Diffusion-Weighted Imaging of the Prostate

OBJECTIVE: To compare image quality and geometric distortion between readout-segmented diffusion-weighted imaging (rs-DWI) and selective-excitation accelerated reduced-field of view (FOV) DWI (sTX-DWI) of the prostate. MATERIALS AND METHODS: Sixty-five patients underwent 3-T MRI of the prostate including rs-DWI and sTX-DWI (b values, 0, 50, and 1000 seconds/mm²; FOV, 150 × 150 mm² and 77 ×150 mm for rs-DWI and sTX-DWI; slice thickness, 3 mm; acquisition time, 8:18 min and 1:37 min for rs-DWI and sTX-DWI). Two readers evaluated aspects of image quality and geometric distortion on a 5-point Likert scale. Quantitative analysis of geometric distortion was assessed by measurements of anteroposterior and left-right diameters and compared to T2-weighted turbo-spin echo sequence using intraclass correlation coefficient (ICC). RESULTS: There was no significant difference in resolution, capsule demarcation, and zonal anatomy (P = 0.111-0.866). Overall image quality was rated "above average" by reader 1 (4.09 ± 0.66 and 4.03 ± 0.79; P = 0.433) and reader 2 (3.86 ± 0.66 and 3.80 ± 0.74; P = 0.465) for rs-DWI and sTX-DWI. Reader 1 (0.74 ± 0.67 and 1.17 ± 0.84; P < 0.001) and reader 2 (0.55 ± 0.64 and 1.09 ± 0.95; P < 0.001) rated the level of geometric distortion significantly lower for rs-DWI than sTX-DWI. Readout-segmented DWI (0.9 ± 2.2 mm) and sTX-DWI (2.1 ± 3.8 mm) overestimated the anteroposterior diameter of the prostate compared to T2-weighted turbo-spin echo sequence (P < 0.001), the difference being more pronounced for sTX-DWI [ICC, 0.89 (95% confidence interval, 0.83-0.93)] compared to rs-DWI [ICC, 0.96 (95% confidence interval, 0.94-0.96)]. CONCLUSION: Selective-excitation accelerated reduced-FOV DW images (sTX-DWI) of the prostate can be acquired more than 5 times faster than rs-DWI with comparable image quality, at the expense of significantly increased geometric distortion.