Association between maternal urinary speciated arsenic concentrations and gestational diabetes in a cohort of Canadian women (Abstract)

Ashley-Martin J, Dodds L, Arbuckle TE, Bouchard MF, Shapiro GD, Fisher M, Monnier P, Morisset AS, Ettinger AS. Environment International. 2018 Dec 121(Pt 1):714-720. doi: 10.1016/j.envint.2018.10.008.

Background

Epidemiological and toxicological evidence suggests that maternal total arsenic (As) levels are associated with an elevated risk of gestational diabetes (GDM). Uncertainty remains regarding the metabolic toxicity of specific arsenic species, comprised of both organic and inorganic sources of arsenic exposure.

Objectives

We assessed associations between speciated As and GDM using data from the Maternal-Infant Research on Environmental Chemicals (MIREC) Study.

Methods

Concentrations of speciated As [(inorganic (trivalent, pentavalent)), methylated arsenic species metabolites (monomethylarsonic acid (MMA), dimethylarsinic acid (DMA)), and organic (arsenobetaine)] were measured in first trimester maternal urine samples. GDM cases were identified in accordance with Canadian guidelines. Multivariable regression models were used to estimate associations between speciated As and GDM, evaluate potential interaction between speciated As exposures, and assess fetal sex-specific findings.

Results

Among 1243 women who had a live, singleton birth and no previous history of diabetes, 4% met the diagnostic criteria for GDM. Our analyses focused on DMA and arsenobetaine as these were the subtypes with detectable concentrations in at least 40% of samples. Compared to women in the lowest tertile of DMA (<1.49 μg As/L), women with concentrations exceeding 3.52 μg As/L (3rd tertile) experienced an increased risk of GDM (aOR = 3.86; 95% CI: 1.18, 12.57) (p-value for trend across tertiles = 0.04). When restricted to women carrying male infants, the magnitude of this association increased (aOR 3rd tertile = 4.71; 95% CI: 1.05, 21.10).

Conclusions

These results suggest a positive relation between DMA and GDM; potential differences in risk by fetal sex requires further investigation.