作者：徐慧，刘鸿，管玉珍，陆真

单位：

210029江苏省南京市，江苏省人民医院心脏大血管外科

单位（英文）：

Department of Cardiovascular Surgery, Jiangsu Provincial People's Hospital, Nanjing 210029, China

关键词：

冠心病；冠状动脉旁路移植术；肺部感染；影响因素分析；列线图

关键词（英文）：

Coronary heart disease; Coronary artery bypass grafting; Lung infection; Root cause analysis; Nomograms

中图分类号：

R 541.4

DOI：

10.12114/j.issn.1008-5971.2023.00.171

基金项目：

江苏省科教能力提升工程项目（ZDXK202230）

摘要：

　目的　探讨老年冠心病患者冠状动脉旁路移植术（CABG）后发生肺部感染的影响因素，构建其风险预测列线图模型并进行验证。方法　选取2020—2021年于江苏省人民医院行CABG的老年冠心病患者464例为建模集，选取2022年于江苏省人民医院行CABG的老年冠心病患者320例为验证集。收集所有患者的临床资料，根据CABG后住院期间是否发生肺部感染将建模集患者分为肺部感染组和非肺部感染组。老年冠心病患者CABG后发生肺部感染的影响因素采用多因素Logistic回归分析；采用R 4.1.2软件包及rms程序包建立老年冠心病患者CABG后发生肺部感染的风险预测列线图模型；采用Bootstrap法重复抽样1 000次对该列线图模型进行内部验证，计算其一致性指数；采用HosmerLemeshoe拟合优度检验评价该列线图模型的拟合程度；绘制校准曲线以评估该列线图模型预测老年冠心病患者CABG后发生肺部感染的性能；采用ROC曲线分析该列线图模型对老年冠心病患者CABG后发生肺部感染的预测价值。结果建模集464例老年冠心病患者CABG后发生肺部感染69例（14.87%）。两组年龄、术前红细胞分布宽度（RDW）、手术方式、手术时间、输注悬浮红细胞者占比、术后呼吸机通气时间比较，差异有统计学意义（P＜0.05）。多因素Logistic回归分析结果显示，年龄、术前RDW、手术方式、手术时间、输注悬浮红细胞、术后呼吸机通气时间是老年冠心病患者CABG后发生肺部感染的独立影响因素（P＜0.05）。基于多因素Logistic回归分析结果，构建老年冠心病患者CABG后发生肺部感染的风险预测列线图模型。采用Bootstrap法分别在建模集与验证集中重复抽样1 000次对该列线图模型进行内部验证，结果显示，其一致性指数分别为0.794〔95%CI（0.766，0.822）〕和0.759〔95%CI（0.737，0.782）〕。Hosmer-Lemeshoe拟合优度检验结果显示，在建模集和验证集中，该列线图模型拟合较好（χ2=1.294，P=0.255；χ2=0.326，P=0.568）。校准曲线分析结果显示，该列线图模型预测建模集与验证集老年冠心病患者CABG后发生肺部感染的校准曲线接近于理想曲线。ROC曲线分析结果显示，该列线图模型预测建模集与验证集老年冠心病患者CABG后发生肺部感染的AUC分别为0.801〔95%CI（0.771，0.831）〕和0.762〔95%CI（0.734，0.790）〕。结论年龄≥70岁、术前RDW≥14.5%、手术方式为体外循环、手术时间≥5 h、输注悬浮红细胞、术后呼吸机通气时间≥24 h是老年冠心病患者CABG后发生肺部感染的独立危险因素，基于上述因素构建的列线图模型有助于预测老年冠心病患者CABG后发生肺部感染的发生风险。

英文摘要：

Objective To analyze the influencing factors of pulmonary infection after coronary artery bypass grafting(CABG) in elderly patients with coronary heart disease, and to construct and validate the nomogram model for predicting itsrisk. Methods A total of 464 elderly patients with coronary heart disease who underwent CABG in Jiangsu Provincial People'sHospital from 2020 to 2021 were selected as the modeling set, 320 elderly patients with coronary heart disease who underwentCABG in Jiangsu Provincial People's Hospital in 2022 were selected as the validation set. Clinical data of patients were collected,the patients in the modeling set were divided into pulmonary infection group and non-pulmonary infection group based onwhether pulmonary infection occurred during hospitalization after CABG. Multivariate Logistic regression analysis method wasused to analyze the influencing factors of pulmonary infection after CABG in elderly patients with coronary heart disease. Thenomogram model for predicting the risk of pulmonary infection after CABG in elderly patients with coronary heart disease was constructed by the R 4.1.2 software package and rms package. Bootstrap method was used to repeatedly sample 1 000 times forinternal verification of the nomogram model, and its consistency index (CI) was calculated. Hosmer-Lemeshow goodness of fittest was used to evaluate the fitting degree of the nomogram model. Calibration curve was drawn to evaluate the effectiveness ofthe nomogram model for predicting pulmonary infection after CABG in elderly patients with coronary heart disease, and the ROCcurve was used to analyze the predictive value of the nomogram model for pulmonary infection after CABG in elderly patientswith coronary heart disease. Results Among 464 elderly patients with coronary heart disease in the modeling set, 69 (14.87%)had pulmonary infection. There were significant differences in age, preoperative red cell distribution width (RDW) , surgicalmethod, surgical time, proportion of patients suspended red blood cell infusion, and postoperative ventilation time between thetwo groups (P < 0.05) . Multivariate Logistic regression analysis showed that age, preoperative RDW, surgical method, surgicaltime, proportion of patients suspended red blood cell infusion, and postoperative ventilation time were the influencing factors ofpulmonary infection after CABG in elderly patients with coronary heart disease (P < 0.05) . The nomogram model for predictingpulmonary infection after CABG in elderly patients with coronary heart disease was constructed based on the above influencingfactors. Bootstrap method was used to repeated by sample 1 000 times in modeling set and in validation set respectively forinternal verification of the nomogram model, the results showed that the CI was 0.794 [95%CI (0.766, 0.822) ] and 0.759 [95%CI(0.737, 0.782) ] , respectively. The results of Hosmer-Lemeshow goodness of fit test showed that the nomogram model fit well inmodeling set (χ2=1.294, P=0.255) and in validation set (χ2=0.326, P=0.568) . The results of calibration curve analysis showedthat the calibration curve of nomogram model for predicting pulmonary infection after CABG in elderly patients with coronary heartdisease was close to the ideal curve. The results of ROC curve analysis showed that the AUC of the nomogram model for predictingpulmonary infection after CABG in elderly patients with coronary heart disease in modeling set and validation set was 0.801[95%CI (0.771, 0.831) ] , 0.762 [95%CI (0.734, 0.790) ] , respectively. Conclusion Age ≥ 70, preoperative RDW ≥ 14.5%,extracorporeal circulation of surgical, surgical time ≥ 5 h, suspending method red blood cell infusion, and postoperative ventilationtime ≥ 24 h are the risk factors of pulmonary infection after CABG in elderly patients with coronary heart disease. The nomogrammodel constructed based on the above factors is helpful to predict the risk of pulmonary infection after CABG in elderly patientswith coronary heart disease.

参考文献：