Résumé
Recently, six randomized controlled trials (RCTs) of endovascular treatment (EVT) versus medical management (MM) in anterior circulation large vessel occlusion with large core documented significant benefit of EVT on functional outcome. Moreover, one trial reported benefit of EVT in the very large core category (Alberta Stroke Program Early CT Score<p>[ASPECTS] 0-2). These results are considered paradoxical by some as they contradict the prevailing view that the presence of large core precludes the possibility of good outcome following reperfusion. They in turn led some investigators to question the applicability of the core/penumbra model in the case of large core stroke, and even its overall validity, specifically regarding the notion that the core reliably predicts tissue infarction. Here we discuss the trial results and propose alternative explanations for the large core paradox. First, although EVT does improve outcome as compared to MM, overall outcome remains poor in ~80% of the treated population. Second, the assessment of core extent on imaging, particularly with CT, is potentially inaccurate, especially in the early time-window. Third, consistent with observational studies, some RCT sub-studies suggest that the benefit of EVT in this population derives at least in part from salvage of penumbra, which appears to have been present in a large percentage of enrolled patients. Fourth, the markedly reduced perfusion that prevails within large cores facilitates the early development of vasogenic edema. This heterogeneity of tissue injury may in turn lead to an overestimation of true core/neuronal death as estimated with CT and MRI.</p><p>Assessing patients with apparent large core should consider these notions when discussing eligibility for EVT. Early reperfusion of large core patients is expected to both target any residual penumbra and prevent the development of vasogenic edema within the severely hypoperfused areas. These considerations underscore the need for more reliable methods to identify irreversible neuronal injury inside the imaging-based estimated core.