Thomas E. Conturo

Magnetic resonance imaging with contrast enhanced phase angle reconstruction

The magnetic resonance image of a specimen is encoded by the phase angle response of the volume elements in a slice or volume illuminated by a pulsed radio frequency source, instead of encoded by the magnitude response. The specimen can be imaged before and during intrinsic perturbations such as caused by external stimuli or execution of cognitive or motor tasks. Preferably the specimen is perfused with a paramagnetic contrast agent such as Gadolinium or Dysprosium, slowly or by bolus injection, after recording one or more baseline images. The phase angle response of the specimen can then be recorded one or more times as perturbation subsists or as the bolus traverses the area of illumination. Fast Fourier transformation converts gradient spin echo response data to phase angles for a spatial distribution of volume elements in the illuminated slice. The baseline phase angle image is subtracted from the image taken after the bolus injection, providing a high contrast image showing the concentration of the contrast agent. The change Δφ in phase angle between the images is proportional to the concentration of the contrast agent, enabling accurate measurements of localized blood volume and flow rate. The invention is particularly applicable to visualization of localized ischemia caused by cerebral vascular disease, such as stroke.

AUR memorial award--1988. MRI enhancement of perfused tissues using chromium labeled red blood cells as an intravascular contrast agent

It has been demonstrated that chromium (Cr) labeling significantly decreases the relaxation times of packed red blood cells (RBCs). In this study, the spin-lattice relaxation time (T1) of human red cells was shortened from 836 ms to 29 ms and the spin-spin relaxation time (T2) shortened from 134 ms to 18 ms, when the cells were labeled at a Cr incubation concentration of 50 mM. Labeling of canine cells at 50 mM resulted in a T1 of 36 ms and a T2 of 26 ms. A labeling concentration of 10 mM produced similar relaxation enhancement, with uptake of 47% of the available Cr, and was determined to be optimal. The enhancement of longitudinal and transverse relaxation rates (1/T1,-1/T2) per amount of hemoglobin-bound Cr are 6.9 s-1 mM-1 and 9.8 s-1 mM-1 respectively, different from those of a pure Cr+3 solution. Labeling cells at 10 mM decreased the survival half-time in vivo from 16.6 days to 4.7 days in dogs. No difference in red cell survival was found with the use of hetero-transfusion versus auto-transfusion of labeled RBCs. Significant shortening of the T1 (912 ms to 266 ms, P = .03) and T2 (90 ms to 70 ms, P = .006) of spleen and the T1 (764 ms to 282 ms, P = .005) and the T2 (128 ms to 86 ms, P = .005) of liver occurred when 10% of the RBC mass of dogs was exchanged with Cr labeled cells. Liver and spleen spin density changes (P greater than 0.23) and muscle spin density and relaxation changes (P greater than 0.4) were insignificant. The in vivo T1 of a canine spleen which had been infarcted did not change following transfusion with labeled cells, where the T1 of liver did shorten. We believe this preliminary study suggests that Cr labeled red cells may have the potential to become an intravascular magnetic resonance imaging contrast agent.

Enhancement of Red Blood Cell Proton Relaxation with Chromium Labeling

Nuclear medicine has utilized chromium (Cr) for decades to label red blood cells (RBCs). The purpose of this project was to determine whether sufficient paramagnetic Cr could be bound to red cells to influence proton relaxation significantly. We demonstrated that the T1 and T2 of RBCs can be substantially shortened by labeling them with paramagnetic Cr. Proton relaxation enhancement occurs when red cells are incubated with sodium chromate (VI) over a concentration range of 0.10 mM to 31.6 mM. Labeling with Cr at a concentration of 31.6 mM shortened the T1 of packed cells from 714 msec to 33 msec, and the T2 from 117 msec to 24 msec, as compared with nonlabeled red cells. In vitro hemolysis was significantly increased after labeling at 31.6 mM, but not at lower concentrations. Cr-induced proton relaxation enhancement varied with RBCs from different species, temperature, pH, and length of incubation. T1 values of kidneys containing labeled red cells (303 msec), or labeled cells diluted 10-fold with nonlabeled cells (479 msec), were decreased compared with kidneys containing only nonlabeled cells (600 msec). Finally, preliminary data indicate that the signal intensity of perfused renal tissue is significantly influenced in vivo by infusion of Cr-labeled RBCs. This study demonstrated that Cr labeling of RBCs sufficiently enhances red cell proton relaxation to provide excised organs containing red cells, of which 10% have been Cr-labeled, with shorter T1 and T2 values than organs containing nonlabeled cells. In addition, the ability of labeled cells to alter signal intensity in vivo suggests that Cr may have the potential to become an MRI contrast agent.