Accuracy of fetal growth indicators as surrogate measures of steroid hormone levels during pregnancy.

This study evaluates the use of fetal growth characteristics as surrogate measures for steroid hormone exposures during pregnancy. The validity of using birth weight, birth length, ponderal index, and placental weight as indicators of pregnancy hormone exposures was assessed using third-trimester serum samples from 568 pregnant women who participated in the Child Health and Development Studies, Berkeley, California (1959-1966). The magnitude of the associations between birth characteristics and hormone concentrations was assessed using geometric means, Pearson's correlations, and linear and logistic regression. Accuracy was evaluated using sensitivity, specificity, and receiver operating characteristic curve analyses. The strongest and most consistent association observed was between birth weight and estriol levels. Despite a positive correlation (r = 0.32) and strong associations with high estriol levels (odds ratio for highest compared with lowest birth weight quartile = 6.63, 95% confidence interval: 3.20, 12.5), the predictive performance of birth weight as a proxy for estriol levels was poor (area under the receiver operating characteristic curve = 0.66, 95% confidence interval: 0.61, 0.71). Likewise, all fetal growth measures revealed little discriminatory ability as indicators of estriol, estrone, estradiol, or progesterone levels. Thus, observed associations with these surrogate measures may not be a reflection of pregnancy hormone exposure and should be interpreted with caution.