Bell, Erin M.; Hertz-Picciotto, Irva; Beaumont, James J.
Department of Epidemiology
School of Public Health
University of North Carolina
Chapel Hill, NC
Occupational Epidemiology Branch
National Cancer Institute
6120 Executive Boulevard
EPS 8111, MSC 7240
Bethesda, MD 20892-7240
(address correspondence to: Erin M. Bell)
Department of Epidemiology and Preventive Medicine
University of California, Davis, CA
(current affiliation California Environmental Protection Agency)
To the Editor:
In a recent case-control study, we examined the association between residential proximity to applications of agricultural pesticides and late fetal death due to birth defects. 1 Elevated risks were observed for all five pesticide categories examined, with the greatest risk occurring when applications took place during the 3rd–8th week of pregnancy and within the same square mile as the maternal residence.
After publication, we determined that the pyrethroid categorization in our analysis was incorrect. The corrected pyrethroid category now includes the following pesticides: cypermethrin, pyrethrin, permethrin, fenvalerate, and flucythrinate. 2
Fenvalerate and permethrin were the most highly used pesticides within the pyrethroid category, with 46 and 30 individuals, respectively, who were potentially exposed during the 3rd–8th weeks of pregnancy. Of the pesticides erroneously included in our original pyrethroid category, paraquat dichloride was the most highly used, with 34 individuals exposed during the same period. With the exception of piperonyl butoxide (22 exposed), fewer than 12 individuals were exposed to each of the non-pyrethroid pesticides.
To determine the correct associations for pyrethroids, we repeated the analyses with only the true pyrethroid pesticides. We also examined the risk associated with exposure to paraquat dichloride.
As previously described, 1 exposure was determined for three gestational time periods (1st–20th, 1st–13th and 3rd–8th weeks of pregnancy) and exposed women were compared to those not exposed to the class of pesticides being evaluated. A fourth exposure definition compared those exposed to specific pyrethroids during the 3rd–8th weeks of pregnancy with those not exposed to any of the pesticide classes during the same time period.
Within the same or eight surrounding Township, Range, and Sections (TRSs) (one square mile), the ORs for exposure to pyrethroids are larger in this new analysis, with ORs ranging from 2.2 (95% CI = 1.0–4.7) to 3.8 (95% CI = 1.6–9.1) for exposure during gestational weeks 1–20 and 3–8, respectively (Table 1). Risk increases as the time window for exposure narrows in on the period of organogenesis and also when the non-exposure category is restricted to those not exposed to any of the pesticide classes examined; analysis under these two conditions produces an OR of 4.9 (95% CI = 1.9–12.9)). The ORs in this new analysis are less stable, however, with the 95% CI width increasing substantially for all estimates. The associations are stronger when analyses are restricted to pesticide applications within the same square mile as the maternal residence, although small numbers prevent examination of exposure during the 3rd–8th weeks of pregnancy. 2
In contrast, potential exposure to paraquat dichloride in the same or eight surrounding TRSs show no association, with ORs of 0.8 (95% CI = 0.4–1.4) for exposure during the first 20 weeks and 0.9 (95% CI = 0.5–1.7) for exposure during the first trimester of pregnancy. The risk increases, however, when exposure occurred during the 3rd–8th week of pregnancy (OR = 1.8 (95% CI = 0.9–3.9) compared to those not exposed to any of the pesticide classes examined during the same time period. Small numbers prevent analysis of exposures occurring only in the same TRS.
Overall, the conclusions from our previous work remain unchanged. This corrected analysis shows a stronger association between fetal death due to congenital anomalies and residential proximity to applications of pyrethroids; risk was elevated four- to five-fold, depending on the definition of “non-exposed,” when exposure occurred during the 3rd–8th week of pregnancy. Pyrethroids are often applied to crops in combination with other pesticides such as organophosphates and carbamates. 3–5 In our study, 90% of those exposed to pyrethroids were also exposed to these other pesticide classes. Thus, we did not have the power to examine whether the observed associations are the result of exposure to pyrethroids, other pesticides, or both.
Erin M. Bell
James J. Beaumont
1. Bell EM, Hertz-Picciotto I, Beaumont JJ. A case-control study of pesticides and fetal death due to congenital anomalies. Epidemiology 2001; 12: 148–156.
2. Bell EM, Hertz-Picciotto I, Beaumont JJ. Erratum: A case-control study of pesticides and fetal death due to congenital anomalies. Epidemiology 2001; 12: 596.
3. Royal Society of Chemistry: The Agrochemicals Handbook. Cambridge, England: The Royal Society of Chemistry, 1991.
4. Extension Toxicology Network (Extoxnet). Pesticide Information Profiles. http://www.ace.orst.edu
, accessed 4/2001.
5. United States Environmental Protection Agency, Office of Prevention, Pesticides and Toxic Substances: Recognition and Management of Pesticide Poisonings, 5th ed. Washington, DC: United States Environmental Protection Agency, 1999.
© 2001 Lippincott Williams & Wilkins, Inc.