PFAS and semen quality: small associations, and only for two measures

Experimental work has long suggested that PFAS interfere with male reproductive biology. The epidemiology has been less obliging: human studies have been variably consistent about whether exposure shows up in the semen of actual men [s1]. An updated systematic review and meta-analysis published on 4 August tried to put a number on the disagreement.

What was pooled

The review followed PRISMA reporting guidelines, was registered on PROSPERO as CRD420251267798, and searched four databases up to 13 July 2025 [s1]. Studies qualified if they assessed the association of PFAS with sperm outcomes [s1]. Fourteen studies, including 7,465 participants, met the criteria [s1].

The primary effect measure was the Pearson correlation coefficient. Outcomes covered sperm concentration, semen volume, total sperm count, motility, morphology, pH, and sperm DNA fragmentation [s1].

What was found, and what was not

The overall picture is described by the authors as heterogeneous [s1]. PFAS mixtures showed limited associations with the conventional semen parameters — the ones a fertility clinic reports on a standard analysis [s1].

Two exceptions were consistent. The first is progressive motility, the fraction of sperm moving purposefully forward, where the pooled correlation was -0.058 (95% CI -0.100 to -0.015) [s1]. The second concerns sperm DNA. The pooled correlation was 0.141 (95% CI 0.087 to 0.195), reported by the authors as an inverse relationship with sperm DNA integrity — that is, higher PFAS corresponding to more DNA fragmentation, which is the same statement made from the other side [s1].

Individual congeners exhibited distinct association profiles, which the authors read as evidence of potential congener-specific effects rather than a uniform class effect [s1].

How big is a correlation of 0.141

Small. A correlation coefficient of 0.141 means the exposure measure accounts for around 2% of the variation in the outcome across these studies. For progressive motility, at -0.058, the share is smaller still.

That is not a reason to dismiss it, but it is a reason to be careful about the sentence people will take away. This is not a finding that PFAS exposure explains an individual's fertility problem. It is a finding that, averaged across 7,465 men in 14 studies, a weak but statistically consistent association exists for one functional measure and one measure of DNA damage [s1].

The authors' own framing is appropriately narrow: the consistent association with increased DNA fragmentation across PFAS metrics suggests a possible role in unexplained male infertility, and prompts further investigation [s1]. "Possible role" and "prompts further investigation" are the operative words.

The design limits underneath the numbers

A meta-analysis inherits the weaknesses of what it pools. Studies of PFAS and semen parameters are overwhelmingly cross-sectional: a blood or serum sample and a semen sample collected at roughly the same time. That design cannot establish which came first, and it cannot rule out that men who differ in exposure differ in other ways — occupation, diet, body weight, age — that also bear on semen quality.

There is also a selection question that pooling cannot fix. Men who provide semen samples are often recruited through fertility clinics, which makes them unrepresentative of men in general in exactly the dimension being measured.

And the congener-specific pattern, while biologically plausible, is also what analysing many compounds against many outcomes tends to produce. Distinguishing a real congener effect from the statistical noise of a large comparison grid requires studies designed for that question.

The wider evidence it sits in

PFAS comprise thousands of anthropogenic fluorinated compounds, held together as a class by the carbon–fluorine bond rather than by any shared biological action [s2]. Effects established for one compound do not transfer automatically to the rest, which is why the congener-specific result above is more interesting than the pooled one.

On the reproductive tract specifically, the strongest existing human signal comes from elsewhere: the C8 Science Panel's work in a mid-Ohio Valley cohort of 32,254 highly exposed residents identified testicular cancer among its probable links with PFOA exposure [s3]. Testicular cancer and semen quality are different outcomes with different mechanisms, and one does not license inference about the other. But they mark the same organ system as the one where the human evidence has been least quiet.

What would move this question forward is a prospective design — exposure measured first, semen parameters measured later, in men not selected for infertility. Until that exists, the honest summary is a weak and consistent association with sperm DNA fragmentation, a weaker one with progressive motility, and not much for anything else.

Sources

• [s1] Associations between per- and polyfluoroalkyl substances and semen parameters: An updated systematic review and meta-analysis. Journal of Hazardous Materials, 4 August 2026. https://doi.org/10.1016/j.jhazmat.2026.143157
• [s2] From forever to fluoride: A critical review of PFAS occurrence, fate, and the transition from sequestration to destruction-based remediation in groundwater. Science of The Total Environment, 31 August 2026. https://doi.org/10.1016/j.scitotenv.2026.182278
• [s3] Mortality study of a PFOA-exposed cohort in the mid-Ohio Valley. Environment International, 30 August 2026. https://doi.org/10.1016/j.envint.2026.110492