Altered metabolites in newborns with persistent pulmonary hypertension

• Published in: Pediatric research 2018-08, Vol.84 (2), p.272-278
• Main Authors: Steurer, Martina A., Oltman, Scott, Baer, Rebecca J., Feuer, Sky, Liang, Liang, Paynter, Randi A., Rand, Larry, Ryckman, Kelli K., Keller, Roberta L., Jelliffe-Pawlowski, Laura L.
• Format: Article
• Language: English
• Subjects: Area Under Curve; Birth Weight; Case-Control Studies; Female; Gestational Age; Humans; Infant, Newborn; Logistic Models; Male; Medicine; Medicine & Public Health; Metabolism; Metabolites; Multivariate Analysis; Neonatal Screening; Newborn babies; Ornithine - blood; Pediatric Surgery; Pediatrics; Persistent Fetal Circulation Syndrome - blood; Persistent Fetal Circulation Syndrome - physiopathology; Phenylalanine - blood; Pulmonary hypertension; Respiration, Artificial; Thyrotropin - blood; Treatment Outcome; Tyrosine - blood; Ventilation
• ISSN: 0031-3998; 1530-0447
• DOI: 10.1038/s41390-018-0023-y

Background There is an emerging evidence that pulmonary hypertension is associated with amino acid, carnitine, and thyroid hormone aberrations. We aimed to characterize metabolic profiles measured by the newborn screen (NBS) in infants with persistent pulmonary hypertension of the newborn (PPHN) Methods Nested case–control study from population-based database. Cases were infants with ICD-9 code for PPHN receiving mechanical ventilation. Controls receiving mechanical ventilation were matched 2:1 for gestational age, sex, birth weight, parenteral nutrition administration, and age at NBS collection. Infants were divided into derivation and validation datasets. A multivariable logistic regression model was derived from candidate metabolites, and the area under the receiver operator characteristic curve (AUROC) was generated from the validation dataset. Results We identified 1076 cases and 2152 controls. Four metabolites remained in the final model. Ornithine (OR 0.32, CI 0.26–0.41), tyrosine (OR 0.48, CI 0.40–0.58), and TSH 0.50 (0.45–0.55) were associated with decreased odds of PPHN; phenylalanine was associated with increased odds of PPHN (OR 4.74, CI 3.25–6.90). The AUROC was 0.772 (CI 0.737–0.807). Conclusions In a large, population-based dataset, infants with PPHN have distinct, early metabolic profiles. These data provide insight into the pathophysiology of PPHN, identifying potential therapeutic targets and novel biomarkers to assess the response.