行胸腔镜手术的肺癌患者发生术中低血压的风险动态列线图预测模型的建立

Nomogram model for prediction of the risk of intraoperative hypotension in lung cancer patients undergoing thoracoscopic surgery

摘要:

目的 分析行胸腔镜手术的肺癌患者发生术中低血压的危险因素，并构建行胸腔镜手术的肺癌患者发生术中低血压的列线图预测模型。方法 选取2018年5月－2023年1月该院诊治的310例行胸腔镜手术的肺癌患者作为研究对象。采用多因素Logistic回归模型，分析肺癌患者行胸腔镜手术发生术中低血压的危险因素，采用R软件构建发生术中低血压的列线图预测模型，并进行验证。结果 310例行胸腔镜手术的肺癌患者中，有73例患者发生术中低血压，发生率为23.55%（73/310）。多因素Logistic回归分析结果显示，糖尿病、肺叶切除术、输液量 > 1 000 mL、出血量 > 1 000 mL、低蛋白血症、贫血、术前新辅助治疗和冠心病，是行胸腔镜手术的肺癌患者发生术中低血压的独立危险因素（P < 0.05）。构建行胸腔镜手术的肺癌患者发生术中低血压的列线图预测模型，校正曲线的预测值和实际值具有较好的一致性；绘制受试者操作特征曲线（ROC curve），曲线下面积（AUC）为0.791（95%CI：0.729～0.853）；决策曲线显示，阈值概率为5%～100%时，列线图预测行胸腔镜手术的肺癌患者发生术中低血压的净获益率较高。结论 糖尿病、肺叶切除术、输液量 > 1 000 mL、出血量 > 1 000 mL、低蛋白血症、贫血、术前新辅助治疗和冠心病，是行胸腔镜手术的肺癌患者发生术中低血压的独立危险因素，构建的行胸腔镜手术的肺癌患者发生术中低血压的列线图预测模型，临床应用价值较高。

Abstract:

Objective To analyze the risk factors of intraoperative hypotension in lung cancer patients undergoing thoracoscopic surgery, and construct a nomogram model for intraoperative hypotension in lung cancer patients undergoing thoracoscopic surgery.Methods 310 patients underwent thoracoscopic surgery for lung cancer from May 2018 to January 2023 were selected as the study objects. Multivariate Logistic regression analysis was used to screen the risk factors for intraoperative hypotension in patients underwent thoracoscopic surgery for lung cancer. R software was used to build the intraoperative hypotension model in patients underwent thoracoscopic surgery for lung cancer, and the intraoperative hypotension model was verified.Results Among the 310 lung cancer patients underwent thoracoscopic surgery, 73 patients had intraoperative hypotension, and the incidence of hypotension was 23.55% (73/310). Multivariate Logistic regression analysis showed that diabetes mellitus, lobectomy, infusion volume > 1 000 mL, blood loss > 1 000 mL, hypoproteinemia, anemia, preoperative neoadjuvant therapy and coronary heart disease were independent risk factors for intraoperative hypotension in lung cancer patients underwent thoracoscopic surgery (P < 0.05). The predicted value of the correction curve of the nomogram model for patients with intraoperative hypotension of lung cancer undergoing thoracoscopic surgery was consistent with the actual value. The area under receiver operator characteristic curve (ROC curve) of the model was 0.791 (95%CI: 0.729 ~ 0.853). When the decision curve showed that the threshold probability was 5% to 100%, the net benefit value of the nomogram predicting intraoperative hypotension for lung cancer patients undergoing thoracoscopic surgery was higher.Conclusion Diabetes mellitus, lobectomy, infusion volume > 1 000 mL, blood loss > 1 000 mL, hypoproteinemia, anemia, preoperative neoadjuvant therapy and coronary heart disease are independent risk factors for intraoperative hypotension in patients with lung cancer undergoing thoracoscopic surgery. The clinical application value of the nomogram of intraoperative hypotension in patients with lung cancer undergoing thoracoscopic surgery is good.