A randomised controlled study was undertaken to assess the analgesic efficacy of continuous lumbar plexus block for the first 48 hours after total knee replacement surgery. Boluses of 0.5% bupivacaine with adrenaline 1 in 200,000 (0.3 ml/kg) were administered through a cannula inserted into the neurovascular sheath of the femoral nerve. Thirteen patients who received this block required significantly less morphine than a control group of 16 patients. Pain scores were similar and there were no complications related to this technique.