Statistical analysis

The software “Review Manager” (RevMan, version 5.3; Nordic Cochrane Centre [Cochrane Collaboration], Copenhagen, Denmark; 2014) was used for the meta-analysis. We summarized pain scores at different time intervals with the mean difference (MD) and 95% confidence intervals (CI). Since there was dissimilarity between studies on the type of rescue analgesic, data for the same were summarized using standardized mean difference (SMD). When data were only presented graphically, Engauge Digitizer version 12.1 was used to extract numerical data. Median, range, and interquartile range data were converted into mean and standard deviation (SD) when required using the method of Wan et al. [20]. Data on adverse events were to be pooled using risk ratios (RR). The random-effects model was used for all the meta-analyses. Heterogeneity was assessed using the I² statistic. I² values of 25–50% represented low, values of 50–75% medium, and more than 75% represented substantial heterogeneity. Due to a limited number of studies in the meta-analysis (less than 10), funnel plots were not used to assess publication bias. We conducted a sensitivity analysis for the meta-analysis of pain scores. In the analysis, individual studies were excluded one at a time and the effect size was recalculated for the remaining studies in the meta-analysis software itself.