AUTHOR=Peng Gang , Tang Yishuo , Cowan Tina M. , Zhao Hongyu , Scharfe Curt TITLE=Timing of Newborn Blood Collection Alters Metabolic Disease Screening Performance JOURNAL=Frontiers in Pediatrics VOLUME=8 YEAR=2021 URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2020.623184 DOI=10.3389/fped.2020.623184 ISSN=2296-2360 ABSTRACT=

Blood collection for newborn genetic disease screening is preferably performed within 24–48 h after birth. We used population-level newborn screening (NBS) data to study early postnatal metabolic changes and whether timing of blood collection could impact screening performance. Newborns were grouped based on their reported age at blood collection (AaBC) into early (12–23 h), standard (24–48 h), and late (49–168 h) collection groups. Metabolic marker levels were compared between the groups using effect size analysis, which controlled for group size differences and influence from the clinical variables of birth weight and gestational age. Metabolite level differences identified between groups were correlated to NBS data from false-positive cases for inborn metabolic disorders including carnitine transport defect (CTD), isovaleric acidemia (IVA), methylmalonic acidemia (MMA), and phenylketonuria (PKU). Our results showed that 56% of the metabolites had AaBC-related differences, which included metabolites with either decreasing or increasing levels after birth. Compared to the standard group, the early-collection group had elevated marker levels for PKU (phenylalanine, Cohen's d = 0.55), IVA (C5, Cohen's d = 0.24), MMA (C3, Cohen's d = 0.23), and CTD (C0, Cohen's d = 0.23). These findings correlated with higher false-positive rates for PKU (P < 0.05), IVA (P < 0.05), and MMA (P < 0.001), and lower false-positive rate for CTD (P < 0.001) in the early-collection group. Blood collection before 24 h could affect screening performance for some metabolic disorders. We have developed web-based tools integrating AaBC and other variables for interpretive analysis of screening data.