Baveno Update: New Place for Beta-Blockers

Non-selective beta-blockers have been used for over 40 years to treat portal hypertension, traditionally in the setting of primary and secondary prevention of variceal bleeding. Recent data support an expanded and earlier use of these drugs in patients with compensated disease and clinically significant portal hypertension to prevent first decompensation and improve survival. Selection of candidates to NSBBs might benefit from the use of non-invasive tests in the future.

Rationale of Use of Non-Selective Beta-Blockers in Cirrhosis

Portal hypertension is a severe clinical syndrome characterized by an increased pressure in the portal venous system, and consequently an increased pressure gradient between the portal venous circulation and the systemic circulation (“portal pressure gradient”).¹ Its complications are a major cause of morbidity and mortality in patients with cirrhosis, and the most common include the formation and rupture of gastroesophageal varices, ascites, and hepatic encephalopathy due to porto-systemic shunting.

From a pathophysiological point of view, portal pressure depends on two distinct factors, namely resistance and flow, as per the Ohm’s law: P = R * Q

Portal hypertension in cirrhosis is initiated by an increase in hepatic vascular resistance, in turn due to fibrosis – a “mechanical component” – and to an increased vascular tone caused by endothelial dysfunction – a “dynamic component”. Later on, portal pressure increase leads to a compensatory splanchnic vasodilatation, porto-systemic collaterals formation and hyperdynamic circulation which increase porto-collateral blood flow and further increasing pressure.²

Portal hypertension in cirrhosis is best assessed by measuring its clinical equivalent hepatic venous pressure gradient (HVPG) by minimally invasive hepatic vein catheterization.¹ HVPG normal value is <5mmHg, portal hypertension is considered subclinical when HVPG values are 6–9mmHg, and HVPG ≥10mmHg is termed “clinically significant portal hypertension” (CSPH), since above this threshold all the complications of the syndrome can appear. Variceal bleeding can occur in patients showing HVPG ≥12mmHg. On the other hand, a reduction in HVPG ≥20% from the baseline value or below 12mmHg is associated with protection from the risk of variceal bleeding, and this group of patients is referred as hemodynamic “responders”.³

Non-selective beta-adrenergic blockers (NSBB) act on the flow component, by reducing porto-collateral blood flow as shown in Figure 1. In addition to this well known effect, NSBB foster intestinal transit and inhibit both bacterial overgrowth and translocation,⁴ protect against spontaneous bacterial peritonitis (SBP)⁵ and have anti-angiogenic effects which lead to a reduction of hepatocellular carcinoma (HCC) risk.⁶

Since the first publication of the efficacy of propranolol in preventing recurrent variceal bleeding back in 1981,⁷ a large amount of data confirmed that classic NSBB (propranolol and nadolol, titrated at the maximal tolerated dose) are able to prevent variceal bleeding.⁸ This has been proven both in the setting of primary prophylaxis (patients at high risk of bleeding, who did not experience variceal bleeding yet) and in secondary prophylaxis/elective treatment (after a bleeding has taken place). On the other hand, “responders” to classic NSBB do not exceed 40–65% of patients receiving this therapy.³

Carvedilol is a peculiar NSBB, which in addition to the beta-1 and beta-2 receptor blockade shows an alfa-1 adrenergic receptor blockade. The latter is able to reduce intrahepatic vascular tone, so acting on the dynamic component of vascular resistance (Fig. 1). Furthermore, carvedilol has been shown to increase the synthesis of nitric oxide in the sinusoidal endothelial cells, as well as to have anti-inflammatory and anti-fibrotic effects.^9–11 By acting on both porto-collateral flow and intrahepatic vascular resistance, carvedilol induces a more marked reduction in portal pressure.¹² Its potential drawback is related to the reduction in arterial blood pressure, which in decompensated patients might worsen sodium and free water reabsorption favoring edemas and ascites. Nonetheless, to optimally treat portal hypertension carvedilol is given at a low dose (starting dose: 3.125mg twice per day, followed by 6.25mg twice per day) which usually has no or minimal impact on systemic hemodynamic.

Given to its superior efficacy and to its simple dosing carvedilol has become the preferred NSBB used in patients with portal hypertension.¹⁰

New Evidence Supporting an Early Use of NSBBs

As stated above, NSBBs in cirrhosis are a cornerstone of the prevention of portal hypertensive bleeding or re-bleeding.

The two most recent Baveno Consensus Workshops on portal hypertension (Baveno VI in 2015 and Baveno VII in 2021)^13,14 changed this paradigm, and underlined that clinically significant portal hypertension is a major driver for any type of clinical decompensation in cirrhosis. Given this, treatment of portal hypertension should aim not only at preventing bleeding, but at preventing all complications. The focus should therefore put on compensated patients who did not experienced complications yet.

The PREDESCI study, a multicentric, investigator-initiated, placebo controlled randomized trial in Spain provided strong evidence in this regard.¹⁵ Patients with compensated cirrhosis were included in the study if they showed CSPH and had no varices or small varices, and were randomized to receive NSBB (mostly carvedilol. n=100) or placebo (n=101). Patients on NSBB showed a dramatic reduction in the risk of a first decompensation (HR: 0.51; 95% CI: 0.26–0.97; p=0.41 vs. placebo); ascites was the most common first decompensating event, occurring much less frequently in patients in the NSBB arm (HR: 0.44; 95% CI: 0.20–0.97; p=0.037 vs. placebo).¹⁵

An individual patient data competing-risk meta-analysis of 4 RCTs using carvedilol in 352 compensated patients confirmed these observations.¹⁶ The risk of developing decompensation was lower with carvedilol vs. controls (subdistribution HR: 0.506; 95% CI: 0.289–0.887; p=0.017; I2=0.0%; Q-statistic-p=0.880), mainly due to a reduced risk of ascites (SHR: 0.491; 95% CI: 0.247–0.974; p=0.042; I2=0.0%; Q-statistic-p=0.384). Importantly, carvedilol was associated with a lower risk of death (SHR: 0.417; 95% CI: 0.194–0.896; p=0,025; I2=0.0%; Q-statistic-p=0.989).

Data from an elegant national cohort study performed in the United States confirms this observation,¹⁷ further supporting the use of carvedilol in the setting of compensated cirrhosis.

Selection of Patients Requiring NSBB – Past, Present and Future

On first assessment, clinicians should assess whether patients had already developed decompensating episodes (in the past or at present). Patients with decompensated cirrhosis have per definition CSPH and treatment should be urgently considered. However, given the complexity of decompensated patients, upper gastrointestinal endoscopy is mandatory to better tailor therapy in this population.

If the patient had no decompensating episode, the scenario is very different. CSPH – which is the substrate for the development of varices and complications – is present only in 50–60% of cases of compensated cirrhosis.^18–20 Patients without CSPH are at a negligible risk, while patients with CSPH are at risk of developing complications that go beyond variceal bleeding and could be ideally treated irrespective of the endoscopic findings.

In the past, there was no sensitive non-invasive tool helping in risk stratification for portal hypertension in compensated patients with cirrhosis, and universal endoscopic screening of varices was recommended.²¹ In patients with high risk varices (HRV), primary prophylaxis with NSBB was started.

Liver elastography emerged in the last 15 years as a sensitive and reliable tool to stratify the risk of portal hypertension in cirrhosis. The Baveno VI consensus in 2015 proposed that this non-invasive test combined with platelet count could be used to skip endoscopic screening in patients at very low risk of having HRV.¹³ These so-called “Baveno VI Criteria” (liver stiffness <20kPa + platelet count >150G/L) have been largely validated,²² and can be used in clinical practice. Spleen stiffness measurement can be added to these criteria to further improve risk stratification and avoid endoscopy in a higher proportion of cases.^23,24

The observation that carvedilol is able to prevent decompensation in patients with compensated cirrhosis and CSPH led the Baveno VII consensus to propose NSBB in an earlier stage of the disease, when varices might be absent.¹⁴

The recommendation is as follows: “Treatment with non-selective beta-blockers (NSBBs) (propranolol, nadolol or carvedilol) should be considered for the prevention of decompensation in patients with CSPH (B1)”.¹⁴

Despite CSPH remains a hemodynamic definition which requires HVPG measurement, the use of non-invasive tests to identify patients who carry CSPH and are at risk of decompensation is very attractive. The likelihood of CSPH in patients with liver stiffness ≥25kPa is very high and these patients have a high risk of developing a first episode of decompensation.²⁵ Whether non-invasive tests can be pragmatically used to start carvedilol remains to be validated, and is one of the focuses of current research.²⁶

Figure 2 summarizes how the field has evolved in the last 15 years.

Literature: