Twenty-eight trials (n=1,538) were included in the review. Sample sizes ranged from 20 to 262 patients. Three trials scored 5 on the Jadad scale, four trials scored 4 and seven scored 3; the other 14 trials scored 2 or less.
One trial (n=262) assessed mortality. It compared epidural anaesthesia and continuous epidural analgesia with general anaesthesia and intravenous opioid. There were no differences in the number of deaths over an eight-week postoperative period (one death in each group).
For cardiovascular morbidity (nine trials), there was no difference between regional anaesthesia and general anaesthesia in terms of postoperative myocardial infarction and pulmonary embolism (three trials). Postoperative hypotension was higher following epidural analgesia than with systemic or with other methods of regional analgesia (three trials).
No differences were reported between regional anaesthesia and general anaesthesia in terms of incidence of DVT (four trials), perioperative blood loss or transfusion requirements (five trials), duration of surgery (13 trials) and cognitive function (two trials).
Pain scores and/or morphine consumption was reduced in 21 trials; the effects were long-lasting (up to 10 days) in certain cases. Specifically, epidural analgesia, single injection femoral nerve block (FNB) with or without sciatic nerve block, continuous catheter-based femoral nerve block and continuous psoas plexus block were found to be more effective than systemic analgesia (18 trials).
Ten trials reported no differences in adverse effects between general anaesthesia and regional anaesthesia. However, opioid-related adverse effects (postoperative nausea and vomiting) were reduced in the regional anaesthesia groups following FNB plus obturator block, FNB plus sciatic block and continuous FNB (six trials).
Three trials indicated that length of hospital stay could be reduced by between one and 13 days. Six trials reported that rehabilitation could be improved with regional anaesthesia, specifically in terms of range of motion and ambulation following FNB or continuous FNB.