Abstract
Pregnant women are regularly exposed to a variety of environmental toxicants in daily life, posing a potential threat of kidney injury before presence of clinical manifestations. As there is a paucity of studies employing an exposome-based approach of kidney health in pregnant women, this study utilizes the above-mentioned strategy to identify the most significant environmental toxicants associated with early kidney injury in pregnant women in the cohort of TMICS (Taiwan Maternal and Infant Cohort Study). A total of 1,139 third-trimester pregnant women (weeks 29-40) were recruited between 2012 and 2015, and one-spot urine samples were successfully collected for study. Sixteen biomonitoring chemicals were measured in urine, including exposure measurements of melamine, 9 phthalate metabolites, nonylphenol (NP), bisphenol A (BPA), methylparaben (MP), ethylparaben (EP), propylparaben (PP), and butylparaben (BP), and outcome measurements of NAG (N-acetyl-β-d-glucosaminidase) and albumin-to-creatinine ratio (ACR). A two-tier strategy of statistical analyses was employed and data was randomly and evenly split to both training (n = 569) and validation (n = 570) sets. Using a weighted quantile sum (WQS) regression in the training set and subsequently a multivariate regression in the validation set, we found that NP was the most important chemical to link with early markers of kidney injury, both ACR and NAG. Our findings indicate that short-term exposure to NP is associated with markers of subclinical kidney injury in pregnant women in Taiwan. Further research is warranted to determine whether NP exposure is linked to clinically relevant kidney outcomes.