Toumpoulis Ioannis K, Anagnostopoulos Constantine E, DeRose Joseph J, Swistel Daniel G
Department of Cardiac Surgery, University Hospital of Ioannina, Ioannina, Greece.
Eur J Cardiothorac Surg. 2004 Jan;25(1):51-8. doi: 10.1016/s1010-7940(03)00651-1.
To evaluate the accuracy of predicting long-term mortality in patients with coronary artery bypass grafting (CABG) by using the European system for cardiac operative risk evaluation (EuroSCORE).
Medical records of patients with CABG (n=3760) between January 1992 and March 2002 were retrospectively reviewed and their predicted surgical risk was calculated according to the standard (study A) and logistic (study B) EuroSCORE. In study A the patients were divided into six groups: 0-2 (n=610), 3-5 (n=1479), 6-8 (n=1099), 9-11 (n=452), 12-14 (n=103) and >14 (n=17). In study B the patients were divided into seven groups: 0.00-2.00 (n=447), 2.01-5.00 (n=1190), 5.01-10.00 (n=890), 10.01-20.00 (n=686), 20.01-30.00 (n=234), 30.01-60.00 (n=254) and >60.00 (n=59). Long-term survival was obtained by the National Death Index and Kaplan-Meier curves were constructed and compared employing the log-rank test. Multivariate Cox regression analysis was performed in order to control for pre, intra and postoperative factors and adjusted hazard ratios were calculated for standard and logistic EuroSCORE groups. The receiver operating characteristic (ROC) curves were plotted to assess the discrimination ability of the EuroSCORE.
In study A there were differences among the six groups in 30-day mortality (0.7%, 1.0%, 3.1%, 4.6%, 13.6% and 23.5%; P<0.001), in major complications (8.5%, 10.4%, 16.2%, 20.4%, 31.1% and 35.3%; P<0.001) as well as in actuarial long-term survival (86.2%, 79.6%, 53.6%, 37.9%, 24.9% and 0% from EuroSCORE 0-2 to >14; P<0.001). In study B there were differences among the seven groups in 30-day mortality (0.9%, 1.1%, 1.2%, 3.6%, 3.4%, 8.7% and 15.3%; P<0.001), major complications (8.5%, 10.1%, 12.1%, 18.4%, 16.2%, 26.0% and 30.5%; P<0.001) as well as in actuarial long-term survival (89.5%, 79.9%, 66.9%, 51.0%, 40.3%, 38.4% and 13.7% from EuroSCORE 0.00-2.00 to >60.00; P<0.001). Multivariate Cox regression analysis confirmed that EuroSCORE (standard or logistic) was a statistically significant predictor for long-term mortality, while the area under the ROC curve was 0.72 for either standard or logistic EuroSCORE.
The predicted surgical risk in CABG patients as calculated by standard or logistic EuroSCORE is a strong predictor for long-term survival in addition to predicting operative survival for which it was originally designed.
通过使用欧洲心脏手术风险评估系统（EuroSCORE）来评估冠状动脉旁路移植术（CABG）患者长期死亡率的预测准确性。
回顾性分析1992年1月至2002年3月期间行CABG的患者（n = 3760）的病历，并根据标准（研究A）和逻辑（研究B）EuroSCORE计算其预测手术风险。在研究A中，患者分为六组：0 - 2分（n = 610）、3 - 5分（n = 1479）、6 - 8分（n = 1099）、9 - 11分（n = 452）、12 - 14分（n = 103）和＞14分（n = 17）。在研究B中，患者分为七组：0.00 - 2.00分（n = 447）、2.01 - 5.00分（n = 1190）、5.01 - 10.00分（n = 890）、10.01 - 20.00分（n = 686）、20.01 - 30.00分（n = 234）、30.01 - 60.00分（n = 254）和＞60.00分（n = 59）。通过国家死亡指数获得长期生存率，并构建Kaplan - Meier曲线，采用对数秩检验进行比较。进行多变量Cox回归分析以控制术前、术中和术后因素，并计算标准和逻辑EuroSCORE组的调整风险比。绘制受试者工作特征（ROC）曲线以评估EuroSCORE的辨别能力。
在研究A中，六组患者的30天死亡率（0.7%、1.0%、3.1%、4.6%、13.6%和23.5%；P < 0.001）、主要并发症发生率（8.5%、10.4%、16.2%、20.4%、31.1%和35.3%；P < 0.001）以及精算长期生存率（从EuroSCORE 0 - 2分到＞14分分别为86.2%、79.6%、53.6%、37.9%、24.9%和0%；P < 0.001）存在差异。在研究B中，七组患者的30天死亡率（0.9%、1.1%、1.2%、3.6%、3.4%、8.7%和15.3%；P < 0.0- ）、主要并发症发生率（8.5%、10.1%、12.1%、18.4%、16.2%、26.0%和30.5%；P < 0.001）以及精算长期生存率（从EuroSCORE 0.00 - 2.00分到＞60.00分分别为89.5%、79.9%、66.9%、51.0%、40.3%、38.4%和13.7%；P < 0.001）存在差异。多变量Cox回归分析证实，EuroSCORE（标准或逻辑）是长期死亡率的统计学显著预测因子，而标准或逻辑EuroSCORE的ROC曲线下面积均为0.72。
标准或逻辑EuroSCORE计算出的CABG患者预测手术风险不仅是其最初设计用于预测手术生存率的有力指标，也是长期生存的有力预测因子。
