Purpose: To assess the diagnostic performance of diffusion-weighted magnetic resonance (MR) imaging (DWI) for prostate cancer detection, using different b-values.
Methods: A total of 201 patients who underwent MR imaging before total prostatectomy were evaluated. MR images were independently assessed by three radiologists. Three combinations of sequences were separately evaluated, as follows: group 1 [T2-weighted images (T2WI) alone], group 2 (T2WI and DWI with a b-value of 1,000 s/mm2), group 3 (T2WI and DWI with a b-value of 2,000 s/mm2). Whole-mount-section histopathological examination was the reference standard. Areas under the receiver operating characteristic curve (AUCs) and diagnostic performance parameters were determined.
Results: The sensitivity, specificity, and AUC for the detection of prostate cancer were as follows: 52.2%, 80.7%, and 0.694 in group 1; 61.2%, 82.6%, and 0.755 in group 2; 73.2%, 89.7%, and 0.842 in group 3. Group 3 achieved the highest diagnostic performance, followed by group 2 (P<0.05). In the transition zone, the specificity was lower (P<0.001) for group 2 (82.2%) than for group 1 (86.2%).
Conclusion: The addition of diffusion-weighted images with a b-value of 2,000 s/mm2 to T2WI can improve the diagnostic performance of MR imaging in prostate cancer detection.