Owens Colum, McClelland Anthony, Walsh Simon, Smith Bernie, Adgey Jennifer
Regional Medical Cardiology Centre, Royal Victoria Hospital, Belfast, Northern Ireland.
Am J Cardiol. 2008 Aug 1;102(3):257-65. doi: 10.1016/j.amjcard.2008.03.046. Epub 2008 May 24.
We aimed to develop 12-lead electrocardiographic (ECG) models testing ST-elevation criteria with QRST variables and compare their performance with the 80-lead body surface map (BSM) in detection of acute myocardial infarction (AMI). Because the prevalence of non-ST-elevation AMI is increasing worldwide, advances in early ECG detection of AMI are urgently needed. The study population was 755 consecutive patients presenting with ischemic chest pain from January 2002 to June 2004. All patients had electrocardiography and body surface mapping performed at initial presentation. AMI occurred in 519 patients (69%, cardiac troponin T or I level > or =0.1 ng/ml). Of these 519 patients, 303 (58%) had no ST-elevation on the initial 12-lead electrocardiogram. Ten patients were classified as having an "aborted AMI" and were included in the AMI analysis. The American College of Cardiology/European Society of Cardiology criteria for ST-elevation on 12-lead electrocardiogram identified 236 patients with AMI (sensitivity 45%, specificity 92%). Additional QRST features improved sensitivity (51% to 68%) but with decreased specificity (71% to 89%), with the optimal multivariate ECG model having a c-statistic of 0.75. The optimal BSM model identified 402 patients as having AMI (sensitivity 76%, specificity 92%, c-statistic 0.84). This improvement in sensitivity over the 12-lead electrocardiogram was due mainly to detection of ST-elevation in the high right anterior, posterior, and right ventricular territories and AMI in the presence of left bundle branch block. In conclusion, QRST variables added to criteria for ST-elevation result in improvement in sensitivity of the 12-lead electrocardiogram, although with decreased specificity. The BSM is superior in detecting AMI and demonstrates the importance of electroanatomic evaluation of patients with acute coronary syndromes.
我们旨在开发利用QRST变量测试ST段抬高标准的12导联心电图（ECG）模型，并将其在检测急性心肌梗死（AMI）方面的性能与80导联体表电位图（BSM）进行比较。由于非ST段抬高型AMI在全球的患病率正在上升，因此迫切需要在AMI的早期心电图检测方面取得进展。研究人群为2002年1月至2004年6月期间连续就诊的755例缺血性胸痛患者。所有患者在初次就诊时均进行了心电图检查和体表电位图检查。519例患者发生了AMI（69%，心肌肌钙蛋白T或I水平≥0.1 ng/ml）。在这519例患者中，303例（58%）在初始12导联心电图上无ST段抬高。10例患者被归类为“顿挫型AMI”并纳入AMI分析。美国心脏病学会/欧洲心脏病学会12导联心电图ST段抬高标准识别出236例AMI患者（敏感性45%，特异性92%）。额外的QRST特征提高了敏感性（从51%提高到68%），但特异性降低（从71%降低到89%），最佳多变量心电图模型的c统计量为0.75。最佳BSM模型识别出402例AMI患者（敏感性76%，特异性92%；c统计量0.84）。与12导联心电图相比，敏感性的提高主要归因于检测到高右前、后和右心室区域的ST段抬高以及存在左束支传导阻滞时的AMI。总之，在ST段抬高标准中加入QRST变量可提高12导联心电图的敏感性，尽管特异性有所降低。BSM在检测AMI方面更具优势，并证明了对急性冠状动脉综合征患者进行电解剖评估的重要性。
