The principal purpose of MRI in the evaluation of a patient with acute stroke is to determine the extent of tissue damage and to identify additional tissue at risk that is potentially salvageable. The combination of diffusion- and perfusion-weighted MRI is a frequently used protocol for this purpose. The evaluation is based on the premise that DWI would delineate the tissue that suffered permanent damage whereas the areas that do not cause signal change on DWI but have abnormal signal on perfusion-weighted images represent tissue at risk, the so-called ischemic penumbra (Hakim, 1998; Srinivasan et al, 2006). If a mismatch occurs between the extent of DWI changes and perfusion deficits (Sorensen et al, 1999), the latter being larger, reperfusion treatment with IV, or, beyond 3 hours, intraarterial thrombolytics or other intravascular techniques is justified to salvage the brain tissue at risk (Warach, 2002). If the extent of diffusion and perfusion abnormalities is similar or the same, the tissue is thought to be irreversibly injured with no penumbra; therefore, the benefit from reperfusion treatment is low or none.
These are useful guiding principles in general; however, several caveats must be kept in mind. Restriction of diffusion, DWI positivity, is not an infarct (ie, definite, permanent damage or loss of tissue) but is an ongoing ischemia. Once seen on DWI, the damage is most often irreversible, but not in every case, and there is chance for rescuing some tissue in the DWI-positive region as well. With regard to the perfusion images, a few words of caution need to be said. While DWI, are easy to read and the extent of deficit is usually very clear, perfusion MRIs are often more difficult to judge. Significant noise problems on calculated transit time, blood volume and blood flow images, the variable blood flow around stroke lesions, and artifact from large vessels may interfere with interpretation (de Crespigny, 2003). In one study, poor interobserver reliability was found in estimating diffusion-perfusion mismatch by visual inspection (Coutts et al, 2003). This may lead to underestimation or overestimation of the amount of tissue at risk. Another difficulty is that no standard perfusion measurements are available with which to compare the obtained data, although in human and animal stroke models PET and SPECT data have been used for validation (de Crespigny, 2003). Repeated measurements in human volunteers using PET and MRI perfusion, however, still showed the superiority of PET (Carroll et al, 2002).
Having said all this, the MR diffusion/perfusion protocols in acute stroke imaging are very useful in clinical decision making, and with further improvements in imaging methodology and image interpretation they will be essential for state-of-the-art acute stroke imaging.
