Expanding horizons in the endovascular treatment of stroke: larger cores and adjunct thrombolytics Ain A. Neuhaus and Alastair M. Buchan* Acute Stroke Programme, Radcliffe Department of Medicine, University of Oxford, Oxford OX3 9DU, UK Keywords Stroke • Endovascular thrombectomy • Penumbra • Thrombolysis • Intra-arterial thrombolysis Endovascular thrombectomy (EVT) has been firmly established as the gold standard of treatment in emergent large vessel occlusion (ELVO) acute ischaemic stroke. As of 2015, efficacy of EVT has been corrobo- rated by randomized clinical trials and subsequent meta-analyses 1 con- firmed by ongoing real-world registry data. The emphasis has been on preventing viable but ischaemic brain tissue (penumbra) from becoming irreversibly infarcted and therefore recruited to the ‘core’ by recanaliz- ing the affected large vessel and restoring microvascular reperfusion. As such, the patient selection criteria have focused on those with small core infarcts, often defined as Alberta Stroke Programme Early CT Score 2 (ASPECTS; Figure 1) ≥6 on non-contrast computed tomography (NCCT) based on the initial clinical trial criteria. Alternative approaches, derived from the late-window EVT trials, have included a mismatch be- tween core volume and clinical severity on the National Institutes of Health Stroke Scale (NIHSS) with small absolute core volumes or a penumbra-core ratio of >1.8 and absolute core volume <70 mL based on perfusion imaging thresholds. 3 In addition to maximizing treatment benefit, restricting treatment to small cores is thought to reduce the risk of symptomatic intracranial hemorrhage (sICH). There have been comparatively little data on those patients with large cores as they were largely excluded from the initial trials and remain under-represented in registries. Meta-analysis of seven EVT trials found that in patients with ASPECTS 0–4, EVT was associated with an overall favorable odds ratio for functional independence, defined as modified Rankin scale (mRS) 0–2; however, this was limited by small sample sizes and wide confidence intervals and a four-fold increase in sICH. 4 A fur- ther meta-analysis including randomized controlled trials, prospective, and retrospective cohort studies further supported the efficacy of EVT in low ASPECTS patients, with increased odds of functional inde- pendence and reduced mortality after EVT despite a large infarct core. 5 RESCUE-Japan LIMIT was a randomized clinical trial of thrombec- tomy in 203 patients with ELVO stroke and a large core infarct, with ASPECTS 3–5 on NCCT or MRI. 6 The study did not include ASPECTS 0–2 patients, as they were presumed to have very poor out- comes regardless of treatment. The odds ratio of achieving mRS 0–3 with EVT compared to medical management was 2.43 (31% vs. 12.7%), and there was an overall significant beneficial shift on the mRS scale. EVT patients were also more likely to exhibit early improvement in NIHSS, with a comparable safety profile. It is worth noting that the primary functional outcome in RESCUE-Japan LIMIT included mRS 3 (moderate disability requiring help but independently mobile), in con- trast to the typical cut-off of mRS 0–2; their secondary analyses for mRS 0–1 and 0–2 did not reach statistical significance. However, this is to be expected given the severity of strokes on initial presentation and still reflects a clinically meaningful effect as further supported by their post-hoc utility-weighted mRS analyses. These findings challenge certain assumptions about clinicopathologi- cal and radiological correlations in the core infarct. First, in the case of ASPECTS, we are classifying relatively large regions of brain on a binary system, which does not consider the volume of ischaemic changes within a region. Second, our ability to classify the severity of ischaemia is limited, and it is likely that a region considered as infarct on ASPECTS will have heterogenous degrees of neuronal and glial death between patients. Even with perfusion imaging, the commonly used threshold of <30% relative cerebral blood flow does not necessarily equate to irreversible pan-necrosis. 7 As such, there may be salvageable tissue within the areas that we would radiologically consider as core. We expect that there will be heterogeneity in the viability thresholds and mechanisms of cell death affecting different cell populations—not only neurons but also the broader neurovascular unit including astrocytes, pericytes, microglia, and others—but our understanding of how to optimize reperfusion based on these remains very limited. A major limitation of RESCUE-Japan LIMIT is its use of diffusion-weighted magnetic resonance imaging (DWI) in >85% of included patients. DWI is significantly more sensitive to ischaemic changes, particularly at early time points, 8 and it is worth noting that >55% of patients in the trial were ran- domized within 4.5 hours of onset with a median onset to imaging time of approximately 3 hours. It is therefore plausible that many of the patients in- cluded in this study would have qualified as ASPECTS ≥6 on NCCT, which limits its external validity as most centres use on CT-based approaches. Further data are now required to validate these initial findings, especially with low ASPECTS on NCCT, and to this end there are ongoing trials in- cluding SELECT-2 (NCT03876457), TENSION (NCT03094715), TESLA (NCT03805308), and LASTE (NCT03811769). Nonetheless, this result is a promising step forward for the management of larger infarcts and the associated enormous morbidity and mortality. * Corresponding author. Tel. +44 (0)1865 220346, Email: alastair.buchan@medsci.ox.ac.uk © The Author(s) 2022. Published by Oxford University Press on behalf of the European Society of Cardiology. All rights reserved. For permissions, please email: journals.permissions@oup.com. Cardiovascular Research (2022) 00,1–5 https://doi.org/10.1093/cvr/cvac145 CLINICAL COMMENTARIES Downloaded from https://academic.oup.com/cardiovascres/advance-article/doi/10.1093/cvr/cvac145/6722304 by guest on 30 September 2022