Development, validation and visualization of risk prediction model for postoperative shivering in patients undergoing video-assisted thoracoscopic lobectomy: a real-world retrospective study

To develop, validate, and visualize a risk prediction model for postoperative shivering in patients undergoing video-assisted thoracoscopic (VATS) lobectomy, addressing the lack of individualized tools for this high-risk population. This retrospective study analyzed 530 patients undergoing VATS lobectomy from a tertiary hospital in Wuhan (January 2022-December 2023). The data were randomly divided into training set and validation set at a ratio of 7∶3. Patients were stratified into postoperative shivering (n = 198) and non-postoperative shivering (n = 332) groups based on Bedside Shivering Assessment Scale (BSAS) criteria. Logistic regression identified independent risk factors, and a nomograph was developed. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, Hosmer-Lemeshow test, and decision curve analysis (DCA). An online visualization tool was created for clinical implementation. Postoperative shivering occurred in 198 of 530 patients undergoing VATS lobectomy (37.36%). The logistic regression analysis identified age < 60 years, intraoperative hypothermia, delayed anesthesia recovery, absence of postoperative analgesia, operation duration > 180 min and lower preoperative temperature as significant risk factors for postoperative shivering (P < 0.05). The resulting nomograph demonstrated strong discrimination with AUCs of 0.847 (95%CI: 0.809-0.885) in the training cohort (n = 424) and 0.836 (95%CI: 0.747-0.925) in validation (n = 106), while calibration curves and the Hosmer-Lemeshow test (χ² = 13.123, P = 0.108) confirmed model reliability. DCA demonstrated clinical utility across threshold probabilities of 0.20-0.98. In order to promote clinical implementation, deployment of the visualization tools online ( https://shivering.shinyapps.io/dynnomapp/ ), which supports dynamic risk assessment. This study established the first risk prediction model for postoperative shivering in patients undergoing VATS lobectomy, integrating six perioperative variables into a clinically applicable nomograph and online tool. The model facilitates identification of high-risk patients, enabling targeted interventions to mitigate shivering-related complications.