Ehrlich · Human reproduction (Oxford, England) 2012 · prospective cohort study · n=174

Urinary bisphenol A concentrations and early reproductive health outcomes among women undergoing IVF.

Cited 246 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective non-randomized cohort study

PubMed 23014629 · doi:10.1093/humrep/des328 · record verified 2026-08-29

What was done

In a prospective preconception cohort study at Massachusetts General Hospital Fertility Center (2004–2010), researchers followed 174 women aged 18–45 years across 237 fresh autologous IVF cycles. Cycle-specific urinary bisphenol A (BPA) concentrations were measured using online solid-phase extraction-HPLC-tandem mass spectrometry. Statistical models (mixed-effects, Poisson, and multivariable logistic regression) adjusted for age, Day 3 serum FSH, smoking, and BMI assessed associations between urinary BPA and ovarian response, oocyte maturation, normal fertilization, and blastocyst formation.

What was found

The geometric mean (SD) urinary BPA concentration was 1.50 (2.22) µg/l. In adjusted models, higher urinary BPA showed a linear dose-response association with decreased peak estradiol (mean decreases of 40, 253, and 471 pg/ml for quartiles 2, 3, and 4 vs quartile 1, respectively; P-trend = 0.001). The highest versus lowest BPA quartile was associated with a 24% decrease in total oocytes retrieved (P-trend < 0.001) and a 27% decrease in normally fertilized oocytes (P-trend = 0.002). Decreased blastocyst formation was observed above the lowest BPA quartile (P-trend = 0.08).

Why it matters

This study provides prospective evidence linking elevated BPA exposure to reduced ovarian response and lower fertilization rates in women undergoing fertility treatment.

Limits

Urinary BPA has a short elimination half-life and high within-person variability, introducing potential exposure misclassification. The cohort was modest (174 women from a single fertility center), blastocyst reductions did not achieve statistical significance, and results may not generalize to natural conception in the general population.

Cited by