Postoperative regional analgesia for total knee replacement can provide excellent pain control and speedy rehabilitation compared with systemic opioid analgesia but the optimal technique to provide best analgesia with minimal adverse effects remains unclear. We carried out an observer-blinded randomized trial of side-directed epidural infusion with lumbar plexus infusion after total knee arthroplasty.

Sixty patients scheduled for total knee replacement were randomized to receive epidural or lumbar plexus infusions of levobupivacaine and clonidine. Pain, sensory and motor block were assessed at 0, 6, 24 and 48 h postoperatively. Range of knee movement and mobility were assessed on the first and second postoperative days.

No significant differences were detected between the epidural and lumbar plexus groups in 24-h pain scores at rest (median visual analogue scale, 30 mm (interquartile range, 10–45) vs. 39 mm (17–51), P = 0.286), and on movement (48 mm (20–66) vs. 60 mm (47–81), P = 0.068). The only statistically significant difference in pain scores in favour of the epidural groups was at 6 h postoperatively (P < 0.001). Median morphine usage in the epidural group was 0 mg (interquartile range, 0–35) compared with 14.5 mg (0–44) in the lumbar plexus group (P = 0.33). Range of movement (epidural: median 70° (interquartile range, 58–75) vs. lumbar plexus: 70° (50–75), P = 0.79) or mobility was similar between groups. Adverse effects were also similar between groups, apart from a higher incidence of bladder catheterization in the epidural group (37.9% vs. 12.5%, P = 0.04). Conclusions: Lumbar plexus infusion is a reasonable alternative to epidural anaesthesia for total knee arthroplasty.