A Prediction Model for the Length of Stay after Single-Port Thoracoscopic surgery of lung cancer:Based on Preoperative Physical Activity Level

Background With 2.2 million new cases annually, lung cancer remains the leading cause of cancer mortality globally. Preoperative physical activity (PA) may optimize postoperative recovery, but its role in predicting Length of stay (LOS) after VATS is understudied. Methods The clinical data of 231 patients with lung cancer who underwent surgery in the Department of Thoracic Surgery of National Cancer Center from August 1, 2023, to April 1, 2024 were prospectively collected. The preoperative physical activity levels were assessed by the International Physical Activity Questionnaire (IPAQ), the residence, operation duration and complications were recorded. Binary logistic regression with bootstrapping (231 resamples) identified predictors of LOS, adjusting for age, sex, and comorbidities. Model performance was assessed via ROC analysis (AUC). Results A total of 231 lung cancer patients were enrolled, including 90 males and 141 females, with an average age of 53.1 years. The median preoperative physical activity value (MET) was 1086 (rang: 0-3312) and the mean LOS was 3.69 (1.58). Preoperative PA ( OR: 7.98, 95% CI: 1.65–38.64 ), urban residence ( OR: 4.01, 95% CI: 1.81–8.89 ), shorter operation time ( OR: 0.40, 95% CI : 0.18–0.91) and complications ( OR: 0.32, 95% CI: 0.12–0.86 ) independently predicted reduced LOS (all P < 0.05 ). The model achieved an AUC of 0.85 ( 95% CI: 0.79–0.91) , the sensitivity was 80.0% and the specificity was 74.9%. Conclusion Preoperative PA is a modifiable predictor of LOSin patients with lung cancer. Integration of PA assessment into prehabilitation programs may optimize resource allocation and recovery pathways.