Construction of a predictive model for prolonged length of stay in patients undergoing non-intubated thoracoscopic resection of lung cancer

Background: Lung cancer is among the most prevalent and lethal malignancies worldwide. Non-intubated video-assisted thoracoscopic surgery (VATS) has demonstrated advantages in reducing hospital length of stay (LOS). However, clinical practice indicates that a substantial proportion of patients still experience prolonged length of stay (PLOS). Currently, no risk prediction model exists specifically for PLOS following non-intubated VATS in lung cancer patients. This study aims to analyze clinical data to identify risk factors associated with PLOS and to develop a predictive model. Methods: -tests in univariate analyses. Variables with statistical significance in univariate analysis were entered into multivariable logistic regression to identify independent predictors and construct the prediction model. A nomogram was created for visualization. Model discrimination was assessed using the area under the receiver operating characteristic (ROC) curve, and calibration was evaluated with calibration plots. Results: was protective (OR =0.043). The resulting predictive nomogram demonstrated excellent discrimination with an area under the curve (AUC) of 0.943. Conclusions: This prediction model shows robust accuracy in identifying lung cancer patients at high risk of PLOS after non-intubated VATS. It provides a theoretical basis for early identification and timely intervention by clinical staff.