Hepatic venous pressure gradient is a useful predictor in guiding treatment on prevention of variceal rebleeding in cirrhosis

The best therapy to prevent esophageal variceal (EV) rebleeding in cirrhotic patients who are non-responsive to pharmacological therapy have not been determined. To evaluate efficacy of a strategy to assign different treatments according to hepatic vein pressure gradient (HVPG) values to prevent EV rebleeding in non-responders. This study is a non-randomized controlled prospective study. 109 cirrhotic patients with EV bleeding who were non-responders based on two HVPG measurements were enrolled and divided two groups: 55 patients (EVL+β-blocker group) were treated with endoscopic variceal ligation (EVL) and nonselective β-blocker; 54 patients (HVPG-guided group) were treated with EVL and nonselective β-blocker if HVPG ≤ 16 mmHg (low-HVPG), with percutaneous transhepatic variceal embolization (PTVE) if HVPG > 16 mmHg and ≤ 20 mmHg (medium-HVPG), or with transjugular intrahepatic portosystemic shunt (TIPS) if HVPG > 20 mmHg (high-HVPG). Patients were followed up for rebleeding and mortality. The mean follow-up period was 17.0 months; rebleeding was higher in the EVL+β-blocker group than HVPG-guided group (25.5%, 9.3%, P = 0.026); 3-year probability of rebleeding in the EVL+Beta-blocker group increased with elevated levels of HVPG (12.5% vs 46.4% vs 64.9%, χ(2) = 11.551, P = 0.003), and 3-year probability of survival was no difference (96.6% vs 85.7% vs 90.9%, χ(2) = 2.638, P = 0.267). Rebleeding rate in PTVE group (7.7%) was lower than that in EVL+β-blockergroup with medium-HVPG (35.7%), but there was no difference. Rebleeding rate in TIPS group (7.7%) was lower than that in EVL+β-blockergroup with high-HVPG (45.5%), but there was no difference. HVPG measurement was useful for making decisions to select EVL and Beta-blocker, PTVE or TIPS in secondary prophylaxis. HVPG-guided treatment is feasible and effective in preventing esophageal varices rebleeding.