ORIGINAL ABSTRACT

Preterm birth (PTB) is associated with a high risk of infant mortality and long-term adverse health effects. Glyphosate is a broad-spectrum herbicide applied in agricultural and non-agricultural settings. Studies suggested an association between maternal exposure to glyphosate and PTB among mostly racially homogenous populations, though results were inconsistent. The objective of this pilot study was to inform the design of a larger and more definitive study of glyphosate exposure and adverse birth outcomes in a racially-diverse population. Urine was obtained from 26 women with a PTB as cases and 26 women with a term birth as controls, from participants enrolled in a birth cohort in Charleston, South Carolina. We used binomial logistic regression to estimate associations between urinary glyphosate and the odds of PTB, and multinomial regression to estimate associations between maternal racial identity and urinary glyphosate among controls. Glyphosate was unrelated to PTB (odds ratio (OR) = 1.06, 95% CI: 0.61, 1.86). Women who identified as Black had greater odds (OR = 3.83, 95% CI: 0.13, 111.33) of having categorical "high" glyphosate (> 0.28 ng/mL) and lesser odds (OR = 0.79, 95% CI: 0.05, 12.21) of "low" glyphosate (< 0.03 ng/mL) relative to women who identified as white, suggesting a potential racial disparity, although the effect estimates were imprecise and included the null. Given concerns of potential reproductive toxicity of glyphosate, the results merit confirmation in a larger investigation to determine specific sources of glyphosate exposure, incorporating longitudinal urinary glyphosate measures during pregnancy and a comprehensive measure of diet.