Beyond the Screening Cut-off: Measurement Drift, Predictors and Consanguinity in Borderline Neonatal Phenylalanine Among More Than 500,000 Newborns
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Background Newborn screening identifies many infants with phenylalanine above the upper limit of normal (2 mg/dL; 120 µmol/L) but below the treatment threshold (6 mg/dL; 360 µmol/L). This borderline group is far larger than the treatment-range group but has rarely been characterized in unselected populations. Methods We analyzed a regional registry (Mashhad University of Medical Sciences, northeastern Iran) of 543,952 children with a newborn screening phenylalanine value, classified as normal (< 2 mg/dL), borderline (2 to < 6 mg/dL) or treatment-range (≥ 6 mg/dL). We assessed year-to-year measurement drift, identified predictors of borderline status using logistic regression adjusted for birth year and repeated the analyses after harmonizing values across years, compared associations with parental consanguinity between categories, and examined weight- and length-for-age at birth. Results Borderline values were recorded in 10.5% of children, but this proportion ranged from 1.1% to 25.5% across birth years because of a parallel upward shift in the whole phenylalanine distribution that was not explained by child characteristics and was absent for TSH measured in the same blood spot; after harmonization, 1.1% of children had borderline values. Treatment-range values occurred in 63 children (11.6 per 100,000). Preterm birth (OR 1.20, 95% CI 1.15–1.26) predicted borderline status under every definition, and female sex (OR 1.09, 1.07–1.12) under all definitions covering the whole study period. Parental consanguinity was not associated with borderline status (OR 1.00, 0.98–1.03) but was associated with treatment-range values (OR 3.19, 1.92–5.30; difference p < 0.001), including after harmonization. Borderline phenylalanine was associated with slightly lower weight-for-age at birth, also in term singletons (− 0.06 SD, 95% CI − 0.07 to − 0.05). Conclusions Much of the apparent borderline population in this registry reflected measurement drift rather than differences between children. After accounting for it, borderline elevation had modest perinatal predictors and no association with consanguinity, unlike treatment-range hyperphenylalaninemia, suggesting that the two groups are largely distinct. Screening programs should monitor analytical drift, and registry-based studies should account for it.
Source: Research Square
Author: Veda Vakili
Institution: Mashhad University of Medical Sciences
Language: English
Subfield: Clinical Biochemistry
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7225912841
