Ding Tingting, Hao Shanhu, Wang Zhiguo, Zhang Wenwen, Zhang Guoxu
Northern Theater Command General Hospital, Shenyang, China.
Front Med (Lausanne). 2024 Oct 16;11:1428456. doi: 10.3389/fmed.2024.1428456. eCollection 2024.
To analyze the risk factors for pneumothorax after particle implantation in the treatment of advanced lung cancer and to construct and validate a nomogram prediction model.
A retrospective analysis was conducted on 148 patients who underwent I particle implantation for advanced lung cancer at the *** from December 2022 to December 2023. Potential risk factors were identified using univariate logistic regression analysis, followed by a multivariate logistic regression analysis to evaluate the predictive factors for pneumothorax. Interaction effects between variables were studied and incorporated into the model construction. ROC curves and nomograms were generated for visualization. Calibration analysis was performed, and the corresponding net benefit was calculated to adjust the predictive model.
Among the 148 patients, 58 (39.19%) experienced pneumothorax, with a mean age of 62.5 (55.25, 70) years. Multivariate analysis showed that the angle between the puncture needle and the pleura < 50° ( = 0.002, : 3.908, : 1.621-9.422), preoperative CT suggesting emphysema ( = 0.002, : 3.798, : 1.600-9.016), atelectasis ( = 0.009, : 3.156, : 1.331-7.481), and lesion located in the left lung fissure ( = 0.008, : 4.675, : 14.683) were independent risk factors for pneumothorax after particle implantation in the treatment of advanced lung cancer. Preoperative CT suggesting lesions in the left lung fissure or suggesting emphysema had a significant impact in the nomogram, with probabilities of pneumothorax occurrence at 40% and 38%, respectively. The predictive AUC for the above four risk factors for pneumothorax after particle implantation in the treatment of lung adenocarcinoma was 0.837 (: 0.767-0.908). When the Youden index was 0.59, the sensitivity was 85.56%, specificity was 74.13%, accuracy was 81.01%, positive predictive value was 83.69%, and negative predictive value was 76.78%.
The angle between the puncture needle and the pleura < 50°, preoperative CT suggesting emphysema, atelectasis, and lesion located in the left lung fissure are independent risk factors for pneumothorax after particle implantation in the treatment of advanced lung cancer. Preoperative planning of the puncture path should avoid lung bullae, interlobar fissures, areas of severe emphysema, and atelectasis.
分析晚期肺癌粒子植入术后气胸的危险因素，并构建和验证列线图预测模型。
回顾性分析2022年12月至2023年12月在***接受I粒子植入治疗晚期肺癌的148例患者。采用单因素logistic回归分析确定潜在危险因素，随后进行多因素logistic回归分析以评估气胸的预测因素。研究变量之间的交互作用并纳入模型构建。生成ROC曲线和列线图进行可视化。进行校准分析，并计算相应的净效益以调整预测模型。
148例患者中，58例（39.19%）发生气胸，平均年龄62.5（55.25，70）岁。多因素分析显示，穿刺针与胸膜夹角<50°（P = 0.002，OR：3.908，95%CI：1.621 - 9.422）、术前CT提示肺气肿（P = 0.002，OR：3.798，95%CI：1.600 - 9.016）、肺不张（P = 0.009，OR：3.156，95%CI：1.331 - 7.481）以及病变位于左肺叶间裂（P = 0.008，OR：4.675，95%CI：1.4683）是晚期肺癌粒子植入术后气胸的独立危险因素。术前CT提示病变位于左肺叶间裂或提示肺气肿在列线图中有显著影响，气胸发生概率分别为40%和38%。上述四个肺癌粒子植入术后气胸危险因素的预测AUC为0.837（95%CI：0.767 - 0.908）。当约登指数为0.59时，灵敏度为85.56%，特异度为74.13%，准确度为81.01%，阳性预测值为83.69%，阴性预测值为76.78%。
穿刺针与胸膜夹角<50°、术前CT提示肺气肿、肺不张以及病变位于左肺叶间裂是晚期肺癌粒子植入术后气胸的独立危险因素。术前穿刺路径规划应避开肺大疱、叶间裂、严重肺气肿区域和肺不张区域。
