Non-Selective Beta-Blockers in Patients with Cirrhosis: Current Evidence and Controversy

Abstract

Non-selective beta-blockers (NSBBs) are the cornerstone of management of liver cirrhosis and its complications. They decrease portal vein blood flow and portal vein pressure via antagonism of β₁ and β₂ receptors, thereby improving the hyperdynamic circulatory status and reducing the risk of decompensated events in cirrhotic patients with portal hypertension (PH). Currently, NSBBs have been recommended for the primary and secondary prevention of variceal bleeding in patients with cirrhosis. Nevertheless, it has been noted that the clinical risk-benefit ratio appears to deteriorate for NSBBs in patients with advanced liver cirrhosis. Considering that the use of NSBBs in the clinical practice of cirrhotic patients remains controversial, this chapter aims to summarize the evidence of current recommendations and controversies regarding the use of NSBBs in patients with cirrhosis.

1. Introduction

Liver cirrhosis, characterized by progressive fibrosis and liver injury, is the end stage of several liver diseases [1]. Portal hypertension (PH), a primary consequence of cirrhosis, may cause serious complications, including variceal bleeding (VB), ascites, and hepatic encephalopathy (HE) [2, 3]. Currently, hepatic venous pressure gradient (HVPG) measurement is considered the gold standard test for the diagnosis of PH in clinical practice [4]. HVPG represents the actual liver portal perfusion pressure, and its value greater than 5 mmHg indicates PH. Portal pressure increases with progressive disease, ultimately surpassing cut-off values of 10 mmHg, which marks the occurrence of clinically significant PH (CSPH). Patients with CSPH and cirrhosis exhibit splanchnic and peripheral vasodilatation, resulting in increased heart rate and cardiac output, which is characteristic of the hyperdynamic circulation state of advanced PH [5, 6].

Non-selective beta-blockers (NSBBs) are the cornerstone of management of PH [7]. NSBBs reduce peripheral blood flow via antagonism of β₁ receptor, thereby reducing the heart rate and cardiac output, as well as splanchnic vasoconstriction through inhibition of β₂ receptor, thereby decreasing portal venous blood flow and portal pressure [8, 9]. Accordingly, NSBBs could improve the hyperdynamic circulatory status of CSPH patients and decrease the development of decompensated events in patients with compensated cirrhosis. Additionally, NSBBs may reduce bacterial translocation and inflammatory responses through non-hemodynamic mechanisms [10]. The traditional NSBBs commonly used in clinical practice include propranolol, nadolol, and timolol. Carvedilol is a novel NSBBs with an intrinsic anti-α-adrenergic activity, which stimulates the release of nitric oxide and causes intrahepatic vasodilatation, thereby further decreasing portal pressure [11].

According to the updated consensus and guidelines, NSBBs are specifically recommended for primary or secondary prophylaxis of VB in patients with high-risk varices [12, 13, 14]. It has been recently shown that NSBBs are not always beneficial to all cirrhotic patients [15]. Tonon et al. identified that the use of NSBBs may increase the risk of acute decompensation and non-acute decompensation in compensated patients [16]. In the window hypothesis, as proposed by Krag et al., it was suggested that the efficacy and safety of NSBBs may vary based on the spectrum of cirrhosis [17]. They questioned the use of NSBBs in early cirrhosis, which is free of medium or large varices, and warned to avoid NSBBs in patients with advanced cirrhosis and refractory ascites. This chapter aims to summarize the existing evidence on the indications and controversies for NSBBs in patients with cirrhosis, in order to provide a reference for the use of NSBBs in the clinical practice.

2. NSBBs in cirrhosis: indications

2.1 Clinically significant portal hypertension

CSPH is defined as HVPG ≥10 mmHg, as decompensation may develop after achieving this threshold [18]. Indeed, CSPH is the strong predictor of PH-associated complications such as VB, ascites, and HE in patients with compensated cirrhosis [19]. NSBB-induced reduction of HVPG at least 10% from baseline may prevent the first episode of VB and ascites [12]. A randomized, double-blind, multicenter trial (PREDESCI) in Spain included 201 compensated cirrhotic patients with CSPH and without high-risk varices. Patients were randomized to treatment with placebo or NSBBs (propranolol or carvedilol depending on hemodynamic response). Hemodynamic response to NSBBs was determined as a decrease of HVPG of ≥10% after 20 minutes of intravenous propranolol (0.15 mg/kg). They found that primary outcomes such as ascites, VB, or HE occurred in 16% of patients in the NSBBs group vs. 27% in the placebo group (P = 0.041). The difference was due to a reduced incidence of ascites (P = 0·0297) [20]. An individual participant data meta-analysis of four randomized controlled trials (RCTs) revealed that the risk of decompensation in the carvedilol group was significantly lower than in the control group, which included endoscopic variceal ligation (EVL) and placebo among compensated patients with CSPH (P = 0.017) [21]. Kumar et al. randomly assigned 136 acute-on-chronic liver failure patients who were definitively diagnosed with CSPH and had no or small varices to carvedilol or placebo groups, which demonstrated that the use of carvedilol improved 28-day survival and reduced the risk of acute kidney injury (AKI) [22]. Thus, it appears to suggest that patients may benefit from the initiation of NSBBs once the CSPH has been diagnosed, whatever the status of the varices.

Notably, it is suggested that carvedilol is more effective in decreasing HVPG than propranolol or nadolol, including propranolol non-responders, with about 75% of patients achieving an HVPG response [23]. Besides, the Baveno VII consensus indicated that administration with NSBBs should be considered to prevent decompensation in patients with CSPH. Carvedilol is recommended as the preferred NSBBs for compensated cirrhosis compared to the traditional NSBBs (i.e., propranolol and nadolol) [12]. Overall, screening patients with CSPH in compensation cirrhosis for prompt initiation of NSBBs (preferred to carvedilol) may improve prognosis.

2.2 Primary prophylaxis

Currently, international guidelines suggest that NSBBs and EVL are similarly effective for primary prophylaxis, which are both recommended for patients with high-risk varices such as medium to large varices or small varices with red signs or Child-Pugh C [12, 13].

In a study by McDowell et al., 152 patients were included to evaluate the efficacy of carvedilol for the primary prevention of VB and survival. It showed that the median survival was significantly higher in the carvedilol group than in the EVL group (7.8 vs. 4.2 years, P = 0.03) [24]. A meta-analysis of 32 RCTs showed that monotherapy with NSBBs may reduce the risk of the first VB and mortality in cirrhotic patients with large varices, with a lower incidence of serious complications than EVL [25]. However, both these studies included patients with compensated cirrhosis and those with decompensated cirrhosis. Notably, the hemodynamic response of patients to NSBBs in cirrhosis varies across the spectrum of cirrhosis [26, 27]. In addition, the primary management goals for patients with cirrhosis also depend on the stage of cirrhosis. In patients with compensated cirrhosis, the major objective of treatment is not limited to the prevention of VB, which must focus on the prevention of decompensation events (ascites and HE). Thus, it is now preferred to choose treatments based on the stage of the disease. A competing-risk meta-analysis of 11 RCTs evaluated the efficacy of NSBBs vs. EVL on survival in patients with high-risk varices, stratifying risk based on the compensated/decompensated stage of cirrhosis. It showed that treatment with NSBBs had significantly improved survival compared with EVL in compensated cirrhotic patients, but in patients with decompensated cirrhosis, the survival was similar between the two therapies [28].

It is worth noting that the development of adverse events and the economic costs should be also considered while effective treatments are selected. Treatment with NSBBs has a lower risk of serious complications such as esophageal ulcers and esophageal perforation compared to EVL, which is more easily obtained, less depending on specialized endoscopic equipment like EVL, and has a higher economic efficiency.

2.3 Secondary prophylaxis

Nowadays, updated international guidelines and consensus recommend the use of NSBBs combined with EVL for secondary prophylaxis of VB [12, 13]. In addition, a combination therapy of NSBBs and isosorbide-mononitrate (IsMn) is also recommended as secondary prophylaxis of VB by the Japanese Society of Gastroenterology [14]. Kong et al. included 43 RCTs to compare the efficacy of medications, endoscopic options, transjugular intrahepatic portosystemic shunts (TIPS), and mainstay combination therapies in cirrhotic patients with at least one episode VB. It showed that nadolol combined with ISMN may be a preferable option to reduce mortality associated with rebleeding in patients with cirrhosis [29]. Another network meta-analysis of 26 RCTs showed that nadolol combined with EVL may be the preferred option for preventing rebleeding in cirrhotic patients who had one previous episode of VB, while NSBBs combined with ISMN may be potential alternatives to improve mortality [30]. Moreover, Jachs et al. identified that the cumulative incidences of rebleeding (P = 0.027) and liver-related death (P = 0.036) were significantly lower in patients administered with carvedilol combined with EVL versus propranolol combined with EVL [31].

3. NSBBs in cirrhosis: controversies

3.1 Preprimary prophylaxis

Currently, there is inadequate evidence to support the use of NSBBs for the prevention of the development or growth of varices in cirrhotic patients with no varices. Groszmann et al. randomly assigned 213 patients with HVPG ≥6 mm Hg without varices to the timolol group and the placebo group. During a median follow-up of 54.9 months, the incidence of the primary endpoint (development of gastroesophageal varices or VB) was not significantly different between the timolol and the placebo groups (39 vs. 40%, P = 0.89). Besides, severe adverse events were reported with a higher proportion of patients in the timolol group compared with the placebo group (18 vs. 6%, P = 0.006) [32]. It’s worth noting that most patients (59%) without varices had only subclinical PH (SPH) in this study. However, NSBBs exert a stronger effect for reduction of portal pressure in CSPH patients than in SPH patients who do not develop increased cardiac output and hyperdynamic circulation, which is a prerequisite for NSBBs to exert their activity [26]. This explains why NSBBs are ineffective generally in preprimary prophylaxis. Similarly, a meta-analysis of six RCTs including patients with no or small varices also showed no significant benefit of NSBBs compared with placebo for the prevention of large varices, first VB, and death [33].

In contrast, Bhardwaj et al. showed that a lower proportion of patients in the carvedilol group had a progression of small varices to large varices compared to the placebo group [34]. Di Pascoli et al. conducted a long-term observation of patients who started with NSBBs when diagnosed with small varices versus those treated with NSBBs after the confirmation of large varices by endoscopy. They also found that the small-varices group had a significantly lower rate of adverse outcomes (progression of varices or VB or death) than the large-varices group and had lower economic costs [35]. In general, the evidence is conflicting regarding the effect of NSBBs on slowing the development of small varices into large varices in cirrhotic patients. Thus, treatment should be focused on elimination or control of the cause in liver disease at this stage. Additionally, it is also important to monitor the magnitude of hemodynamic changes to determine a hyperdynamic circulatory state and prevent decompensated events in patients with no or small varices.

3.2 Ascites and refractory ascites

In 2010, a milestone study by Serste et al. revealed that treatment with propranolol was associated with poor survival in patients with refractory ascites. This study created a lively debate in the hepatology field [36]. Notably, the NSBBs group had a higher proportion of Child-Pugh class C and varices, higher bilirubin levels, and lower serum sodium in this study, which might justify these differences. Furthermore, nearly half of the patients were treated with a high-dose propranolol of 160 mg reported by Serste et al. However, a real-world multicenter study found that low-dose (≤ 80 mg/d) NSBBs treatment was associated with improved survival in primary and secondary prophylaxis of VB among advanced cirrhosis [37]. Park et al. also showed that low-dose propranolol (40–120 mg/d) decreased the risk of rebleeding of varices and mortality in patients with refractory ascites [38].

The window hypothesis, as proposed by Krag et al., suggests that the benefit windows of NSBBs may be closed once patients develop refractory ascites [17]. It may be explained that proinflammatory cytokines are increased in the splanchnic compartment among patients with refractory ascites, in whom chronic inflammation may cause decreased expression of β-adrenergic receptors and sympathetic nerve fiber loss, thereby leading to diminished effects of NSBBs [39]. However, several recent studies to investigate the efficacy of NSBBs in patients with ascites have reported opposite results. Tergast et al. enrolled 624 decompensated cirrhotic patients with ascites to evaluate the impact of NSBBs on survival. It showed that the administration of NSBBs is associated with improved 28-day transplant-free survival [40]. Leithead et al. included 322 patients with ascites who were listed for liver transplantation and identified significantly lower waiting-list mortality in NSBBs group [41]. Additionally, Onali et al. and Facciorusso et al. did not demonstrate that NSBBs affected the overall survival in cirrhotic patients with refractory ascites [42, 43].

Notably, the guideline for the management of ascites in cirrhosis by the British Society of Gastroenterology suggests that refractory ascites should not be considered a contraindication of NSBBs [44]. Patients with refractory ascites who are taking NSBBs should be closely monitored, and if they develop hypotension or acute/progressive renal dysfunction, the dosage may be reduced or discontinued appropriately. Similarly, the Baveno VII consensus proposed that NSBBs should be discontinued or decreased in patients with ascites who develop sustained hypotension (systolic blood pressure < 90 mmHg or mean arterial blood pressure < 65 mmHg) and/or hepato-renal syndrome-acute kidney injury (HRS-AKI). Once blood pressure returns to baseline and/or HRS-AKI resolves, NSBBs should be restarted [12]. In conclusion, the current evidence would not support discontinuation of NSBBs in patients with ascites or refractory ascites. NSBBs should be used with careful attention and reduced or discontinued in those, if appropriate.

3.3 NSBBs and renal dysfunction

Serste et al. included 139 patients with severe hepatitis to estimate the effect of NSBBs on the risk of development of AKI, in which the cumulative AKI incidence was significantly higher in the NSBBs group than in the non-NSBBs group (89.6 vs. 50.4%, P = 0.0001). They concluded that NSBBs were an independent risk factor for AKI in those patients [45]. Kim et al. identified that the impact of NSBBs on renal injury varied according to the severity of hepatic decompensation, with an increased risk of AKI by more than three-fold in cirrhotic patients with ascites [46]. Meanwhile, Singh et al. evaluated the effect of NSBBs on the development of AKI and 12-month transplant-free-survival in patients with ≥grade 2 ascites compared to EVL. They observed that the use of NSBBs is associated with poor survival and increased risk of AKI in those patients [47]. Finally, a meta-analysis including 14 studies, as reported by Xu et al., found that NSBBs significantly increased the risk of developing renal dysfunction [48]. The explanation for this apparent discrepancy lies in the fact that renal perfusion and its function are dependent on excessive activation of the sympathetic nervous system in patients with advanced cirrhosis, while NSBBs blunt the sympathetic overdrive, reduce cardiac output, and decrease renal perfusion pressures below critical values, thereby impairing renal function [49]. Notably, Xu et al. failed to demonstrate the association of NSBBs with an increased risk of development of renal dysfunction after covariate adjustment. Bang et al. and Scheiner et al. also found that treatment with NSBBs was associated neither with impairing of kidney function nor with increased incidence of AKI [50, 51].

However, Li et al. showed that the administration of NSBBs may reduce the occurrence of AKI and marginally alleviate chronic renal function declination [52]. Maiwall et al. observed consecutive cirrhotic patients with a complete record of baseline hemodynamics. During a median follow-up of 379 days, they found that treatment with NSBBs could prevent the frequency of AKI and the development of chronic kidney disease (CKD) [53]. It may be attributed to the fact that PH and severe vasodilation predispose cirrhotic patients to repeatedly develop AKI and progression to CKD, while NSBBs could improve hemodynamic response and reduce portal pressure, thereby decreasing the occurrence of AKI and CKD. Thus, the current evidence regarding the impact of NSBBs on renal dysfunction remains controversial. According to updated guidelines and consensus, NSBBs should be discontinued or dose-reduced in patients who develop AKI and CKD [12, 13].

3.4 NSBBs and spontaneous bacterial peritonitis

A meta-analysis included six studies to evaluate the possible role of NSBBs in spontaneous bacterial peritonitis (SBP) among cirrhotic patients with ascites, which found that NSBBs reduced the risk of SBP in the whole cohort, and the effect is still observed when compared between hemodynamic responders and non-responders [54]. It may be attributed to the non-hemodynamic mechanisms of NSBBs in reducing bacterial translocation and inflammatory response [55]. A large-scale retrospective cohort study from Denmark included 361 patients with SBP and revealed that the use of NSBBs was associated with improved survival. Additionally, in patients with severe decompensated cirrhosis, the use of NSBBs was associated with a lower risk of occurrence of SBP [50]. A prospective cohort study evaluated the possible factors associated with the prediction of recurrent SBP, which found that nonuse of prophylactic NSBBs was strongly associated with an increased rate of SBP recurrence [56].

However, Mandorfer et al. showed that treatment with NSBBs may be associated with poor transplant-free survival, increased risk of HRS and AKI, and extended periods of hospitalization in patients with SBP [57]. Notably, the NSBBs group had a higher bilirubin level, and a higher proportion of Child-Pugh class C and varices among SBP patients for subgroup analysis in this study might justify these differences. Similarly, according to the European Association for the Study of the Liver (EASL) clinical practice guideline, NSBBs should be discontinued in patients with acute complications such as VB, sepsis, SBP, or AKI. After recovery, it may be reinitiated appropriately [13]. Currently, controversy remains regarding the definitive efficacy of NSBBs in patients with SBP, but it is recommended to discontinue the use of NSBBs once the patients develop SBP. In future, large-scale prospective studies are needed to further explore the relationship between NSBBs and SBP.

3.5 NSBBs and portal vein thrombosis

Portal vein thrombosis (PVT) is mainly associated with decreased portal vein blood flow and velocity, and the portal vein blood flow <15 cm/s is considered an independent risk factor of PVT [58, 59]. The development of PVT is strongly associated with poor prognosis in cirrhotic patients [60, 61]. A significant higher risk of VB was observed in patients with PVT due to further increased portal pressure (47.33 vs. 19.63%) [62]. Currently, NSBBs are the first-line treatment for the prevention of VB, which significantly decrease the portal blood velocity and portal vein pressure [12]. However, Xu et al. indicated that a higher incidence of PVT was observed in the NSBBs group compared to the non-NSBBs group in univariate logistic regression analyses (P = 0.035), but no significant difference was found between the two groups after covariate adjustment (P = 0.194) [63]. Pan et al. demonstrated that decreased portal flow velocity and use of NSBBs were associated with the occurrence of PVT in liver cirrhosis [64]. Meanwhile, a meta-analysis included nine studies to evaluate whether NSBBs increase the risk of PVT, which found that the use of NSBBs increased a 4.62-fold risk of PVT in cirrhotic patients, but the results are limited by poor quality of included studies and significant heterogeneity across studies [65]. Therefore, there is inconclusive evidence to support the association of PVT formation with NSBBs, which requires large-scale RCTs to further confirm in the future.

4. Conclusion

NSBBs are the cornerstone for the management of PH in clinical practice, which has been specifically recommended for the primary and secondary prophylaxis of VB. In addition, the patient should be initiated on NSBBs promptly after the diagnosis of CSPH, regardless of the status of the varices. Considering the hemodynamic response of patients to NSBBs in cirrhosis varies based on the spectrum of cirrhosis, physicians should comprehensively assess the disease conditions of patients in clinical practice to choose the appropriate time and duration of administration with NSBBs.

Additionally, patients with ascites and refractory ascites should not be considered to be contraindications for NSBBs, which should be used under careful monitoring. Notably, if patients persistently develop hypotension and/or HRS-AKI, NSBBs should be discontinued or decreased until their blood pressure returns or HRS-AKI resolves, then restarted.