Alimentary Pharmacology & Therapeutics | 2021
Editorial: when to start carvedilol in cirrhosis—time to reconsider?
Abstract
There is increasing evidence that nonselective betablockers (NSBB), particularly carvedilol, have benefits reaching beyond the prevention of variceal bleeding. A recent long term followup study of a previous randomised controlled trial comparing carvedilol versus endoscopic band ligation for primary prophylaxis of variceal bleeding demonstrated a significant survival benefit for patients on carvedilol.1 Importantly, this could not all be explained by a reduction in liverrelated mortality and further study into the underlying mechanisms is needed. The PREDESCI study examined the role of NSBB in preventing decompensation, with patients allocated to propranolol, or carvedilol in HVPG nonresponders, versus placebo.2 They demonstrated a reduced incidence of cirrhosis decompensation in the NSBB group with no difference in adverse events. Prompted by a study by Sersté et al showing increased mortality in patients with refractory ascites on NSBB, the “window hypothesis” suggests that NSBB are deleterious in advanced liver disease with refractory ascites, with additionally no benefit seen in those with early cirrhosis.3,4 More recent evidence has disputed this however, with NSBB, including carvedilol, shown to be safe and conferring a survival benefit even in patients with significant ascites and endstage liver disease.5,6 Karn Wijarnpreecha et al have conducted a retrospective populationbased cohort study of over 100,000 patients investigating the risk of hepatocellular carcinoma (HCC) in cirrhotic patients on NSBB.7 They demonstrated a consistently lower incidence of HCC in patients treated with any of propranolol, nadolol or carvedilol versus no betablocker. Significant effort was made to minimise confounding factors and the protective effect of a NSBB was maintained after propensity score matching, multivariate analysis and additional subgroup analysis. The authors discuss evidence for a number of mechanisms that may be responsible including carvedilolinduced inhibition of tumour angiogenesis and a reduction in oxidative stress/IL6 pathway. A previous study has also demonstrated that carvedilol has antifibrotic activity, raising the possibility that it could slow the progression of the underlying liver disease.8 There are some limitations to this study which are acknowledged by the authors in that it was retrospective with the possibility of disease coding variability and the inability to control for factors not electronically recorded. Of note, the characteristics of the carvedilol group were significantly different to both the nadolol and propranolol groups; only 23% of the carvedilol group had documented portal hypertension, with only 14% reported varices compared with >70% for both in the other two groups. It would therefore be interesting to understand the indication for commencing carvedilol in the majority of this cohort. With the restrictions in endoscopy capacity during the COVID19 pandemic, some units have reevaluated the criteria for when to start carvedilol, from endoscopic to criteria including fibroscan, platelets, previous variceal size and Child Pugh score. If carvedilol slows down disease progression, starting before decompensation seems sensible in those with progressive disease. However, the results from this recent paper regarding potential protection from HCC suggests that perhaps carvedilol should be commenced in all patients with cirrhosis. Ideally, this should be the subject of a large randomised controlled clinical trial.