Dual-tract Transhepatic U-shaped Hemodialysis Inferior Vena Cava Catheter: A Feasibility Study in a Swine Model Wael E. A. Saad, MBBCh, Lawrence G. Sahler, MD, Karin E. Westesson, MD, Paul F. Della Pietra, MS, Wael M. Darwish, MBBCh, Takashi Kitanosono, MD, and David L. Waldman, MD, PhD PURPOSE: To evaluate the feasibility of establishing a U-shaped inferior vena cava (IVC) catheter entirely from a transhepatic approach and to determine the catheter caliber that would provide adequate flow for hemodialysis. MATERIALS AND METHODS: Three pigs (weight, 45–50 kg) were used. A peripheral right hepatic vein was accessed transhepatically by using a 22-gauge needle, and a 0.018-inch wire was passed into the hepatic veins and IVC. An accessory right hepatic vein was accessed from the IVC. A snare was deployed in the accessory vein and used as a target for a second transhepatic 22-gauge needle pass. A wire was snared through the second transhepatic tract, around into the IVC, and through the first transhepatic tract. The 0.018-inch wire was upsized to a 0.035-inch platform. Measurements where made to tailor a U-shaped catheter from simple 10.2- and 12-F tubes by cutting them longitudinally (single long side hole) along the length of the IVC segment. The U-shaped hemodialysis catheter was placed over the wire and positioned so that the catheter opening lay in the IVC. With use of a dialysis machine, pressures and flow tolerance at set flow rates (100, 200, 300, 350, and 400 mL/min) were tested. RESULTS: All pigs underwent and survived successful catheter placements. All 10.2- and 12-F catheters tolerated flow rates up to 350 and 400 mL/min, respectively. CONCLUSIONS: Establishing a U-shaped hemodialysis catheter with an entirely transhepatic approach is technically feasible. The 10.2-F U-shaped dialysis catheters provided a flow rate (>350 mL/min) that is appropriate for hemodi- alysis in human clinical settings. J Vasc Interv Radiol 2009; 20:1625–1631 Abbreviations: AUD  Amplatz Universal Drain, IVC  inferior vena cava INDWELLING hemodialysis catheters cause venous stenoses and occlusions in the accessed veins in the long-term ( 1– 6). This cumulatively reduces options of sub- sequent venous access over time ( 1–3). Ul- timately, it is not uncommon to en- counter chronic terminal hemodialysis patients with occluded upper and lower extremity veins ( 1–3). This leaves higher-risk venous access op- tions such as translumbar or transhe- patic inferior vena cava (IVC) catheter access (2,3,6,7). Transhepatic hemodi- alysis catheters are plagued with un- stable access (inadvertent partial/ complete transhepatic hemodialysis catheter dislodgment) (6–8). Transhepatic U-shaped catheters are used in the biliary tract for two reasons: ( a) when there is no room to place a pigtail catheter inside the biliary lumen in an oc- cluded system and ( b) when there is a high rate of complete inadvertent discon- tinuation of the biliary catheter ( 9 –12). A U-shaped catheter is considered more sta- ble because it is tethered at both ends and each end (sutures, catheter hub, and con- nections included) anchors against the other to prevent complete, if not major, dislodgment of the catheter. The purpose of this study was to eval- uate the technical feasibility of establish- ing a U-shaped IVC catheter entirely from a transhepatic approach. The secondary purpose was to evaluate the hemody- namic parameters measured during he- modialysis by using the transhepatically placed U-shaped hemodialysis catheter and help establish a catheter caliber size that would provide adequate flow and pressures for hemodialysis. MATERIALS AND METHODS Study Design A swine model (mixed breed do- mestic swine, Sus Scrofa) was chosen From the Department of Imaging Sciences (W.E.A.S., L.G.S., W.M.D., T.K., D.L.W.) and the Department of Medicine, Division of Nephrology (P.F.D.P.), Uni- versity of Rochester Medical Center, 601 Elmwood Ave, Box 648, Rochester, NY 14642; and the Univer- sity of Rochester School of Medicine and Dentistry, Rochester, New York (K.E.W.). Received November 13, 2008; final revision received June 20, 2009; ac- cepted August 26, 2009. Address correspondence to W.E.A.S.; E-mail: wspikes@yahoo.com None of the authors have identified a conflict of interest. © SIR, 2009 DOI: 10.1016/j.jvir.2009.08.022 1625