Can we ignore minimal hepatic encephalopathy any longer?


  • Potential conflict of interest: Nothing to report

  • See Article on Page 549

The interesting article by Prasad and colleagues in this issue of HEPATOLOGY may potentially change the overall management of hepatic encephalopathy (HE).1 The authors confirm a significant reduction in many of the domains related to quality of life (QOL) in patients with cirrhosis who meet diagnostic criteria for minimal HE (MHE). Whereas the reduction in QOL had been reported,2 it was argued that this reflects a nonspecific effect of advancing liver disease.3 However, this concept cannot be defended in light of the significant improvement in most of the measures of QOL with 3 months of lactulose therapy in the study by Prasad et al. Previous reports had already shown improvement in performance in various psychomotor tests used to diagnose MHE with lactulose treatment.4, 5 The study by Prasad et al. not only confirmed that lactulose improves MHE, but that this improvement is associated with a significant improvement in health-related QOL. Indeed, the results of this study strongly suggest MHE is the cause of reduced QOL in patients with cirrhosis.

The diagnostic criteria for MHE were formally standardized in 2002.6 Patients require a normal mental status exam and a significant reduction (>2 standard deviations from appropriate normative population) in performance of at least 2 of the battery of psychometric tests for a diagnosis of MHE. Details on what constitutes an adequate neurological evaluation have been published.7 In the study by Prasad et al., more than 60% of the patients with cirrhosis chosen for the study had MHE. These high percentages of patients with MHE have been noted in other studies.8, 9 However, one wonders if the prevalence of MHE is inflated in some studies because of the use of more or different psychometric tests10 compared to the original psychometric tests used to characterize MHE.11 Another important factor in determining the prevalence of MHE may be the selection of patients. A major factor increasing the probability of finding MHE is a previous episode of overt HE.12 However, patients with a history of overt HE were specifically excluded in this study. As we observe this high prevalence of MHE in patients with advanced liver disease, the question arises as to whether or not all patients with cirrhosis may eventually develop this problem. If a natural history study confirms this, may we eventually abandon screening for MHE and just treat all patients from the time they are diagnosed with cirrhosis? Indeed once a patient with cirrhosis learns that their QOL can be improved by HE treatment, we may no longer have a choice in the matter. However, the major objection to this approach would be the annoying side effects of lactulose. Perhaps treatments that are better tolerated, such as antibiotics and/or probiotic diets, may change this perspective.

There was no placebo treatment arm in this study, and the physicians were not blinded to who was on lactulose. Lack of blinding in HE studies is usually considered unacceptable because of the highly subjective nature of assessment of overt HE. The authors, however, are correct in stating that the psychometric test performances were unlikely to be affected by the patients potentially knowing that they were on active treatments. However, the QOL survey might have been compromised by the knowledge that active treatment was being employed. When the authors initiate their larger trial of lactulose therapy of MHE (as mentioned in the discussion section), they should include a placebo arm in order to blind all participants in the study. The difficulty of blinding with a treatment that causes diarrhea can be overcome by instituting a probiotic diet9 or an antibiotic like rifaxamin.14 Some interesting issues that could be addressed in the proposed larger follow-up study are self-evident, such as if HE treatment in this early phase of HE postpone overt HE for longer periods of time or if it will affect survival. As with any good study, the results seem to provoke more questions than answers.

Als-Nielsen and colleagues, in their systematic review, categorically stated that the evidence in favor of using nonabsorbable disaccharides for the treatment of HE was unconvincing and did not meet current minimum criteria for adequacy in this evidence-based era. They strongly suggested that lactulose should not be considered the standard of care as treatment or be the mandatory comparator in randomized controlled trials of HE treatment.13 However, this is not to say that lactulose doesn't have its place in the treatment of HE. The study by Prasad et al. supports the ability of lactulose to reverse milder HE. One definite result of this study will be a return of faith in lactulose as a viable treatment of HE. As the authors state, the failure of so many previous studies to prove that lactulose was an effective therapy for HE may have been due to the presence of many confounding variables when patients with advanced liver disease and severe HE are studied.14

As potentially exciting as these results may be, it is important to remember there are significant barriers to widespread diagnosis of MHE in the clinic. A recent survey of AASLD members showed that the majority of respondents already felt MHE was a significant medical problem and should be treated.15 However, the additional time and overall difficulty in attempting to diagnose MHE in a clinical setting was considered a major hurdle in more than 75% of those surveyed. Clearly, something has to be done to simplify the diagnosis of MHE now that we have evidence that treating MHE improves most measures of QOL. We await development of a simple neurophysiological HE detection device validated against a battery of psychomotor tests. Alternatively, we may discover that all patients with cirrhosis will eventually develop MHE, and therefore treatment of MHE will become the standard of care without the need to provide evidence that MHE is present. In any event, it appears that we cannot ignore MHE any longer.