AUTHOR=Wang Chenhong , Ma Xiaolu , Xu Yanping , Chen Zheng , Shi Liping , Du Lizhong 

TITLE=A prediction model of pulmonary hypertension in preterm infants with bronchopulmonary dysplasia

JOURNAL=Frontiers in Pediatrics

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2022.925312

DOI=10.3389/fped.2022.925312

ISSN=2296-2360

ABSTRACT=Objective: Pulmonary hypertension (PH) is a severe cardiovascular complication of bronchopulmonary dysplasia (BPD) that contributes to the high mortality rates for preterm infants. The objective of this study is to establish a prediction model of BPD associated PH (BPD-PH) by integrating multiple predictive factors for infants with BPD. 
Method: A retrospective investigation of the perinatal clinical records and data of echocardiography in all the preterm infants with BPD was performed from January 2012 to December 2019. A prediction model of BPD-PH was established based on the univariate and multivariate logistic regression analysis of the clinical data, and evaluated by using area under receiver operating characteristic curve (ROC), Hosmer-Lemeshow test. Internal validation was performed with bootstrap resampling.
Result: A total of 268 infants with BPD were divided into BPD-PH group and no-PH group. Multivariate logistic regression analysis showed that the independent predictive factors of BPD-PH were moderate to severe BPD, small for gestational age, duration of hemodynamically significant patent ductus arteriosus ≥28d and early PH. A prediction model was established based on the β coefficients of the four predictors. Area under ROC of the prediction model was 0.930. Hosmer-Lemeshow test (P=0.976) and calibration curve showed good calibration. 
Conclusion: The prediction model based on the four risk factors predicts the development of BPD-PH with high sensitivity and specificity, and might help clinicians to make individualized intervention to minimize the disease risk.