Background: Neonatal screening programs for congenital adrenal hyperplasia (21-CAH) using an immunoassay for 17-hydroxyprogesterone (17-OHP) generate a high rate of positive results due to physiological reasons and to cross-reactions with steroids other than 17-hydroxyprogesterone especially in preterm neonates and in critically ill newborns.

Methods: In order to increase the specificity of the screening process we applied an LC-MS/MS method quantifying 17-hydroxyprogesterone, 11-deoxycortisol, 21-deoxycortisol, cortisol and androstenedione. The steroids were eluted in aqueous solution containing d₈ 17-hydroxyprogesterone and d₂ cortisol and quantified in multiple reaction mode.

Results: Detection limit was below 1 nmol/L, recovery ranged from 64% (androstenedione) to 83% (cortisol). Linearity was proven within a range of 5-100 nmol/L (cortisol: 12.5-200 nmol/L), total run time was 6 min. Retrospective analysis of 6,151 blood samples and 50 blood samples from newborns with clinically confirmed 21-CAH, as well as prospective analysis of 1,609 samples out of a total of 242,500 testing positive in our routine 17-OHP immunoassay allowed clear distinction of affected and non affected newborns. High levels of 21-deoxycortisol were only found in children with 21-hydroxylase deficiency. Calculating the ratio of 17-hydroxyprogesterone + 21-deoxycortisol divided by cortisol further increased the sensitivity of the method.

Conclusion: Our LC-MS/MS procedure as a second-tier test can be used to reduce false positive results of standard 21-CAH screening. The short total run time of 6 minutes allows for immediate reanalysis of all immunoassay results above the cut-off.