A predictive model of pediatric postoperative pulmonary complications following one-lung ventilation

Abstract

Abstract Since the disparities in respiratory anatomy and physiology between children and adults, children are more susceptible to postoperative pulmonary complications (PPCs) after one-lung ventilation (OLV). Hypothesizing that the incidence of PPCs could be predicted using easily accessible perioperative variables, we aimed to develop a nomogram specifically for children receiving thoracic surgery with OLV. The outcome of this study was the incidence of PPCs. Univariate analysis and the least absolute shrinkage and selection operator regression model were applied to select the most relevant prognostic predictors. Multivariable logistic regression was used to develop a nomogram based on the selected prediction factors. Internal validation was conducted to evaluate its performance. Following screening, a total of 249 children were ultimately included in the study. Among them, 89 (35.7%) presented PPCs. Four predictive factors were ultimately chosen for nomogram development: preoperative neutrophil-to-lymphocyte ratio, intraoperative ventilation mode, maximum peak airway pressure, and minimum oxygenation index during OLV. By incorporating of these four factors, the nomogram achieved an area under the curve of 0.846 [95% confidence interval, 0.797-0.894) with well-fitted calibration curves. In conclusion, this nomogram, based on four factors, predicts pediatric PPCs after OLV, enabling early risk assessment and interventions for better outcomes. This study is registered at the Chinese Clinical Trial Registry (Registration number: ChiCTR2300072042, Date of Registration: 1/6/2023)