Diffusion- and Perfusion-Weighted MRI

Abstract

Background and Purpose-Diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) have been used increasingly in recent years to evaluate acute stroke in the emergency setting. In the present study, we compared DWI and PWI findings in acute stroke patients with and without severe extracranial internal carotid artery (ICA) disease. Methods-Twenty-seven patients with nonlacunar ischemic stroke were selected for this analysis. DWI, PWI, and conventional MRI were performed in all patients within 24 hours of symptom onset and after 1 week. To exclude patients with partial or complete reperfusion, we included only patients with a PWI deficit larger than the DWI lesion. Severe ICA disease (Ͼ70% stenosis) was present unilaterally in 9 and bilaterally in 2 patients. Acute DWI lesion volume, the size of the acute PWI/DWI mismatch, and final infarct size (on T2-weighted images) were determined. Results-The PWI/DWI mismatch was significantly larger in patients with severe ICA disease than in patients without extracranial carotid stenosis, both when time-to-peak and mean transit time maps (PϽ0.01) were used to calculate the mismatch. Quantitative analysis of the time-to-peak delay in the mismatch indicated that a relatively smaller fraction of the total mismatch was critically ischemic in patients with carotid stenosis than in those without. Average lesion volume increased less in the stenosis group (Pϭ0.14), despite the larger PWI/DWI mismatch, and final infarct size was smaller in the stenosis group (PϽ0.05). In the 2 patients with bilateral ICA disease, variable hemodynamic involvement of the contralateral hemisphere was found in addition to the ipsilateral PWI deficit. Conclusions-In most acute stroke patients with severe ICA stenosis, a considerably smaller fraction of the total PWI/DWI mismatch is at risk than in patients without carotid disease.

Significance of the perfusion‐diffusion mismatch in chronic cerebral ischemia

“…Thus, multiparametric MRI of brain perfusion appears to be suitable for differentiating between delayed perfusion resulting from chronic hemodynamic impairment in the large cerebral arteries and manifest ischemia threatening brain tissue. Our findings agree with and extend an earlier study that showed that brain perfusion is delayed in the presence of occlusion of the ICA and additional arteries of the circle of Willis (19). The current evidence suggests that in chronic cerebral ischemia with recurring ischemic events over weeks the perfusion-diffusion mismatch may be as severe as in acute stroke.…”

“…Recent observations have shown that brain infarcts may become as large as predicted by the perfusion abnormalities in acute MR images (7,14), they may regress (15,16), or they may deteriorate despite arterial recanalization (17,18). Moreover, it is well established that chronic internal carotid artery (ICA) occlusion is accompanied by a prominent perfusion impairment in the downstream brain region (19). We encountered a number of patients who presented with recurring transient ischemic attacks and a severe perfusion-diffusion mismatch but remained clinically stable and did not develop structural brain lesions as evident from T2-weighted (T2W) images.…”

The Performance of MRI-Based Cerebral Blood Flow Measurements in Acute and Subacute Stroke Compared With 15O-Water Positron Emission Tomography

“…According to previous studies of TTP maps in patients with high-grade ICA stenosis, vessel pathology was supposed to influence the reliability of PWI maps. 37,50 However, we found the performance of CBF MRI not related to the presence of extracranial vessel pathology because an ipsilateral stenosis or occlusion did not significantly change the C-ratio. This discrepancy to the previous studies can be explained by the fact that CBF maps are less sensitive to collateral flow than TTP maps.…”

“…TTP maps without AIF have been frequently studied, because they are easy to obtain, robust, and well comparable among different imaging facilities. 8,12,37 Regarding the underlying principles, however, AIF-based maps may be superior and maps of the mean transit time or the CBF should be favored. Because MRI-based CBF was seldom used in clinical trials and because a clinically relevant volumetric approach is not yet available, our study adds important evidence to the validation of MRI-based CBF measurement.…”

“…44 These findings differ from studies of acute ischemia in which AIF selection does affect quantitative parameters. 37,50 A manually chosen AIF is highly observer dependent, whereas automatic algorithms could be helpful but are not fully validated yet. 55 We therefore constantly derived the AIF from intravascular voxels 33 of the first 2 segments of the MCA or from the distal ICA of the unaffected hemisphere.…”