Pregnancy per- and polyfluoroalkyl substances (PFAS) and maternal glucose intolerance

Abstract

OPS 18: Cardiometabolic effects of chemical exposures, Room 110, Floor 1, August 26, 2019, 1:30 PM - 3:00 PM Background/Aim: Per- and polyfluoroalkyl substances (PFAS) exposure may alter glucose tolerance (GT). However, research on the association between PFAS and altered GT during pregnancy is limited. Few studies have assessed potential effect modification by maternal factors. Methods: We assessed PFAS exposure and GT in 1,585 pregnant women from the Project Viva cohort in Boston, MA, USA. We quantified six PFAS in maternal plasma samples (median 9.6 weeks gestation; collected 1999-2002). GT was assessed by 50-g non-fasting, oral glucose challenge tests (GCT) (median 28 weeks gestation) and 3-hr fasting, 100-g oral glucose tolerance tests (OGTT). Women with abnormal GCT results (glucose ≥140 mg/dL) were classified as having isolated hyperglycemia (IH) (normal OGTT), impaired GT (IGT) (1 abnormal OGTT value), or gestational diabetes mellitus (GDM) (≥2 abnormal OGTT values) (American Diabetes Association criteria). As a secondary analysis we classified women as having abnormal vs. normal GCT results. We used multinomial logistic regression to assess associations of natural log-transformed continuous PFAS concentrations (ln-PFAS) with GT, adjusting for maternal age, race/ethnicity, pre-pregnancy BMI, education, smoking status, and parity. To explore potential effect modification, we used logistic regression to assess associations of ln-PFAS with abnormal vs. normal GCT, stratified by maternal BMI and age. Results: 265 (17%) women had abnormal GT. PFAS concentrations were not associated with GT in the full cohort [e.g. ln-perfluorononanoate (PFNA): OR (95% CI): IH, 1.23 (0.85, 1.78); IGT, 1.11 (0.62, 1.98); GDM, 1.15 (0.74, 1.78)]. In stratified analyses, ln-PFNA was positively associated with abnormal GCT among overweight (BMI 25 - <30 kg/m2) [OR: 1.76 (1.04, 2.97)] and younger (<35 years) [OR: 1.48 (1.05, 2.09)] women. Conclusion: We observed modest associations between PFNA and increased odds of abnormal GCT in certain population subgroups. However, we found no evidence of an effect of PFAS exposure on glucose tolerance in the full cohort.