Return To Abstract Listing

Scientific Session 25 — SS25: Vascular/Interventional - Interventional Oncology and Portal Venous Interventions

Thursday, April 26, 2018

Abstracts 1150-2823



2815. Predicting coil embolization in patients with GI bleeds and Hepatofugal flow on pre-TIPS imaging

Mukherjee S*,  Kohlbrenner R,  Ring E,  Kohi M. UC San Francisco, San Francisco, CA

Address correspondence to S. Mukherjee (sumukher@ucsd.edu)

Objective: The purpose of this study was to investigate whether hepatofugal flow (HF) in the main portal vein (MPV) prior to transjugular intrahepatic portosystemic shunt (TIPS) creation can serve as a predictor for variceal embolization after TIPS creation in patients with gastrointestinal (GI) bleeding.

Materials and Methods: A retrospective review of consecutive patients with GI bleeding who underwent TIPS from 1/2007 to 1/2017 was conducted. Portosystemic pressure gradients before and after TIPS creation and embolization of varices were reviewed. The directionality of the blood flow in the MPV prior to TIPS creation was determined to be HF or hepatopedal (HP) by a fellowship trained interventional radiologist blinded to the purpose of this study. Variceal embolization data was analyzed using the Fisher exact test and mean post-TIPS portosystemic pressure gradients was analyzed using a t test. A p value less than 0.05 was considered statistically significant.

Results: Forty patients were identified. The predominant causes of cirrhosis were hepatitis C and alcohol abuse. Twenty-four patients demonstrated HP flow and 16 patients had HF flow. Of the 24 patients with HP flow, six (25%) required coil embolization after TIPS. This is compared to the 10 out of 16 patients (62.5%) with HF flow who underwent embolization after TIPS (p = 0.0245). Portosystemic pressure gradient after TIPS in the HP group was 6 mm Hg compared to 8 mm Hg (p = 0.058). Portosystemic pressure gradient before TIPS was not significantly different between the two groups.

Conclusion: In the setting of GI bleeding, HF blood flow in the MPV prior to TIPS creation may serve as a predictor for requiring variceal embolization after TIPS. Portograms before TIPS are an essential part of imaging required for measuring the length of the shunt. Our group continues to derive as much information from this imaging study, including consistent markers, that can be used best prepare interventional radiologists prior to TIPS procedures. In this study, we have demonstrated an association between the presence of HF flow on portograms before TIPS and the probable need for coil embolization in TIPS procedure. Of course, larger scale studies are required to validate the data, including prospective randomized trials. In addition, we have shown a possible trend between HF flow demonstrated on imaging before TIPS predicting a pressure gradient of 6 mm Hg or more after TIPS. Further data is required to validate this trend.