Associations of pesticide residue exposure from fruit and vegetable intake with ovarian reserve

Background: We previously reported that the intake of fruits and vegetables (FV) known to have high pesticide contamination in the US food supply is related to lower sperm counts. Whether the same is true for ovarian reserve is unknown.

Methods: Participants were 633 females, 21-45 years, presenting to an academic fertility center. We combined surveillance data from the US Department of Agriculture and self-reported food intake data to characterize exposure to pesticide residues through FV intake. Poisson and linear regression were used to evaluate associations of high-pesticide residue, low-pesticide residue, and total FV intake with markers of ovarian reserve (antral follicle count [AFC], follicle-stimulating hormone [FSH], anti-Müllerian hormone [AMH]) adjusting for potential confounders.

Results: There was no association of FV intake, overall or according to pesticide residue status, with day 3 FSH or AMH concentrations in multivariable-adjusted models. Regardless of pesticide residue status, FV intake was inversely related to AFC in these models. This pattern was magnified among females who had had a fertility evaluation before joining the study (n=508). Among females who had not had a fertility evaluation before joining the study (n=103), however, there were diverging patterns of association for high- and low-pesticide residue FV intake and markers of ovarian reserve. In this group, day-3 FSH was 71.6% (95%CI: 39.5%, 111.2%) higher among females in the highest quintile of high pesticide residue FV intake than among females in the lowest quintile (p-trend<0.001). Low pesticide residue and total FV intake were unrelated to day 3 FSH in this group, with differences between top and bottom quintile of intake of -8.3% (-25.8%, 13.3%) and 7.5% (-13.8%, 34.0%), respectively.

Discussion: High pesticide residue FV intake may be related to lower ovarian reserve among females without a history of infertility treatment. Replication in populations with larger sample sizes and less susceptible to reverse causation is important.