Insecticides and Autism: Overview and New Results From the Charge Study

Background:   Insecticides are designed to damage living organisms, often by targeting the CNS. Several recent studies have provided evidence of a link between specific classes of pesticides and the risk for autism or for symptoms of pervasive developmental disorders (PDD). These used objective measures of exposure: one linked commercial applications of organochlorine, pyrethroid, and other pesticides from a statewide database to residences during pregnancy; a second attempted to replicate those findings in a sample of cases and controls with more detailed confounding information; a third measured urinary metabolites of organophosphates during pregnancy and followed the children, obtaining the PDD measures through the Child Behavior Checklist. 

Objectives:  1) To examine the relationship of autism spectrum disorder (ASD) risk with household applications of insecticide products and specific active ingredients in those formulations.  2) To evaluate evidence for an interaction between pyrethroid exposures and MAOA genotype.

Methods: Participants (n=783) were enrolled in the CHildhood Autism Risks from Genes and Environment (CHARGE) study beginning in 2003. ASD was confirmed on the ADOS and ADI-R. Controls were recruited from State birth files, using a stratified random sample matched on age, sex, and broad geographic area, and were considered typically developing (TD) if they scored higher than 2 SD below the mean on Mullen’s Scales of Early Learning and Vineland Adaptive Behavior Scales, and below 15 on the Social Communications Questionnaire. Exposures to household products were collected through an extensive interview that obtained product type, use and brand, and associated time periods from preconception through pregnancy and early childhood.  Insecticides were searched in online databases containing active ingredients by brand, type, and date to assign specific chemical exposures. Standard PCR was used for genotyping the variable number tandem repeats in the promoter region of the X-linked MAOA gene.  

Results: Parents of ASD children were more likely than those of controls to report applications of insecticide sprays and foggers in the preconception and pregnancy period, and the products they used were more likely to contain pyrethroids (multivariate adjusted Odds Ratio, aOR=1.88 (95% confidence interval (CI), 1.21, 2.94)). Repeated applications (6 or more months during pregnancy) conferred especially high risk (aOR=3.47, 95% CI=1.48, 8.11). Boys carrying four tandem repeats in the MAOA promoter locus, were at exceptionally high risk (nearly five-fold) if pyrethroids were used. Products containing less toxic pyrethrins carried lower risks, and results were robust to various sensitivity analyses.  

Conclusions: Reporting accuracy could differ for mothers of ASD versus unaffected children, but the synergistic relationship for pyrethroid use and MAOA genotype is incompatible with this explanation. Moreover, participants reported products they used, not exposures: the latter were determined by linkage of product type, brand, and year of use with databases providing active ingredients; these often changed over time. Evidence presented here suggests the pyrethroid class of insecticides comprise a modifiable environmental factor that may increase ASD risk, particularly in genetically susceptible individuals, but replication in a prospective setting should be sought.