The usefulness of b value threshold map in the evaluation of rectal adenocarcinoma.

Abstract

To investigate the usefulness of b value threshold (bThreshold) map in the evaluation of rectal adenocarcinoma by comparing it with diffusion-weighted images and ADC maps regarding lesion detection and the prediction of pathological features. Thirty-five patients with rectal tumors were enrolled and underwent axial DWI using a 3-Tesla MRI system. Contrast-to-noise ratio (CNR) between the lesions and normal tissues were assessed on the diffusion-weighted images and bThreshold maps. Reproducibility for ADC and bThreshold values were assessed. Significant differences between different groups for pathological prognostic factors were evaluated. Diagnostic performance of ADC and bThreshold values for those factors were assessed. Reproducibility was excellent for the ADC and bThreshold values (ICC 0.985 and 0.992; CV 3.8% and 4.0%) measurements. The CNR between lesions and normal tissues on bThreshold maps was significantly higher than that on diffusion-weighted images (9.91 ± 5.35 vs. 7.68 ± 3.08, p = 0.012). There were significant differences in the ADC and bThreshold values between different pathologic differentiation degrees and T stages; significant difference was observed in the bThreshold values between the different N stage groups (all p values < 0.050). No significant differences were observed between the ROC curves of ADC and the bThreshold values of rectal lesions for pathologic differentiation and T stage. bThreshold maps showed good diagnostic performance for N stage. Both ADC and bThreshold values can differentiate between degrees of pathologic differentiation and T1-2 versus T3-4. Potential added advantages however of the bThreshold map include a higher CNR compared with DWI images, thereby improving lesion visualization detection, and better diagnostic performance for end staging than ADC. Thus, the bThreshold map may compliment DWI and ADC to evaluate pathologic features of rectal primary tumors and metastatic lymph nodes.