[Prospective study to compare high-resolution computed tomography and magnetic resonance imaging in the detection of pancreatic neoplasms: use of intravenous and oral MR contrast media].

PURPOSE

To compare thin-section hydro-CT and MRI in the detection of pancreatic neoplasms. Evaluation of an oral, superparamagnetic contrast agent (OMP) for pancreatic MRI.

MATERIAL AND METHODS

45 patients with suspected pancreatic neoplasms were examined with consecutive thin-section helical CT (Hydro-CT, 3/6/3, 130 ml Ultravist, Schering) and MRI (1.0 T, breath-holding, T1w-GRE, T1w-GRE fat-sat, T2w-TSE). The MRI protocol included precontrast studies, studies after oral administration of OMP (Abdoscan, Nycomed Amersham) and studies after administration of OMP and Gadodiamide i.v. (Omniscan, Nycomed Amersham). All images were assessed by two independent radiologists in a blinded fashion. Radiologic diagnosis was correlated with histology or clinical follow-up (> or = 3 month).

RESULTS

39 patients could be included in analysis. In 13 cases a pancreatic neoplasm was proven by histology. Thin-section hydro-CT and MRI showed no statistically significant differences for the detection of pancreatic neoplasia. The sensitivity of helical hydro-CT was superior compared to MRI (88.5% vs. 65.44/73.1%/76.9%). The specificity of MRI was superior compared to helical hydro-CT (86.5% vs. 94.2%/90.4%/88.5%).

CONCLUSION

Thin-section hydro-CT and MRI show similar results for the detection of pancreatic neoplasms. Compared to thin-section helical CT, MRI still has the drawbacks of much more time consumption and, still, lower spatial resolution. The use of an oral, superparamagnetic contrast agent added no advantage for pancreatic MRI.