Related Experiment Video
Updated: Jan 13, 2026
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Development and validation of a multi-variable prediction model for major postoperative complications after lung
Xiang Li1, Dongze Chen2, Shi Yan1
1Department of Thoracic Surgery II, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital & Institute, Beijing, China.
Background:
Lung cancer predominantly affects elderly patients, in whom curative thoracic surgery is often complicated by potentially fatal postoperative complications. This study aimed to identify preoperative risk factors and develop a prediction model for major postoperative complications (MPCs) to better select elderly patients for lung cancer surgery.
Methods:
We retrospectively reviewed medical records of elderly lung cancer patients treated surgically at Peking University Cancer Hospital from 1995 to 2019. Postoperative MPC occurring within 30 days was rigorously documented and defined according to the Clavien-Dindo grading system. Independent preoperative risk factors of MPC were determined using multivariable logistic regression. Candidate predictors were selected through a two-stage process combining logistic regression with minimization of the Akaike information criterion. Model performance was validated using the area under the receiver operating characteristic curves (AUC), calibration plots, and decision curve analysis (DCA). The model was internally validated using bootstrapping method. A nomogram was also constructed. Additional risk stratification and sensitivity analyses were performed to validate the effectiveness and reliability of the model.
Results:
Among 989 patients enrolled, 6.67% experienced MPC. After adjustment in the multivariable logistic regression analysis, thoracotomy emerged as the strongest independent risk factor for MPC [odds ratio (OR) =4.84, 95% confidence interval (CI): 2.53-9.27]. The prediction model incorporating nine preoperative variables achieved an AUC of 0.815 (95% CI: 0.759-0.871). The final model demonstrated robust discrimination after internal validation (bootstrapped AUC =0.779, 95% CI: 0.723-0.836), and DCA confirmed its clinical utility. Risk stratification analysis revealed a 10.5-fold increase in the incidence of MPC among patients classified as high-risk compared with those at low-risk. Finally, an easy-to-use online tool was developed to potentially assist physicians in the clinic.
Conclusions:
Thoracotomy significantly increased the risk of MPC. This newly developed model provides valuable support for surgical decision-making and facilitates tailored perioperative care strategies for elderly lung cancer patients.
