MRI measurements of coronary flow and flow reserve were in good agreement with US measurements. In addition, MRI measurements of coronary flow reserve successfully discriminated stenotic from normal vessels. These results indicate that MRI is a useful method for the noninvasive assessment of coronary flow and stenosis.

Five venous thrombi were induced in the external jugular veins of three laboratory dogs, and were repeatedly imaged over 3 weeks using a 0.35-T magnetic resonance (MR) imager. MR magnitude and phase images, T1 and T2 relaxation times, venography, and histologic sections of these thrombi were evaluated to determine the changes in appearance on MR images with time. Venous thrombi appeared hyperintense compared with muscle on both relatively T1- and T2-weighted spin-echo sequences regardless of the age of the clot. Organization of the thrombus beyond 1 week was manifested as increased prominence of flow signal void in and around the clot. Distinction between intraluminal thrombus and flow-related artifacts was aided by phase image reconstruction. Nineteen venous thrombi locations in 13 patients revealed an MR appearance similar to that of the experimental animal model. Three patients (six thrombi locations) had serial examinations over 4 weeks. No significant change in thrombus signal characteristics was noted with time. It is concluded that MR imaging at 0.35 T cannot be used to predict the age of thrombus (up to 3 weeks) but may be helpful in following its resolution.

Phase-contrast magnetic resonance imaging (PC-MRI) can be used to produce multiframe cine flow images of the coronary arteries. Accurate coronary flow measurement requires the elimination of respiratory motion artifacts using k-space segmentation to acquire the data in a single breath-hold. However, the duration of the breath-hold is proportional to the number of cine frames. In the present study, the number of cine frames was varied and the accuracies of the coronary flow measurements were assessed using perivascular US. For the range of flows studied (2 ml/min to 147 ml/min), the correlation coefficients for PC-MRI and US increased (.70-.98) and the limits of agreement improved (+/-45 ml.min-1 to +/-10 ml.min-1) as the number of cine frames increased from one to six. The results suggest that the accuracy of breath-hold cine PC-MRI measurements of coronary artery flow improves as the number of cine frames increases.

No abstract