BACKGROUND: Combined spinal-epidural anesthesia (CSEA) for cesarean section has gained an increasing interest as it combines a reliability of spinal anesthesia and the flexibility of epidural anesthesia. The aim of this study is to compare the surgical analgesia and the frequency of side effects for cesarean section produced by CSEA using subarachnoid fentanyl or placbo. METHODS: The study was performed in a randomized, double-blined fashion in 40 (20 per group) healthy, full-term parturients presenting for elective cesarean section. We compared the effects of intrathecal fentanyl (20 microgram), and placebo when administered together with 0.5% hyperbaric bupivacaine 7 mg in combined spinal-epidural anesthesia (CSEA) for cesarean section. Patients' anesthetic levels, vital signs and intraoperative pain were recorded. If anesthetic level achieved by intrathecal injection was not sufficient for cesarean section (T4), additional 2% lidocaine 2 ml per segment was administered epidurally. Patients were asked to rate their severity of pain on a visual analog scale (VAS) score intraoperatively and intravenous fentanyl was administered if the patient experienced intraoperative discomfort. The quality and side effects of anesthsia and neonatal Apgar scores were compared between two groups. RESULTS: The number (percent) of patients achieved sensory block level above T4 by subarachnoid injection alone was significantly higher in the fentanyl group (17/18, 94.4%) than the control grop (10/16, 62.5%). The dose of epidural lidocaine was significantly less in the fentanyl group (p<0.05). The frequency of intraoperative pain was significantly less in the fentanyl group (17%) than in the control group (50%). CONCLUSION: We conclude that adding fentanyl into subarachnoid injection in CSEA for cesarean section significantly decreases the additional epidural local anesthetics and intraoperative pain.