Eight patients were studied. All patients were scheduled to undergo hysterectomy with a Pfannenstiel incision. The patients were excluded if they had any pre-existing cardiopulmonary, hepatic, renal or endocrine disease. After identifying the epidural space using a loss of resistance technique, a 20-gauge epidural catheter was placed between the third and fourth lumbar vertebrae. All patients received an initial dose of 10 ml of bupivacaine 0.33%. The patients then underwent a general anesthetic. Induction was accomplished with 4 mg/kg of thiopental and 1 mg/kg of succinylcholine. Relaxation was induced with 0.08 mg/kg of vecuronium and general anesthesia was maintained with enflurane and nitrous oxide. Five minutes after induction, 5 ml bupivacaine 0.33% was again administered via the epidural catheter and at 10 minutes after induction, an additional 3-5 ml of bupivacaine was administered. All patients then received between 1/2 and 1/3 of the initial dose (15 ml) at one-hour intervals until the end of surgery. At the time of closure, each patient was given 2 mg morphine epidurally. A continuous epidural infusion was then initiated. The patients were randomly allocated into one of four groups. Group 1 received a 80 µg morphine and 2 ml of bupivacaine 0.125%/hr. Group 2 received the same mixture with the addition 0.083 µg/kg/hr of naloxone. Group 3 and 4 were identical to group 2 except that the naloxone infusion rate was 0.125 µg/kg/hr for group 3 and 0.167 µg/kg/hr for Group 4. Visual analog scales were employed to assess postoperative pain at 2, 4, 8, 16, 32 and 48 hours postoperatively. Nausea, itching, somnolence and respiratory depression were assessed using evaluation scores. All assessments were carried out by anesthesiologists who had not been involved in the care of the patients and who were blinded to the group assignment.

All four groups experienced good pain control with the highest VAS scores at four hours after the end of surgery. Group 4, the highest naloxone dose, had lower VAS scores compared with group 1 at 8, 16 and 32 hours postoperatively (p < 0.05). Groups 3 and 4 had less itching than did groups 1 and 2 with the difference persisting throughout most of the study (p < 0.05 at 8, 16 and 32 hours for Groups 3 and 4 versus group 1). Groups 3 and 4 reported less nausea at 16 hours than did Group 1 (p < 0.05). The incidence of vomiting or retching at some point in the postoperative course was 40%, 25%, 15% and 5% for Groups 1-4, respectively. There was no difficulty with alertness or respiratory depression.

The authors concluded that epidural naloxone reduced morphine induced side effects in a dose-dependent fashion without reversal of the analgesic effect.

All physicians involved in providing acute perioperative pain control using epidural analgesia are confronted with a significant number of side effects. These side effects may limit the total dose of opiate administered and may, in some cases, interfere with providing adequate pain control. The search for a consistently effective means of controlling intraspinal opiate side effects without reversal of analgesia or associated sedation has been ongoing. These authors have demonstrated an effective means of controlling the side effects associated with the administration of intraspinal opiates. The ability to incorporate this medication within the epidural infusion is appealing from an ease-of-use standpoint. The major concern in reviewing this study is that there is little experience with naloxone in the epidural space. The absence of any form of neurotoxicity testing for epidural naloxone is of some concern, however, there were no side effects reported in this study of 80 patients and the concept remains alluring. Further demonstration of the success of this technique and appropriate neurotoxicity testing may encourage more wide-spread use.