Ning Zhou,1,* Nan Feng,2,* Zichen Jiao,3 Xiaoming Shi,1 Tao Wang,1,3 Gefei Zhao3 

1Department of Thoracic Surgery, Nanjing Drum Tower Clinical College of Nanjing Medical University, Nanjing, People’s Republic of China; 2Department of Radiology, Nanjing Drum Tower Hospital, Medical School, Nanjing University, Nanjing, People’s Republic of China; 3Department of Thoracic Surgery, Nanjing Drum Tower Hospital, Medical School, Nanjing University, Nanjing, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Gefei Zhao, Department of Thoracic Surgery, Nanjing Drum Tower Hospital, No. 359 Puzhu Middle Road, Pukou District, Nanjing, Jiangsu, 210000, People’s Republic of China, Email zgf6160@sina.com Tao Wang, Department of Thoracic Surgery, Nanjing Drum Tower Hospital, No. 359 Puzhu Middle Road, Pukou District, Nanjing, Jiangsu, 210000, People’s Republic of China, Email wangtao_pumc@live.cn

Background: Pleural reaction (PR) frequently occurs during computed tomography (CT) -guided lung puncture procedures, and its development is influenced by various factors. This study aims to identify the risk factors associated with PR in CT-guided percutaneous lung nodule localization (CT-PLNL) procedures.
Methods: This retrospective study included 467 patients who underwent video-assisted thoracic surgery (VATS) at Nanjing Drum Tower Hospital between January 2022 and December 2023, all of whom had received CT-PLNL. Clinical data, including medical records, imaging findings, and laboratory results, were collected. Univariate analysis and Least Absolute Shrinkage and Selection Operator (LASSO) regression identified independent risk factors for PR. Binary logistic regression was performed to further analyze these factors. Receiver Operating Characteristic (ROC) curves were plotted to assess model performance, and Bootstrap validation evaluated discriminative ability. Calibration curves and decision curve analysis (DCA) were conducted to compare predicted versus actual probabilities and assess clinical applicability.
Results: The incidence of PR was 5.35% (25/467). Significant variables from univariate analysis and LASSO regression were analyzed by logistic regression. Age, intrapulmonary needle path adjustment, inadequate anesthesia, and a history of diabetes were identified as independent risk factors for PR. ROC curves showed Area Under the Curve (AUC) values indicating excellent discriminative ability. Calibration curves showed appropriate fit, and DCA demonstrated high clinical applicability.
Conclusion: Younger age groups, intraprocedural needle adjustments, inadequate anesthesia, and diabetes were independent risk factors for PR after CT-PLNL. Optimizing anesthesia, avoiding unnecessary needle manipulations, and perioperative glucose monitoring in diabetic patients may mitigate PR risks and enhance procedural safety.

Keywords: pleural reaction, lung nodule, puncture localization, risk factors, LASSO