Decompensation of cirrhosis is a turning point in cirrhosis course, as associated with a marked decrease in life expectancy. Thus, prevention of decompensation is crucial.

The usefulness of carvedilol to prevent decompensation of cirrhosis in patients with TE-LSM ≥ 25 kPa as a surrogate marker for clinically significant portal hypertension, has never been evaluated in a clinical trial.

Decompensation of cirrhosis is a turning point in cirrhosis course, as associated with a marked decrease in life expectancy. Thus, prevention of decompensation is crucial.

Portal hypertension (PH) is the strongest predictor of decompensation. Hepatic venous pressure gradient (HVPG) is the reference standard for the evaluation of PH. HVPG ≥10 mm Hg, called "clinically significant portal hypertension", identifies a population with a high risk of decompensation. HVPG measurement is an invasive procedure, only routinely available in expert centers. Liver stiffness measurement (LSM) using transient elastography (TE) (referred as TE LSM) using Fibroscan can provide an indirect estimate of HVPG. TE-LSM ≥ 25 kPa can rule-in HVPG ≥10 mm Hg with a specificity >90%.

Nonselective beta-blockers (NSBBs) lower portal pressure by decreasing portal venous inflow. Carvedilol also decreases intrahepatic vascular resistance, and thereby achieves a greater reduction in portal pressure than propranolol. At low-dose (≤12.5 mg/day), carvedilol is safe in patients with compensated cirrhosis.

In patients with asymptomatic cirrhosis, NSBBs were recommended when medium or large varices (high-risk varices) are present for prophylaxis of variceal bleeding. In a recent randomized controlled trial, the PREDESCI study (NCT01059396), NSBBs reduced incidence of decompensation or death in patients with compensated cirrhosis with clinically significant portal hypertension. In the PREDESCI study, the diagnosis of clinically significant portal hypertension was based on invasive HVPG measurement, so that its results are not applicable in clinical practice.

The usefulness of carvedilol to prevent decompensation of cirrhosis in patients with TE-LSM ≥25 kPa as a surrogate marker for clinically significant portal hypertension, has never been evaluated in a randomized controlled trial.

Inclusion Criteria:

• Male or female≥ 18 years of age
• Cirrhosis related to hepatitis C or hepatitis B virus without viral replication for at least 2 years.

Or Cirrhosis related to alcohol consumption (active or abstinent) Or Cirrhosis related to metabolic syndrome or cryptogenic with BMI < 30 kg/m2

• 2 TE-LSM (Fibroscan®) performed in fasting conditions, using either the M or the XL probe >=25 kPa, within 12 months before inclusion
• Absence of medium or large varices or small varices with red signs at endoscopy within 3 months before inclusion
• Child-Pugh A5 to B8
• Affiliation to a French social security system.
• Written informed consent obtained from the participant or participant's legal representative
• For child-bearing aged women, contraception using oral contraceptive, or intrauterine device or mechanical contraception

Exclusion Criteria:

• History of overt ascites or encephalopathy <12 months before inclusion

• Treatment with either diuretics or lactulose or rifaximin <3 months before inclusion

• Any history of portal hypertension related bleeding

• Baseline heart rate <65/min or systolic blood pressure <100 mm Hg

• Previous transjugular intrahepatic portosystemic shunt (TIPSS) or liver transplantation

• Previous history or active hepatocellular carcinoma

• Glomerular filtration rate (CKD-Epi) < 30 mL/min

• Strict indication to selective or nonselective beta-blockers: history of acute myocardial infarction, congestive heart failure

• Strict contraindication to selective or nonselective beta-blockers:

  • decompensated congestive heart failure
  • grade 2 or 3 atrioventricular block
  • sinus node dysfunction without pacemaker
  • severe asthma according to WHO classification [63]
  • severe Chronic Obstructive Pulmonary Disease, defined stage 3 or 4 of the GOLD classification, i.e.FEV1<50% of the predicted value (https://goldcopd.org/)
  • severe Raynaud disease, defined as repetitive episodes of biphasic colour (at least two) of pallor, cyanosis, erythema, in addition to paresthesia or numbness, occurring in both cold and normal environments [64].

• Known hypersensitivity to carvedilol

• Concomitant use of Cimétidin

• Concomitant use of class I antiarythmic agents (except lidocaïn) (i.e.cibenzoline, disopyramide, flécaïnide, hydroquinidine méxilétine, propafenone, quinidine)

• Concomitant use of calcium antagonists: diltiazem, vérapamil and bépridil

• Concomitant use of clonidine, méthyldopa, guanfacine, moxonidine, rilménidine

• Concomitant use of fingolimod

• Concomitant use of potent inhibitors (e.g. ketoconazole, HIV protease inhibitors) or inductors (e.g. rifampin, carbamazepine, phenytoin) of CYP3A4 (see appendix 7)

• Pregnancy or breastfeeding

• Non ability for participant to comply with the requirements of the study

• Life expectancy <12 months