Current Risk Factors for Male Infertility and Semen Parameters: An Umbrella Review of Systematic Reviews and Meta-Analyses
67 risk factors linked to male infertility catalogued — but 78% of the underlying evidence rated low or very low quality
Journal: Asian Journal of Andrology | Published: 2026-01-13 | Type: Umbrella Review / Meta-Analysis | PMID:41527944Authors: Wang Qi-Hao et al. (Department of Urology, West China Hospital, Sichuan University, Chengdu)
Funding/COI: Not listed
Summary
This umbrella review from West China Hospital synthesized 43 systematic reviews covering 67 risk factors for male infertility and abnormal semen parameters. The headline finding is damning for the field: of 249 individually graded effect sizes, 78.3% received "very low" or "low" quality ratings under GRADE. Diabetes, metabolic syndrome, obesity, SSRIs, lead exposure, and SARS-CoV-2 all showed associations with worse semen quality — but "associated with" is doing heavy lifting when the underlying evidence is this thin.
Disease associations: type 1 diabetes, metabolic syndrome, hyperthyroidism, systemic lupus erythematosus, chronic prostatitis, and leukocytospermia linked to abnormal semen parameters
Lifestyle factors: obesity, sleep disorders, and smoking associated with decreased semen quality
Occupational/environmental exposures: carbon disulfide, organophosphates, and lead associated with worse parameters
Medications: sulfasalazine, mesalazine, and SSRIs associated with reduced semen quality
Viral infections: SARS-CoV-2, HPV, and hepatitis viruses associated with decreased semen quality
Protective associations: regular physical exercise, nut consumption, and healthy dietary patterns linked to improved semen quality
Study Quality
Umbrella reviews sit at the top of the evidence hierarchy by design — they synthesize the synthesizers. This one searched Web of Science, MEDLINE, and Embase from January 2000 to February 2025 and applied GRADE scoring to each effect size individually, which is methodologically sound and relatively rare in this literature. The problem isn't the review's methodology; it's what it found underneath: 78% of the evidence base is low or very low quality, meaning the review has done rigorous work to map a weak foundation.
The study also covers a heterogeneous mix of endpoints — infertility diagnosis, semen volume, sperm concentration, count, morphology, motility, and progressive motility. An association with "abnormal semen parameters" is not the same as an association with clinical infertility, and collapsing these endpoints conflates surrogate markers with outcomes that actually matter to patients.
Red Flags
Zero effect sizes out of 249 rated "high" quality — the top tier of the GRADE scale is completely empty
Causal relationships unconfirmed for most factors; the authors acknowledge this directly in the abstract
Funding and conflict of interest not listed — a significant gap for a paper with commercial implications in supplements and occupational health
"Nut consumption" listed alongside "healthy dietary pattern" as a protective factor — one is a single food item, the other a lifestyle construct; grouping them obscures what the evidence actually supports
The January 2000–February 2025 search window spans enormous changes in semen analysis methodology, population demographics, and diagnostic criteria, compounding heterogeneity
Collapsing multiple semen parameters into aggregate analyses risks obscuring which specific outcomes are affected by which exposures
Strengths
GRADE applied to all 249 effect sizes individually — unusually rigorous for this literature
Broad scope covering diseases, lifestyle factors, environmental and occupational exposures, medications, and viral infections in a single synthesis
Includes SARS-CoV-2 effects, a relatively recent and clinically relevant addition to this evidence base
Authors are explicit about causal uncertainty in the abstract rather than overclaiming
Verdict
This umbrella review is most useful as a map of how little we reliably know about male infertility risk. The methodology is solid — GRADE applied systematically to 249 effect sizes is meaningful work — but the signal it returns is mostly noise: 78% of the evidence underpinning known "risk factors" is low or very low quality. That list of 67 risk factors will inevitably get condensed in health media to "67 things that harm your sperm," which is precisely the wrong takeaway. What this paper actually demonstrates is that the field has generated substantial associations and almost no confirmed causation.