Response to Letter by Hsia et al, Titled “Advanced Imaging for Mild Stroke Patients–Selection for Acute Interventions”

Abstract

We appreciate the interest by Dr. Hsia and colleagues in our article discussing potential advanced imaging-based selection paradigm for patients presenting with mild strokes. 1 The management and selection for reperfusion therapies is challenging in this cohort because of significant heterogeneity in presentation, natural history supporting higher likelihood of a good clinical outcome, and potential risks associated with reperfusion therapies. In our study, 2 the EXTEND-IA trials contributed 40 patients with NIHSS 0-5. The focus of our research was to evaluate mild strokes presenting with large vessel occlusions that are potentially amenable to endovascular therapy. While the overall incidence of patients presenting with mild strokes is high (~50% of all strokes), 3 patients with mild strokes and large vessel occlusion (LVO) represent a much smaller population (~5% of all strokes). 4,5 This population differs from patients with mild stroke without LVO in their higher risk of neurological deterioration, resulting in poorer outcomes. The procedural risks are also higher, 2, 6-8 thus the heightened need for patient selection. Most endovascular thrombectomy (EVT) centers are equipped with advanced imaging modalities, making it feasible to adapt advanced imaging-based selection paradigms for EVT in these patients.AHA/ASA guidelines recommend using MR DWI/FLAIR mismatch in extended time window (beyond 4.5 h) to determine eligibility for intravenous (IV) thrombolytics. 9 In Australian guidelines 10 and routine practice at both comprehensive stroke centers (CSCs) and primary stroke centers (PSCs), CT perfusion mismatch is used to identify patients for IV thrombolytic beyond 4.5 h and EVT beyond 6 h, with MR perfusion-diffusion mismatch an alternative if accessible within an acceptable timeframe. We agree with Dr. Hsia and colleagues that identifying imaging targets certainly helps in triage and decision making for IV thrombolysis in patients with mild strokes, even for patients presenting within 4.5 h of stroke onset. 11 Given that potential targets for IV thrombolytic include lacunar strokes that comprise an important proportion of non-LVO mild strokes, MRI would be ideal. However, this may lead to logistical challenges-especially in the emergency rooms of PSCs where MRI availability may be limited. Further systematic evaluation of such approaches, including outcomes in patients excluded from receiving IV thrombolytics based on such approaches and potential costeffectiveness analyses may be required to support their wider implementation.