2020 年第 11 期 第 15 卷

感染性心内膜炎患者围术期死亡的危险因素分析

Risk factors of perioperative death in patients with infective endocarditis

作者：董柱张本曹一秋刘莹郑燕纯杨博王晓武张卫达

英文作者：Dong Zhu Zhang Ben Cao Yiqiu Liu Ying Zheng Yanchun Yang Bo Wang Xiaowu Zhang Weida

单位：中国人民解放军南部战区总医院心脏外科中心，广州510010

英文单位：Cardiac Surgery Center General Hospital of Southern Theatre Command Chinese People′s Liberation Army Guangzhou 510010 China

关键词：感染性心内膜炎；危险因素；围术期

英文关键词：Infectiveendocarditis;Riskfactors;Perioperativeperiod 

• 摘要：

• 目的 探讨感染性心内膜炎（IE）患者围术期死亡的危险因素及各指标的预测效能。方法 收集2012年4月至2019年4月中国人民解放军南部战区总医院收治的91例IE患者的临床资料，围术期死亡6例（死亡组），顺利出院85例（存活组）。2组患者术前均应用抗菌药物控制感染，体温正常、血常规正常2~4周后手术治疗。其中行单纯主动脉瓣置换13例，主动脉瓣置换+二尖瓣置换7例，主动脉瓣置换+二尖瓣成形5例，主动脉瓣置换+二尖瓣成形+三尖瓣成形5例，主动脉瓣置换+肺动脉瓣置换+三尖瓣成形1例，单纯二尖瓣成形20例，二尖瓣置换+三尖瓣成形11例，二尖瓣置换+三尖瓣置换1例，单纯三尖瓣成形4例，单纯心内赘生物清除24例。合并冠状动脉多支病变行冠状动脉旁路移植术4例，合并先天性心脏病行一期修复，包括室间隔缺损9例，动脉导管未闭2例，卵圆孔未闭1例。比较2组患者的基线资料和术中、术后资料，应用多因素Logistic回归方法分析围术期死亡的危险因素，绘制受试者工作特征曲线评价各指标的检验效能。结果 死亡组合并糖尿病、既往心肌梗死病史、术前肝功能不全、围术期肾功能不全和术后神经系统症状、肺部并发症比例均高于存活组，术前纽约心脏病协会心功能分级劣于存活组，升主动脉阻断时间和气管插管时间、重症监护病房（ICU）停留时间、住院时间均长于存活组［(113±38)min比（75±33）min、（3.8±1.0）d比(1.9±1.4)d、（12±8）d比(5±4)d、(52±29)d比（29±14）d］（均P＜0.05）。多因素Logistic回归分析结果表明，ICU停留时间长、围术期肾功能不全是IE患者术后死亡的独立危险因素（均P＜0.05）,二者预测IE患者术后死亡的曲线下面积分别为0.861（95%置信区间:0.760~0.962，P=0.003）和0.792（95%置信区间:0.563~1.000，P=0.017），二者联合预测IE患者术后死亡的曲线下面积为0.922(95%置信区间:0.851~0.992，P=0.001)。结论 ICU停留时间长和围术期肾功能不全是IE患者术后死亡的独立危险因素，缩短患者ICU停留时间及维持患者围术期肾功能和循环的稳定，有助于降低患者围术期的死亡风险。

• Objective To explore the risk factors of perioperative death in patients with infective endocarditis (IE) and the predictive efficacy of each index. Methods Totally 91 patients with IE admitted to General Hospital of Southern Theatre Command, Chinese People′s Liberation Army from April 2012 to April 2019 were enrolled. Six patients died during the perioperative period （the death group） and 85 patients survived（the survival group）. Both groups were treated with antibiotics to control the infection before surgery. Surgery was done in patients with normal body temperature and blood routine for 2-4 weeks. There were 13 patients of aortic valve replacement（AVR）, 7 patients of AVR+mitral valve replacement（MVR）, 5 patients of AVR+ mitral valvuloplasty（MVP）, 5 patients of AVR+MVP+tricuspid valvuloplasty（TVP）, 1 patient of AVR+pulmonary valve replacement（PVR）+TVP, 20 patients of MVP, 11 patients of MVR+TVP, 1 patient of MVR+tricuspid valve replacement（TVR）, 4 patients of TVP and 24 patients of endocardial vegetations removal. Coronary artery bypass grafting was performed in 4 patients with coronary multi-branch lesions. One-stage repair was performed in patients combine with congenital heart disease, including 9 patients with ventricular septal defect（VSD）, 2 patients with patent ductus arteriosus（PDA） and 1 patient with patent foramen ovale（PFO）. Baseline data, intraoperative and postoperative data of the two groups were compared. Logistic regression analysis was used to analyze the risk factors of perioperative period death. The receiver operating characteristic（ROC） curve was used to evaluate the effectiveness of each index. Results The rates of combination with diabetes, history of myocardial infarction, preoperative hepatic insufficiency, perioperative period of renal insufficiency, postoperative nervous system symptoms, lung complications in the death group were higher than those in the survival group. The preoperative New York Heart Association cardiac function grade of the death group was lower than that of the survival group. The ascending aorta blocking time, endotracheal intubation time, intensive care unit (ICU) stay time and hospitalization time in the death group were longer than those in the survival group［(113±38)min vs (75±33)min, (3.8±1.0)d vs (1.9±1.4)d, (12±8)d vs (5±4)d, (52±29)d vs (29±14)d］(all P<0.05). Multivariate Logistic regression analysis showed that long ICU stay time and perioperative period of renal insufficiency were independent risk factors for perioperative period death. According to ROC curve analysis, the area under the curve obtained by ICU stay time was 0.861 ［95% confidence interval （CI）: 0.760-0.962， P=0.003］; the area under the curve obtained by perioperative period of renal insufficiency was 0.792（95%CI: 0.563-1.000， P=0.017）; the area under the curve obtained by combining ICU stay time long and perioperative period of renal insufficiency was 0.922(95%CI: 0.851-0.992, P=0.001). Conclusions Long ICU stay time and perioperative period of renal insufficiency were independent risk factors for postoperative death in IE patients. Reducing the length of stay in the ICU and maintaining the stability of perioperative renal function and circulation can help patients reduce the risk of perioperative death.