休克指数对急性心肌梗死危重患者院内死亡的预测价值

Predictive value of shock index on in-hospital mortality of patients with critical acute myocardial infarction

目的 探讨休克指数对急性心肌梗死危重患者院内死亡的预测价值.方法 选择2005年7月至2013年7月首都医科大学附属北京安贞医院心内科收治的危重急性心肌梗死需置入主动脉内球囊反搏（IABP）辅助的患者572例.以受试者工作特征（ROC）曲线分析休克指数对急性心肌梗死患者院内预后的预测价值,根据其最佳界值将患者分为2组.比较2组一般资料和入院时临床指标、住院诊治情况和置入IABP前后生命体征的差异.结果 ROC曲线下面积为0.719,即休克指数对急性心肌梗死危重患者院内死亡具有良好的预测价值.根据最佳界值1.21将所有患者分为休克指数≥1.21组（151例）和休克指数＜1.21组（421例）.2组一般资料比较,差异均无统计学意义（均P＞0.05）.休克指数≥1.21组发病至首次就诊时间、入院时心率、心功能Ⅲ～Ⅳ级患者比例明显高于休克指数＜1.21组[6（4,12）h比5（3,10）h、100（77,110）次/min比80（70,90）次/min、67.5％（102/151）比34.9％（147/421）],收缩压、舒张压明显低于休克指数＜ 1.21组[95 （80,120） mmHg（1 mmHg=0.133 kPa）比106 （90,125） mmHg、60（50,70） mmHg比70（60,80） mmHg],差异均有统计学意义（均P＜0.05）.2组均以前壁心肌梗死为主.休克指数≥1.21组行急诊经皮冠状动脉介入患者比例、左心室射血分数明显低于休克指数＜1.21组[61.6％ （93/151）比77.0％ （324/421）、44％（37％,53％）比48％ （40％,56％）],IABP支持时间长于休克指数＜1.21组[2（1,3）d比1（1,2）d],接受药物治疗患者比例、院内死亡发生率、心源性休克发生率明显高于休克指数＜1.21组[27.8％（42/151）比14.7％（62/421）、29.8％ （45/151）比8.6％（36/421）、49.7％ （75/151）比18.3％ （77/421）],差异均有统计学意义（均P＜0.05）.置入IABP后,休克指数≥1.21组收缩压、舒张压、心率变化值明显多于休克指数＜1.21组[29（12,40） mmHg比21（10,31） mmHg、13（4,25） mmHg比9（0,19） mmHg、-15（-30,-4）次/min比0（-9,10）次/min],差异有统计学意义（P＜0.05）.结论 休克指数是预测急性心肌梗死危重患者院内死亡风险的简单、有效的临床指标.当休克指数≥1.21时,患者的院内死亡风险明显升高且对IABP的支持治疗效果欠佳.

Objective To evaluate the predictive value of shock index on in-hospital mortality of patients with critical acute myocardial infarction.Methods Totally 572 adult patients with critical acute myocardial infarction （AMI） who underwent intra-aortic balloon pump （IABP） from July 2005 to July 2013 were retrospectively analyzed.The receiver operating characteristic （ROC） curve was used to evaluate the predictive value of shock index,then the patients were divided into two groups based on the cut-off value.The general informations,clinical indicators,status of diagnosis and treatment,pre and post-IABP vital life signs were compared between groups.Results The area under ROC curve was 0.719,indicating a good predictive value of shock index on in-hospital mortality of patients with critical acute myocardial infarction.According to the cut-off value,the patients were divided into shock index ≥1.21 group （151 cases） and shock index 〈 1.21 group （421 cases）.The general informations were not significantly different between groups （P 〈 0.05）.The first visit time,heart rate on admission,proportion of heart function degree Ⅲ-Ⅳ in shock index ≥ 1.21 group were significantly higher than those in shock index 〈1.21 group [6 （4,12） h vs5 （3,10） h,100 （77,110） times/min vs 80 （70,90） times/min,67.5% （102/151） vs 34.9% （147/421）],the systolic blood pressure and diastolic blood pressure in shock index ≥1.21 group were significantly lower than those in shock index 〈 1.21 group[95 （80,120） mmHg vs 106 （90,125） mmHg,60 （50,70） mmHg vs 70 （60,80） mmHg] （P〈0.05）.Antetheca myocardial infarction took majority in both groups.The ratio of patients undergoing emergency percutaneous coronary intervention,left ventricular ejection fraction in shock index ≥ 1.21 group were significantly lower than those in shock index 〈 1.21 group [61.6% （93/151） vs 77.0% （324/421）,44% （37%,53%） vs 48% （40%,56%）];the IABP support duration in shock index ≥ 1.21 group was significantly longer than that in shock index 〈 1.21 group [2 （1,3） vs 1 （1,2） d];the proportion of patients receiving single drug treatment,the in-hospital mortality,the incidence of cardiac shock in shock index ≥ 1.21 group were significantly higher than those in shock index 〈 1.21group [27.8% （42/151） vs 14.7% （62/421）,29.8% （45/151） vs 8.6% （36/421）,49.7% （75/151） vs 18.3% （77/421）] （P 〈 0.05）.After IABP placement,the variations of systolic and diastolic blood pressure,heart rate in shock index ≥ 1.21 group were significantly greater than those in shock index 〈 1.21 group [29 （12,40） mmHg vs21 （10,31） mmHg,13 （4,25） mmHg vs9 （0,19） mmHg,-15 （-30,-4） times/min vs 0 （-9,10） times/min] （P 〈 0.05）.Conclusions Shock index is a simple and effective clinical index for predicting death risk of severe acute myocardial infarction;when shock index is ≥ 1.21,the in-hospital mortality increases significantly and the effect on IABP support is poor.