The transjugular intrahepatic portosystemic shunt (TIPS) is a well-established procedure for the treatment of portal hypertensive bleeding, refractory ascites and vascular diseases of the liver. The major drawbacks of this procedure are shunt dysfunction and portosystemic encephalopathy (PSE). The availability of self-expandable polytetrafluoroethylene-covered stentgrafts (PTFE-SGs) has dramatically improved the long-term patency of TIPS. However, the incidence of PSE remains a threatening complication in about 50% of patients.

The Investigators hypothesized that under-dilated PTFE-SGs would not self-expand to nominal diameter and their under-dilation would be safe and could reduce the rate of post-TIPS encephalopathy, while maintaining clinical efficacy.

Aim of this proof-of-concept exploratory study is to determine whether "under-dilated TIPS" is a feasible procedure that reduces the incidence of PSE while maintaining clinical efficacy.

The transjugular intrahepatic portosystemic shunt (TIPS) is a well-established procedure for the treatment of portal hypertensive bleeding, refractory ascites and vascular diseases of the liver. TIPS is no longer viewed solely as a salvage therapy or a bridge to liver transplantation and is currently indicated for a number of conditions related to portal hypertension with positive results regarding survival, particularly when used to treat high-risk patients with acute variceal hemorrhage. The major drawbacks of this procedure are shunt dysfunction and portosystemic encephalopathy (PSE), reported in 30-70% and 23-55% of patients with cirrhosis within the first year, respectively. The availability of self-expandable polytetrafluoroethylene-covered stent grafts (PTFE-SGs) has dramatically improved the long-term patency of TIPS. However, the incidence of PSE, that has apparently decreased with a more careful selection of patients, remains a threatening complication in about 50% of patients.

Current guidelines for TIPS placement recommend that the post-TIPS porto-caval pressure gradient (PSPG) should be reduced to less than 12 mmHg, particularly in patients with variceal hemorrhage as an indication. However, there is not enough evidence to support the use of available 10 mm rather than 8 mm nominal diameter PTFE-SG aiming to achieve this hemodynamic goal and the best control of variceal rebleeding and/or ascites. A step-wise procedure based on the progressive dilation of 10 mm diameter PTFE-SG by using balloon catheters of increasing diameter (i.e., from 8 to 10 mm) at the time of TIPS positioning has been proposed. However, if a larger diameter, leads to a higher decrease in PSPG, the higher amount of portal blood diverted to the systemic circulation and the lower residual portal perfusion of the parenchyma markedly increase the probability of post-TIPS encephalopathy.

It is conceivable that balloon dilation of TIPS to diameters smaller than those currently indicated (to 7 mm or lower) would allow for a sufficient PCG decrease in patients without a high post-TIPS portal inflow (i.e., relatively small spleen, lack of extensive portal collateralization, relatively hypodynamic circulation). However, currently no tools allow a pre-procedural definition of the ideal stentgraft diameter in individual patients. Moreover, PTFE-SGs that are specifically designed for TIPS are presently available as 8 or 10 mm and are not considered permanently under sizeable/adjustable for their intrinsic tendency to expand to nominal diameter. On the other hand, dilating PTFE-SGs far below the recommended diameter may increase the risk of TIPS thrombosis, a complication that would require prompt re-intervention.

The Investigators hypothesized that, within the cirrhotic parenchyma, under-dilated PTFE-SGs would not self-expand to nominal diameter and their under-dilation would be safe and could reduce the rate of post-TIPS encephalopathy, while maintaining clinical efficacy.

Aim Aim of this study is a) to determine whether "under-dilated TIPS" is a feasible procedure and b) to verify if this strategy reduces the incidence of PSE and other complications while maintaining clinical efficacy.

Study design This study is an exploratory proof-of-concept study analyzing feasibility and clinical outcomes of under-dilated TIPS in unselected consecutive cirrhotic patients in whom TIPS is indicated clinically and who agree to participate in the study. Initially, our strategy will be to under-dilate TIPS to 7 mm in 15 patients and, if the procedure won't lead to shunt occlusion, the rest of the patients will receive a TIPS under-dilated to 6 mm or less.

The group with under-dilated TIPS to less than 7 mm will be compared to a historical control group composed of patients who had standard TIPS placed prior to initiation of the study.

TIPS placement and hemodynamic evaluation PTFE-SGs (Viatorr® and Viatorr CX®, W.L. Gore & Associates Inc., Flagstaff, AZ, USA) will be placed as previously described. Semi-compliant balloon catheters will be used both to pre-dilate the intra-parenchymal tract and to dilate PTFE-SGs after deployment. The intra-parenchymal tract will be pre-dilated to 6 mm or less in all patients. After its deployment, the PTFE-SG will be dilated to 7 (first 15 subjects) or 6 mm (following group) in patients in whom both the procedure and the final TIPS shape will be straightforward. During dilation of the intra-parenchymal tract, balloon pressure will be maintained at the nominal value for no more than 15-30 seconds, even in the lack of a complete flattening of notches at the level of vascular walls. The intra-parenchymal tract and the PTFE-SG will be dilated to 8 mm in patients with a challenging procedure and/or in whom the final TIPS shape will be curved.

For all groups, immediately after TIPS placement, pressures in the portal vein, along the intra-parenchymal tract of TIPS, and in the inferior vena cava will be recorded until a stable tracing will be obtained in each position (45-60 seconds). Permanent tracings will be read with PowerLab computer software (ADInstruments, Inc.). Post-TIPS porto-systemic pressure gradient (PSPG) will calculated by subtracting the inferior vena cava pressure from the portal vein pressure. All procedures will be performed under monitored anesthesia care without intubation and using midazolam and fentanyl as sedative and analgesic, respectively.

Inclusion Criteria:

• diagnosis of cirrhosis determined on the basis of clinical history, histological examination, morphological characteristics of the liver at ultrasound, computed tomography or magnetic resonance imaging;
• TIPS placed to prevent recurrent variceal bleeding or to control refractory ascites, according to current guidelines.

Exclusion Criteria:

• placement of two or more coaxial stent grafts;
• refusal to consent to have TIPS dilated to a small diameter and/or to attend follow-up visits;
• TIPS placed in the setting of acute variceal hemorrhage either as "early" TIPS or as salvage TIPS for continued bleeding or early rebleeding.