Post-stroke depression (PSD) constitutes an important complication of stroke, leading to great disability as well as increased mortality. Since which treatment for PSD should be preferred are still matters of controversy, we aimed to compare and rank these treatments.

We did a network meta-analysis to incorporate both direct and indirect evidence from relevant trials by Bayesian random effects model. We searched PubMed, the Cochrane Library Central Register of Controlled Trials, Scopus, Embase for randomized controlled trials of different PSD treatments. The primary outcomes were efficacy and tolerability. We assessed the quality of evidence using the GRADE framework.

From 1347 citations, 27 randomised trials with a total of 1620 participants were included in this network meta-analysis. In terms of primary outcome, only for reboxetione (standardised mean difference [SMD] −12.84, 95% credible interval [CrI] −23.13 to −2.65) and nortriptyline (SMD −7.95, 95% CrI −14.85 to −1.75) enough evidence existed to support superiority compared with placebo. No significant difference was observed in terms tolerability. Considering patient response rate, repetitive transcranial magnetic stimulation (rTMS) was statistically more effective than sertraline plus nimodipine (Relative risk [RR] 5.53, 95% CrI 1.36 to 23.86) and fluoxetine (RR 10.74, 95% CrI 3.55 to 35.97)

Compared with placebo, reboxetione and nortriptyline offered a clear advantage for PSD patients. rTMS is probably the best option to consider in addition of pharmacological treatment. Nevertheless, doctors need to consider our results together with all known safety and economy information when selecting the strategy for individual patients.

The author has not supplied his declaration of competing interest.