Gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid (Gd-EOB-DTPA)-enhanced MRI has played a very important role in the diagnosis of liver tumors. Diffusion-weighted imaging (DWI) is also used for detection and characterization of liver tumors. DWI with a low b-value is effective in suppressing vascular structures and provides a higher tumor-to-liver contrast (1). When liver MRI is performed using Gd-EOB-DTPA, DWI is often obtained after Gd-EOB-DTPA injection to shorten examination time in the busy clinical practice. To obtain meaningful DWI, robust fat suppression is essential because chemical shift artifacts associated with single-shot echo-planar imaging may interfere with the detection of pathology. Two types of pulse fat-suppression techniques have been widely used for DWI: chemical shift selective (CHESS) and short inversion time inversion recovery (STIR) sequences. However, there are no reports regarding the comparison of between DWI with STIR (STIR low b DWI) and DWI with CHESS (CHESS low b DWI) after Gd-EOB-DTPA injection for visualizing malignant liver tumors. Therefore, the purpose of this study was to compare the conspicuity of malignant liver tumors between STIR low b DWI and CHESS low b DWI after administration of Gd-EOB-DTPA.Forty-seven patients with histologically confirmed 24 hepatocellular carcinomas (HCCs), 3 cholangiocarcinomas, and 20 metastases underwent STIR and CHESS low b DWIs at 20 min after Gd-EOB-DTPA injection. Both low b DWIs were acquired by breath-hold single-shot echo-planar imaging using a b-value of 50 s/mm2. A 5-mm section thickness and an intersection gap of 0.5 mm were used to cover the liver. Inversion time for STIR low b DWI was 180 ms, which was selected to null fat signals. The pulse sequence parameters of both low b DWIs were as follows: repetition time,3500 ms; echo time, 40 ms; flip angle, 90°; field of view, 350 mm; matrix, 168x256; number of excitations, 1; sensitivity encoding acceleration factor, 2; and number of slices, 30. Look-Locker sequences (single slice multiphase imaging using gradient-echo sequence with inversion recovery pulse) were also obtained to calculate T1 relaxation times of the liver and subcutaneous fat at 20 min after Gd-EOB-DTPA injection. Contrast-to-noise ratios (CNRs = [signal intensity of tumor – signal intensity of liver] / standard deviation of liver) were compared between for tumors vs. the liver on STIR and CHESS low b DWIs by using a paired Student’s t test. We also assessed the relationship between CNR and T1 relaxation time of the liver.CNR was significantly higher on STIR low b DWI (25.1±15.8) than on CHESS low b DWI (11.9±7.6) (P<0.001), regardless of the presence of chronic liver disease. Forty of 47 patients (85.1%) showed a higher CNR on STIR low b DWI, compared with CHESS low b DWI. There was no significant difference in T1 relaxation time of the liver at 20 min after administration of Gd-EOB-DTPA between patients with higher CNRs on STIR low b DWIs and those on CHESS low b DWIs. However, the differences in T1 relaxation times between the liver and fat were significantly shorter in patients with higher CNRs on STIR low b DWIs (81±48.2 ms) (P=0.036) than those on CHESS low b DWIs (123±36.9 ms).Malignant liver tumors were more conspicuous on STIR low b DWI than on CHESS low b DWI. It is reasonable to speculate that liver parenchyma with a shot T1 relaxation time from Gd-EOB-DTPA shows lower signal on STIR low b DWI in contrast to high signal malignant liver tumors that do not take up the contrast (2,3).Low b DWI with STIR is useful for visualizing malignant liver tumors after Gd-EOB-DTPA injection.