Value of diffusion-weighted images in diagnosis of locoregionally recurrent rectal tumors

Abstract

Introduction: To assess the value of referring to Diffusion-weighted images in evaluation of T2-weighted images of patients clinically suspicious of locoregional rectal cancer recurrence. Methods: After ethics committee approval and informed consent were obtained, 37 consecutive patients (male/female of 22/15; mean age 56 ± 13.5 SD) clinically suspicious of recurrent rectal tumor were prospectively included in the study over a two-year period. T2-weighted images of the patients were reviewed and the results were recorded. Right after that, the corresponding DWI images were provided for the radiologist and new ratings were given to the patients after taking into account the DWI findings. Finally, the patients underwent tissue biopsy. Receiver Operating Characteristic (ROC) analysis was performed, and Area Under the Curve (AUC) of the “T2-weighted alone” and “T2-weighted þ DWI” methods were calculated and compared. Results: “T2-weighted alone” and “T2-weighted þ DWI” methods had an AUC of 0.64 (95% CI 0.47 to 0.79) and 0.75 (95% CI 0.58 to 0.88), respectively. The Difference between the two AUCs was 0.11 (P ¼ 0.16). In the subgroup of patients having equivocal ratings in T2-weighted images, DWI images correctly identified 81% (13/16) of patients with true tumor recurrence and 66% (8/12) of patients without recurrence.