Thirty-eight RCTs (n=11,918 participants) were included in the review. All trials were double-blind, but few trials described the method of randomisation, blinding or allocation concealment.
Compared with placebo, pregabalin was associated with a statistically significantly greater risk of twenty adverse events (full list of the adverse events results were available in a separate online appendix, see Additional Data URL). The highest risk was for balance disorder (RR 8.22, 95% CI 1.75 to 38.97), followed by euphoria (RR 6.18, 95% CI 2.76 to 13.87), incoordination (RR 4.88, 95% CI 2.18 to 10.95), ataxia (RR 4.77, 95% CI 2.77 to 8.20), and oedema (RR 4.63, 95% CI 2.15 to 9.95). There was no statistically significantly increased risk of serious adverse events with pregabalin. The number needed to harm was reported as a graph, generally showing a dose-dependent pattern in the onset of adverse events, which was especially evident for balance disorder, amblyopia, confusional state, disturbance in attention, asthenia, and constipation.
There was no evidence of statistical heterogeneity in the main analyses, but there was some evidence of statistical heterogeneity in the dose-dependent analyses.