Meta-analysis of major depressive disorder relapse and recurrence with second-generation antidepressants.
Level 1 - systematic review of randomized trials
Systematic review and meta-analysis of randomized trials
PubMed 18832497 · doi:10.1176/ps.2008.59.10.1121
What was done
A systematic review and meta-analysis evaluated the efficacy of second-generation antidepressants for preventing major depressive disorder relapse (continuation phase) and recurrence (maintenance phase). Databases (MEDLINE, EMBASE, PsycINFO, Cochrane Library, International Pharmaceutical Abstracts) were searched from January 1980 through April 2007. Two reviewers independently selected studies and extracted data from 27 trials (23 placebo-controlled trials and 4 head-to-head comparative trials).
What was found
Pooled analysis of second-generation antidepressants versus placebo demonstrated a persistent protective effect against relapse and recurrence, with a number needed to treat (NNT) of 5 (95% CI: 4 to 6). Relapse estimates were not significantly altered by drug type, trial duration, or length of pre-randomization open-label treatment. Adverse event-related discontinuation was 7% with active treatment versus 5% with placebo. Comparative trials found no statistically significant differences between duloxetine and paroxetine, fluoxetine and sertraline, fluvoxamine and sertraline, or trazodone and venlafaxine.
Why it matters
This meta-analysis provides clear quantitative confirmation that continuation and maintenance pharmacotherapy with second-generation antidepressants substantively reduces depressive relapse and recurrence over placebo.
Limits
The abstract does not provide total participant sample size, specific odds ratios/relative risks, or metrics of statistical heterogeneity. Head-to-head comparative data were sparse (only four trials across different drug pairs), limiting conclusions regarding the superiority of any specific second-generation antidepressant.
Cited by
- contradicts Long-term use of SSRIs and SNRIs for depression is only slightly better in terms of outcomes compared to placebo.