Sixty-six RCTs (3,261 participants) were included in the analysis There were 19 trials of epidural analgesia, 15 trials of peripheral local anaesthetic, 7 trials of systemic NMDA receptor antagonists, 17 trials of systemic NSAIDs and 8 trials of systemic opioids.
The validity assessment showed a median quality score of 4 (range: 2 to 7); this was the same for studies that favoured pre-treatment or post-treatment. The proportion of trials favouring pre-treatment and post-treatment did not differ significantly (Mann-Whitney U-test, P=0.7). There was also no significant difference between higher quality trials (range: 5 to 7) and lower quality trials (range: 2 to 4) (P=0.44).
Not all trials measured all three outcome measures. Fifty trials measured pain intensity, 44 measured supplemental analgesic requirement and 28 trials measured time to first analgesic as one of the outcome measures. The authors reported that, whilst epidural analgesia showed improvements in all three outcome measures, local analgesic wound infiltration and the use of NSAIDs were effective only in improving supplemental analgesic use and time to first rescue dose. The results were equivocal for the use of systemic NMDA antagonist and opioids. Further details were supplied in the paper.
The following represents the combined results from each analgesic intervention type for all three outcome measures. Where the point estimate and confidence interval (CI) exceeded 0, the effect was deemed to be statistically significant.
Data from 19 epidural analgesic trials (reflecting 37 combined outcome variables) showed a combined ES of 0.38 (95% CI: 0.28, 0.47, P=< 1E-8), which suggested a highly significant difference in favour of pre-treatment. Fifteen trials of local anaesthetic wound infiltration (26 combined outcome variables) revealed a combined ES of 0.29 (95% CI: 0.17, 0.40, P=0.000001). Seven trials of NMDA antagonist use (16 combined outcome variables) showed a combined ES of 0.09 (95% CI: -0.03, 0.22, P=0.12). Seventeen trials involving the use of NSAIDs (30 combined outcome variables) revealed a combined ES of 0.39 (95% CI: 0.27, 0.48, P=< 1E-8), again reflecting a highly significant difference in favour of pre-treatment. Finally, in 8 trials of opioid use (13 combined outcome variables), the combined ES was -0.10 (95% CI: -0.26, 0.07, P=0.25).