia with midazolam 2 to 4 mg/kg, sodium thiopental 6 to 8 mg/kg, and fentanyl 4 to 8 μg/kg, the ulnar nerve was stimulated at the wrist through subcutaneous needle electrodes at a frequency of 0.15 Hz. The response to stimulation was measured and recorded with a force-displacement transducer applied to the thumb. Patients were randomized to one of three treatment groups. A steady baseline to ulnar nerve stimulation with nitrous oxide-oxygen-opioid-thiopental anesthesia was established. The first study group (Group 1) received a placebo, the second group (Group 2) received 8 mg of ondansetron, and the third group (Group 3) received 16 mg of ondansetron as an intravenous infusion over 5 minutes. Patients were then given incremental doses of atracurium 0.05 mg/kg at 3-minute intervals to establish approximately 95% twitch inhibition so as to construct a dose-response curve. An atracurium infusion was then begun to maintain a constant degree of neuromuscular blockade. At the end of surgery, patients were allowed to recover spontaneously, or pharmacologic antagonism of residual neuromuscular blockade was achieved with neostigmine 0.05 mg/kg and glycopyrrolate 0.01 mg/kg. Mechanomyographic response to train-of four stimuli (2 Hz for 2 seconds) every 20 seconds was monitored during the atracurium infusion and recovery from neuromuscular blockade. Measurements and Main Results: Log dose-response curves were determined for the study groups and compared using analysis of variance (ANOVA). The 50%, 75%, and 95% effective doses (ED 50, ED 75, and ED 95) were calculated from the equation describing the log dose-response. Maintenance infusion rates were determined, and the neostigmine-accelerated recovery index of 25% to 75% was measured for each group. The results were compared using ANOVA. There were no significant differences among the treatment groups with respect to maintenance infusion rate (7.8 ± 1.8 μg/kg/min for Group 1, 7.7 ± 2.5 μg/kg/min for Group 2, and 7.3 ± 2.3 μg/kg/min for Group 3) or neostigmine-accelerated recovery interval of 25% to 75% (4.5 ± 2.3 minutes, 4.4 ± 3.1 minutes, 6.6 ± 3.9 minutes in Groups 1, 2, and 3, respectively). The log dose-response data for Groups 1, 2, and 3 did not differ significantly ( p = 0.068), and the calculated ED95 in each treatment group demonstrated no dose-related change (0.254 ± 0.022, 0.279 ± 0.033, and 0.240 ± 0.022 for Groups 1, 2, and 3, respectively). Conclusions: Ondansetron is an antiemetic drug that can be used in the perioperative period without concern for potentiation of nondepolarizing neuromuscular blockade, change in atracurium maintenance dose, or change in rate of neostigmine-induced recovery from neuromuscular blockade with atracurium.