Diagnostic synergy of non-invasive cardiovascular magnetic resonance and invasive endomyocardial biopsy in troponin-positive patients without coronary artery disease.

Baccouche et al. address a common problem of patients admitted to hospital because of biomarker-positive acute chest pain, who have no evidence of critical stenoses on coronary angiography. In this setting, in the TACTICS-TIMI18 trial the rate of death or reinfarction at six months was 3.1% [1].

The article strengthens the role of LGE-CMR as a noninvasive tool in the differential diagnosis of patients with TnI-positive acute chest pain and non-significant CAD. The method is safe, reproducible, and is becoming widely available. Arguably the most important feature of CMR is late gadolinium enhancement [2].
Different patterns of LGE distribution within the myocardium help to differentiate between ischaemic and non-ichaemic causes of the disease and in many cases provide information on the etiology of non-ischaemic myocardial injury (i.e. myocarditis, hypertrophic or dilated cardiomyopathy, amyloidosis, etc.). The pattern of LGE distribution can be especially useful in the evaluation of myocarditis and myocardial infarction [3].

Endomyocardial biopsy has been a gold-standard technique in the diagnosis of myocarditis for years [4]. Current recommendations for endomyocardial biopsy do not involve patients with troponin positive ACS and normal coronary arteries on coronary angiography, especially when they respond to conventional treatment, have no dysrrhythmia and deteriorating LV function [5]. The majority of patients with myocarditis, presenting with a mask of MI have excellent long-term prognosis [4].

However, late sudden deaths have been reported [1,4,6] and some of the patients have progression to dilated cardiomyopathy with time [7]. That is why, identifying the cause of the disease, explaining its pathogenesis, learning more about the myocardial tissue disease process may be helpful in defining patients’ prognosis [8]. On the other hand, endomyocardial biopsy is an invasive procedure, although safe in experienced hands, but with a small percentage of complications related to the procedure [4]. Death and cardiac tamponade have been reported, although no complications were observed in the presented paper. Other inherent limitations of endomyocardial biopsy are sampling error that may lead to a negative diagnosis with smoldering, patchy process of myocarditis. Interobserver variability in the interpretation of data is another issue.

In this paper, the authors used immunohistological criteria, developed initially for the diagnosis of a chronic inflammatory process in the setting of a clinically acute disease. The immunohistological criteria of active myocarditis applied by Baccouche et al. define the inflammatory infiltrate as ≥14 leukocytes/mm2 (CD3+ T lymphocytes and/or CD68+ macrophages) with additional myocardial damage, while borderline myocarditis – without myocardial damage/necrosis [9]. In order to assess chronic myocardial inflammation, other authors used the criterion of ≥7 leukocytes/mm2 (CD3+ cells) [10,11]. There are few reports on the results of endomyocardial biopsy in the setting of MI-mimicking myocarditis [4,12].

True acurate diagnosis of myocarditis may be life-saving, especially in the setting of rare forms of myocarditis, like giant-cell, eosinophilic myocarditis [13,14]. CMR is by no means helpful in defining the type of inflammatory infiltrate in the myocardium, and this knowledge influences on the way of treatment.
Kuhl et al. identified genomes of Parvovirus B19 in the myocardial tissues of patients presenting with ACS and normal coronary angiography [12]. Parvovirus B19 is known to infect endothelial cells and to cause obstruction of microcirculation, in this way leading to ischemia, micro-areas of necrosis, and elevation of necrotic enzymes [15]. Of interest, Mahrholdt et al. demonstrated that each of two most common agents causing viral myocarditis [parvovirus B19 (PVB19) and human herpesvirus 6 (HHV6)] can lead to a different pattern of myocardial damage and clinical course as demonstrated with LGE-CMR [16]. The identification of viral genome in the myocardium has prognostic implications [12,17].

In the presented paper, endomyocardial biopsy was shown to be superior to LGE-CMR in identifing myocarditis mostly due to its ability to reveal minor forms of the disease (borderline myocarditis). There are at least two causes which could have influenced this finding. First, endomyocardial biopsies performed in the presented study were all CMR-guided, which most likely improved the diagnostic accuracy of this procedure. Secondly, limitations of CMR protocol to cine and LGE-CMR could have underestimated the diagnostic potential of CMR in myocarditis. The study by Abdel-Aty et al. demonstrated that the diagnostic accuracy of CMR in myocarditis increased with the extension of LGE-CMR protocol by addition of two other sequences: T2W triple inversion recovery (to assess myocardial oedema) and T1W spin echo early after gadolinium injection (reflecting increased hyperemia and capillary leak) [18]. The best diagnostic accuracy was achieved with any two positive out of three sequencies yielding 85% diagnostic accuracy.

That study and also others led to the recent development of standardized CMR protocol and diagnostic criteria for myocarditis known as Lake Louise Consensus Criteria [19]. According to those criteria the presence of at least two out of three findings consistent with myocardial inflammation (regional or global myocardial oedema on T2W imaging, global hyperemia and capillary leak on T1W imaging early after gadolinium administration or at least 1 focal lesion with a nonischaemic pattern of necrosis/fibrosis on inversion recovery prepared T1W images late after gadolinium administration) in a patient with clinical suspicion of myocarditis are diagnostic of the disease. The presence of left ventricular dysfunction or pericardial effusion are also supportive [19]. If there is strong clinical evidence of myocardial inflammation and the onset of symptoms has been very recent but none of the criteria are present or if only one of the criteria is present, a repeat CMR study between 1 and 2 weeks after the initial one is recommended [19].

Conclusion:

To sum up, as suggested by the authors of the paper CMR should become the first line method in the diagnosis of unknown causes of TnI-positive acute chest pain without the evidence of CAD, with EMB limited to cases of unknown or uncertain diagnosis after CMR study and clinically important indications, e.g. deteriorating LV function, new atrio-ventricular blocks, complex ventricular arrhythmias. An appropriate use of these methods may help to overcome some limitations of individually applied techniques.