急性ST段抬高型心肌梗死患者急诊介入前心室颤动预测模型构建及验证  

Construction and Verification of a Prediction Model of Ventricular Fibrillation in patients with acute ST-segment elevation myocardial infarction before primary percutaneous coronary intervention

作　　者：翟虎[1,2,3,4] 张纯溪 王禹 刘迎午[1,2,3,4] ZHAI Hu;ZHANG Chunci;WANG Yu;LIU Yingwu(Heart Center,the Third Central Hospital of Tianjin,Tianjin,300170,China;Key Laboratory of Extracorporeal Life Support for Critical Disease;Artificial Cell Engineering Technology Research Center;Tianjin Institute of Hepatobiliary Disease;Department of Emergency,the Third Central Hospital of Tianjin)

机构地区：[1]天津市第三中心医院心脏中心,天津300170 [2]天津市重症疾病体外生命支持重点实验室 [3]天津市人工细胞工程技术研究中心 [4]天津市肝胆研究所 [5]天津市第三中心医院急诊科

出　　处：《临床心血管病杂志》2023年第6期459-465,共7页Journal of Clinical Cardiology

摘　　要：目的:构建急性ST段抬高型心肌梗死(ST-segment elevation mayocardial infarction, STEMI)患者急诊介入前心室颤动(室颤)的预测模型并对预测价值进行内部验证。方法:回顾分析2020年1月—2022年5月连续收治的223例STEMI患者,采用多因素logistic回归筛选独立危险因素并构建预测模型。ROC曲线分析预测价值,并利用2022年7~12月另96例STEMI患者数据验证其预测能力。结果:与对照组(202例)比较,室颤组(21例)进急诊室时心率更快、血压更低、Killip分级更高、罪犯血管为前降支比例更高、主动脉气囊反搏使用率更高、存活出院率更低(均P<0.05);急诊外周血白细胞计数和尿酸水平更高、左室射血分数更低(均P<0.05)。多因素logistic回归表明急诊室心率、舒张压、尿酸水平和Killip分级为早期室颤的独立影响因素。预测模型(1.05×心率+0.947×舒张压+1.008×尿酸+13.207×Killip分级)的ROC曲线下面积为0.703(95%CI 0.569~0.837,P=0.002),截断值为552.8,灵敏度和特异度分别为57.1%和85.6%。后期内部验证的准确度为0.865,灵敏度为71.4%,特异度为87.6%。结论:利用STEMI患者发病早期临床指标构建的预测模型有助于识别急诊介入前室颤的高危人群。Objective: The aim was to construct a prediction model of ventricular fibrillation(VF) in patients with ST-segment elevation myocardial infarction(STEMI) before primary percutaneous coronary intervention(PPCI) and verify its prediction value. Methods: The 223 consecutive patients with STEMI admitted from January 2020 to May 2022 were retrospectively analyzed. Independent risk factors were screened and a prediction model was constructed via multivariate logistic regression. A ROC curve was used to analyze its predictive value and the data of another 96 STEMI patients from July to December 2022 were used for internal validation. Results: Compared with the control group(n=202), the VF group(n=21) had faster heart rate, lower blood pressure, higher Killip grade at admission, higher proportion of culprit vessels as left anterior descending arteries, higher proportion of using intra-aortic balloon pulsation, and lower survival and discharge rate(all P<0.05). As for the VF group, the peripheral blood white blood cell count and uric acid level at admission were higher, the glutamic pyruvic transaminase was higher, and the left ventricular ejection fraction was lower within 24 hours after admission(all P<0.05). Multivariate logistic regression showed that heart rate, diastolic blood pressure, uric acid level and Killip grade were independent factors of VF. The area under the ROC curve of the prediction model(1.05×heart rate+0.947×diastolic blood pressure+1.008×uric acid+13.207×Killip grade) was 0.703(95%CI 0.569-0.837, P=0.002), with sensitivity and specificity of 57.1% and 85.6%, respectively. The accuracy, sensitivity and specificity of the internal verification in the second stage are 0.865, 0.714, 0.876, respectively. Conclusion: prediction model constructed by the easy to obtain variables in the early stage of STEMI patients is helpful to identify the high-risk group of VF before PPCI.

关 键 词：急性心肌梗死 急性冠脉综合征 室颤 预测模型 预后 

分 类 号：R542.2[医药卫生—心血管疾病]