Two-dimensional multiplanar and three-dimensional volume-rendered vascularCT in pancreatic carcinoma: Interobserver agreement and comparison with standard helical techniques

OBJECTIVE. The purpose of this study was to compare two-dimensional curved multiplanar and three-dimensional reconstructions, routine axial presentati ons, and combined techniques in the assessment of vascular involvement by p ancreatic malignancy. MATERIALS AND METHODS. For 44 patients with known pancreatic malignancy a t otal of 56 arterial phase helical CT scans were obtained. Targeted pancreat ic imaging was performed, and reformatted images were generated. Axial sour ce images, reformatted images, and the combination of axial and reformatted images were interpreted independently by three observers. The observers gr aded the celiac axis, common and proper hepatic, splenic, gas troduodenal, and superior mesenteric arteries for tumor involvement. Grades of vascular involvement were compared by intra- and interobserver variability analyses. RESULTS. Intraobserver agreement averaged over five vessels was good betwee n the axial and combined techniques for each individual observer (0.64 less than or equal to kappa less than or equal to 0.66), but intraobserver agre e ment was poor between the axial and reformatted (kappa = 0.17 and kappa = 0.31, respectively) and the re formatted and combined techniques (kappa = 0.31 and kappa = 0.38, respectively) for two observers. For grading of vasc ular involvement in each vessel, intraobserver agreement was good to excell ent between the axial and combined techniques (0.48 less than or equal to k appa less than or equal to 0.82). Interobserver agreement averaged over fiv e vessels was poor for imaging techniques except between observer 2 and obs erver 3 on the axial (kappa = 0.47) and combined techniques (kappa = 0.47). For grading of vascular involvement in each vessel, interobserver agreemen t for reformatted technique was poor (0.09 less than or equal to kappa less than or equal to 0.40). CONCLUSION. Multiplanar and volume-rendered techniques showed the highest i ntra- and interobserver variability in grading vascular involvement by panc reatic malignancy. These images should be used in combination with routine axial images to decrease observer variability.