A Prospective, Randomized, Double-blind Comparison of Epidural and Intravenous Sufentanil Infusions

BACKGROUND: The site of action (spinal vs. central) of epidurally administered lipid-soluble opioids has been the subject of controversy. We compared the efficacy, plasma concentration and side effects of epidural and intravenously administered sufentanil for postoperative pain relief. METHODS: Using a double-blind, prospective design, 50 patients scheduled for intraabdominal operations during combined epidural-general anesthesia were randomized into one of two groups. Patients in group 1 (n = 24) received a 1-microgram/ml sufentanil infusion epidurally at 0.2 microgram.kg-1.h-1 and a saline infusion intravenously at the same rate. Patients in group 2 (n = 26) received a 1-microgram/ml sufentanil infusion intravenously at 0.2 microgram.kg-1.h-1 and a saline infusion epidurally at the same rate. Intravenous morphine sulfate was available in 2-mg increments to all patients in the postanesthesia care unit until visual analogue scale (0-100 mm) pain score was < or = 30. Then, a patient-controlled intravenous pump providing morphine on demand (1 mg with a 10-min lockout) was begun. Blood samples were drawn for sufentanil plasma levels and patients were assessed for pain, sedation and nausea for the 48 h after commencement of the infusions. RESULTS: Similar visual analogue pain, sedation, and nausea scores were found between the patients in the two groups. No differences were found in supplemental morphine requirements and plasma sufentanil concentrations between the patients in the two groups. A higher incidence of excessive sedation requiring infusion decrease was infusion decrease was found in the intravenous group (six vs. one, P < 0.05). CONCLUSIONS: Many clinical similarities were found when epidural and intravenous sufentanil infusions were compared. The higher incidence of excessive sedation in the patients receiving intravenous sufentanil could not be explained on the basis of plasma sufentanil concentrations alone. This study indicates that little clinical difference exists between epidural and intravenous administration of sufentanil.