Development and internal validation of a multivariable nomogram for predicting postoperative complications and prolonged length of hospital stay in patients undergoing video-assisted thoracoscopic surgery: a single-center prospective cohort study

This is the authors' abstract. We don't add key points for this paper.

Objective: To develop and internally validate a nomogram prediction model for postoperative complications and prolonged length of hospital stay in patients undergoing video-assisted thoracoscopic surgery based on the Enhanced Recovery After Surgery concept. Methods: We selected 399 lung cancer patients who underwent thoracoscopic surgery at the Thoracic Surgery Department of the Affiliated Brain Hospital of Nanjing Medical University from June 2022 to April 2023. Postoperative complications and length of stay were recorded. A nomogram prediction model was established using multivariate logistic regression analysis based on optimized features from the LASSO regression model. The model's diagnostic ability was evaluated using the ROC curve and C-index. Results: < 0.05). The nomogram prediction model highlighted abdominal breathing, exercise training, preoperative joint activities, and correct coughing as independent factors. The area under the ROC curve for predicting postoperative complications was 0.833 (95% CI: 0.784-0.936), with a C-index of 0.825. For prolonged hospital stay, the ROC area was 0.856 (95% CI: 0.749-0.941), with a C-index of 0.767. Conclusions: This model has high clinical prediction ability. Nursing interventions based on rapid rehabilitation surgery benefit lung cancer patients post-thoracoscopic surgery by reducing complications and shortening hospital stays. The prediction model, combined with basic clinical data, aids in creating better treatment plans.