Multiplanar CT pancreatography and distal cholangiography with minimum intensity projections.

Abstract

To develop a technique of projectional computed tomographic (CT) cholangiopancreatography (CTCP). Sixty-one patients underwent CT for suspected biliary or pancreatic abnormalities. The axial CT and CTCP techniques included spiral scanning during the portal venous phase and thick-slab minimum intensity projections. Visualization of pancreatic and extrahepatic bile ducts (divided in five duct segments per patient) was graded blindly on a scale of 1-5 by a consensus of two radiologists. Two hundred seventy-seven duct segments were used to compare axial CT and CTCP in the depiction of duct segments and dilatation; 109 segments were used to compare CTCP with ERCP. Fifty-six of 277 duct segments were not visualized on axial CT images; 15 segments were not visualized on CTCP images (P < .001). There was no statistically significant difference between the number of segments missed with ERCP and the number missed with CTCP: nine and three of 109 segments, respectively. Duct visualization was equal on axial CT and CTCP images in 35 of 109 duct segments and was superior on CTCP images in all but one of the remaining segments (P < .001). Duct visualization on CTCP images was equal to that on ERCP images in 35 segments, superior in nine, and significantly inferior in 66 (P < .001). CTCP improves CT depiction of pancreatic and bile ducts with a quality that approaches that of ERCP.