BODY MASS INDEX AND BROMOXYNIL EXPOSURE IN A SAMPLE OF RURAL RESIDENTS DURING SPRING HERBICIDE APPLICATION

Abstract

Bromoxynil (3,5-dibromo-4-hydroxybenzonitrile), a phenolic herbicide, is widely used in production of cereals and other crops. Little is known, however, about bromoxynil exposure in humans. Results of previous research suggest a longer residence time in the body for bromoxynil compared to phenoxy herbicides [e.g., (2,4-dichlorophenoxy)acetic acid (2,4-D), 4-chloro-2-methylphenoxyacetic acid (MCPA)] and that bromoxynil would tend to partition into fatty tissue more so than 2,4-D. In previous research, body mass index (BMI) was found to be an independent predictor of plasma concentrations of 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE), the persistent lipophilic metabolite of the chlorinated pesticide bis(4-chlorophenyl)-1,1,1-trichloroethane (DDT). As part of the Prairie Ecosystem Study, gas chromatography/mass spectrometry analysis was used to measure concentrations of bromoxynil and seven other herbicides (2,4-D, dicamba, fenoxaprop, MCPA, ethalfluralin, triallate, and trifluralin) in plasma from residents (104 men, 88 women, 24 youths age 12–17yr) of a cereal-producing region in Saskatchewan, Canada, during spring herbicide application, 1996. Multiple logistic regression analysis was used to explore whether BMI predicted detection of bromoxynil in plasma from the adults. The prevalence of detection (detection limits: 2–50μg/L) was markedly higher for bromoxynil (men, 44.2%; women, 14.8%; youths, 20.8%) compared to each of the other herbicides including 2,4-D (men, 16.5%; women, 3.4%; youths, 12.5%) and MCPA (men, 6.8%; women, 1.1%; youths, 4.2%), although bromoxynil is commonly formulated or tank mixed with these herbicides. In the multiple logistic regression analysis, the variables BMI, exposure group [bromoxynil applicators, non-applicator family members of bromoxynil applicators, all others (reference group)], and days elapsed since the last use of bromoxynil were found to be independent predictors of detection of bromoxynil, while age, gender, and farm residency were not statistically significant. With adjustment for exposure group [bromoxynil applicators: odds ratio (OR=24.30, 95% confidence interval (CI)=9.59–61.58; nonapplicator family members of bromoxynil applicators: OR=3.53, 95% CI=1.19–10.44; all others (reference group)], the OR for detection of bromoxynil was 2.35 (95% CI=0.87–6.33) for participants in the middle (25.53–29.00kg/m²) tertile (men: OR=2.85, 95% CI=0.75–10.82; women: OR=1.63, 95% CI=0.36–7.40) of BMI and 4.01 (95% CI=1.46–11.03) for participants in the highest (>29.00kg/m²) tertile (men: OR=4.67, 95% CI=1.17–18.58; women: OR=2.20, 95% CI=0.44–10.99) with participants in the lowest (<25.53kg/m²) tertile as the reference group. Similar ORs were observed for BMI with adjustment for days elapsed since the last use of bromoxynil. In conclusion, further research is needed to investigate whether adiposity is an important modifying factor for persistence of bromoxynil in the body.

This research was funded in part by the Eco-Research Program, Tri-Council Secretariat, Canada, through a grant (922-94-0009) for the Study of the Sustainability of the Semi-Arid Prairie Ecosystem (PECOS) and by the National Health Research Development Program, Health Canada through a National Health Research Scholar award (6608-1323-48) to Dr. Karen M. Semchuk. The authors gratefully acknowledge the support and key contributions made to the design and conduct of the study by the residents of the study area and by the following research team members: Maura Gillis-Cipywnyk, Janice Colquhoun, Dr. Margaret F. O. Crossley, Dr. James A. Dosman, Pat Hanke, Leslie Holfeld, Dr. Vic Laxdal, Donna Ledingham, Dr. Alan M. Rosenberg, Phyllis Snodgrass, and Dr. Taylor A. Steeves.

Donald G. Irvine is deceased.

Present address for Allan J. Cessna is National Water Research Institute Environment Canada Saskatoon Saskatchewan Canada