Construction of risk factors and nomogram model related to thoracoscopic postoperative complications of stage I non-small cell lung cancer

The aim of this study is to assess the risk factors associated with thoracoscopic postoperative complications in patients with stage I non-small cell lung cancer (NSCLC) and to develop a preoperative predictive model for evaluating the risk of overall postoperative complications, cardiopulmonary complications, and long-term pulmonary air leakage using an intuitive nomogram model. A retrospective analysis was conducted on clinical data of patients diagnosed with stage I NSCLC who underwent thoracoscopic surgery at our hospital from 2017 to 2022, focusing on risk factors for postoperative complications in this cohort. Develop a nomogram model and assess its accuracy and predictive performance. A total of 694 patients were included in the study, and 113 (16.3%) developed complications. Logistic regression analysis showed that gender (P = 0.047), history of respiratory disease (P = 0.004), thoracic adhesion (P = 0.016), operative time (P = 0.045), and ALI (P = 0.031) were significant risk factors for postoperative complications. Subgroup analysis showed that risk factors for cardiopulmonary complications included gender (P = 0.026), history of respiratory disease (P = 0.003), thoracic adhesion (P = 0.025), and ALI (P = 0.049). Risk factors for prolonged air leakage included type of surgical procedure (P = 0.012), thoracic adhesion (P = 0.026), and ALI (P = 0.036). Based on the above results, the AUC of the overall complications, cardiopulmonary complications, and long-term lung leakage were 0.721, 0.743, and 0.830, respectively. The calibration curve and DCA curve show that the model has good predictive performance and clinical application value. This study investigated the prediction model for thoracoscopic postoperative complications in patients with stage I NSCLC, conducted a subgroup analysis, and produced nomogram models for Cardiopulmonary complications and prolonged pulmonary air leakage. The dependability and prediction efficacy of the three models were assessed. The model demonstrates good predictive performance within this single-center cohort and may have clinical application value.