BACKGROUND Morphine-6-glucuronide (M-6-G), a major metabolite of morphine, is reported to be more potent than morphine when administered intrathecally; however, its efficiency remains under debate when administered intravenously. This study was designed to assess the analgesic efficiency of intravenous M-6-G for the treatment of acute postoperative pain. METHODS After informed consent was obtained, 37 adults (American Society of Anesthesiologists physical status I-II) who were scheduled for elective open knee surgery were enrolled in the study. General anesthesia was induced with thiopental, alfentanil, and vecuronium and was maintained with a mixture of nitrous oxide/isoflurane and bolus doses of alfentanil. At skin closure, patients were randomized into three groups: (1) morphine group (n = 13), which received morphine 0.15 mg/kg; (2) M-6-G group (n = 12), which received M-6-G 0.1 mg/kg; and (3) placebo group (n = 12), which received saline. At the time of extubation, plasma concentration of morphine and M-6-G was measured. Postoperative analgesic efficiency was assessed by the cumulative dose of morphine delivered by patient-controlled analgesia. Opioid-related side effects were also evaluated. RESULTS No difference was noted in patient characteristics and opioid-related side effects. Morphine requirements (mean +/- SD) during the first 24 h in the M-6-G group (41+/-9 mg) and the placebo group (49+/-8 mg) were significantly greater (P<0.05) compared with the morphine group (29+/-8 mg). CONCLUSIONS A single intravenous bolus dose of M-6-G was found to be ineffective in the treatment of acute postoperative pain. This might be related to the low permeability of the blood-brain barrier for M-6-G.