Prediction of respiratory complications after extensive chest wall resections

Objective. To analyze the incidence of respiratory complications following chest wall surgeries with reconstructive approaches and to develop a nomogram for predicting complication risk. Material and methods. From 2013 to 2024, a multicenter ambispective study was conducted including 125 patients who underwent chest wall resection for various indications. Results. The rate of respiratory complications after chest wall resection was 15.2%. Logistic regression analysis identified three independent prognostic factors: intraoperative blood loss (OR=1.006), number of resected ribs (OR=1.561), and FEV1 level (OR=0.935; reciprocal OR=1.069), highlighting their contribution to complication risk. The regression model was statistically significant (p<0.001) and demonstrated high discriminatory ability. The area under the ROC curve was 0.895±0.049 (95% CI 0.798—0.992), with sensitivity of 94.7% and specificity of 78.3%. The optimal cut-off value based on the Youden index was 0.138, above which the risk of respiratory complications was considered high. Nagelkerke’s pseudo-R² was 78.1%, indicating strong explanatory power. Based on these results, a nomogram was developed using machine learning methods to predict respiratory complication risk. Validation confirmed the model’s reliability and clinical applicability for individualized risk assessment. Conclusion. This study identified blood loss, number of resected ribs, and FEV1 as key prognostic factors for respiratory complications after chest wall reconstruction. A predictive nomogram based on logistic regression was developed for clinical use.