Diffusion- and perfusion-weighted MRI response to thrombolysis in stroke

The authors review the theoretical basis of determination of cerebral blood flow (CBF) using dynamic measurements of nondiffusible contrast agents, and demonstrate how parametric and nonparametric deconvolution techniques can be modified for the special requirements of CBF determination using dynamic MRI. Using Monte Carlo modeling, the use of simple, analytical residue models is shown to introduce large errors in flow estimates when actual, underlying vascular characteristics are not sufficiently described by the chosen function. The determination of the shape of the residue function on a regional basis is shown to be possible only at high signal-to-noise ratio. Comparison of several nonparametric deconvolution techniques showed that a nonparametric deconvolution technique (singular value decomposition) allows estimation of flow relatively independent of underlying vascular structure and volume even at low signal-to-noise ratio associated with pixel-by-pixel deconvolution.

This report evaluates several methods to map relative cerebral blood flow (rCBF) by applying both parametric and nonparametric techniques to deconvolve high resolution dynamic MRI measurements of paramagnetic bolus passages with noninvasively determined arterial inputs. We found a nonparametric (singular value decomposition (SVD)) deconvolution technique produced the most robust results, giving mean gray:white flow ratio of 2.7 +/- 0.5 (SEM) in six normal volunteers, in excellent agreement with recent PET literature values for age-matched subjects. Similar results were obtained by using a model-dependent approach that assumes an exponential residue function, but not for a Gaussian-shaped residue function or for either Fourier or regularization-based model-independent approaches. Pilot studies of our CBF mapping techniques in patients with tumor, stroke, and migraine aura demonstrated that these techniques can be readily used on data routinely acquired by using current echo planar imaging technology. By using these techniques, the authors visualized important regional hemodynamic changes not detectable with rCBV mapping algorithms.

Identifying tissue at risk for infarction is important in deciding which patients would benefit most from potentially harmful therapies and provides a way to evaluate newer therapies with regard to the amount of ischemic tissue salvaged. To operationally define and characterize cerebral tissue at risk for stroke progression. We retrospectively selected 25 patients with an acute onset of a hemispheric stroke from our database who had undergone a combination of two diffusion-weighted MRI studies and a perfusion-weighted MRI study. We applied a logistic regression model using maps of the relative mean transit time and relative cerebral blood flow (rCBF) as well as three different maps of the relative cerebral blood volume (rCBV) to predict an operationally defined penumbra (region of mismatch between the diffusion lesion on day 1 and its extension 24 to 72 hours later). Maps of the rCBF and initial rCBV were significant predictors for identifying penumbral tissue. Our operationally defined penumbral region was characterized by a reduction in the initial rCBV (47% of contralateral control region [CCR]), an increase (163% of CCR) in the total rCBV, and a reduction (37% of CCR) in the rCBF, whereas the operationally defined ischemic core showed a more severe reduction in the rCBF (12% of CCR) and in the initial rCBV (19% of CCR). These MR indexes may allow the identification and quantification of viable but ischemically threatened cerebral tissue amenable to therapeutic interventions in the hyperacute care of stroke patients.