Non-selective beta-blockers in the prevention of first variceal bleeding. Is there any definite alternative?

Abstract

Variceal bleeding is one of the more frequent and severe complications for patients with cirrhosis and portal hypertension [1]. The risk of having a variceal bleeding episode is directly related to the size of the varix, the presence of red signs on its wall, and the degree of liver dysfunction evaluated by the Child–Pugh classification [2]. Once variceal bleeding occurs, it is a high resource-consuming event associated with a high mortality approaching 25% in most recent series [3]. During the consensus conference held in Baveno [4] it was established that patients with cirrhosis and large esophageal varices (defined as those larger than 5 mm) should be treated to prevent the first variceal bleeding episode. However, in those with small varices, there is no available data to recommend treatment [4]. Several therapeutic alternatives have been tested in the last two decades. From these studies, including thousands of patients, we know that in patients with cirrhosis the pharmacological treatment with non-selective beta-blockers, but not portacaval shunts or endoscopic sclerotherapy, is effective and the chosen method to prevent the first variceal bleeding [1]. The beneficial effect of beta-blockers is observed in patients with or without ascites or with poor or good liver function [5]. The risk of bleeding is reduced by 40–50% by beta-blockers [6] and mortality is reduced almost significantly. It is therefore mandatory to treat all cirrhotic patients with large varices with non-selective beta-blockers. However, although the residual risk of bleeding is relatively low, around 15% at 2 years, beta-blockers do not protect all treated patients [6]. The addition of isosorbide-5-mononitrate (IS-MN) to non-selective beta-blockers has been suggested to improve its clinical efficacy because of the ability of the combined therapy to promote a greater reduction in portal pressure, almost double, than beta-blockers administered alone [7,8]. Several clinical studies strongly suggest that this is true for the secondary prevention of variceal bleeding and in many centers the combination of non-selective beta-blockers with IS-MN is the pharmacological treatment of choice for the prevention of variceal rebleeding [9–12]. However, the data are controversial regarding primary prophylaxis. Indeed, up to now, three randomized controlled trials (RCTs), only one as a full report, have addressed the role of the combined therapy in the prevention of first variceal bleeding with divergent results [13–15]. The combined analysis of these three studies, including 552 patients, has not shown significant differences in the prevention of first bleeding or in mortality between patients treated with beta-blockers alone or in association with IS-MN [6]. Therefore, the available evidence does not support the use of the combination therapy in primary prophylaxis. However, the relatively low residual risk of bleeding while on beta-blockers makes it difficult to demonstrate a hypothetical improvement in efficacy by the combined therapy. In the future, if patients with a greater variceal bleeding risk can be detected and treated, the possible beneficial effect of the combined therapy should probably be re-evaluated. Fifteen to twenty-five percent of patients with cirrhosis and large varices have contraindications to non-selective beta-blockers or develop severe side effects that preclude its use [16]. In this situation, long-acting vasodilators such as IS-MN, due to its ability to reduce portal pressure [17,18], raised expectations as a possible alternative to beta-blockers in the primary prevention of first variceal bleeding. IS-MN seemed to fulfill these expectations when it was reported, in an open RCT including 118 patients with varices of any size, that IS-MN was as effective as propranolol in the prevention of first variceal bleeding [19]. However, initial concerns about the clinical utility of ISMN, administered alone, appeared when the follow-up was extended to 7 years and a significant increase in mortality was observed in patients over 50 years of age receiving Journal of Hepatology 37 (2002) 393–395