中国冠状动脉旁路移植手术住院死亡危险因素分析

Risk factors for in-hospital mortality in coronary artery bypass grafting： analysis of database from chinese coronary artery bypass grafting registry study

目的分析影响中国冠状动脉旁路移植术（CABG）术后住院死亡的危险因素。方法全国32家心脏外科中心2004-2005两年共行9247例CABG术。确定潜在危险因素后，根据潜在危险因素从临床资料中收集数据，最终数据分为生存组和住院死亡组，对影响住院死亡的潜在危险因素进行单因素分析和logistic多因素回归分析，最终确立影响中国CABG住院死亡的危险因素，并对结果的校准度和分辨能力进行检验。结果全组平均年龄（62．1±9．1）岁，女性占21．5％，冠脉三支病变占76．7％，左主干病变25．8％。总体住院病死率3．3％。logistic多因素回归分析发现，年龄、肾衰史、慢性阻塞性肺疾病、既往心血管手术、不稳定型心绞痛、左心室射血分数、术前危重状态、非择期手术、合并其他手术为CABG住院死亡的独立危险因素。Hosmer-Lemeshowx。检验结果χ2=2．987，P=0．935。受试者工作特征（ROC）曲线下面积为0．75。结论通过logistic多因素回归分析，得出年龄、肾衰史、慢性阻塞性肺疾病、既往心血管手术、不稳定型心绞痛、左心室射血分数、术前危重状态、非择期手术、合并其他手术等9个因素为影响中国病人CABG住院死亡的独立危险因素。分析结果具有良好的校准度和分辨能力。

Objective To assess risk factors for in-hospital mortality in coronary artery bypass grafting （CABG） patients. Methods 9247 coronary artery disease patients undergoing CABG from January, 2004 to December, 2005 in 32 heart centers in China were reviewed. The potential risk factors were identified through literature reviewing and referring other risk models. Data collection proceeded according to the potential risk factors. The final data were divided into two groups： alive group and non-alive group during hospitalization. Univariate analysis and logistic regression were used to analyse the potential risk factors. Independent risk factors for mortality were determined at last. The calibration and discrimination of the result were tested. Results CABG operation in-hospital mortality was 3.3 % （ 302/9247 ）. The mean age was（ 62.1 ± 9.1 ） years and 21.5 % were female. Nine variables including： age, renal failure, chronic obstructive pulmonary disease, previous cardiac surgery, unstable angina, left ventricular ejection fraction （LVEF）, critical preoperative state, non-elective operation and CABG plus other cardiac operations were independently ore＇elated with CABG in-hospital mortality. The results of calibration assessing by Hosmer-Lemeshow χ2 test was χ2 = 2.987 and P =0.935. The results of discrimination assessing by area under receiver-operating characteristic （ROC） curve is 0.75. Conclusion The following risk factors were associated with increased in-hospital mortality： elder age, renal failure, chronic obstructive pulmonary disease, previous cardiac surgery, unstable angina, low LVEF, critical preoperative state, non-elective operation and CABG plus other cardiac precedures.