A Randomized, Multi-Center, Open-Label Study to Evaluate the Efficacy of Carvedilol vs. Propranolol to Reduce Portal Pressure in Patients With Liver Cirrhosis.

OBJECTIVES: Propranolol has been used as prophylaxis for variceal bleeding in patients with cirrhosis. More recent data suggest that carvedilol may be more effective for reducing the hepatic venous pressure gradient (HVPG) than propranolol. The primary aim of this study was to evaluate the hemodynamic response to carvedilol compared with propranolol.

METHODS: A total of 110 patients with a baseline HVPG value >12 mm Hg were allocated randomly to receive either carvedilol or propranolol. The HVPG measurement was repeated after 6 weeks of daily medication. The primary end point was a ≥20% fall in HVPG compared with baseline or <12 mm Hg.

RESULTS: The difference in the proportion of responders in the carvedilol (49.1%) vs. propranolol (30.9%) groups did not reach statistical significance in the intention-to-treat analysis (P=0.08). However, among patients with a model for end-stage liver disease (MELD) score ≥15, carvedilol resulted in a significantly greater response than that of propranolol (7/12, 58.3% vs. 0/10, 0%; P=0.005). Similarly, carvedilol was superior to propranolol in patients with Child-Pugh score ≥9 (46.2 vs. 0%; P=0.046). The presence of ascites also had a significant influence on the response rate (51.5 vs. 24.2%; P=0.042). A MELD score ≥15 was the only significant predictor of response among these post hoc groups after adjusting for multiple comparisons (P=0.005). Severe adverse events were higher in the carvedilol group although drug-associated adverse events were not different.

CONCLUSIONS: Overall, carvedilol offered no clear advantage over propranolol but it may be more effective in advanced cirrhotic patients with a MELD score≥15 in reducing the portal pressure gradient. However, this potential benefit may come with a cost of increased risk of side-effects and outcome data over a longer term is needed to understand the relative risk benefit.