目的分析Grace评分联合Crusade评分对心房颤动合并非ST段抬高型心肌梗死(NSTEMI)患者经皮冠状动脉介入(PCI)术后远期发生全因死亡及主要不良心血管事件的预测价值。方法回顾性分析2010年1月—2015年1月于北京11家三级甲等医院就诊的心房颤动合并NSTEMI行PCI患者的相关资料。绘制Grace评分、Crusade评分及联合评分预测心房颤动合并NSTEMI患者PCI术后远期预后的受试者工作特征曲线(ROC曲线),计算3种方法ROC曲线下面积(AUC)。依据联合评分将患者分为G1组(Grace评分≤118分,Crusade评分≤40分,n=112),G2组(Grace评分≤118分,Crusade评分≥41分,n=20),G3组(Grace评分>118分,Crusade评分≤40分,n=143),G4组(Grace评分>118分,Crusade评分≥41分,n=104)。绘制并比较4组生存曲线。结果共纳入379例患者,随访(36.8±18.3)个月,完成随访362例。Grace评分、Crusade评分及联合评分预测心房颤动合并NSTEMI患者PCI术后远期全因死亡风险的AUC分别为0.660[95%CI(0.609,0.709)]、0.702[95%CI(0.652,0.749)]、0.707[95%CI(0.658,0.754)],预测PCI术后远期主要不良心血管事件风险的AUC分别为0.559[95%CI(0.506,0.611)]、0.593[95%CI(0.540,0.644)]、0.594[95%CI(0.542,0.645)]。G1、G2、G3、G4组的生存曲线比较,差异有统计学意义(χ~2=20.07,P<0.01)。G4组生存率低于G1、G3组(P<0.05)。结论 Grace评分、Crusade评分、联合评分对心房颤动合并NSTEMI患者PCI术后远期全因死亡风险有一定的预测价值,且联合评分更能够筛选出高危患者;3种方法预测主要不良心血管事件风险的价值相当。 Objective To analyze the long-term all-cause mortality and major adverse cardiovascular events after Percutaneous Coronary Intervention (PCI) in patients with atrial fibrillation and non-ST-segment elevation myocardial infarction (NSTEMI) assessed by Grace Score and Crusade score. Methods A retrospective analysis was conducted on the data of patients with atrial fibrillation and NSTEMI in PCI from January 2010 to January 2015 in 11 tertiary level hospitals in Beijing. The ROC curve of subjects with atrial fibrillation predicting the long-term prognosis of patients with atrial fibrillation and NSTEMI after PCI was drawn. The area under the ROC curve (AUC) was calculated by the three methods. The patients were divided into G1 group (Grace score ≤118, Crusade score ≤40, n = 112), G2 group (Grace score ≤118, Crusade score ≥41, n = 20) Grace score> 118, Crusade score ≤40, n = 143), G4 (Grace score> 118, Crusade score ≥41, n = 104). Draw and compare 4 groups of survival curves. Results A total of 379 patients were enrolled in this study. The follow-up was 36.8 ± 18.3 months and 362 patients were followed up. AUC of 0.602 [95% CI (0.609, 0.709)], 0.702 [95% CI (0.652, 0.749)] for predicting long-term all-cause mortality after perioperative PCI for patients with atrial fibrillation and NSTEMI, respectively, using the Grace, Crusade and combined scores , 0.707 [95% CI (0.658,0.754)] respectively. The AUC of predicting long-term adverse cardiovascular events after PCI were 0.559 [95% CI 0.506,0.611], 0.593 [95% CI )], 0.594 [95% CI (0.542, 0.645)]. The survival curves of G1, G2, G3 and G4 groups were statistically significant (χ ~ 2 = 20.07, P <0.01). The survival rate of G4 group was lower than that of G1 and G3 groups (P <0.05). Conclusions The Grace score, Crusade score and combined score have certain predictive value for long-term all-cause mortality after PCI in patients with atrial fibrillation and NSTEMI, and combined score can better screen high-risk patients. Three methods are used to predict the risk of major adverse cardiovascular events The value of the equivalent.