Key messages

• The addition of beta-blockers to endoscopic variceal ligation as secondary prophylaxis of variceal hemorrhage, leads to a substantial reduction in the risk of rebleeding.
• The addition of beta-blockers to endoscopic variceal ligation as secondary prophylaxis of variceal hemorrhage might not reduce mortality, although the certainty of this evidence is low.
• The addition of beta-blockers would be associated to frequent but mild adverse effects, which generally do not lead to discontinuation of therapy, although the certainty of this evidence is low.

What is the evidence.
See evidence matrix  in Epistemonikos later

We found seven [1],[2],[3],[4],[5],[11],[12] systematic reviews including three randomized controlled trials reported in four references addressing the question of interest [13], [14],[15],[16].
Two trials [17],[18] incorporated in most systematic reviews identified, were excluded from this summary because the treatment arm included isosorbide mononitrate.

What types of patients were included

All reviewed trials included patients who had already had a previous episode of variceal bleeding.
Two trials [14],[15] included only cirrhotic patients. One trial [16] did not report whether all patients were cirrhotic.
Only in one trial [14] most patients were Child B. Two trials [15],[16] did not report this data.
In one trial [14], the most common cause of liver cirrhosis was alcohol. In another trial [15], the most common cause was viral and one trial [16] did not report this data.

What types of interventions were included

The interval between ligation sessions was less than three weeks in two trials [14],[15]. One trial [16] did not report this data.
Propranolol was used in two trials [15],[16] and nadolol in one trial [14].
In all of the trials [14],[15],[16] the drug therapy was titrated to achieve a reduction of 25% of baseline heart rate.
The follow-up time was greater than one year in one trial [14]. In two trials [15],[16] follow-up time was less than one year.

What types of outcomes
were measured

The main outcomes were overall mortality and rebleeding in all of the systematic reviews identified. Four reviews [1], [2],[3],[12] included adverse effects in their analysis, but none of them reported the data of adverse effects of the trials included in this summary.

To whom this evidence does and does not apply

• The information presented in this summary applies to adult patients with liver cirrhosis of any etiology, who have already experienced an episode of variceal bleeding.
• It only applies to patients treated with endoscopic variceal ligation-based therapy and does not include sclerotherapy.
• This information does not apply to patients with liver cirrhosis with no history of variceal bleeding, or who have had an episode of upper gastrointestinal bleeding for any reason other than variceal.

• The outcomes presented in the summary of findings table correspond to those critical for decision-making according to the opinion of the authors of this article.

• This is an intervention with a substantial benefit in terms of rebleeding and few adverse effects, which generally do not lead to discontinuation of therapy. While there is uncertainty about the latter, the balance between benefits and risks is favorable.
• There is uncertainty regarding the effect of beta-blockade on mortality in this scenario. On the one hand, it is expected that the decrease in rebleeding events would translate into a decrease in mortality risk resulting from the immediate complications of digestive hemorrhage. On the other hand, available evidence does not show a categorical decline in mortality. This situation can be explained by the fact that neither endoscopic therapy nor beta-blockade are involved in the progression of liver cirrhosis and the appearance of other serious complications of the disease, such as spontaneous bacterial peritonitis, refractory ascites, hepatorenal syndrome. On this regard, although combined acute therapy and prophylactic endoscopic therapy categorically reduce the risk of rebleeding, this does not necessarily translate into a decrease in mortality, since the variceal rebleeding episode would be a marker of poor prognosis and progression of liver failure, which ultimately explains long-term mortality.

• The vast majority of patients and doctors should lean in favor of combination therapy based on the evidence.
• While there is uncertainty about adverse effects, especially on their frequency and magnitude, probably these are not the main drivers for decision-making.

• This is an inexpensive and widely available intervention, so the cost-benefit balance is very favorable.

Differences between this summary and other sources

• The findings of our summary coincide with systematic reviews identified in relation to the risk of rebleeding.
• Clinical guidelines for secondary prophylaxis of variceal bleeding reach to similar conclusion to those reported in this summary. The consensus of BAVENO VI [19] recommended combination therapy as first-line treatment for secondary prophylaxis of variceal bleeding (either propranolol or nadolol). Carvedilol is not recommended by the absence of evidence to support it. On the other hand, the european guideline 2015 [20] provides exactly the same recommendations, except carvedilol is mentioned as an alternative to propranolol or nadolol.

• The probability that future evidence changes the conclusions presented in this summary is low due to the certainty of the evidence.
• According to the records of the International Controlled Trials Registry Platform of the World Health Organization, there is at least one ongoing trial [21] which would complete recruitment by 2018, with an estimate of 212 patients. This trial could add greater certainty, particularly with regard to adverse effects and the magnitude of benefit.