目的探讨GRACE评分对急性冠脉综合征(ACS)患者住院期间心源性死亡风险的预测价值。方法回顾性纳入2009年1月至2010年12月住院治疗的ACS患者160例,住院期间(28 d内)发生心源性死亡患者60例(死亡组),同期住院存活患者100例(对照组)。收集患者入院基线时资料,并进行GRACE评分。采用ROC曲线计算GRACE评分对预测心源性死亡敏感性和特异性。结果死亡组患者的GRACE评分为(179.00±39.74)分,而对照组为(128.93±30.88)分,两组相比具有统计学差异(P〈0.01)。两组患者GRACE评分危险分层构成中,死亡组高危层比率明显高于对照组,而中低危层低于对照组,差异有显著统计学意义(P〈0.01)。对于ACS患者,GRACE评分在158分时,对应ROC曲线下面积最大为0.821(95%CI:0.743~0.899,P〈0.01),预测在院期间心源性死亡的敏感性为0.75,特异度0.85。结论 GRACE危险评分方法可以用于评估ACS患者住院期间心源性死亡风险;当GRACE评分在158分时,预测住院发生心源性死亡的敏感性和特异性均较好。
Objective To study the predictive value of GRACE scores to the risk of cardiac death in patients with acute coronary syndrome (ACS) during hospitalization. Methods ACS patients (n=160) were chosen from Jan. 2009 to Dec. 2010 and divided into death group (n=60) and control group (n=100) during hospitalization (within 28 d). The baseline data was collected from the patients and given GRACE scoring. The sensitivity and specificity of GRACE scores for predicting cardiac death was calculated by using ROC curve. Results GRACE scores were (179.00 ± 39.74) in death group and (128.93 ± 30.88) in control group (P〈0.01). In GRACE score risk stratification structure, the percentage of high-risk stratification was higher and that of middle-lower stratification was lower in death group than those in control group (P〈0.01). For ACS patients, when GRACE scores were 158, the largest area under corresponding ROC curve was 0.821 (95%CI: 0.743~0.899, P〈0.01), and the sensitivity was 0.75 and specificity was 0.85 for prediciting cardiac death during hospitalization. Conclusion GRACE scoring system can be applied for reviewing cardiac death risk in ACS patients during hospitalization. When scores are 158, the sensitivity and specificity are higher in predicting cardiac death during hospitalization.