Background: Serum concentrations of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) were assessed in relation to miscarriage in a population of mid-Ohio River Valley residents highly exposed to PFOA through contaminated drinking water.
Methods: Serum PFOA and PFOS concentrations were measured in 1129 women in 2005-2006 who reported pregnancy outcomes in follow-up interviews between 2008 and 2011. In the analysis, we included 1438 reported live births, stillbirths, and miscarriages with estimated conception dates after the serum measurements. Preconception serum levels of PFOA and PFOS were analyzed in relation to miscarriage using logistic regression and generalized estimating equations.
Results: There was little evidence of association between PFOA and miscarriage. For PFOS, when including all reported prospective pregnancies, the odds ratio of miscarriage per log ng/ml increase was 1.21 (95% confidence interval = 0.94-1.55); in subanalyses restricted to each woman's first pregnancy conceived after the serum measurement, the odds ratio was 1.34 (1.02-1.76). Categorical analyses showed elevated odds ratios for the top 4 quintiles relative to the first quintile, without a monotonic trend. Positive associations between PFOS and miscarriage were strongest among nulligravid pregnancies.
Conclusions: In this prospective study of miscarriage in a population exposed to high levels of PFOA and background levels of PFOS, we found little evidence of association with serum levels of PFOA and limited evidence of association with serum levels of PFOS.