OBJECTIVE: To estimate whether the cesarean delivery rate differs between women with severe preeclampsia who receive intrapartum epidural analgesia versus patient-controlled intravenous opioid analgesia. METHODS: Women with severe preeclampsia at at least 24 weeks' gestation were randomly assigned to receive either intrapartum epidural (n = 56) versus patient-controlled intravenous opioid analgesia (n = 60), and each was administered by a standardized protocol. The sample size was selected to have 80% power to detect at least a 50% difference in the predicted intergroup cesarean delivery rates. Data were analyzed by intent to treat. RESULTS: Selected maternal characteristics and neonatal outcomes were similar in the two groups. The cesarean delivery rates in the epidural group (18%) and the patient-controlled analgesia group (12%) were similar (P = .35). Women who received epidural analgesia were more likely to require ephedrine for the treatment of hypotension (9% versus 0%, P = .02), but their infants were less likely to require naloxone at delivery (9% versus 54%, P < .001). Epidural analgesia provided significantly better pain relief as determined by a visual analogue intrapartum pain score (P < .001) and a postpartum pain management survey (P = .002). CONCLUSION: Compared with patient-controlled intravenous opioid analgesia, intrapartum epidural analgesia did not significantly increase the cesarean delivery rate in women with severe preeclampsia at our level III center, and it provided superior pain relief. © 2002 by the American College of Obstetricians and Gynecologists.