Purpose: To evaluate the benefit of adding diffusion-weighted ( DW diffusion weighted ) imaging to conventional magnetic resonance (MR) imaging in the preoperative detection of small pancreatic adenocarcinoma (≤ 3.0 cm).
Materials and methods: The institutional review board approved this retrospective study and waived the requirement for informed consent. This study included 83 patients (56 men, 27 women) with surgically confirmed pancreatic adenocarcinoma of 3.0 cm or smaller and 12 patients (nine men, three women) with histologically proven focal pancreatitis who underwent gadoxetic acid-enhanced MR imaging and DW diffusion weighted imaging at 3.0 T. Two observers reviewed the two image sets, a conventional MR image set and a combined image set including DW diffusion weighted images, to evaluate tumor conspicuity using receiver operating characteristic curve analysis. Contrast ratios between the tumor and proximal or distal pancreas on T1- and T2-weighted images, arterial and portal phase images, and DW diffusion weighted images obtained with a b value of 800 sec/mm(2) were compared by using one-way analysis of variance or Kruskal-Wallis test.
Results: On DW diffusion weighted images, all tumors but two were delineated as completely (32 of 83, 39%) or partially (49 of 83, 59%) hyperintense. Area under the receiver operating characteristic curve values for the two observers were 0.890 and 0.884 for the conventional image set and 0.976 and 0.978 for the combined image set (P = .01 and P = .04), respectively. In regard to tumor detection, the combined image set (98%, 81 of 83 [observer 1]; 96%, 80 of 83 [observer 2]) yielded better sensitivity than the conventional image set (75%, 62 of 83 [observer 1]; 76%, 63 of 83 [observer 2]) (P < .001). Contrast ratios of tumor to proximal or distal pancreas were significantly higher with DW diffusion weighted images than with other images (P < .001).
Conclusion: In preoperative evaluation of small pancreatic adenocarcinoma, the addition of DW diffusion weighted imaging to conventional MR imaging improves the sensitivity of cancer detection.