ABSTRACT
BACKGROUND: In Korea, 17-alpha-hydroxyprogesterone (17-OHP) neonatal screening for congenital adrenal hyperplasia (CAH) has a high false positive rate. Preterm infants have higher levels of 17-OHP than term infants. We established the separate cutoff values of 17-OHP under the guideline of the Clinical and Laboratory Standard Institute C28-A3 to reduce a false positive rate. METHODS: The 17-OHP enzyme-immunoassay was used in blood spots of 22,601 newborns. To decide whether to partition cutoff values based on sex, sampling date and birth weight was assessed by Z-test and standard deviation (SD) ratio. If the result was significant, we estimated the cutoff value with 90% confidence intervals (CIs) using the nonparametric method. RESULTS: In the subclasses based on sex and sampling date, the results were not significant. However, the birth weight-adjusted subclasses (SD ratio > 1.5) showed that it was necessary to distinguish low-birth-weight infants from the others. We selected the subclass categories to reflect the concept of low- or very-low-birth-weight infant. The maximum percentile to define a 90% CI was chosen in each subclass. After applied the re-estimated cutoff value, the recall rate was decreased from 0.6% to less than 0.2%. CONCLUSIONS: The birth weight-adjusted cutoff value of 17-OHP in neonatal screening for CAH can be reduced the false positive rate of low-birth-weight infants. This approach would decrease unnecessary blood draws, medical evaluation, parental anxiety and burden on health care resources.