Hyperacute T waves are specific for occlusion myocardial infarction, even in the absence of STEMI criteria

Abstract Background Hyperacute T waves (HATWs) are widely recognized as an important electrocardiographic (ECG) marker of acute coronary occlusion myocardial infarction (OMI). The American College of Cardiology (ACC) and European Society of Cardiology (ESC) recommend identifying HATWs as a STEMI equivalent. However, to date there have been no objective criteria for HATW proposed. Purpose We sought to derive and validate a quantitative definition of HATWs as a diagnostic indicator for OMI in a cohort of possible acute coronary syndrome (ACS) patients in the Emergency Department (ED), specifically in patients whose ECGs do not meet STEMI criteria. Methods We derived the HATW score by logistic regression based on expert-labelled HATWs in patients with angiographically proven acute OMI, then validated the HATW score as a diagnostic indicator of OMI on a separate population of ED patients with suspected ACS. The primary analysis was sensitivity (at 98% specificity) of the HATW score for OMI in the validation group patients without STEMI criteria. From 5 PCI centers in USA and Europe, we collected 3,474 ECGs from 2,656 adult patients undergoing ED evaluation for possible ACS. We excluded patients with no ECG available, QRS duration 110 msec, and those with elevated troponin but no angiogram due to inability to determine the primary outcome definition. 1,291 and 1,395 were allocated to the derivation and validation groups. The first available ECG was measured by proprietary software (PMCardio) for all ECG measurements including STEMI criteria (Fourth Universal Definition of Myocardial Infarction). OMI was defined as a culprit lesion with TIMI 0 or 1 flow. We hypothesized HATWs would be best defined not by amplitude, but by "magnitude" (area under the curve of the T wave relative to the QRS amplitude) and T wave symmetry (Time from T wave peak to end, relative to Time from T wave start to peak). We analyzed limb leads, and right and left precordial leads separately. Results The derivation group included 2,079 ECGs from 1,261 patients (21% with OMI), with expert annotations yielding 1,401 leads with HATWs and 18,062 without. Logistic regression of T wave magnitude and symmetry yielded the optimal HATW score presented in Figure 1, example case shown in Figure 2. The validation group included 1,395 patients (10.6% OMI). Among the 1,300 patients without STEMI criteria, the HATW score (using 2 contiguous leads with mean score 0.7) had 98.4% specificity, 20.7% sensitivity, 47.4% PPV, and 94.5% NPV for OMI. Conclusion HATWs can be defined by increased T wave magnitude and symmetry. Our resulting HATW score shows clinical utility as an ECG finding of OMI, even in patients without STEMI criteria. While further validation studies are needed, our findings add further support to the ACC and ESC recommendations that HATWs be identified and treated as STEMI equivalents, or ECG features of acute coronary occlusion myocardial infarction.