Get access

Comparison of SSRIs and SNRIs in major depressive disorder: a meta-analysis of head-to-head randomized clinical trials


Márcio Machado, Research Associate, Toronto Health Economics and Technology Assessment Collaborative (THETA), University of Toronto, 144 College Street, Toronto, ON M5S 3M2, Canada. Tel.: +416 946 3700; fax: +416 946 3719; e-mail:


Context:  Controversy exists whether serotonin–norepinephrine reuptake inhibitors (SNRIs) have improved efficacy compared with selective serotonin reuptake inhibitors (SSRIs).

Objective:  To compare clinical outcomes of adults treated with SSRIs or SNRIs for major depressive disorder (MDD) under ideal clinical condition, research design, and outcome measure.

Data sources:  Electronic databases searched were Medline, Embase and Cochrane Library from inception to July 2007.

Study selection:  Included studies were those head-to-head randomized trials comparing remission (HAMD-17 ≤7–8, MADRS ≤10–12) after 8–12 weeks of therapeutic doses of SSRIs or SNRIs in patients diagnosed with MDD were targeted for analysis. Reviews, letters, commentaries, economic studies, etc. were excluded. Studies were reviewed by two independent researchers. Where disagreements occurred in study selection, a consensus approach was used.

Data extraction and analysis:  Targeted outcome data included number of patients achieving remission, withdrawing from therapy due to lack of efficacy (LoE) and/or adverse drug reactions (ADRs), and total patients in trial. A random effects model combined intent-to-treat (ITT) and per-protocol (PP) odds ratio (OR), and remission and dropout rates. Chi-square assessed heterogeneity. Quality assessment was done using Downs-Black checklist.

Results:  Thirty-three studies were identified; 18 were rejected (patients had co-morbidities in 7, outcomes differed in 5, different follow-up in 3, and three reviews). Fifteen head-to-head trials of 3094 patients, average age was 41·9 ± 11·9 years (for SNRIs) and 41·6 ± 12·1 years (for SSRIs), P = 0·941. All analyses displayed non-heterogeneity (P > 0·05). The OR (under ITT) was 1·27 (1·06–1·52 95% CI) favoring SNRIs. Meta-analytic remission rates were 48·5 ± 3·2% and 41·9 ± 4·2% for SNRIs and SSRIs, respectively. The meta-analytic difference in remission rates between drugs was 5·7% (P = 0·007). Dropout rates due to ADRs were higher with SNRIs than SSRIs (3·2% difference, P < 0·001). Dropout rates due to LoE were non-significant between studied groups (P > 0·05).

Conclusions:  Serotonin and norepinephrine reuptake inhibitors showed statistical but not clinical significance when compared with SSRIs in treating MDD.