Debate has proliferated as to the true site of action of opioids when placed in the epidural space. The aim of this study was to compare the analgesic effects of a bolus of diamorphine given by the epidural or intramuscular route.

Sixty patients having elective primary total knee replacements were recruited and randomized to receive epidural or intramuscular diamorphine. A lumbar epidural catheter was sited and 10 mL of bupivacaine 0.5% wt vol−1 was injected. Patients subsequently received diamorphine 5 mg into the epidural space or as an intramuscular injection. Patient-controlled analgesia with intravenous morphine was used for postoperative analgesia. The primary outcome measures included time to first patient-controlled analgesia use and total morphine consumption in 24 h. Secondary end-points considered possible treatment complications.

All primary end-points showed significant differences in favour of epidural diamorphine. Medians for times to first patient-controlled analgesia use and total 24 h morphine requirements were significantly different (P < 0.001) at 418 vs. 198 min and 11 vs. 39 mg, respectively. There were no significant differences in secondary end-points.

This study has shown the superior analgesic efficacy of epidural diamorphine when compared to intramuscular injection.