Perfusion Imaging in Acute Ischemic Stroke

Abstract

Background: Advancements in stroke imaging have allowed clinicians to more accurately select acute ischemic stroke patients for reperfusion therapies. This has moved patient selection from a time-based to an imaging-based paradigm. With few studies directly assessing the differences between various multimodal CT protocols, namely perfusion versus angiography, debate remains around the ideal imaging requirements for patient selection. Methods: Patients in the Endovascular Treatment for Small Core and Proximal Occlusion Ischemic Stroke (ESCAPE) trial who underwent computed tomography (CT) perfusion imaging were analyzed for their perfusion patterns. The number and outcome of mismatch and nonmismatch patients was analyzed and compared against their angiographic profile. Then a direct comparison between perfusion (CTP) and angiographic CT imaging (CTA), specifically multiphase CTA (mCTA) was performed wherein imaging experts were given only the angiographic images and were tasked with identifying suitable patients for reperfusion therapy. These results were then compared against the semi-automated perfusion data obtained by the Apollo MIStar program. Results: While most ESCAPE patients were found to have a penumbral pattern (90.6%), CTP identified non-penumbral patients that had a significantly poorer prognosis at 90 days. (mRS 0 – 2, 46% - 17%; p=0.041). In the direct comparison between CTP and mCTA, mCTA was relatively insensitive for a penumbral pattern (18.7%) with as many as 62.5% of large core patients being labelled as having moderate-good collaterals. Conclusions: In both the ESCAPE analysis and direct perfusion-angiography comparison, perfusion imaging was superior in identifying target mismatch and non-mismatch patients who were likely to respond well and poorly to reperfusion therapies respectively. Perfusion imaging, despite several limitations, allows for the most accurate and reliable selection of acute stroke patients.