BACKGROUND: A light plane of general anesthesia is chosen for fetal safety during a cesarean section. Therefore an experience of wakefulness and pain perception is not infrequent and can be distressful to patients. The electroencephalogram-derived bispectral index (BIS) is a promising new method to assess anesthetic adequacy. This study was designed to assess the adequacy of general anesthesia and to evaluate the usefulness of the BIS in monitoring the awareness during a cesarean section. METHODS: Twenty one parturients undergoing general anesthesia for an elective caesarean section were examined. Anesthesia was induced with 4 mg/kg thiopental and 1 mg/kg succinylcholine, and then followed with a mixture of 50% N2O, 50% O2 and 0.8% enflurane at a flow of 4 L/min. Before administration of succinylcholine, a tourniquet was applied to the free arm and inflated to 250 mmHg. Responsiveness to verbal commands using the Tunstall isolated forearm technique was detected and the BIS was monitored throughout the study period. RESULTS: The incidence of responsiveness to verbal commands were 33% at the time of skin incision and fetal delivery respectively. BIS values of response to commands were significantly higher than that of no respose at both time points (p < 0.05). Fisher's exact test suggested that the BIS value less than 75 was related to unresponsiveness to verbal commands at time of skin incision (p < 0.05) and the BIS less than 85 at fetal delivery (p < 0.05). CONCLSIONS: These results suggest that adequate anesthesia is not provided with a mixture of 50% N2O, 50% O2 and 0.8% enflurane at a flow of 4 L/min at the time of skin incision and fetal delivery. In addition BIS could be a predictor of awareness during cesarean section under general anesthesia.