Portacaval space anatomy: Potential implications for percutaneous portacaval shunts

PURPOSE: To assess the anatomic feasibility of percutaneous portacaval shunt formation. MATERIALS AND METHODS: One hundred contrast material-enhanced computed tomographic (CT) scans were retrospectively reviewed. The distance and direction from the inferior vena cava (IVC) to the main portal vein (MPV) were measured. Anatomic structures intervening between the IVC and the MPV, and related to the MPV were also noted. RESULTS: The distance between the IVC and the MPV was less than 1 cm in most cases. The direction from the IVC to the upper, middle, and lower thirds of the MPV was predictable and lay within a relatively narrow range. At the middle MPV, there were no intervening structures between the IVC and the MPV in most cases. Intervening structures included the liver at the upper MPV and nodes at the lower MPV. The most common structures related to the MPV included the liver, hepatic artery, and gallbladder (upper MPV); liver, stomach, and pancreas (middle MPV); and pancreas, duodenum, and celiac axis (lower MPV). CONCLUSIONS: These results indicate a short distance and predictable direction between the IVC and MPV. If aberrant vessels and a prominent caudate lobe are excluded, there appears to be a relatively safe cavo-portal route (with respect to neighboring and intervening structures) at the level of the middle MPV. These findings indicate that percutaneous portacaval shunting may be worthy of further experimental study.