Uldbjerg · The Science of the total environment 2022 · prospective cohort study · n=88

Sex-specific associations between maternal exposure to parabens, phenols and phthalates during pregnancy and birth size outcomes in offspring.

Cited 41 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study evaluating biomarker exposure and birth outcomes

PubMed 35508231 · doi:10.1016/j.scitotenv.2022.155565 · record verified 2026-08-31

What was done

In a prospective cohort of 88 pregnant women undergoing amniocentesis between 2012 and 2014, maternal urine samples collected between gestational weeks 12 and 36 (median: 18 weeks) were measured for parabens, phenols, and phthalate diester metabolites using isotope-diluted liquid chromatography-tandem mass spectrometry. Linear regression models estimated sex-stratified associations between medium/high (tertiles 2–3) versus low (tertile 1) exposure to detected chemicals and birth size parameters (weight, length, and head and abdominal circumferences).

What was found

Few overall associations reached statistical significance. However, among male offspring, medium/high exposure to bisphenol A (BPA) was associated with significantly lower birth weight (-428 g, 95% CI -756 to -99.4), birth length (-1.76 cm, 95% CI -3.28 to -0.25), and abdominal circumference (-1.97 cm, 95% CI -3.55 to -0.39). Medium/high methyl paraben exposure in male offspring was also associated with lower birth weight (-661 g, 95% CI -1251 to -70.7) and birth length (-3.11 cm, 95% CI -5.76 to -0.46). No statistically significant associations were identified in female offspring, though effect estimates across individual exposures were consistently more negative in males.

Why it matters

The findings add preliminary evidence that prenatal exposures to common endocrine-disrupting chemicals like BPA and methyl paraben may impair fetal growth in a sexually dimorphic pattern, with male fetuses appearing more vulnerable.

Limits

The study is severely limited by a very small sample size (n = 88 total, further reduced by sex stratification), leading to wide confidence intervals and a high risk of both type I and type II errors. The cohort was drawn exclusively from women undergoing amniocentesis, which may limit generalizability to lower-risk pregnancies, and exposure was characterized primarily from single spot urine samples despite known short biological half-lives of non-persistent chemicals.

Cited by