| Sumario: | Objective: To analyze the risk factors of lung cancer complicated with atelectasis after thoracoscopic segmental resection, and to construct a prediction model. Methods: 585 patients with lung cancer who underwent thoracoscopic segmental resection in our hospital from February 2021 to February 2023 were retrospectively selected as the study objects, and they were divided into atelectasis group and non-atelectasis group according to whether they developed atelectasis after surgery. Clinical data of patients with lung cancer were collected, and multi-factor Logistic regression was used to screen the risk factors affecting atelectasis after thoracoscopic segmental resection of lung cancer. R software was used to establish a random forest model for predicting atelectasis, and the effectiveness of the model was verified. Results: The incidence of atelectasis was 8. 55% in 585 patients with lung cancer after thoracoscopic segmental resection. Multivariate results showed that BMI≥24 kg/m², smoking history, underlying diseases, poor preoperative pulmonary function, operative time ≥2 h and adenocarcinoma were independent risk factors for atatasis after thoracoscopic lung resection (P<C0. 05). The relative important predictors of lung atatasis after thoracoscopic segmental resection in random forest were preoperative lung function, BMI, smoking history, operation time, underlying disease, and pathological type of tumor. ROC results showed that the AUC of random forest algorithm in predicting the occurrence of atatasis was slightly higher than that of multivariate Logistic regression model(0. 841 vs 0. 834). Conclusion: BMI ≥24 kg/m², smoking history, underlying diseases, poor preoperative pulmonary function, operative time ≥2 h, and adenocarcarcinoma are independent risk factors for atectasis after thoracoscopic segmental resection. Based on these factors, the random forest model for predicting atectasis has good risk prediction efficacy.
|
|---|