Multiple 17-OHP Cutoff Co-Variates Fail to Improve 21-Hydroxylase Deficiency Screening Accuracy

Abstract

To improve the positive predictive value (PPV) of newborn screening for 21-hydroxylase deficiency (21OHD), co-variates have been used to modify 17-hydroxyprogesterone (17OHP) cutoffs. The objective of this study is to evaluate whether 17OHP screening cutoffs adjusted for both collection time (CT) and birth weight (BW) improved the sensitivity and PPV of 21OHD screening. Unaffected newborn screening samples were stratified based on BW and CT to establish 17OHP concentration cutoffs at the 95th and 99th percentile. These cutoffs were applied to a cohort of confirmed cases of 21OHD to determine the sensitivity and PPV of the modified screening parameters. 17OHP cutoffs at the 99th percentile, adjusted for BW and CT, had a sensitivity of 96.3% and a specificity of 98.9%, but a relatively low PPV (0.130) for the identification of 21OHD and did not detect all cases. Use of the 95th percentile further increased sensitivity to 98.1% but resulted in a notably lower PPV (0.027). Alternative approaches that do not rely exclusively on 17OHP are needed to improve newborn screening accuracy for 21OHD.