Sexual dysfunction associated with selective serotonin reuptake inhibitors in adults with depression: a systematic review and meta-analysis

High-certainty meta-analysis of 13 RCTs, but all data is from 1990–2014

Journal: European Journal of Clinical Pharmacology | Published: 2026-02-21 | Type: Systematic Review, Meta-Analysis | PMID: 41721013 Authors: Dagostin Ferraz S, Kuyunga L, Rech P, et al. (Translational Biomedicine Laboratory, Universidade do Extremo Sul Catarinense, Brazil; Universidade Estadual de Mato Grosso do Sul) Funding/COI: Funding not listed. Authors declare no competing interests.

Summary

A 2026 systematic review and meta-analysis of 13 RCTs (5,941 adults with major depressive disorder) found that SSRIs are associated with a 3.28-fold increased risk of orgasmic dysfunction compared to placebo — a finding rated high-certainty by GRADE. Reduced sexual satisfaction was also statistically significant. The catch: every included study was published between 1990 and 2014, meaning this is a rigorous synthesis of old data, not a window into modern prescribing.

Claims

• Orgasmic dysfunction: RR = 3.28 (95% CI: 2.33–4.60, p < 0.00001; I² = 8%) — low heterogeneity, high-certainty evidence per GRADE
• Reduced sexual satisfaction: RR = 1.21 (95% CI: 1.11–1.32, p = 0.0001; I² = 0%) — GRADE rated moderate certainty due to risk-of-bias concerns
• Decreased sexual desire: RR = 1.40 (95% CI: 0.92–2.12, p = 0.12; I² = 54%) — trend only, did not reach statistical significance; high heterogeneity
• CSFQ total scores: No significant difference vs. placebo across the 6 studies included in meta-analysis
• 13 RCTs met qualitative inclusion criteria; only 6 were pooled in the meta-analysis
• Total enrolled: 5,941 participants — 2,651 men (44.6%), 3,290 women (55.4%)
• Studies published 1988–2014; treatment duration ranged 6–52 weeks (median: 8 weeks)
• GRADE certainty: high for orgasmic dysfunction; moderate for satisfaction, desire, and CSFQ

Study Quality

This review followed PRISMA guidelines, was pre-registered on PROSPERO (CRD42023448691), and used the revised Cochrane RoB 2 tool with two independent reviewers — inter-rater agreement was strong (κ = 0.82 at abstract screening, κ = 0.91 at full-text). The multi-database search (PubMed/MEDLINE, LILACS, Embase, Cochrane Library) with no language restrictions is a genuine methodological strength. GRADE ratings were applied across outcomes, lending transparency to confidence estimates.

The meta-analysis itself is narrower than the narrative review: 7 of 13 RCTs were excluded from pooling due to clinical heterogeneity, differing outcome measurement methods, or insufficient data. That's more than half the included studies sitting out the quantitative synthesis, which limits the meta-analysis's power and scope. Heterogeneity on the desire endpoint (I² = 54%) further complicates interpretation there.

Red Flags

• Stale evidence base: All 13 included RCTs were published between 1990 and 2014. The body of evidence is over a decade old at minimum — contemporary SSRIs, dosing strategies, and patient populations may differ substantially
• More than half the RCTs excluded from meta-analysis: 7 of 13 studies couldn't be pooled, cited for clinical heterogeneity and methodological inconsistency — the meta-analysis rests on 6 trials
• Funding opacity: No funding source is listed for the review itself, and the funding of the 13 individual RCTs is not reported in the abstract or extracted in the results — at least some of those trials were likely industry-sponsored, which the review does not address
• Conclusion language drifts toward advice: The authors write that "healthcare providers should engage in open discussions... consider alternative treatments when appropriate" — this is in the paper, not reproduced as our guidance
• Depression confounding: Sexual dysfunction is a symptom of depression itself; separating SSRI effect from underlying illness effect is methodologically difficult, and the placebo comparison only partially controls for this
• No SSRI-specific breakdown: The results pool across multiple SSRIs (fluoxetine, sertraline, paroxetine, etc.) with meaningfully different pharmacological profiles — no drug-level subgroup analysis is reported

Strengths

• Pre-registered on PROSPERO — reduces post-hoc outcome manipulation risk
• PRISMA-compliant with a well-documented PRISMA flow diagram
• RoB 2 applied by two independent reviewers with strong inter-rater agreement (κ ≥ 0.82)
• GRADE assessment applied per outcome — provides explicit certainty ratings rather than leaving readers to guess
• Low heterogeneity on the two significant outcomes (I² = 8% and 0%), making the pooled RR estimates more interpretable
• Large total sample across included studies (n = 5,941) with both sexes represented
• Multi-database search with no language restriction reduces publication and language bias

Verdict

This is a competently executed meta-analysis of a genuinely important question, and the signal on orgasmic dysfunction is hard to dismiss — a 3.28 relative risk with I² near zero across six RCTs is not noise. But the evidence base is frozen in the early 2000s, more than half the eligible trials couldn't be pooled, and the funding landscape of those original industry-era RCTs goes unreported. What you get is a rigorous synthesis of old data that confirms what clinicians have known for decades. It contributes a GRADE-certified confidence level to an established finding, which has value — but don't mistake "high certainty" for "the whole picture."