Department of Respiratory Medicine, The Second Affiliated Hospital of Anhui Medical University, Hefei, Anhui230032, China , medicalpaper203@gmail.com
Abstract: (30 Views)
Background:To investigate the independent risk factors of pneumothorax after CT-guided pulmonary puncture, and to construct a nomogram model to predict the occurrence of pneumothorax. Materials and Methods The relevant factors of 257 patients with pulmonary nodule puncture in the lung tumor ward of a tertiary hospital from 2021 to 2023 were collected, and the logistic regression analysis method of backward elimination was used to obtain independent predictors. Finally, the multivariate logistic regression results were used. By drawing forest charts, exploreing independent risk factors for pulmonary puncture complicated by pneumothorax, risk prediction models were established and general nomograms and develop more intuitive multivariate dynamic nomograms were constructed to evaluate the model by calculating the AUC value of the area under the curve through the ROC curve performance; and to use Bootstrap's internal resampling method to perform internal validation of the model. Results Multivariate logistic regression analysis showed that lung nodule size (every 1 mm increase) (OR=0.82), education (OR=2.73) and age (OR=1.08) were independent risk factors for pneumothorax after lung puncture. Conclusion The nomogram constructed in this study can effectively predict the occurrence of postoperative pneumothorax in patients undergoing CT-guided pulmonary puncture. The model has a good degree of discrimination and consistency, and clinical medical staff can accurately and quickly predict the nomogram through pulmonary puncture complicated with pneumothorax, quickly identify patients with high risk of pulmonary puncture complicated with pneumothorax, and provide a reference for the development of targeted intervention measures.