<p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" class="first" dir="auto"
id="d5475293e204">Increasing evidence suggests a potential role of endocrine disrupting
chemicals (EDCs)
in inducing gestational diabetes mellitus (GDM). However, as far as we know, no study
has examined the associations between GDM and exposure to parabens, a kind of EDCs.
In this study, we explored the association between urinary parabens of pregnant women
and GDM and studied the modification effect of prepregnancy body mass index (BMI).
Urine samples were collected from 696 pregnant women and parabens were measured, including
four alkyl side chain substituted para-hydroxybenzoic acid ester, substituents varying
from methyl to butyl (abbreviates as MeP, EtP, PrP and BuP), and benzyl substituted
para-hydroxybenzoic acid ester (BzP). Logistic regression models adjusting for potential
confounders were used to study the association of parabens and GDM in the overall
population, and further stratified analysis by prepregnancy BMI categories was also
performed. The detection rates for the five parabens in the urine samples were 97.70%
(MeP), 71.26% (EtP), 96.55% (PrP), 15.80% (BuP) and 2.73% (BzP). No significant association
was found between parabens and GDM among the overall population. However, significant
non-linear associations of PrP and the summed estrogenic activity of parabens with
GDM were found in the stratified analysis by prepregnancy BMI in the overweight/obese
population, with adjusted odds ratios (aORs) of 3.47 (95% CI: 1.28, 9.42) and 2.87
(95% CI: 1.07, 7.73) for GDM in the second tertile of urinary PrP and the summed estrogen
activity, respectively, when compared to the first tertile. Although no statistically
significant association between parabens and GDM was found in the overall population,
we found that among the overweight/obese pregnant women, who represent a subgroup
more prone to GDM, moderately higher levels of PrP and summed estrogenic activity
of parabens were significantly associated with an increasing GDM prevalence.
</p>