CT Perfusion Imaging versus MR Diffusion Weighted Imaging: Prediction of Final Infarct Size in Hyperacute Stroke
Background: Both CT perfusion imaging (CTP) and MR diffusion weighted imaging (DWI) have been applied clinically to predict tissue outcome in patients with hyperacute stroke, however these techniques have not been directly compared in the same patient population. In CTP, whole brain CT scanning is performed during the steady state administration of a contrast bolus, creating both perfused blood volume weighted and CT angiographic images. Purpose: To determine the value of CTP imaging, compared to DWI, in predicting final infarct size in hyperacute stroke patients. Materials and Methods: CTP followed by DWI imaging was performed in 23 consecutive patients presenting within 12 hours of stroke onset. Mean time from stroke onset to imaging was 4.6 hours for CTP, 5.4 hours for DWI, and 6.5 days for conventional nonenhanced CT and/or MR follow-up. No patients had secondary events between initial and follow-up scanning. Initial and follow-up ischemic volumes were computed from the CTP, DWI, and follow-up scans using image segmentation software, and compared using linear and multiple regression. Results: CTP and DWI volumes were independent predictors of final infarct size (p=0.02). The regression line for CTP volume vs. final infarct size had slope 1.02 and r2 = 0.88 (p<0.0001). The regression line for DWI volume vs. final infarct size had slope 1.41 and r2 = 0.92 (p<0.0001). Overall sensitivity and specificity for parenchymal stroke detection were 83% and 100% for CTP, and 100% and 100% for DWI, respectively. Conclusion: Although DWI is more sensitive than CTP for parenchymal stroke detection, both DWI and CTP are highly accurate predictors of final infarct volume. DWI tends to underestimate final infarct size, whereas CTP more closely approximates final infarct size.