BACKGROUND AND AIMS: Studies of the efficacy of gabapentin for treating alcohol use disorder (AUD) have yielded mixed findings. The aims of our study were to estimate gabapentin's effects on six alcohol-related outcomes, test potential moderators, examine publication bias and evaluate the quality of the studies.

METHODS: Meta-analysis of placebo-controlled randomized controlled trials (RCTs). Using PubMed and ClinicalTrials.gov, we selected RCTs of gabapentin's effects on alcohol consumption or a biochemical correlate of it, excluding studies limited to other primary outcomes or that combined gabapentin with other medications. We assessed study quality and used a random-effects model to analyze each outcome measure and the Egger regression test and funnel plots to assess publication bias.

RESULTS: We identified seven RCTs of gabapentin that met study criteria. The quality of the studies overall was good, and there was no evidence of publication bias. Four to seven studies contributed to the analysis of the six outcome measures. For all outcome measures the effect estimates were in a direction that favored gabapentin over placebo. However, only for percentage of heavy drinking days was there good evidence of a benefit (g = -0.64, 95% confidence interval = -1.22 to -0.06).

CONCLUSIONS: Although gabapentin appears to be more efficacious than placebo in treating AUD, the only measure on which the analysis clearly favors the active medication is percentage of heavy drinking days. Additional studies are needed to define more clearly the role of gabapentin in AUD treatment.