Journals / Turkish Journal of Medical Sciences / 2017 / Cilt: 47 - Sayı: 5
Mean ADC values discriminate rectal mucinous carcinomafrom rectal nonmucinous adenocarcinoma
- Pages
- 1520–1525
- DOI
- —
Abstract
Background/aim: This study aimed to differentiate rectal mucinous carcinoma (MC) from nonmucinous rectal adenocarcinoma (AC) using mean apparent diffusion coefficient (mADC) values obtained with diffusion-weighted imaging. Materials and methods: Sixty-two pathologically confirmed rectal AC (n = 44) and MC (n = 18) patients were included in this study. The two groups underwent pelvic MRI to determine the local staging baseline for rectal tumors. Once the region of interest (ROI) was determined, a border was drawn around each hyperintense tumor (b = 1000 s/mm2images). Following a repeat of this procedure for each patient, the ROIs were recorded to apparent diffusion coefficient (ADC) maps, and mADC values were measured. The mADC was determined per slice, followed by a calculation of whole tumor volume ADC mean using the individual mADC values. The MannWhitney test was performed to compare mADCs for the two groups. A receiver operating characteristic (ROC) curve was generated to determine the differentiating capacity of ADCs from MC to AC. Results: The mADC was higher in MC (1.631 ± 0.375 × 10 3 mm2 /s) (range: 0.95 2.36 × 10 3 mm2 /s) than in AC (0.921 ± 0.157 × 10 3 mm2 /s) (range: 0.6 1.48 × 10 3 mm2 /s) (P < 0.001). mADCs were effective in distinguishing MC from AC (area under the ROC curve, 0.972 (95% CI : 0.928 1.00)). A threshold of 1.27 × 10 3 mm2 /s was set that corresponded with high sensitivity (94.4%) and specificity (97.7%). Twelve MCs (67%) were predominantly hypointense, and 6 MCs (33%) were seen as mixed signal intensity lesions. Forty ACs (91%) were observed as hyperintense lesions, and 4 ACs (9%) had mixed signal intensity. There was a significant difference in the signal intensities between MC and AC (c2= 54.7, P < 0.001). Conclusion: MCs and ACs show different diffusion characteristics, which can be distinguished with high sensitivity and specificity and can help to improve prognostic treatment options.