Luo Jing-guang, Yang Ming, Han Ling, Jia Xiao, Chen Li-wei, Zhao Yan
Department of Cardiology, Fu Xing Hospital, Capital Medical University, Beijing, China.
Coron Artery Dis. 2013 Nov;24(7):537-41. doi: 10.1097/MCA.0000000000000023.
The Global Registry of Acute Coronary Events (GRACE) risk score provides a prediction of the probability of death and myocardial infarction within 6 months after acute coronary syndrome. Our aim was to detect the validity of the GRACE risk score in predicting 6-month death or reinfarction in Chinese acute myocardial infarction (AMI) inpatients 80 years of age and older between 2003 and 2012.
Using the GRACE risk score to predict the combined endpoints of all-cause death or reinfarction within 6 months of AMI, we evaluated its discrimination and calibration by the C-statistic and the Hosmer-Lemeshow test, respectively.
In total, 370 patients presenting with ST segment elevation myocardial infarction (STEMI) and non-STEMI were included. The GRACE risk score ranged between 181 and 325; death or reinfarction within 6 months of AMI was 49.2%. The GRACE model showed good discrimination overall (C-statistic 0.708, 95% confidence interval: 0.655-0.760, P<0.001) and in patients with non-STEMI (C-statistic 0.756, 95% confidence interval: 0.690-0.823, P<0.001). However, the discriminatory capacity was lower in patients with STEMI (C-statistic 0.645, P=0.001). The calibration was optimal overall (Hosmer-Lemeshow, P=0.398) and in the subgroups (STEMI, P=0.098; non-STEMI, P=0.822). There was a high correlation (R=0.926) between the predicted and the observed 6-month death or reinfarction after AMI.
The GRACE score is accurate for determination of 6-month death or reinfarction in Chinese AMI inpatients 80 years of age and older; however, the discrimination and calibration performs less well in patients with STEMI.
全球急性冠状动脉事件注册研究(GRACE)风险评分可预测急性冠状动脉综合征后6个月内的死亡和心肌梗死概率。我们的目的是检测GRACE风险评分对2003年至2012年间80岁及以上中国急性心肌梗死(AMI)住院患者6个月死亡或再梗死的预测效度。
使用GRACE风险评分预测AMI后6个月内全因死亡或再梗死的联合终点,我们分别通过C统计量和Hosmer-Lemeshow检验评估其区分度和校准度。
共纳入370例ST段抬高型心肌梗死(STEMI)和非STEMI患者。GRACE风险评分在181至325之间;AMI后6个月内死亡或再梗死发生率为49.2%。GRACE模型总体显示出良好的区分度(C统计量0.708,95%置信区间:0.655 - 0.760,P<0.001),在非STEMI患者中也是如此(C统计量0.756,95%置信区间:0.690 - 0.823,P<0.001)。然而,STEMI患者的区分能力较低(C统计量0.645,P = 0.001)。总体校准度最佳(Hosmer-Lemeshow,P = 0.398),各亚组也是如此(STEMI,P = 0.098;非STEMI,P = 0.822)。AMI后6个月预测死亡或再梗死与观察到的情况之间存在高度相关性(R = 0.926)。
GRACE评分对于确定80岁及以上中国AMI住院患者6个月死亡或再梗死是准确的;然而,STEMI患者的区分度和校准度表现较差。