Predictive potential of age‐group cut‐off values of N‐terminal pro‐brain natriuretic peptide in Kawasaki disease

• Published in: Pediatrics International Vol. 64; no. 1; pp. e15371 - n/a
• Main Authors: Hirai, Seiko, Nakamura, Takahiro, Misawa, Masahiro
• Format: Journal Article
• Language: English
• Published: Australia Wiley 01.01.2022; Blackwell Publishing Ltd
• Subjects: Age; Biomarkers; Brain natriuretic peptide; Coronary artery; coronary artery lesion; Electronic medical records; Humans; Kawasaki disease; Mucocutaneous Lymph Node Syndrome; Mucocutaneous Lymph Node Syndrome - complications; Natriuretic Peptide, Brain; N‐terminal pro‐brain natriuretic peptide; Patients; Pediatrics; Peptide Fragments; Peptides; Regression analysis; Retrospective Studies; ROC Curve; Treatment resistance; N-terminal pro-brain natriuretic peptide; coronary artery lesion; Kawasaki disease
• ISSN: 1328-8067; 1442-200X; 1442-200X
• DOI: 10.1111/ped.15371

Background The relationship between serum N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) and the clinical features of patients with Kawasaki disease (KD) has been the subject of research. Recent studies have revealed that serum NT‐proBNP levels vary with age. We therefore aimed to determine the utility of age‐stratified cut‐off values for NT‐proBNP in predicting coronary artery lesions (CALs) in patients with KD. Methods We retrospectively assessed the electronic medical records of patients who were hospitalized for KD or incomplete KD between January 1, 2015, and August 31, 2019. The patients were divided into high and normal NT‐proBNP groups using age‐stratified cut‐off based on serum NT‐proBNP levels measured immediately before KD treatment initiation. Results The study comprised 242 cases, including 71 and 171 cases in high and normal NT‐proBNP groups, respectively. Thirty‐seven of them (15.3%) were resistant to treatment; 15 (6.2%) had CALs, with a higher incidence in the high NT‐proBNP group than in the normal NT‐proBNP group. On multivariate logistic regression analysis, high serum NT‐proBNP levels were significantly correlated to CALs (OR, 9.76; 95% CI, 2.64–36.2). On logistic regression analysis to compare the predictive accuracy of the age‐stratified and fixed cut‐off for CALs, the age‐stratified cut‐off values showed a larger area under the receiver operating characteristic curve than fixed cut‐off value. Conclusions Based on age‐stratified cut‐off values for serum NT‐proBNP, high NT‐proBNP levels at the time of diagnosis were significantly associated with CALs in patients with KD. Higher predictive accuracy for CALs of the age‐stratified cut‐off values was also suggested.