The results of the present study can be summarized as follows. First, ECG underestimates the extent of myocardial injury among patients with infarct-like myocarditis; overall topographic agreement between the site of LGE and site of STE was only 68% in the present study. Second, some ECG parameters, including sumSTE, normalization of STE >24 hours, and development of negative T wave, are significantly and independently related to the extent of myocardial damage.
Electrocardiography in Infarct-Like Myocarditis and Relation to Myocardial Damage
Several studies have previously shown that a subset of AM patients may have an infarct-like presentation, with chest pain, elevated Tn levels, and ECG findings suggestive of STEMI.2–5 This pattern of clinical presentation has been related to parvovirus B19 infection of the endothelial cells of myocardial vessels, causing endothelial dysfunction and coronary vasospasm, migration of inflammatory cells into the myocardial interstitium, and subsequent myocyte damage.21,22
Cardiac MRI with the LGE technique has emerged as an important tool for the diagnostic workup of these patients, noninvasively distinguishing AM from ischemic injury and other nonischemic conditions with infarct-like presentation, such as Takotsubo cardiomyopathy.7 Following a myocardial infarction, LGE with subendocardial or transmural pattern is typical.7 Among patients with AM, LGE is typically epicardial, in the mid-wall, or patchy.7 Among patients with Takotsubo cardiomyopathy, which is characterized by a typical pattern of LV dysfunction, significant LGE is conversely usually absent.23,24
Electrocardiographic repolarization changes in infarct-like myocarditis have been described5,25,26; STE, attributed to epicardial inflammatory injury, is followed by gradual ST-segment normalization and subsequent variable occurrence of T-wave inversion, attributed to epicardial damage. Finally, the ECG normalizes, with resolution of T-wave abnormalities; of note, development of Q wave is rarely observed. Recent cMRI studies have extensively investigated the relation between repolarization changes observed in STEMI and infarct size and extent of myocardium salvaged by reperfusion therapies.9–14 Conversely, little is known about the clinical meaning of repolarization changes in the setting of infarct-like myocarditis. Karjalainen and Heikkila evaluated the relation between the amount of myocardial damage, assessed as peak creatine kinase MB value, and ECG changes in 18 young men with infarct-like myocarditis. They observed that the peak creatine kinase MB value was significantly higher in patients with conspicuous STE and with subsequent T-wave inversion.27 More recently, Deluigi et al28 and Di Bella et al29 evaluated the relation between the site of repolarization abnormalities and location of myocardial injury assessed by cardiac MRI in patients with AM (including 20 and 46 patients with infarct-like myocarditis, respectively); both groups found a weak correlation between ECG leads showing repolarization abnormalities and location of LGE. The results of the present study, which used cMRI as a reference technique for the evaluation of myocardial damage, confirm and extend these previous observations. First, we observed that the site of STE is not a perfect predictor of the region of myocardial injury among both groups of patients presenting with anterolateral or inferolateral STE; topographic agreement between the site of LGE and the site of STE was only 59% and 46%, respectively. Second, the amount of STE (sumSTE), late normalization of STE (ie, >24 hours), and development of negative T wave were found to be significantly and independently related to the extent of LGE, suggesting that these ECG indexes could be used for a fast bedside estimation of the extent of myocardial damage in this group of patients. This finding is novel in the setting of infarct-like myocarditis and parallels recent cMRI observations in the setting of STEMI, which documented a relation between sumSTE and the extent of myocardial injury and between ST-segment resolution and improvement of myocardial damage owing to reperfusion.9–14 Interestingly, Naruse and colleagues recently observed a relation between time to ECG normalization and LGE among patients with Takotsubo cardiomyopathy as well.30 Third, none of the patients included in the present study developed pathologic Q waves. Moon et al have shown that the presence of Q wave in ischemic heart disease is mainly related to the extent of infarct size and, secondarily, to its transmural extent31; accordingly, the scattered nature and the absence of transmurality of myocardial damage in AM may explain the lack of development of Q waves in this group of patients.
This study has some limitations that should be acknowledged. First, the study population was relatively small; however this reflects the low incidence of the disease.5 Second, endomyocardial biopsy, which is considered the gold standard for the diagnosis of myocarditis,32 was not performed; however, all included patients had cMRI findings consistent with AM, in agreement with current criteria.6 Moreover, according to current guidelines, endomyocardial biopsy is not indicated in patients with suspected AM presenting without heart failure symptoms and preserved LV systolic function.33