Does Diffusion weighted imaging add diagnostic confidence in characterizing focal lesions if included in contrast-enhanced magnetic resonance imaging (MRI) of the liver?

P urpose: To investigate whether DWI improves the diagnostic confidence in focal lesions characterization when added to contrast-enhanced liver MRI. M aterial and Methods: One experienced abdominal radiologist, blinded to clinical information, reviewed thirty-six focal liver lesions (11 benign, 25 malignant) in 23 patients who underwent MRI on a 1.5T magnet. Pre- and post-i.v. Gd-BOPTA sequences were reviewed without (set 1) and with single-shot echo-planar DW images and apparent diffusion coefficient (ADC) maps (b = 50, 400, 800 sec/mm2) (set 2) in separate sessions. The reader recorded on a 1-5 score: (i) the confidence in lesion diagnosis (malignant/benign nature and type), for both sets; (ii) the impact of ADC vs. signal intensity (DWI images/ADC map) in achieving the diagnosis, for set 2. R esults: No significant difference (p > 0.01; Wilcoxon test) was found between the confidence scores of sets 1 and 2, both for lesions nature (mean:4.25 ± 0.8 vs. 4.11 ± 0.7, respectively) and type (4.11 ± 0.97 vs. 4.02 ± 084). Reader attributed significantly greater impact to signal intensity than ADC in achieving the diagnosis both for lesions nature and type (scores of 4.08 ± 1.15 vs. 3.2 ± 1.42 and 3.9 ± .1.13 vs. 2.9 ± 1.39, respectively) (p < 0.01; Wilcoxon test). Accuracy in characterizing lesions nature was similar without or with DWI (83.3%). Conclusion: Expert reader relied on DWI signal intensity to characterize focal liver lesions. Nonetheless, DWI adds neither diagnostic confidence nor accuracy to contrast-enhanced MRI. Work-in-progress results will show whether DWI is more useful for less-experienced readers.