High b Value (2,000 s/mm2) Diffusion-Weighted Magnetic Resonance Imaging in Prostate Cancer at 3 Tesla: Comparison with 1,000 s/mm2 for Tumor Conspicuity and Discrimination of Aggressiveness

Abstract

ObjectiveThe objective of our study was to investigate tumor conspicuity and the discrimination potential for tumor aggressiveness on diffusion-weighted magnetic resonance imaging (DW-MRI) with high b value at 3-T.Materials and MethodsThe institutional review board approved this study and waived the requirement for informed consent. A total of 50 patients with prostate cancer (69 cancer foci; 48 in the PZ, 20 in the TZ, and one in whole prostate) who underwent multiparametric prostate MRI including DW-MRI (b values: 0, 1000 s/mm2 and 0, 2000 s/mm2) on a 3-T system were included. Lesion conspicuity score (LCS) using visual assessment (1 = invisible for surrounding normal site; 2 = slightly high intensity; 3 = moderately high; and 4 = very high) and tumor-normal signal intensity ratio (TNR) were assessed, and apparent diffusion coefficient (ADC, ×10−3 mm2/s) of the tumor regions and normal regions were measured.ResultsMean LCS and TNR at 0, 2000 s/mm2 was significantly higher than those at 0, 1000 s/mm2 (p<0.001 for both). In addition, ADC at both 0, 1000 and 0, 2000 s/mm2 was found to distinguish intermediate or high risk cancer with Gleason score ≥7 from low risk cancer with Gleason score ≤6 (p<0.001 for both). Furthermore, ADC of tumor regions correlated with Gleason score at both 0, 1000 s/mm2 (ρ = −0.602; p<0.001) and 0, 2000 s/mm2 (ρ = −0.645; p<0.001).ConclusionsFor tumor conspicuity and characterization of prostate cancer on DW-MRI of 3-T MRI, b = 0, 2000 s/mm2 is more useful than b = 0, 1000 s/mm2.

Efficient experimental designs for isotropic generalized diffusion tensor MRI (IGDTI)

“…The integral of Equation [3] over all orientations on the unit sphere becomes (Eq. [5]): [6] which is related to the generalized mean diffusivity (17) of order n: D ðnÞ ¼ TrD ðnÞ nþ1 . Equation [5] describes the orientation-averaged signal at arbitrary diffusion sensitization obtained from a very large number of DWIs acquired with orientations uniformly sampling the unit sphere.…”

“…[5]) and for the computation of the HOT generalized Traces (Eq. [6]) from a small total number of DWIs.…”

“…[6] combinations that achieve orientation-averaged diffusion weightings in different b-value ranges suggested based on the diffusivities of fixed and live brain tissue.…”

“…To derive orientationally averaged DWIs and measure the mADCs at each b-value in the GDTI data sets, we averaged the log signal attenuations from DWIs acquired with multiple orientations densely and uniformly sampling the unit sphere. In addition, we measured the HOTs by fitting the densely sampled GDTI data sets to Equation [3] using the method described in Liu et al (10) and computed the HOT Traces and generalized mean diffusivities D ðnÞ up to order 6 using Equation [6] and, subsequently, the mean tkurtosis using Equation [9]. From the IGDTI data sets, we measured the orientation-averaged DWIs and mADCs at each diffusion sensitization using the linear combinations of log signal attenuations described in Table 2 for the corresponding b-value ranges and compared the results to those derived from the GDTI data.…”

“…Clinical studies have shown that the mADC measured at higher b-values improves conspicuity and discrimination in prostate tumors (6), breast lesions (7), and cerebral gliomas (8) and provides a better marker for cell viability (9).…”