Experimental Hepatic Tumor Necrosis Comparison of Spin-Echo and Pulsed Magnetization Transfer Contrast Magnetic Resonance Imaging

Abstract

We compared the effectiveness of pulsed magnetization transfer contrast (MTC) magnetic resonance imaging (MRI) and spin-echo MRI in detecting tumor necrosis. Adenocarcinoma cells were transplanted in the livers of 12 syngenic BDIX rats. To induce various degrees of tumor necrosis, the rats were randomly assigned to the following groups: 1) control; 2) localized hyperthermia; 3) intralesional cisplatin; and 4) hyperthermia plus intralesional cisplatin. At day 7 after treatment, the rats were imaged using a 1.5-T imager with 1) multiplanar gradient-recalled echo sequence (MPGR) 500/8/20 degrees with and without magnetization transfer contrast (MTC); 2) spin-echo 2500/20,80, and 3) spin-echo 300/20 pulse sequences. The rats were then sacrificed and pathologic specimens were prepared using MR images as guidance. T2 and ratios of signal intensity after saturation to signal intensity before saturation (Ms/Mo ratios) of the necrotic and granulation tissues and viable tumors were determined in 10 rats. Compared with standard MPGR images, MPGR images with MTC provided better contrast between the pathologic tissues and normal liver. However, T2 values were more useful than Ms/Mo ratios in distinguishing necrotic areas from viable tumor. The T2 values of coagulative necrosis and granulation tissue were significantly different from that of viable tumor. No significant difference between the Ms/Mo ratios of the different pathologic tissues and normal liver was found. Pulsed magnetization transfer contrast MRI was inferior to spin-echo MRI in distinguishing necrotic from viable tumors in rat livers using the pulse sequences described, and none of the sequences studied was thought to be reliable enough for this purpose.