Urinary biomonitoring of glyphosate exposure among male farmers and non-farmers in the Biomarkers of Exposure and Effect in Agriculture (BEEA) study

Abstract

BACKGROUND AND AIM: Glyphosate is the most widely applied herbicide worldwide. Despite potential health effects of glyphosate, biomonitoring data of glyphosate exposure in agriculture and other settings are limited. We measured urinary glyphosate concentrations and assessed their determinants among male farmers and non-farmers from Iowa and North Carolina in the BEEA study. METHODS: Based on questionnaire-assessed pesticide exposure histories, we selected four groups of BEEA participants: recently exposed farmers with occupational glyphosate use in the last 7 days (n=98), farmers with high cumulative lifetime days of glyphosate use (>80th percentile) but no use in the last 7 days (n=70), farmers with minimal lifetime glyphosate use (n=100), and non-farmers with no occupational pesticide exposure (n=100). We quantified glyphosate in first-morning void urine samples using ion chromatography isotope-dilution tandem mass spectrometry. We estimated geometric mean urinary glyphosate concentrations across groups and evaluated associations with potential determinants using linear regression. RESULTS: Glyphosate was detected in urine of most recently (91%), high-lifetime (93%), and low-lifetime (88%) exposed farmers, as well as non-farmers (81%); geometric mean concentrations were 0.86, 0.58, 0.44, and 0.37 µg/L, respectively. Compared to non-farmers, urinary glyphosate concentrations were significantly elevated among recently exposed farmers (P<0.0001), particularly those who used glyphosate within 1 day of urine collection [age- and creatinine-adjusted geometric mean ratio (GMR)=6.4, 95% confidence interval (CI)=4.3-9.3] and those not wearing gloves during glyphosate application (GMR=2.6, 95% CI=2.0-3.4). CONCLUSIONS: Preliminary results suggest that a high proportion of both farming and non-farming populations may be exposed to glyphosate. Substantial exposure contrast between those with and without recent occupational glyphosate use was detectable in urine. Given the short elimination half-life of glyphosate, the exposure differences between groups may be even greater than those reflected by measured urine concentrations. Continued biomonitoring of glyphosate in various settings is warranted. KEYWORDS: Glyphosate, exposure assessment, pesticide biomonitoring