Associations between prenatal bisphenol S with gestational age and preterm birth: A systematic review and meta-analysis.

Abstract

Bisphenol S (BPS), an endocrine-disrupting chemical used in thermal paper and hard plastics, is a common replacement for bisphenol A. Human exposure to BPS is widespread, raising concern about potential effects on pregnancy and fetal development. We conducted a meta-analysis of studies published before July 2025 that evaluated associations between BPS exposure and two outcomes: preterm birth and gestational age. Eligible studies used cohort, case-control, or cross-sectional designs; measured BPS in humans; reported adjusted associations; and detected BPS concentrations in over 50% of participant samples. We standardized published effect estimates to reflect changes in the odds of preterm birth or difference in gestational age per one ng/mL increase in natural log-transformed BPS. We used mixed-effects meta-analysis to estimate trimester-specific pooled odds ratios (for preterm birth) and beta coefficients (for gestational age). Nine studies met inclusion criteria. Most showed low-to-medium threats of bias, primarily due to potential misclassification of BPS exposure. Pooled associations for BPS measured between preconception and second trimester produced a statistically significant finding of -0.04 weeks (95% CI: -0.05, -0.02) for gestational age as well as a non-significant increase in the odds of preterm birth. No pooled estimates for BPS measured between third trimester were statistically significant, however the associations changed direction. Variation in median BPS concentrations and detection rates contributed to heterogeneity across studies. These results suggest that timing of gestational BPS exposure may influence pregnancy outcomes. Future research should incorporate repeated exposure measurements to better capture sensitive windows of vulnerability.