Recent studies have produced conflicting results regarding whether the addition of epidural fentanyl improves postoperative analgesia from epidural morphine. Therefore, we prospectively determined the dose-response relationship and the minimum effective combination dose of epidural morphine and fentanyl (fentanyl given after morphine) for posthysterectomy analgesia. We studied 120 patients undergoing radical abdominal hysterectomy. All patients had epidural lidocaine 1.5% with epinephrine (1:200,000) for surgical anesthesia followed by light general anesthesia with endotracheal intubation. They were assigned randomly into six groups according to the combination of each narcotic dose: morphine 2 mg, morphine 2 mg/fentanyl 50 micrograms, morphine 2 mg/fentanyl 100 micrograms, morphine 4 mg, morphine 4 mg/fentanyl 50 micrograms, and morphine 4 mg/fentanyl 100 micrograms. Morphine and fentanyl were given epidurally in a double-blind manner approximately 60 and 15 min, respectively, before the completion of surgery. For 2 mg of morphine, the addition of 50 or 100 micrograms of fentanyl improved pain relief during the first 6 h postoperatively (P < 0.05), provided longer duration of analgesia (P < 0.05), and required less analgesic supplement (P < 0.05), but did not alter the incidence of side effects. For 4 mg of morphine, the same conclusion was drawn, except that vomiting occurred more frequently with addition of 100 micrograms of fentanyl (P < 0.05). Among fentanyl groups, there was no significant difference in pain scores, duration of analgesia, and analgesic requirements. Therefore, we conclude that epidural fentanyl given after morphine improves early postoperative analgesia from epidural morphine, and the minimum effective combination dose is morphine 2 mg/fentanyl 50 micrograms for posthysterectomy surgery analgesia.