Stroke magnetic resonance imaging within 6 hours after onset of hyperacute cerebral ischemia

Abstract

We studied the diagnostic and prognostic value of diffusion- and perfusion-weighted magnetic resonancce imaging (DWI and PWI) for the initial evaluation and follow-up monitoring of patients with stroke that had ensued less than 6 hours previously. Further, we examined the role of vessel patency or occlusion and subsequent recanalization or persistent occlusion for further clinical and morphological stroke progression so as to define categories of patients and facilitate treatment decisions. Fifty-one patients underwent stroke magnetic resonance imaging (DWI, PWI, magnetic resonance angiography, and T2-weighted imaging) within 3.3 ± 1.29 hours, and, of those, 41 underwent follow-up magnetic resonance imaging on day 2 and 28 on day 5. In addition, we assessed clinical scores (on the National Institutes of Health Stroke Scale, Scandinavian Stroke Scale, Barthel Index, and Modified Rankin Scale) on days 1, 2, 5, 30, and 90 and performed volumetric analysis of lesion volumes. In all, 25 patients had a proximal, 18 a distal, and 8 no vessel occlusion. Furthermore, 15 of 43 patients exhibited recanalization on day 2. Vessel occlusion was associated with a PWI-DWI mismatch on the initial magnetic resonance imaging, vessel patency with a PWI-DWI match (p < 0.0001). Outcome scores and lesion volumes differed significantly between patients experiencing recanalization and those who did not (all p < 0.0001). Acute DWI and PWI lesion volumes correlated poorly with acute clinical scores and only modestly with outcome scores. We have concluded on the basis of this study that early recanalization saves tissue at risk of ischemic infarction and results in significantly smaller infarcts and a significantly better clinical outcome. Patients with proximal vessel occlusions have a larger amount of tissue at risk, a lower recanalization rate, and a worse outcome. Urgent recanalization seems to be of utmost importance for these patients. Ann Neurol 2001;49:460–469