A nomogram to predict the risk of death during hospitalization in Chinese neonates with respiratory failure

Background: Neonatal respiratory failure (NRF) is a critical condition with high morbidity and mortality rates. This study aimed to develop a nomogram prediction model to early predict the risk of death in Chinese neonates with NRF. Methods: A retrospective analysis was conducted on NRF neonates from 21 tertiary neonatal intensive care units (NICUs) across 13 prefecture-level cities in Jiangsu Province, China, from March 2019 to March 2022. NRF neonates from one random NICU were selected as the external validation set, while those from the remaining 20 NICUs were divided into the training set and the internal validation set at a 7:3 ratio. Death was the primary outcome. LASSO regression and multivariate logistic regression were used to identify the predictive factors from the training set and then the nomogram was constructed. Results: A total of 5387 neonates with NRF were included in the analysis. Among them, 3444 were in the training set, 1470 were in the internal validation set, and 473 were in the external validation set. The nomogram was constructed based on the eight predictors of the 1-min Apgar score, birth weight, gestational age, the relationship between birth weight and gestational age, mode of first respiratory support, inhaled nitric oxide, antenatal corticosteroids, and vasoactive drugs. The area under the curve of the nomogram in the training set, internal validation set, and external validation set was 0.763, 0.733, and 0.891, respectively. The P-values of the Hosmer-Lemeshow goodness of fit test were 0.638, 0.273, and 0.253, respectively. Brier scores were 0.066, 0.072, and 0.037, respectively. The decision curve analysis demonstrated a significant net benefit in all cases. These data indicate the good performance of the nomogram. Conclusions: This nomogram can serve as a reference for clinicians to identify high-risk neonates early and reduce the incidence of neonatal mortality.