We evaluated the efficacy and safety of preoperative intravenous administration of dezocine for comfortable birth by cesarean section under epidural anesthesia. Sixty primigravida women with full-term singletons underwent elective cesarean section, and were randomly divided into three groups. Patients in groups A and B were intravenously injected with 5 and 10 mg of dezocine, 10 min before skin incision, whereas patients in group C were intravenously injected with saline. We recorded data on visceral traction responses and intraoperative adverse reactions, such as nausea and vomiting. While the neonate was being delivered, the umbilical arterial and venous blood gas values were determined. The Apgar scores at 1, 5, and 10 min after delivery, as well as the Neurologic Adaptive and Capacity Score (NACS) at 15 min, 2 h, and 24 h, were recorded. The "fineness rate" required to relieve the traction reaction in group B was higher than that in group C (P < 0.05). The intraoperative Modified Observer's Assessment of Alertness/Sedation (MOAA/S) scores of groups A and B were lower than that of group C, and the time period that the intraoperative MOAA/S score was <= 4 points was longer in group B (P < 0.05). The umbilical blood gas values, as well as the NACS and Apgar scores, were not statistically significantly different among the groups at any time points. The preoperative administration of 10 mg of dezocine further improved the comfort level of women during cesarean section under epidural anesthesia. Adverse maternal and neonatal effects, such as respiratory depression, were not observed.