Background: Previous studies have indicated that statins can reduce the severity of depressive symptoms. However, the optimal choice of statin remains unclear. Therefore, we conducted a network meta-analysis to determine the optimal statin for treating depression.
Method: We performed a pairwise and network meta-analysis by searching the PubMed, Embase, and Cochrane Library databases on October 29th, 2020. Eligible studies were randomized controlled trials that reported on changes in depressive symptoms. The Cochrane Collaboration tool was used to assess risk of bias. We tested for possible inconsistency globally by using a χ2-test and locally by calculating inconsistency factors for each comparison in closed loops. The ranking probabilities of being at each possible rank for each intervention were estimated. Comparison-adjusted funnel plots were obtained to assess publication bias. Sensitivity analysis was also performed.
Results: We identified 13 studies that matched our inclusion criteria. The risks of bias were mostly low. None of the global or local tests found significance. Compared with placebo, atorvastatin significantly reduced the severity of depressive symptoms (mean difference -3.46, 95% confidence interval -5.26 to -1.67). Atorvastatin had the first and second rank with probabilities of 44.9% and 39.0%, respectively. Comparison-adjusted funnel plots revealed no significant publication bias.
Limitations: Low similarity of included studies and a relative large treatment effect of a single study were observed.
Conclusions: In this first network meta-analysis, atorvastatin, with high intensity and a lipophilic effect, was identified as the optimal choice of statin for treating depression.
Keywords: depression; network meta-analysis; randomized controlled trial; statin.
Copyright © 2021. Published by Elsevier B.V.