• Imaging and diagnostics
• CT scanning
• MRI
• chagas' disease
• endomyocardial fibrosis

Sudden occlusion of a coronary artery initiates an expanding array of functional, metabolic and structural abnormalities, ultimately leading to myocyte necrosis, which extends from the subendocardial to the subepicardial layers of the perfusion bed—what Reimer called the ‘wavefront phenomenon’.1 In this setting, the myocardial area at risk (AAR) is defined as the myocardial tissue within the vascular territory that is distal to the culprit lesion of the infarct-related artery. If not timely reperfused, this area is at definite risk of ischaemic death. In humans, the portion of the AAR, that is irreversibly injured (ie, infarcted) ranges from 0% (aborted infarction) to as much as 88%.2 The proportion of the AAR that ultimately survives—that is, the salvaged myocardium—is dependent on multiple factors, including time to reperfusion, ischaemic preconditioning, collateral flow, distal embolisation, reperfusion injury and microvascular dysfunction. Since the extent of myocardial salvage is an important determinant of final infarct size, the quantitative assessment of myocardial salvage offers tremendous potential to aid in the development of novel therapeutic strategies targeted at reducing ischaemic/reperfusion injury in the setting of acute myocardial infarction (MI).

The extent of myocardial salvage is defined as the difference between the total AAR and final infarct size. Therefore, in recent years there has been a renewed interest in techniques used for the assessment of the myocardial AAR in patients with acute MI. However, in contrast to myocardial delayed enhancement MRI (MDE-MRI), which represents a robust and validated technique that allows for an accurate assessment and detailed characterisation of the infarcted regions,3 quantification of the AAR still remains challenging.

In animal studies, fluorescein staining and fluorescent microspheres are the reference standard for measuring the AAR.4–6 In humans, a commonly used technique is single-photon emission tomography, which requires the radioisotope to be injected during coronary occlusion and before reperfusion. Contrast echocardiography has also been used in the catheterisation laboratory before reperfusion. However, due to obvious logistic issues, both methods have had very limited clinical applicability. Recently, cardiovascular magnetic resonance (CMR) has emerged as a promising imaging modality that could allow for the ‘retrospective’ quantification of the AAR several days, and even weeks after reperfusion.7 8 Two CMR techniques have gained most attention: T₂-weighted (T₂w) imaging, and circumferential endocardial surface area (ESA) of MDE with transmural projection. T₂w imaging is sensible to myocardial oedema, and it is thought that the area of oedema can mark the original AAR. On the other hand, quantification of the AAR by the ESA method is based on the accepted concept that there is no wavefront circumferentially and, therefore, there should be no meaningful salvage at the lateral borders. This concept is based on the fact that the microcirculation of the human heart is composed of end-capillary loops that supply discrete perfusion fields.9 Numerous studies evaluating both CMR techniques have been published in the past few years and the results reported so far have been very encouraging.

In a recent issue of Heart, Versteylen et al10 present data from 78 patients with a first ST elevation AMI in whom the AAR… was quantified using five different approaches: the Aldrich electrocardiographic score; two angiographic scores (BARI and APPROACH); and two CMR techniques (T₂w imaging and the ESA technique). They compared the performance of these approaches using established physiological concepts based on the ‘wavefront phenomenon’ as the standard of reference. These concepts were: (1) AAR should always be larger than or equal to infarct size (IS); (2) in nearly transmural infarcts, AAR should approach IS; (3) as infarct transmurality increases, the correlation between AAR and IS should increase; and (4) myocardial salvage should be inversely correlated with infarct transmurality. The authors were able to demonstrate that CMR techniques agreed best with the physiological concepts, followed by the angiographic methods, with the electrocardiographic score exhibiting the worse agreement. Between both CMR techniques, the ESA method demonstrated a better degree of agreement than T₂w oedema imaging.

Recently, these CMR techniques, particularly T₂w imaging, have gained considerable interest as diagnostic tools for the quantification of the AAR. In fact, T₂w CMR imaging for delineating myocardial AAR is being increasingly considered to be ready for prime time. Not only is it being used to help determine patient management decisions in the clinical setting, it is also being used to provide primary or secondary endpoints in numerous trials including thousands of patients worldwide. However, there has been some controversy in this area. Some groups have brought several unsettling issues into discussion regarding the available evidence and questioned the status of these CMR techniques as accurate and validated methods for the assessment of the AAR and salvaged myocardium.11 In this context, the current study represents a welcome contribution,10 providing valuable new information about the relationship between the different approaches to quantify the AAR and important physiological premises based on the ‘wavefront phenomenon’ that are known to be true about the AAR.

It is important to recognise, however, that the current study suffers from a limitation, which was also present in most of the previous clinical studies that examined the value of CMR for the assessment of the AAR: the lack of an appropriate standard of reference. The majority of these previous reports simply compared one CMR technique against another,7 12–14 against angiographic scores10 12–14 and/or against pre-established physiological concepts.7 10 However, these are not adequate standards of reference and, thus, did not validate the CMR technique that was being tested. Limitations of angiographic scores are quite obvious due to complex cross-registration between coronary artery territories and myocardial segments. In addition, there is also an important issue related to the physiological concepts used to define the AAR. It is frequently assumed that, if a coronary artery were permanently occluded, the final infarct size/shape would be the same (or at least almost the same) as the myocardial AAR. Based on this concept, the wavefront of necrosis would progress towards the subepicardium until the entire transmural AAR was irreversibly infarcted. However, there is one fundamental aspect of the ‘coronary occlusion/myocardial injury pathophysiology’ that has often been neglected: the collateral blood flow. In practice, it is not uncommon in the clinical scenario to see a patient with a total LAD occlusion and only a small subendocardial infarct of the anterior wall with an area of peri-infarct myocardial ischaemia. This area of peri-infarct ischaemia may affect the entire territory supplied by the occluded LAD but, in some cases, the region of non-infarcted myocardium within the ‘AAR’ does not even demonstrate any signs of ischaemia. In such a situation, which would be the ‘true AAR’? Would it be the entire transmural myocardial territory supplied by the occluded coronary artery? Or should the ‘true AAR’ exclude the myocardial tissue that, despite being located within the territory of the occluded vessel, never suffered enough ischemic insult to result in myocardial necrosis due to the presence of collateral blood flow?

In theory, we could have an ‘anatomical AAR’ and a ‘functional AAR’ that possibly would have differences in size directly related to the magnitude of collateral flow. The problem is that using the ‘anatomical AAR’ instead of the ‘functional AAR’ to index salvaged myocardium would result in a higher variability and, therefore, would be problematic for evaluating a new therapy for AMI in a clinical scenario such as the one previously described. The importance of coronary collateral flow should not be minimised on the evaluation of the AAR and myocardial salvage. In practice, however, the assessment of coronary collateral blood flow is very difficult. It can only be assessed during the occlusion of the collateral-receiving artery by the measurement of intracoronary occlusive pressure or velocity derived collateral flow index, expressed as a fraction of flow during vessel patency.

T₂w oedema imaging is the CMR technique currently being used not only as a diagnostic tool in the clinical setting, but also as an endpoint in numerous ongoing clinical trials. However, has this technique been properly and conclusively validated? There are three small experimental animal studies that are cited as the demonstration that hyperintense regions on T₂w images delineate the AAR after acute MI.4–6 However, the conclusions of these studies were based primarily on size comparisons between CMR images and pathology, and none showed any images or data directly comparing the shape and contour of the T₂ abnormality with the shape and contour of the AAR as delineated by pathology. In fact, in the usual T₂w images the limits between hyperintense and non-hyperintense myocardium are extremely imprecise, leading to significant difficulties for manual or automatic quantification of theses areas and also for the contour delineation that would allow comparison with other techniques and/or pathology. This might be the result of a myriad of technical pitfalls related to T₂w images in the moving heart, leading to a debate even on the most basic aspect of this technique—that is, whether it does or does not accurately represent myocardial tissue oedema.11 Interestingly, there are also several small experimental studies suggesting that T₂w images, in fact, delineate the areas of infarction rather than the myocardial AAR.15–17 This apparent conflicting evidence illustrates the point that, based on our current knowledge, it is still precipitate to consider CMR-based techniques ready for prime time for the assessment of the AAR after acute MI. It is true that the evidence accumulated so far has demonstrated very interesting and encouraging results. It is also true that numerous clinical studies evaluating the value of CMR for the assessment of the AAR after acute MI have been published in recent years. It is very important, however, to acknowledge that further larger and well designed validation studies are urgently needed before the quantification of the AAR by CMR can be definitively ratified. In this regard, such a study should include measurements of collateral flow, that can affect not only the infarct size but also the true area at risk, and an improved T₂w image technique, such as T₂ mapping. Sometimes, we need to give one step back in order to achieve two steps forward.