Prevalence of Female Sexual Dysfunction Among Women with Systemic Lupus Erythematosus: A Systematic Review and Meta-Analysis

58.8% of lupus patients screen positive for FSD — 2.6x healthy controls, though extreme heterogeneity muddies the estimate

Journal: Frontiers in Immunology | Published: 2026-05-04 | Type: Systematic Review, Meta-Analysis | PMID: 42158876 Authors: Dai Xiaowei (Jilin University, Reproductive Medicine); Ding Hongxiang, Mao Dikai, Huang Jiaguo (Hangzhou Normal University, Urology) Funding/COI: Funding not listed; no commercial conflicts declared

Summary

Thirteen studies covering 1,511 women with SLE and 2,246 healthy controls found a pooled FSD prevalence of 58.8% and an odds ratio of 2.64 versus healthy controls. The direction of the finding is consistent, but the evidence base is undermined by extreme heterogeneity (I² = 96.8%), with individual study estimates ranging from 22% to 85.9%. The pooled figure is a statistical artifact of very different studies — not a reliable prevalence estimate.

Claims

• Pooled FSD prevalence in SLE: 58.8% (95% CI: 46.1%–71.6%)
• Odds ratio for FSD in SLE vs. healthy controls: OR 2.64 (95% CI: 1.27–5.47)
• Individual study estimates ranged from 22% to 85.9%
• Regional variation was statistically significant (p = 0.003): Africa 77.7%, Asia 65.8%, Europe 61.6%, Americas 34.9%
• 12 of 13 studies used the Female Sexual Function Index (FSFI) as the assessment instrument
• For comparison: a prior meta-analysis found 49.1% FSD prevalence in rheumatoid arthritis

Study Quality

Thirteen studies were included — 8 case-control and 5 cross-sectional, 12 of them single-center. Quality was assessed with AHRQ (cross-sectional, scores 7–9) and Newcastle-Ottawa Scale (case-control, scores 6–8); 9 studies rated high quality and 4 moderate. The review followed PRISMA guidelines and was pre-registered (INPLASY 202570092). Random-effects modeling was correctly applied given the heterogeneity.

The central problem is the I² of 96.8%. This is not mild or even substantial heterogeneity — it is extreme, meaning the underlying studies are reporting incompatible findings. Subgroup analyses by region, study design, and quality did not resolve heterogeneity within subgroups. Planned subgroup analyses by FSD assessment tool and number of study centers were abandoned for lack of data. When studies disagree this fundamentally, pooling them into a single prevalence number produces a figure that obscures more than it reveals.

Red Flags

• I² = 96.8% — extreme heterogeneity; the pooled 58.8% should not be taken at face value
• Only 13 studies, 1,511 patients, 12 of 13 single-center — thin evidence base
• African subgroup estimate (77.7%) rests on a single study — not a valid regional comparison
• Wide OR confidence interval (1.27–5.47) — the true effect size is poorly constrained
• Funding not disclosed — cannot rule out undeclared support
• Conclusion advocates updating clinical guidelines — an overreach given the data quality; this is the authors' recommendation, not a finding
• Urology-affiliated authors conducting a review on a rheumatological condition without rheumatology co-authors — not disqualifying, but notable

Strengths

• Multi-database search (PubMed, Cochrane, Web of Science, Embase) without language restrictions
• Pre-registered protocol (INPLASY 202570092)
• Dual independent reviewers for screening, extraction, and quality assessment
• Geographically diverse included studies (Asia, Europe, Americas, Africa)
• Consistent FSD measurement: 12/13 studies used FSFI
• Formal quality assessment using validated tools (AHRQ, NOS)

Verdict

The association between lupus and elevated FSD rates is biologically plausible and directionally consistent across studies — chronic inflammation, pain, fatigue, and corticosteroid use all offer mechanistic pathways. What this meta-analysis cannot credibly deliver is a single prevalence number. An I² of 96.8% means the included studies are not measuring the same underlying reality, and presenting "58.8%" as if it were a stable fact is misleading. Read this as a signal that FSD is common and understudied in SLE patients, and that the field lacks methodologically consistent data to say much more than that.