Use of MRI to Estimate the Therapeutic Window in Acute Stroke

Abstract

Marc Fisher MD Kennedy Lees MD Section Editors: The fundamental goal of many acute stroke therapies is rapid arterial recanalization leading to reperfusion of critically hypoperfused brain tissue. Clinical benefits of this therapeutic approach require the presence of a salvageable penumbra. It is estimated that salvageable tissue is present in up to 80% of patients with stroke who present rapidly, but these capricious zones of potentially recoverable parenchyma typically disappear within the first 6 to 12 hours after symptom onset.1 The rate of disappearance of the penumbral tissue appears to vary considerably between individuals based on a variety of physiological factors, predominantly the availability of adequate collateral circulation. How to quickly and reliably identify these fortunate patients who may have a prolonged therapeutic window has been the ultimate quest of a large volume of modern cerebrovascular neuroimaging research. A leading approach to this challenge has been to estimate the ischemic penumbra based on the difference between the volume of tissue that exhibits a disturbance in cerebral blood flow, as assessed by perfusion-weighted MRI (PWI), and the volume of tissue that has already developed evidence of advanced ischemic injury reflected by cytotoxic edema on diffusion-weighted MRI (DWI).2 The “mismatch” regions (areas of PWI abnormality that do not have corresponding DWI lesions) have been considered likely to benefit from reperfusion therapies. This hypothesis has been assessed in a variety of recent clinical trials, including DIAS, DEDAS, DIAS II, DEFUSE, and now, most recently, EPITHET.3–7 EPITHET was a randomized, double-blind, placebo-controlled trial designed to determine whether intravenous tissue plasminogen activator, administered 3 to 6 hours after stroke onset, would reduce infarct growth in patients with PWI/DWI mismatch.7 Because technology to rapidly and accurately determine which patients have a PWI/DWI mismatch was not available, both mismatch and nonmismatch patients were …