Carvedilol as the new Non-Selective Beta-Blocker of Choice in Patients With Cirrhosis and Portal Hypertension

Laura Turco; Thomas Reiberger; Giovanni Vitale; Vincenzo La Mura

Liver International. 2023;43(6):1183-1194.

Abstract and Introduction

Abstract

Portal hypertension (PH) is the most common complication ofcirrhosis and represents the main driver of hepatic decompensation. The overarching goal of PH treatments in patients with compensated cirrhosis is to reduce the risk of hepatic decompensation (i.e development of ascites, variceal bleeding and/or hepatic encephalopathy). In decompensated patients, PH-directed therapies aim at avoiding further decompensation (i.e. recurrent/refractory ascites, variceal rebleeding, recurrent encephalopathy, spontaneous bacterial peritonitis or hepatorenal syndrome) and at improving survival. Carvedilol is a non-selective beta-blocker (NSBB) acting on hyperdynamic circulation/splanchnic vasodilation and on intrahepatic resistance. It has shown superior efficacy than traditional NSBBs in lowering PH in patients with cirrhosis and may be, therefore, the NSBB of choice for the treatment of clinically significant portal hypertension. In primary prophylaxis of variceal bleeding, carvedilol has been demonstrated to be more effective than endoscopic variceal ligation (EVL). In patients with compensated cirrhosis carvedilol achieves higher rate of hemodynamic response than propranolol, resulting in a decreased risk of hepatic decompensation. In secondary prophylaxis, the combination of EVL with carvedilol may prevent rebleeding and non-bleeding further decompensation better than that with propranolol. In patients with ascites and gastroesophageal varices, carvedilol is safe and may improve survival, as long as no impairment of the systemic hemodynamic or renal dysfunction occurs, with maintained arterial blood pressure as suitable safety surrogate. The target dose of carvedilol to treat PH should be 12.5 mg/day. This review summarizes the evidence behind Baveno-VII recommendations on the use of carvedilol in patients with cirrhosis.

Introduction

Portal hypertension (PH) is the most common complication of cirrhosis and represents the main driver of hepatic decompensation, that is the transition from an asymptomatic compensated to a symptomatic decompensated disease.^[1,2] Decompensation is defined by the development of ascites, variceal bleeding and/or hepatic encephalopathy and constitutes a more advanced stage of cirrhosis characterized by distinct disease-driving pathophysiological mechanisms associated with a high risk for liver-related death.^[3]

Cirrhosis has traditionally been used as a pathological term to describe advanced fibrosis of the liver, which was mostly diagnosed by liver biopsy. Nowadays, the broad availability of hepatic elastography for non-invasive assessment of liver fibrosis has led to the concept of using liver stiffness measurement (LSM) for risk stratification.^[2] Subsequently, the term advanced chronic liver disease (ACLD) based on LSM is used to define the risk/target population at risk for liver-related events.^[4]

In absence of decompensating events, patients with ACLD are compensated (cACLD) but may suffer from mild portal hypertension (i.e. a portal pressure gradient of 6–9 mmHg), or even clinically significant portal hypertension (CSPH, i.e. portal pressure gradient ≥10 mmHg). The presence of CSPH represents an important step in the progression of cACLD as these patients are at risk of developing clinical decompensation.^[5] The diagnosis (and concept) of CSPH has been historically based on the measurement of hepatic venous pressure gradient (HVPG),^[6] an indirect measure of portal pressure, best validated in patients with alcohol-related and viral cirrhosis^[7] and to a lesser extent in patients with metabolic liver disease,^[8,9] or on the presence of gastroesophageal varices (GEV) at upper endoscopy or the presence of collaterals at cross-section images (patients with GEV and/or collaterals have CSPH by definition).

Non-selective beta-blockers (NSBBs) have been used as the cornerstone of treatment of portal hypertension. Their use was, however, limited to the prophylaxis of variceal bleeding (both of first and recurrent bleeding, that is primary or secondary prophylaxis).^[2,10] In earlier stages of cirrhosis, that is in patients with no or only small varices, NSBBs have not found to be universally effective in the past.^[11] Nowadays, the advancement of knowledge allowed to expand the target population of patients that can benefit from NSBBs treatment to all those with CSPH. Importantly, the hyperdynamic portal hypertensive circulation, that is the main target of hemodynamic action of NSBBs, occurs only with CSPH^[12] and patients can benefit from NSBBs treatment not only in terms of lowering the risk of variceal bleeding but more broadly by a considerable reduction in the risk of developing any decompensating event.^[13]

Carvedilol, an NSBB with both β- and α1-receptors blocking activity, seems to be more effective than traditional NSBBs in reducing PH in cirrhosis. In recent years, new evidence showed its beneficial effect in patients with cirrhosis and CSPH.^[12,13] The last Baveno VII Consensus Conference^[2] took into great account those evidence, recommending carvedilol as the potential NSBB of choice to treat PH in cirrhosis.

In this review, we summarized the evidence behind the Baveno-VII recommendations on the use of carvedilol in patients with cirrhosis and CSPH including a critical analysis of them and an update of the recent literature.

1 2 3 4 5 6 7 8

Next Section

Abstract and Introduction
Pathophysiology of Portal Hypertension
The Paradigm Shift in the Treatment of Portal Hypertension
Carvedilol
Compensated Cirrhosis (cACLD)
Decompensated Cirrhosis
Safety Issues
Conclusions
References
Sidebar

Table 1. Characteristics of the studies showing the efficacy of carvedilol in patients with cirrhosis.
Author, year, journal	Key finding related to carvedilol	Study setting	Number of patients	Cohort characteristics	Comments
Studies on carvedilol with a focus on HVPG
Banares R, 1999, Hepatology (1)	Acute (1 h) HVPG decrease is greater after carvedilol (HVPG−20%) than with propranolol (HVPG−13%). Higher rate of acute HVPG response (≥20% decrease/HVPG≤12 mmHg) with carvedilol (64%) than with propranolol (14%)	RCT	14 carvedilol 14 propranolol 7 placebo	Cirrhosis and GEV, with and without previous bleeding	Some patients were decompensated (ascites 50% in carvedilol and 50% in propranolol). Carvedilol had stronger effect on arterial pressure (−23%) than propranolol (−11%)
Banares R, 2002, Hepatology (2)	Carvedilol (HVPG −19%) has greater portal hypertensive effects than propranolol (HVPG −12%). Higher rate of HVPG-response (≥20%/≥12 mmHg) with carvedilol (54%) than with propranolol (23%)	Prospective, RCT	26 carvedilol 25 propranolol	Cirrhosis and GEV, primary prophylaxis	Some patients were decompensated (ascites in carvedilol: 39% and propranolol 24%). Hypotension and water retention may occur with the use of carvedilol (in pts with ascites)
Tripathi D, 2002, AP&T (3)	12.5 mg of carvedilol decreased HVPG by −23% already after 1 h	Prospective, non-randomized	10 carvedilol	Cirrhosis and GEV, primary prophylaxis	A further decrease of HVPG was found after 4 weeks of chronic intake of 12.5 mg carvedilol
Reiberger T, 2013, Gut (4)	Carvedilol decreases HVPG stronger than propranolol. Carvedilol can rescue hemodynamic response in propranolol non-responders. Carvedilol-induced haemodynamic response prevents variceal bleeding, decompensation and death.	Prospective, non-randomized	38 carvedilol 37 propranolol 29 EVL	Cirrhosis and GEV, primary prophylaxis	Only patients with propranolol non-response or intolerance were treated with carvedilol. Only carvedilol non-responders were treated with EVL
Gupta V, 2017, Hepatol Int (5)	HVPG response (decrease ≥20% or to <12 mmHg) 1 months after variceal bleeding was more common with carvedilol (75%) than with propranolol (50%)	Prospective, RCT	30 carvedilol 29 propranolol	Cirrhosis, secondary prophylaxis	The relative increase in MAP was more pronounced with carvedilol (−11%) than with propranolol (−8%)
Kirnake V, 2016, J Clin Exp Hep (6)	HVPG response ≥10% to carvedilol is maintained over a long period and is associated with less variceal bleeding, better ascites control, and improved MELD and Child scores.	Prospective, non-randomized	69 patients, carvedilol responders (67%) received 6 months carvedilol	Cirrhosis, primary (36%) and secondary (64%) prophylaxis	Response to a single oral 25 mg carvedilol test dose may predict chronic HVPG response. Similar survival in carvedilol-treated responders and in non-carvedilol-treated non-responders
Schwarzer R, 2018, APT (7)	Carvedilol 12.5 mg/day is more effective than 6.25 mg/day in decreasing HVPG in primary prophylaxis. 76% achieve ≥10% HVPG decrease with 12.5 mg/day	Retrospective, dose-escalation design	72 carvedilol	Cirrhosis, small or large varices, primary prophylaxis	Arterial hypotension may occur, especially in patients with ascites
Zacharias, 2018, Cochrane DB Syst Rev (8)	Carvedilol is more effective at reducing HVPG than other beta-blockers	Systematic review of RCT	Overall 810 patients considered	Cirrhosis and GEV	No beneficial or harmful effects of carvedilol vs. traditional NSBBs on mortality, variceal bleeding, or adverse events
Jindal A, 2020, AJG (9)	HVPG response (≥20% decrease or HVPG <12 mmHg) is achieved in 45% of pts with HVPG 12–19 mmHg and in 22.2% of pts with HVPG ≥20 mmHg.	Retrospective, non-randomized	105 measured on carvedilol after a mean of 9.3 months	Cirrhosis and GEV, includes pts with previous bleeding and with ascites	Risk of (further) decompensation is higher with HVPG ≥20 mmHg than with HVPG 12–19 mmHg. The effect size of carvedilol-HVPG-response was not assessed in either HVPG strata
Jachs M, 2022, CGH (10)	Carvedilol results in higher rates of HVPG response than propranolol in secondary prophylaxis combined with EVL. The combination of EVL + carvedilol (vs. EVL + propranolol) seems to be also superior in reducing the risk of non-bleeding further decompensation and mortality.	Retrospective, non-randomized	54 carvedilol+EVL, 42 propranolol+EVL	Cirrhosis, secondary prophylaxis	Median HVPG decrease: carvedilol −20% vs. propranolol −11%; 20% HVPG response rates: carvedilol: 53% vs. 29%; Lower rate of rebleeding, non-bleeding further decompensation and liver-related death with carvedilol
Clinical studies on carvedilol (not focusing on an HVPG endpoint)
Tripathi D, 2009, Hepatology (11)	Carvedilol was superior to EVL to prevent the first variceal bleeding episode. Mortality was similar in patients receiving carvedilol or EVL for primary prophylaxis.	Prospective, RCT	77 carvedilol, 75 EVL	Cirrhosis and GEV, randomized to carvedilol vs. EVL	HVPG was not assessed in this study, but it has to be assumed that the lower risk for first variceal bleeding is due to HVPG decrease in the carvedilol (while HVPG is not expected to decrease with EVL)
Shah HA, 2014, J Hepatol (12)	Carvedilol may not be superior to EVL to prevent variceal bleeding in patients with viral cirrhosis	Prospective, RCT	82 carvedilol, 86 EVL	Cirrhosis, GEV, primary prophylaxis	89.9% viral cirrhosis with unknown treatment status, Ulcer bleeding occurred in n = 1 patient
Stanley AJ, 2014, J Hepatol (13)	Carvedilol is not superior to EVL in the prevention of variceal rebleeding	Prospective, RCT	33 carvedilol, 31 EVL	Cirrhosis, secondary prophylaxis.	68% compliance with carvedilol, 65% compliance with EVL. Similar mortality rate in carvedilol vs. EVL
Bhardwaj, 2017, Gut (14)	Carvedilol delays progression from small to large varices	Prospective, RCT	70 carvedilol, 70 placebo	Cirrhosis, small GEV	Mean carvedilol dose to achieve target heart rate was 12 mg/day, reduction of HVPG after 1 year of carvedilol was only −9%
Villanueva, 2019, Lancet (15)	Carvedilol prevents decompensation in patients with cACLD and CSPH	Prospective, RCT (PREDESCI)	33/201 with carvedilol	Compensated cirrhosis with CSPH without high-risk GEV	Carvedilol was only given to iv-propranolol non-responders; positive outcomes were also observed with propranolol (67 patients)
McDowell HR, 2021, APT (16)	Carvedilol was associated with a survival benefit in patients with cirrhosis and varices	Retrospective analysis of prospective RCT	77 carvedilol of 152 patients	Cirrhosis and GEV, randomized to carvedilol vs. EVL	The difference in all-cause and liver-related mortality suggests that the survival benefit with carvedilol may not be entirely liver-related
Dunne PDJ, 2022, APT (17)	Carvedilol use in secondary bleeding prophylaxis is associated with survival benefit, fewer liver-related deaths, and fewer decompensation-related admissions	Follow-up analysis of prospective RCT	26 carvedilol, 28 EVL	Cirrhosis, after variceal bleeding	No significant differences in the rebleeding rates between carvedilol vs. EVL in secondary prophylaxis
Villanueva C, 2022, JHEP (18)	Long-term carvedilol therapy reduces decompensation rate and improves survival in patients with compensated cirrhosis and CSPH	Systematic review of RCT	181 carvedilol, 79 EVL, 92 placebo	Compensated cirrhosis, CSPH, without previous bleeding	The main benefit of carvedilol was to reduce the risk of developing ascites
Non-invasive monitoring of carvedilol (NSBB) response
Kim, 2019, J Hepatol (19)	Decrease in spleen stiffness (measured by ARFI) predicted hemodynamic response to carvedilol	Prospective	106 carvedilol	Cirrhosis and high-risk GEV	Hemodynamic response to carvedilol was achieved in 55.7% (derivation group) and 42.4% (validation group)
Marasco, 2020, Hepatol Int (20)	Decrease in spleen stiffness ≥10% (measured by TE) could predict hemodynamic response to NSBB	Prospective	11 carvedilol, 9 propranolol	Cirrhosis and high-risk GEV	Only 20 of 59 initially included patients completed both paired HVPG and SSM measurements
Ongoing, large prospective trials on carvedilol
Tripathi, 2019, BMJ Open Gastro (21)	CALIBRE Primary endpoint: First variceal bleeding rate with carvedilol (12.5 mg/day) vs. with EVL after 1 year	Prospective, RCT	2630 pts, 4 year recruitment, 66 hospitals	Cirrhosis, medium/large GEV, primary prophylaxis	Secondary endpoints: decompensation, (transplant-free) survival, adverse events, quality of life, healthcare resources (costs)
BOPPP Trial, 2019 (22)	BOPPP Primary endpoint: First variceal bleeding rates with carvedilol (12.5 mg/day) vs. placebo after 3 years, and cost-effectiveness	Prospective, RCT	1200 pts, 5-year recruitment, 25 hospitals	Cirrhosis, small varices, primary prophylaxis	Secondary endpoints: all-cause mortality, progression to medium/large varices, decompensation, MELD increase, HCC, myocardial infarction

Note: References to the table are reported in supplementary material.
Abbreviations: ARFI, acoustic radiation force impulse; cACLD, compensated advanced chronic liver disease; CSPH, clinically significant portal hypertension; EVL, endoscopic variceal ligation; GEV, gastroesophageal varices; HCC, hepatocellular carcinoma; HVPG, hepatic venous pressure gradient; MAP, mean arterial pressure; MELD, model for end-stage liver disease; NSBB, non-selective beta-blockers, RCT, randomized controlled trial; TE, transient elastography.