Thirty-one trials (n=1,523 patients) were included in the review; four were of both opioids and muscle relaxants. Thirteen trials had Jadad scores of two points, four had three points, 10 had four points, and four had five points.
Local anaesthetics: Nine trials assessed local anaesthetics (n=516). There were no differences between lidocaine, prilocaine and ropivacaine in the onset of sensory block (seven trials), and the onset of motor block (six trials). Three trials found significant postoperative benefits with ropivacaine, compared with lidocaine, with lower pain scores, and a longer time to first analgesia, but there were no differences in one trial. There were no significant side-effects. Ropivacaine caused delays in sensory recovery (five trials) and motor recovery (three trials). Similar tourniquet tolerance times were found across all local anaesthetics.
Additional opioids: Sixteen trials assessed opioids (n=761). Compared with no additional opioid, there were no clinically significant benefits with morphine (two trials; n=57), fentanyl (seven trials; n=345), and meperidine (one trial; n=20) and there were no significant increases in side-effects. Sufentanil (two trials; n=125) appeared to provide significantly faster onset of sensory block. Tramadol provided faster onset of sensory block and tourniquet tolerance, but the postoperative benefits were not consistent and the risk of minor side-effects was increased.
Additional muscle relaxants: Ten trials assessed muscle relaxants (n=494). Two out of six trials reported faster onset of sensory block and motor block when muscle relaxants were added to local anaesthetics. There were no significant changes in tourniquet tolerance and intra-operative analgesia. Two trials found that less postoperative analgesia was administered. One trial reported signs of toxicity and delays in motor recovery with mivacurium in addition to local anaesthetic.