Underdilated VCX-TIPS Versus EVL Plus NSBB for Secondary Prophylaxis of Variceal Bleeding in Cirrhosis

This multicenter randomized controlled trial is designed to compare an underdilated VCX-TIPS strategy with standard secondary prophylaxis using endoscopic variceal ligation plus nonselective beta-blockers in patients with cirrhosis who have experienced esophageal variceal bleeding or type 1 gastroesophageal variceal bleeding.

Patients with cirrhosis aged 18 to 75 years who have achieved successful control of acute variceal bleeding and are clinically stable will be screened. Eligible participants will be randomized after written informed consent has been obtained, generally between 5 and 21 days after successful hemostasis. Patients with ongoing bleeding, hemodynamic instability, a need for immediate rescue TIPS, or a strong indication for early or pre-emptive TIPS that makes randomization to EVL plus NSBB inappropriate will be excluded.

Participants assigned to the experimental group will undergo TIPS using a commercially available 8-10 mm VIATORR Controlled Expansion stent. After stent deployment, the stent will be initially dilated only with a 6-mm balloon to achieve an initial effective shunt diameter of approximately 6 mm. Further dilation to 8 mm will not be performed solely because the portal pressure gradient remains above 12 mmHg. Additional dilation will be allowed only under predefined rescue conditions, such as persistent active bleeding, inadequate reduction of portal pressure gradient with persistent high-risk collateral perfusion despite embolization, or clinically significant rebleeding during follow-up. Selective embolization of responsible variceal inflow vessels may be performed according to standardized local procedures.

Participants assigned to the control group will receive standard secondary prophylaxis with serial endoscopic variceal ligation and nonselective beta-blockers. Endoscopic variceal ligation will be repeated approximately every 4 weeks until variceal eradication or conversion to a low-risk status. Carvedilol will be the preferred nonselective beta-blocker, starting at 6.25 mg once daily and increasing to 12.5 mg once daily after 1 week if tolerated. Propranolol may be used if carvedilol is not tolerated or is clinically inappropriate.

The primary endpoint is the time from randomization to the first occurrence of all-cause death or clinically significant upper gastrointestinal rebleeding within 365 days. Key secondary outcomes include transplant-free survival, overall survival, liver-related mortality, clinically significant rebleeding, variceal-related rebleeding, overt hepatic encephalopathy, severe overt hepatic encephalopathy, acute-on-chronic liver failure, liver function deterioration, TIPS dysfunction, actual stent diameter at 3 months, unplanned rehospitalization, length of hospital stay, transfusion requirement, direct medical costs, and adverse events. Participants will be followed for 12 months for the primary analysis and up to 24 months for extended secondary outcome analyses.

Inclusion criteria

• Age 18 to 75 years, regardless of sex.
• Diagnosis of liver cirrhosis based on previous histology, imaging, laboratory tests, and/or clinical manifestations.
• Hospitalization for acute upper gastrointestinal bleeding, with endoscopic confirmation that the bleeding is related to esophageal varices or type 1 gastroesophageal varices.
• Successful control of acute bleeding after standard acute-phase treatment, with stable vital signs and entry into the secondary prophylaxis phase.
• Child-Pugh score of 5 to 13.
• The investigator judges that both underdilated VCX-TIPS and EVL plus NSBB are clinically feasible and that randomization is appropriate.
• The participant or legally authorized representative understands the study procedures and voluntarily signs written informed consent.

Exclusion criteria

• Hemodynamic instability, persistent active bleeding, or need for immediate rescue TIPS, surgical hemostasis, or interventional radiologic hemostasis.
• A strong current indication for early or pre-emptive TIPS that makes randomization to EVL plus NSBB clinically inappropriate.
• Child-Pugh score greater than 13.
• Previous TIPS, surgical portosystemic shunt, BRTO, CARTO, PARTO, or other procedures that substantially alter portosystemic blood flow.
• Isolated gastric varices, type 2 gastroesophageal varices, ectopic varices, or a bleeding source other than esophageal varices or type 1 gastroesophageal varices.
• Non-cirrhotic portal hypertension.
• Cavernous transformation of the portal vein or portal venous thrombosis that makes standardized TIPS technically infeasible.
• Previous recurrent or refractory overt hepatic encephalopathy, or a history of West Haven grade II to IV overt hepatic encephalopathy unrelated to gastrointestinal bleeding.
• Severe heart failure, severe pulmonary hypertension, severe tricuspid regurgitation, right heart failure, or other contraindications to TIPS.
• Uncontrolled severe infection, sepsis, or multiple organ failure.
• Hepatocellular carcinoma beyond the Milan criteria or other advanced malignancy.
• Severe renal insufficiency requiring long-term renal replacement therapy.
• Pregnancy or breastfeeding.
• Absolute contraindication to EVL, NSBB, or TIPS.
• Any condition that, in the investigator's judgment, makes the participant unsuitable for the study or unable to complete follow-up.