Monitoring chemical exposure using breast milk: a methodological review.

Retrospective exposure assessments are often a major weakness in environmental epidemiologic studies. Many environmentally important chemicals are persistent lipophilic compounds which partition into fat and breast milk. The analysis of these chemicals in breast milk provides a noninvasive means of collecting large volumes of a biologic fluid that correlates well with body burden. This may provide an integrated measure of cumulative absorption over several years which can be used to assess exposure in several different epidemiologic study designs. However, consideration of toxicokinetic principles in the study design is important in order to allow for maternal and other factors unrelated to exposure which may influence breast milk concentrations. These factors include maternal age, parity, maternal body weight, time of sampling during the lactation period, and fat content of the breast milk. Failure to use standardized entrance criteria and to allow for the above factors in the analysis and interpretation of the study results may lead to invalid conclusions regarding past exposures. Allowance must also be made for the restrictions on the sampling frame available in any epidemiologic study using breast milk analysis. Published studies to date often have had several methodological deficiencies. With the introduction of biologic specimen banks, breast milk collection and analysis should play an important role in future epidemiologic studies, especially those investigating health outcomes in infants.