Risk prediction model for long-term atelectasis in children with pneumonia

BACKGROUND: This study aimed to develop a risk prediction model for long-term atelectasis in children with pneumonia. METHODS: A retrospective study of 532 children with atelectasis was performed at the Children’s Hospital of Chongqing Medical University from February 2017 to March 2020. The predictive variables were screened by LASSO regression analysis and the nomogram was drawn by R software. The area under the Receiver Operating Characteristic (ROC) curve, calibration chart and decision curve were used to evaluate the predictive accuracy and clinical utility. 1000 Bootstrap resampling was used for internal verification. RESULTS: Multivariate logistic regression analysis showed that clinical course before bronchoscopy, length of stay, bronchial mucus plug formation, age were independent risk factors for long-term atelectasis in children. The area under the ROC curve of nomogram was 0.857(95% CI = 0.8136 ~ 0.9006) in training set and 0.849(95% CI = 0.7848–0.9132) in the testing set. The calibration curve demonstrated that the nomogram was well-fitted, and decision curve analysis (DCA) showed that the nomogram had good clinical utility. CONCLUSIONS: The model based on the risk factors of long-term atelectasis in children with pneumonia has good predictive accuracy and consistency, which can provide a certain reference value for clinical prevention and treatment of long-term atelectasis in children.