ESPE Abstracts (2016) 86 P-P1-552

Congenital Adrenal Hyperplasia Newborn Screening: Improving the Effectiveness of the Neonatal 17OH-Progesterone and Serum Confirmatory Tests

Daniel Carvalhoa, Giselle Hayashia,b, Mirela Mirandaa, Helena Valassia, Atecla Alvesa, Andresa Rodriguesa, Larissa Gomesa, Guiomar Madureiraa, Berenice Mendoncaa & Tânia Bachegaa


aLaboratório de Hormônios e Genética Molecular LIM42, Disciplina de Endocrinologia, Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo, São Paulo, SP, Brazil; bLaboratório da APAE SÃO PAULO, São Paulo, SP, Brazil


Background: Main concerns of Congenital Adrenal Hyperplasia Newborn Screening (CAH-NBS) are the high false-positive results (FPR) rate, low positive predictive value (PPV) and heterogeneity of confirmatory tests. Considering the CAH-NBS implementation in our country, our.

Objectives: Objectives are to optimize the Neonatal 17OH-Progesterone (N17OHP) cutoffs and to evaluate the best serum confirmatory test.

Methods: 473 983 newborns were evaluated; N17OHP was measured by IFMA (AutoDelfia) and cutoffs (99th and 99.8th) adjusted to birth-weight (BW1: <1500 g; BW2: 1500–2000 g; BW3: 2001–2500 g; BW4:>2500 g), and to age at sample collection (before/after 72 h of life). For confirmatory tests, 17OHP (RIA and LC-MS/MS) and 21-deoxicortisol (21DF), Δ4 and cortisol (LC-MS/MS) were analyzed. Asymptomatic newborns with persistently increased 17OHP levels had the CYP21A2 gene sequenced.

Results: The recall rate was 0.05% using the P99th of N17OHP levels and 0.03% using the P99.8th; PPV increased from 11% (P99th) to 17% (P99.8th). N17OHP cutoffs in samples collected earlier (<72 h) were significantly lower than those collected later; consequently, different N17OHP cutoffs according to BW and age were determined. Twenty-six newborns were diagnosed (22SW/12M), confirmed by sequencing. Confirmatory tests were performed in 149 newborns and FPR persisted in 70% (by RIA) and 13% (LC-MS/MS); PPV of LC-MS/MS methodology was significantly higher than RIA (52 vs 27%). Serum 21DF and steroid ratios (17OHP/cortisol; 17OHP+Δ4/cortisol; 17OHP+21DF/cortisol) presented similar FPR and PPV values in comparison to 17OHP by LC-MS/MS. Among asymptomatic newborns with persistently increased 17OHP levels, genotype identified 2 with NC-form.

Conclusion: N17OHP levels adjusted to P99.8th (birth-weight and age) improved the CAH-NBS by reducing the FPR rate without missing the classical form diagnosis. Although serum 17OHP by RIA is widely used as confirmatory test in our country, 17OHP dosage by LC-MS/MS significantly reduced recall rate. The 21DF and steroid ratio measurements did not provide higher accuracy than serum 17OHP by LC-MS/MS. Molecular analysis could be restricted for asymptomatic newborns with persistently increased 17OHP levels.