Estimation of the false-negative rate in newborn screening for congenital adrenal hyperplasia

Abstract

Newborn screening based on measurement of 17alpha-hydroxyprogesterone (17-OHP) in a dried blood spot on filter paper is an effective tool for early diagnosis of congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency. Its most important rationale is prevention of a life-threatening salt-wasting (SW) crisis; in moderate forms of CAH, early diagnosis and treatment may prevent permanent negative effects of androgen overproduction. Our target was to analyse if all CAH patients who had been identified clinically before puberty would have been detected by the newborn screening. Newborn screening cards of 110 CAH patients born between 1988 and 2000 in five Middle-European countries and diagnosed prior to puberty (77 SW and 33 moderate) and cards from 920 random, healthy newborn controls were analysed. CAH screening had not yet been introduced during this time. The diagnosis was based on clinical and laboratory signs and, in most cases, on CYP21 gene mutation analysis. All 17-OHP measurements in dried blood spots were carried out using a time-resolved fluoroimmunoassay kit. In the newborn screening blood spots, the median of 17-OHP levels was 561 nmol/l (range 91-1404 nmol/l) in subjects with the SW form and 40 nmol/l (4-247 nmol/l) in the moderate form. All 77 SW patients would have been detected by newborn screening using the recommended cut-off limits (30 nmol/l). However, 10 of 33 patients with moderate CAH would have been missed. 17-OHP levels of all controls were below the cut-off. Newborn screening is efficient for diagnosing the SW form of CAH, but is inappropriate for identifying all patients with a moderate form of CAH. It appears that the false-negative rate is at least one-third in children with the moderate form of CAH.