Efficacy of Liquid-chromatography and Radioimmunoassay in False-positives’ Drop-off in CAH Newborn Screening

CONTEXT: The main challenge in congenital adrenal hyperplasia newborn screening (NBS) is the high rate of false-positive results, emphasizing the need for specific serum confirmatory tests. Data on the effectiveness of serum 17-hydroxyprogesterone (17OHP) measurements using available methodologies remain limited. OBJECTIVES: To evaluate the efficacy of confirmatory tests using measurements of serum steroids by radioimmunoassay (RIA) and liquid chromatography tandem mass spectrometry (LC-MS/MS) methodologies. DESIGN AND SETTING: A prospective longitudinal cohort study. PATIENTS: Data of 708 437 newborns (NBs) were evaluated. METHODS: Neonatal-17OHP (N17OHP) results ≥ 2 × 99.5th were recalled for serum dosages by RIA (17OHP), by LC-MS/MS (17OHP, 21-deoxycortisol-21DF, androstenedione-Δ4, and cortisol) and positive predictive values (PPV) were determined. Altered hormonal results were submitted to CYP21A2 genotyping. Nonparametric tests and receiver operating characteristic (ROC)-curves were used. RESULTS: Recall rate of N17OHP (first tier) was 0.03% and serum 17OHP levels remained increased in 26% (RIA) and 11% (LC-MS/MS) of NBs, yielding PPV of 30% and 49%, respectively. Fifty-eight NBs were diagnosed with classical forms, and 32 asymptomatic NBs persisted with increased serum 17OHP levels. ROC-curves between false and true-positive tests disclosed highest diagnostic performance for 17OHP/LC-MS/MS [cutoff 48.3 ng/mL (146.3 nmol/L) = 100%-sensitivity]. The (17OHP+Δ4)/cortisol ratio presented the highest discriminative ability (cutoff 24.9 = 100%-PPV); serum 21DF presented a good performance and was comparable to 17OHP measured by LC-MS/MS (99%-specificity). CONCLUSION: Measurement of 17OHP by LC-MS/MS provides the best performance for serum confirmatory tests and may be complemented by the (17OHP+Δ4)/cortisol ratio analysis. Genotyping is useful mainly to resolve persistently indeterminate biochemical results.