• advanced cardiac imaging
• cardiac imaging and diagnostics
• cardiac magnetic resonance (CMR) imaging

Patients with acute chest pain and elevated troponin concentrations usually undergo invasive coronary angiography unless there are obvious non-coronary causes of this constellation. It thus became apparent that up to 10% of such patients do not have obstructive coronary artery disease. In women, this group may even be larger. The catchy name for this phenomenon is MINOCA (Myocardial Infarction with Non-Obstructed Coronary Arteries). Although the definition seems straightforward, MINOCA is a heterogeneous entity comprising takotsubo syndrome (TTS), myocarditis, plaque rupture with embolisation, coronary dissection, etc. Scientific information about MINOCA patients suffers from variable inclusion and exclusion criteria. For instance, studies differ with respect to the proportion of patients with smooth coronary arteries and those with plaque up to 50% diameter stenosis. In many studies, the cause of MINOCA has not been systematically identified, for instance, by using cardiac magnetic resonance (CMR). Consequently, recent data on the prognosis of MINOCA patients show a huge variation in annual mortality from 0.6% up to 8.1%.

CMR is helpful to elucidate the underlying causes leading to a MINOCA presentation. It is the method of choice for the non-invasive diagnosis of myocarditis which is difficult to make on clinical grounds alone. Furthermore, if performed as early as possible following coronary angiography (which is nowadays often done without a left ventricular angiogram), it is ideally suited to detect the typical wall motion abnormalities and absence of late gadolinium enhancement associated with TTS. CMR can also detect minute subendocardial areas of acute myocardial infarction. Surprisingly, such areas have been found in a rather constant proportion of 20%–25% in all-comer studies of MINOCA patients using this technique.1–3

Our knowledge of the diagnostic spectrum found in MINOCA patients who were systematically studied by CMR within <4 days following the acute event is broadened by the study by Bhatia et al from the Mayo Clinic published in Heart.4 In a comparatively large cohort of 215 patients fulfilling MINOCA criteria, myocarditis was found in 32%, small areas of myocardial infarction in 22%, non-ischaemic cardiomyopathy in 20% and TTS in 9%. Another strength of the study is the short period between the onset of symptoms and the performance of CMR. Thus, TTS which may disappear quickly was still captured in a sizeable number of patients. No diagnosis was found in 17% of patients.

How do these findings compare with other CMR studies evaluating the spectrum of pathologies associated with MINOCA? Table 1 shows the studies which are currently available. A consistent pattern of findings can be seen in these studies which is related to the selection of patients. The studies by Assomull et al,5 Laraudogoitia Zaldumbide et al,6 Monney et al 7 and Leurent et al 8 focused on younger patients below the age of 50 years with a predominance of males. Young males are most often affected by myocarditis of the infarct-like type with a presentation of chest pain, raised troponin and normal coronary arteries. Hence, it is not surprising that these authors came to the conclusion that the overwhelming majority of MINOCA patients (up to 81%) have myocarditis as the underlying disorder. One might call this a self-fulfilling prophecy. However, the diagnosis of myocarditis has important implications for the patient: while the prognosis of such patients is generally good—which is reassuring—the patient should be advised to abstain from vigorous exercise for a period of approximately 3 months. At this time, it is recommended to repeat CMR in order to see whether tissue injury has regressed. Often, neither T2 prolongation nor late gadolinium enhancement is still to be found at this point in time. This is even more reassuring since it is very unlikely that such a patient will suffer from malignant ventricular arrhythmias in the future.

The study by Mahmoudi et al 9 is difficult to interpret since CMR examinations were performed at a mean of 2 months after the acute event. Therefore, given the transient nature of TTS and T2 prolongation/signal hyperintensity in myocarditis it is not surprising that 73% of studies were normal.

The study by Bhatia et al 4 assessed an all-comer population similar to the studies published by Chopard et al, Collste et al and Dastidar et al.1–3 The patient population was older (>50 years, mean age close to 60 years) with a balanced male/female ratio or even a female preponderance. All four studies are consistent in reporting a percentage of 19%–25% of MINOCA patients who were ultimately found to have gone through a myocardial infarct despite the absence of epicardial coronary obstructions. Given that many of the patients included in the studies had diffuse coronary artery disease (although this was not stenosing, see figure 2 in the paper by Bhatia et al 4), this had to be expected. Myocardial infarctions may be caused by the rupture of unstable plaques with subsequent embolisation of cholesterol material and superimposed thrombi to the periphery of the coronary artery. However, detection of such plaque ruptures may be difficult by conventional invasive coronary angiography which makes CMR a very helpful additional tool in the workup of such patients. The consequences of the finding are obvious: these patients deserve the same intensive modification of risk factors as their counterparts with stenosing coronary disease and non-ST-elevation myocardial infarction. Furthermore, this finding should trigger a search for coagulation abnormalities which may be associated with MINOCA.

Some of the studies3 5 reported the unexpected CMR diagnosis of a non-ischaemic cardiomyopathy in MINOCA patients but others did not. It is unclear whether this was due to excluding such patients in some studies or whether such patients were simply not in the cohort. In the study by Bhatia et al, 20% were found to have a non-ischaemic cardiomyopathy but this was not further detailed in the paper. Presumably, these patients had dilated or hypertrophic cardiomyopathy. Both entities are known to be associated with elevated troponin values. It would have been interesting to learn why these cardiomyopathies were not diagnosed during cardiac catheterisation or prior echocardiography.

Depending on how early after the acute clinical event CMR was performed, patients with TTS were more or less frequent (table 1). This is not surprising because the wall motion abnormalities associated with TTS tend to resolve quickly. Although the debate continues, a latent microvascular abnormality may be the fertile ground on which the psychological stressor can trigger microvascular spasm ultimately causing the peculiar reversible wall motion abnormalities characteristic of the syndrome.

What about those MINOCA patients (around 25%, table 1) in whom CMR cannot find an underlying abnormality? In the study by Bhatia et al, this group comprised 17% of MINOCA patients. One can imagine that thromboembolism secondary to plaque rupture may lead to microvascular obstruction not extensive enough to lead to even small areas of late gadolinium enhancement by CMR. However, many observations point to coronary functional abnormalities as another frequent cause of MINOCA. In order to prove the presence of coronary functional abnormalities, one needs to perform intracoronary flow/resistance measurements at rest and after the application of adenosine and/or intracoronary acetylcholine spasm provocation testing. The presence of microvascular dysfunction can also be ascertained non-invasively by using myocardial blood flow measurements by positron emission tomography or by CMR. As the study by Bhatia et al was retrospective and patients were included back to the year 2000, these novel and sophisticated CMR measurements were not available. The next step in the CMR evaluation of MINOCA patients would be to include such measurements.

The study by Bhatia et al reminds us that much has to be done before we will be able to finally unravel all the complexities of MINOCA. The study should stimulate us to perform CMR in all MINOCA patients, although in the study by Bhatia et al only 3% of all patients with MINOCA underwent further testing by CMR.

In summary, the study by Bhatia et al 4 confirms in a large number of carefully studied patients that myocarditis and small myocardial infarcts are frequent in patients with MINOCA. Furthermore, CMR if performed early after the event as in this study will still capture cases of TTS. The mechanism of MINOCA in patients with cardiomyopathies and those with normal CMR findings remains unclear. A combination of coronary function testing and CMR will most likely yield the best diagnostic and therapeutic results.