Limited versus Whole-Brain Perfusion for the Indication of Thrombolysis in the Extended Time Window of Acute Cerebral Ischemia

Background Perfusioncomputed tomography (PCT) has emerged as alternative to magnetic resonance imaging (MRI) for assessment of patients clinically qualifying for off-label thrombolysis within 4.5 to 9 hours after onset of ischemic stroke. However, disadvantage of PCT is its often limited anatomic coverage with only 2 or 3 sliceswhen using a 4- to 64-section scanner. Our purpose was therefore to evaluate the value of 2- and 3- slice perfusioncompared to whole-brain perfusion. Methods One hundred twenty-five patients undergoing MRI beyond 4.5 hours after symptom onset with supratentorial perfusiondeficit were selected retrospectively. Accordingly to PCT slicepositioning, 2 or 3 slicesof the whole-brain perfusionweighted imaging data set were depicted. Volumes of infarct (using cerebral blood volume) and penumbra(using time-to-peak and cerebral blood volume) were calculated, and results were compared with 2- and 3- slice-derived volumes, respectively. Results Whole-brain imaging revealed a mismatch of more than 20% in 68.8% of patients (defined as 100%). Two- sliceimaging detected a perfusiondeficit in 72% and a mismatch in 48.8% (sensitivity = 70.9%). Three- sliceimaging detected a perfusiondeficit in 76% and a mismatch in 50.4% (sensitivity = 73.3%). Although there was no significant difference between 2- and 3- sliceimaging ( P > .23), both techniques revealed significantly less patients with mismatch compared to whole-brain coverage ( P Conclusions Two- and 3- sliceimaging like obtained with PCT on most installed CT systems to assess perfusiondeficits with subsequent mismatch calculation in acute stroke outside the 4.5-hour time window is significantly inferior to whole-brain coverage and, hence, has to be considered as a less-than- ideal solution.