Objective: In newborn screening programs for congenital adrenal hyperplasia, 17-alpha-hydroxyprogesterone (17OHP) cutoff levels are based on birth weight (BW) or on gestational age (GA). We investigated which approach would result in the greatest specificity and sensitivity.
Study design: For the determination of 17OHP, a neonatal 17OHP assay was used in filter paper blood of 9492 newborns. The relationships between 17OHP and BW and between 17OHP and GA were studied by regression analysis. Reference curves with a specificity of 99.95% were constructed with the method that summarizes the distribution by three smoothed curves representing the skewness (L curve), the median (M curve), and the coefficient of variation (S curve). Median cutoff levels for BW and for GA according to the 99.95% reference curves were calculated.
Results: Regression analysis showed that GA is a better predictor of 17OHP than BW (R(2) was 50.6 vs. 35.8%, respectively). At a specificity of 99.95%, the calculated median 17OHP cutoff level was lower for GA [12.6 microg/liter (38 nmol/liter)] than for BW [17.6 microg/liter (54 nmol/liter)], thus leading to a greater sensitivity.
Conclusion: This study demonstrates that GA is a better predictor of 17OHP in newborns and will result in greater specificity than BW despite the fact that the determination of GA might be less reliable than BW.