Polychlorinated biphenyls, TEQs, children, and data analysis.

Abstract

The data analyses in the publications reviewed in Kimbrough et al (1) and Kimbrough and Krouskas (2) are discussed. All of the studies involve infants exposed in utero to small amounts of polychlorinated biphenyls (PCBs) and/or other chlorinated hydrocarbons. Because of missing sampling data, attrition of the cohorts over time, the use of TEQvalues (toxic equivalence) for some congeners of PCBs (polychlorinated biphenyls), PCDDs (polychlorinated dibenzo-p-dioxins) and PCDFs (polychlorinated dibenzofurans), the correlations reported by the authors between chemical exposure and adverse outcomes are not convincing. For the most part the exposures in these cohorts in the US and in Europe were within the ranges observed in the general population, and in many biological samples the level of organochlorines were below the limit of detection of the analytical method. Appropriate evaluation of such data is difficult given the many existing confounders. Many comparisons were made in the different studies. The observed differences were not consistent across studies and most likely occurred by chance. None of the observed differences reported in the children represent clinical disease.