Effect of clot formation and retraction on spin-echo MR images of blood: an in vitro study

Phantoms were constructed that contained red blood cell (RBC)-free clots in varying stages of clot retraction. MR images of these samples were compared with those of retracted whole venous blood clots and a fresh rat brain standard. Images were obtained at 0.3 T, 0.5 T, 1.0 T, 1.5 T, and 2.4 T with T1-, spin-density- and T2-weighted spin-echo pulse sequences. The presence or absence of venous blood cells in the clot caused only minor differences in T2- and spin-density-weighted images of the clots at or below 1.5 T. On T2-weighted scans, the retraction of the RBC-free clot resulted in a progressive decrease in signal intensity at 2.4 T. Fully retracted RBC-free clots were markedly hypointense relative to serum and ranged from slightly hyperintense to isointense with brain and venous clots at 0.5-1.5 T. There were no striking concomitant signal intensity changes on the spin-density- or T1-weighted scans, which could have caused the changes seen on the T2-weighted images of the clots. Our results indicate that the physical basis of these MR effects in the RBC-free clots is the concentration of plasma protein. The combined concentration of plasma protein and the tightly packed RBC proteins in the venous clots causes the strikingly similar MR appearance of venous and RBC-free clots on clinical images. These results do not demonstrate the presence of the previously postulated selective T2 relaxation of intracellular paramagnetic deoxyhemoglobin in these in vitro venous clots.