Variceal bleeding is a major complication of cirrhosis, associated with a hospital mortality rate of 10%-20%. Surviving patients are at high risk for recurrent hemorrhage. For these reasons, management should be directed at its prevention. Endoscopic variceal band ligation (EBL) in combination with non-selective β-blocker (NSBB) therapy is the recommended first line therapy. Transjugular intrahepatic portosystemic stent-shunt (TIPS) is the most effective method to prevent rebleeding, however, it is burdened with increased hepatic encephalopathy and deterioration of liver function in patients with advanced cirrhosis. So TIPS placement forms an alternative if first line therapy fails.

Hepatic venous pressure gradient (HVPG) is currently the best available method to evaluate the presence and severity of portal hypertension. Patients who experience a reduction in HVPG of ≥20% or to <12mmHg in response to drug therapy are defined as 'responders'. The lowest rebleeding rates are observed in patients on secondary prophylaxis who are HVPG responders. A recent meta-analysis has demonstrated that combination therapy is only marginally more effective than drug therapy. This suggests that pharmacological therapy is the cornerstone of combination therapy. Adding EBL may not be the optimal approach to improve the outcome of HVPG nonresponders and HVPG non-responders are a special high-risk population that may benefit from a more aggressive approach, such as an early decision for TIPS. It recently was shown that TIPS placement within 72 hours after acute bleeding not only prevented recurrent bleeding but also improved survival. These raise the question of whether ligation together with NSBB should remain the first choice for elective secondary prophylaxis.

Therefore, the purpose of the study is to compare whether HVPG-guided therapy is superior to standard combination therapy for the prevention of variceal bleeding in patients with decompensated cirrhosis.

Inclusion Criteria:

• Confirmed diagnosis of liver cirrhosis
• History of esophageal variceal bleeding confirmed by endoscopy
• Time interval between index bleeding and randomization > 5 days.
• Child-Pugh score < 12，MELD score<19

Exclusion Criteria:

• Clinical manifestation of active bleeding
• Gastric variceal bleeding: GOV2，IGV1 or IGV2
• Degree of main portal vein thrombosis > 50%
• Refractory ascites
• Contraindications of TIPS
• Contraindications of NSBB
• Budd-Chiari Syndrome
• Malignancy tumor
• Uncontrolled infections
• History of portal-systemic shunt surgery
• HIV
• Pregnancy or breastfeeding woman
• Poor incompliance