Source: Gluud LL, Dam G, Les I, et al. Branched-chain amino acids for people with hepatic encephalopathy. Cochrane Database Syst Rev. 2017;5:CD001939. doi: 10.1002/14651858.CD001939.pub4
Study Population: Adults with hepatic encephalopathy and cirrhosis, mostly due to alcoholic liver disease or viral hepatitis
Efficacy Endpoints: Mortality, Hepatic encephalopathy
Harm Endpoints: Gastrointestinal symptoms, including nausea and diarrhea; and any other reported adverse effects
Narrative: Cirrhosis, where scar tissue replaces normal hepatic tissue, is the most common cause of hepatic encephalopathy,1 a brain dysfunction that can be mild with minimal confusion, or overt and severe with coma.1, 2, 3 Branched-chain amino acids (BCAAs) may reduce hepatic encephalopathy by helping skeletal muscle detoxify blood4, 5, 6 and are therefore recommended by some as routine treatment.2, 3
The systematic review summarized here identified 16 randomized trials of 827 subjects (97% with cirrhosis) with hepatic encephalopathy.1 Primary outcomes were mortality, hepatic encephalopathy (number with improvement), and harms. Subjects were followed for varying time periods ranging from 1 to 104 weeks.
BCAAs had benefits in improving signs and symptoms of hepatic encephalopathy assessed mainly by West haven criteria (RR 0.7, 95% CI 0.6 to 0.9 ARD: 17%, NNT: 6): for every 6 subjects treated, one measurably improved. However, no effect was found on mortality (RR 0.9, 95 % CI 0.7 to 1.1). BCAAs also increased nausea and diarrhea (RR 3.4, 95% CI,0.7 to 16.5, ARD: 5%, NNH: 20); though, no serious adverse events were reported.
The West-Haven criteria classifies the degree of mental status disturbance in encephalopathy by 4-point scoring system ranging from reversal of sleep patterns and mild alteration in cognition to deep coma.2
The portal-systemic encephalopathy (PSE) index may also objectively describe the overall clinical severity of HE.7 It is calculated following assessment of five elements; Mental status (Evaluated by West-Haven criteria), presence and intensity of asterixis, time taken to complete psychometric tests of intellectual function (such as number connection test), venous ammonia level and electroencephalogram (EEG) abnormalities.7
In this metanalysis, Six studies (Cerra 1985, Hwang 1988, Muto 2005, Strauss 1986, Vilstrup 1990, Rossi-Fanelli 1986) assessed improvement of hepatic encephalopathy strictly only by West haven scoring only.1 However, the other 10 studies (Fiaccadori 1984, Horst 1984, Michel 1985, Egberts 1985, Calvey 1985, Marchesini 1990, Hayashi 1991, Plauth 1993, Les 2011, Marchesini 2003) not only included West haven criteria to evaluate degree of encephalopathy, but they also included some or all components of PSE Index.1
Caveats: The authors of the Cochrane review graded the evidence as high quality for hepatic encephalopathy. However, there was moderate heterogeneity of their results, which raises concerns about validity and applicability. In addition, most trials were non-blinded, small, and judged to be at “high risk of bias.” The ‘high quality’ grade here seems highly debatable; and a large, well-done clinical trial may easily upend these findings.
The review also pooled trials using different forms of BCAA.1 Nine assessed oral and seven assessed intravenous administration, with only oral showing a statistical benefit. In addition, assessing hepatic encephalopathy is fraught with subjectivity and disagreement. For instance, many studies in this review used the PSE index, which the Food and Drug Administration has rejected as inadequate.8 This is mainly due to the inclusion of blood ammonia levels and severity of asterixis.8 The utility of ammonia level is controversial given that ammonia concentration is not useful for screening for hepatic encephalopathy since their levels vary if they are arterial or venous.9, 10 In addition too, due to the fact that these levels are significantly affected by collection techniques and can be falsely elevated if the sample was collected after fist clenching, using tourniquet, or if the sample was not placed on ice.10
Also, asterixis is not specific to HE as it can also be observed in patients with other forms of metabolic encephalopathies such as in uremia and respiratory failure.10
Without blinded assessors, a method not used in most trials here, and a validated scale, it is difficult to have confidence in these findings. This is particularly true when the only objective outcome, mortality, showed no difference between groups.
The yellow color recommendation (unclear benefits) is based on the inconclusive data supporting benefits of BCAA for hepatic encephalopathy.
Author: Ahmed Hamed MD; Amira Hamed MD; Karissa Lambert MD
Supervising Editors: Michael Ritchie, MD; James McCormack, MD
Published/Updated: July 1, 2019
The title bar is color-coded with our overall recommendation.