Review article: the modern management of hepatic encephalopathy


  • J. S. BAJAJ

    1. Division of Gastroenterology, Hepatology and Nutrition, Virginia Commonwealth University and McGuire VA Medical Center, Richmond, VA, USA
    Search for more papers by this author

  • This commissioned article was subject to full peer-review.

Dr J. S. Bajaj, Division of Gastroenterology, Hepatology and Nutrition, Virginia Commonwealth University and McGuire VA Medical Center, 1201 Broad Rock Boulevard, Richmond, VA 23249, USA.


Aliment Pharmacol Ther31, 537–547


Background  Hepatic encephalopathy, both overt and minimal, forms a continuum of cognitive change in cirrhosis. Strategies to diagnose and treat hepatic encephalopathy have evolved considerably.

Aim To examine the updated diagnostic and treatment strategies for hepatic encephalopathy.

Methods  Techniques for the clinical, psychometric and neurophysiological evaluation of hepatic encephalopathy are reviewed. The methods reviewed include pure clinical scales (West-Haven), psychometric tests (PSE-syndrome test), neurophysiological tests (EEG, Critical flicker frequency, CFF) and computerized tests (Inhibitory control test, ICT).

Results  Clinical scales are limited, whereas psychometric tests (specifically PSE-syndrome test), CFF and ICT can be used to diagnose minimal hepatic encephalopathy. However, there is no single test that can capture the entire spectrum of cognitive impairment. Treatment options and goals depend on the acuity of hepatic encephalopathy. In-patient management should concentrate on supportive care, precipitating factor reversal and lactulose and/or rifaximin therapy. Out-patient therapy should aim to prevent recurrences, and both lactulose and rifaximin have evidence to support their use.

Conclusions  Diagnostic techniques for hepatic encephalopathy range from simple scales to sophisticated tools. Treatment options depend on the stage of hepatic encephalopathy. The future challenge is to evaluate cognitive function as a continuum with clinically relevant outcomes and to develop well-tolerated and inexpensive treatments for hepatic encephalopathy.


Hepatic encephalopathy reflects a spectrum of neuro-psychiatric abnormalities seen in patients with liver dysfunction, after exclusion of other known brain disease.1 HE is divided into two primary components: overt HE (OHE) and minimal HE (MHE).1 OHE can be diagnosed clinically through a constellation of signs and symptoms, whereas MHE requires specialized testing. It has been estimated that OHE is present in 30–45% of patients, with an annual risk of development in 20% of patient with cirrhosis. MHE is manifested by impairment in specialized testing and is considered by most of the clinicians to be a preclinical stage of OHE.2 Approximately 60–80% of patients with cirrhosis tested have evidence of cognitive dysfunction or MHE.3 This condition is associated with increased progression to OHE, poor quality of life and a high risk of traffic violations and accidents.4

The burden of OHE and MHE is immense considering their wide-ranging effects on the patients, family and society. A recent report showed that although there is a reduction in hospital stay for patients with OHE, the costs are likely to increase over the coming years.2 Considering that these syndromes affect the overall survival, quality of life, ability to work and drive and are associated with traffic accidents, the study and diagnosis of these conditions are of utmost concern to practicing gastroenterologists and hepatologists. A review of diagnosis and treatment of HE from 2000 to 2009, including book chapters, reviews and abstracts, was conducted with the terms ‘hepatic encephalopathy’, ‘minimal hepatic encephalopathy’ and ‘subclinical hepatic encephalopathy’, and key references outside this date range were also included.

Updated nomenclature of HE types

The Working Group on Hepatic Encephalopathy recommended updating the nomenclature of HE according to clinical presentation and aetiology.1 These definitions were formulated with the aim to simplify the diagnosis of HE clinically and also to include the HE associated with congenital shunts, urea-cycle disorders and acute liver failure into the entire spectrum of HE as encountered by the clinician. The group recommended dividing HE into three broad categories: A (acute liver failure), B (porto-systemic bypass without intrinsic liver disease) and C (cirrhosis). This review will be restricted to HE associated with cirrhosis. The nomenclature is shown in Table 1, and a graphical representation of the time course of HE type C or in cirrhosis is shown in the figure. It is clinically divided into episodic or persistent HE depending on their chronicity and clinical behaviour, as shown in Figure 1: Episodic HE remains clinically undetectable in between HE episodes, persistent HE patients never become free of HE and patients with MHE remain below the clinical detection level.5

Table 1.   Classification of the subtypes
AEncephalopathy associated with acute liver failure
BEncephalopathy with porto-systemic bypass and no intrinsic hepatocellular disease
CEncephalopathy associated with cirrhosis or portal hypertension/porto-systemic shuntsEpisodic HE• Precipitated
• Spontaneous
• Recurrent
Persistent HE• Mild
• Severe
• Treatment dependent
Figure 1.

 Temporal and clinical detection relationship of HE subtypes. Hepatic encephalopathy is clinically divided into episodic or persistent HE depending on their chronicity and clinical behaviour, as shown in Figure; Episodic HE remains clinically undetectable in between HE episodes, persistent HE patients never become free of HE and patients with MHE remain below the clinical detection level. HE, hepatic encephalopathy; MHE, minimal hepatic encephalopathy.

Clinical diagnostic criteria for HE

There are several systems that have been used to diagnose the severity of HE clinically.6–9 However, most clinical systems are mired in subjectivity until the end stage of coma is reached and are not reproducible among examiners.10 This is because of the lack of objective criteria for the clinical diagnosis of HE. The most widely used are the West-Haven criteria (Table 2); the most reproducible stages are stages 0, i.e. no abnormality, and stages 3 and 4, which are coma.8 There is a large area of uncertainty in between.

Table 2.   West-Haven criteria for HE
StageConsciousnessIntellect and behaviourNeurological findings
0NormalNormalNormal examination; if impaired psychomotor testing, then MHE
1Mild lack of awarenessShortened attention span; impaired addition or subtractionMild asterixis or tremor
2LethargicDisoriented; inappropriate behaviourObvious asterixis; slurred speech
3Somnolent but arousableGross disorientation; bizarre behaviourMuscular rigidity and clonus; Hyper-reflexia
4ComaComaDecerebrate posturing

Only low-stage HE, i.e. pre-coma stages 1 and 2 in the West-Haven criteria, and MHE are the components of HE that require a detailed neuropsychological evaluation. The rest can be diagnosed clinically. However, there is no current system that can diagnose a patient with the specific stages of HE objectively through the entire spectrum of normal, minimal and overt HE.11

There are several important points to be considered while making a diagnosis of HE clinically.

  • (i) Does the patient have altered mental status as defined by disorientation?
  • (ii) Does the patient have any other aetiology other than HE that can explain the alteration in mental status?

There are at least two levels of cognitive function in patients with HE that need to be specifically investigated.

  • (i) Deterioration in psychometric or neurophysiological function
  • (ii) Deterioration in mental status

The existence of these two levels of functioning is important to acknowledge, as only mental status changes can be diagnosed by clinical interview and examination. Mental status testing can be achieved by evaluation of alertness and orientation using clinical judgement. Specific questionnaires such as clinical hepatic encephalopathy staging scale (CHESS) have been used in a small study but require further analysis.12 The Hepatic Encephalopathy Scoring Algorithm (HESA), a blended scale of psychometric and clinical assessments, has also been used for the evaluation of advanced HE.10, 13 Cognitive problems in patients with normal mental status, which constitute MHE, are difficult to diagnose in clinic unless special efforts are made.

Approach to a patient with cirrhosis who has alteration in mental status

Although HE is present in most cirrhotic patients with altered mental status, it is definitely not the only reason for change in mentation in these patients.5 There are several other differential diagnoses for the development of cognitive dysfunction in cirrhotics, especially intra-cranial events, electrolyte abnormalities and sepsis.8 Therefore, the overall susceptibility of the brain towards alteration of higher mental function is present in cirrhosis, and HE should only be diagnosed after the exclusion of other potential causes.

Physical Examination: A detailed evaluation of the vitals and airway should be performed at the outset and these should be managed accordingly. Once these pressing issues have been managed, it is then important to perform a detailed neurological examination.1 The physical examination should also concentrate on placing the patient in the West-Haven Criteria.8

Motor examination: In most cases, the presence of a previously unknown focal motor deficit is not typical of OHE, which tends to be a global rather than a focal process. If a new focal deficit is found, alternatives to HE such as a subdural haematoma or ischaemic processes should be considered.

Patients with OHE can have hyper-reflexia, positive Babinski’s sign, and in stages 2 and 3, have asterixis.14 Asterixis is defined as a flapping tremor caused by the disturbance in the oscillatory networks in the brain.15 It can be demonstrated in the tongue, and the upper and lower extremities. Patients who are too obtunded to raise their hands up, ‘as they are stopping traffic’, should be instructed to grip the examiner’s hands. The grip in patients with asterixis is never constant and oscillates between tight and loose.

Care should be taken not to confuse asterixis with tremulousness associated with alcohol abuse or withdrawal. Asterixis is not specific for HE and can also been seen in carbon dioxide intoxication and uraemia.

Motor examination in HE patients can also demonstrate Parkinsonian symptoms with the attendant rigidity and tremors. In a small subgroup, spinal cord involvement with spastic paraparesis, resulting from hepatic myelopathy, can also occur, but this syndrome is distinct from HE.16

If the above signs are evident, then the diagnosis of HE is fairly secure clinically and there is no need for further expensive imaging, and psychometric or neurophysiological testing. The further medical management is discussed in the treatment section.

Approach to a patient with cirrhosis with normal mental status

Patients with this problem are the focus of several recent investigations because the diagnostic methods used for cognitive evaluation in this population are varied and there is no consensus regarding the optimum strategy. The study of the cognitive deficits as a continuum, the spectrum of neuro-cognitive impairment in cirrhosis (SONIC), may be necessary to avoid these pitfalls in classifying patients into normal, minimal and overt HE.17

Strategies to divide patients with normal mental status into normal and minimal HE are predominantly psychometric or neurophysiological.

Minimal Hepatic encephalopathy has a specific deficit profile on psychometric testing and the major abnormalities in HE consist of attention deficits.18–20 The attentional hierarchy, as described by Posner, is impaired at all levels of vigilance, orienting and executive functions.18–21 Attention deficits also results in learning impairment and difficulty in working memory. There is also a defect in visuo-motor coordination and construction ability and in speed of mental processing. Underlying most of these deficits is the impairment of response inhibition.4

Neuropsychometric testing

Several batteries for the diagnosis of HE have been studied, all of which are based on detecting attention deficits and processing speed.12

The PSE-syndrome test used by Weissenborn et al. has been validated for MHE diagnosis in Germany, Italy and Spain.19 Five tests, number connection test-A (NCT-A), number connection test-B (NCT-B), line tracing test errors and time, serial dotting test and digit symbol test (DST), comprise the PSE-syndrome test. Test results within the ±1 s.d. range is scored 0, between 1 and 2 s.d. −1 point, between 2 and 3 s.d. −2 points and beyond 3 s.d. −3. Results better than mean plus 1 s.d. are given 1 point; therefore, the score ranges from +6 to −18 points. The cut-off value between normal and pathological results was found to be −4 points, which resulted in a sensitivity of 96% and specificity of 100%. This test battery has been validated in Spain and Italy, and there are population norms available (; whereas there are some regional variations in the subscores for line tracing test.22, 23 The Working Group on Hepatic Encephalopathy has recommended this test; however, validation of this in the US has not been performed till date.1

The Working Group also recommended that if the PSE-syndrome test is not available, a combination of two of the following four tests; NCT-A, NCT-B, DST or BDT, could be used. The convention is impairment in at least two of these tests two standard deviations beyond age and education-matched healthy controls indicates cognitive impairment.1

A recent consensus statement of the International Society for the Study of Hepatic Encephalopathy and Nitrogen Metabolism (ISHEN) recommended the use of repeated battery for the assessment of neuropsychological status (RBANS) or PSE-syndrome test.8 RBANS has been used for the evaluation of Alzheimer’s disease, schizophrenia, traumatic brain injury and in a selected population of patients with cirrhosis awaiting liver transplantation. However, it has not been specifically validated in HE.24 RBANS is a copyrighted set of tests consisting of five domains. It is divided into cortical and sub-cortical domains as well and as expected, HE patients predominantly perform worse in the sub-cortical component.8

Neurophysiological testing

Another type of tests used in HE is neurophysiological tests, which are offered under the supervision of a neurologist and require specialized equipment, personnel and time. The lack of learning effects and the relative specificity of the response are the relative advantages of neurophysiological tests.1 These range from a simple electroencephalogram (EEG) to sophisticated techniques of automated evoked potentials. Spectral EEG, mean dominant frequency and peak power frequency of EEG have been studied but have demonstrated subjectivity. The need for expensive equipment, low sensitivity and the lack of accompanying behavioural information are drawbacks of these methods.25‘Triphasic waves’ in advanced HE are noted on EEG.26 In the earlier stages, the mean dominant frequency and spectral EEG analysis can predict the OHE. However, EEG studies only cortical activity, which reduces its concordance with batteries such as PSE-syndrome test that rely on cortical and sub-cortical components.23 Evoked potentials are the latency between application of a stimulus and the brain’s ability to sense it, which can be visual, somatosensory and auditory. Visual evoked potentials are insensitive to change in HE and require active patient cooperation, making them useful only for the detection of early HE stages.26 Abnormalities in somatosensory evoked potential were seen in 48% with abnormal inter-peak latencies N20-N65, but there was no correlation with psychometric tests in later studies. Brain electrical responses to visual or auditory stimulus are also important means of gauging HE. An auditory P300 is when a patient receives an infrequent stimulus embedded in other frequent stimuli (‘oddball’ paradigm). The brain shows a response typically 300 milliseconds after the oddball stimulus; delay in brain response (i.e. after a 300-millisecond gap) signifies dysfunction.11 Auditory P300 does have good diagnostic potential and can be used when available for evaluation of neurophysiological function in cirrhosis. Visual P300 responses were found to be abnormal in 78% compared with psychometric abnormalities in only 41% of patients, but were related to decline in hepatic metabolic capacity; these are not recommended because of their inconsistent results.27

Limitations of currently available psychometric and neurophysiological tests

These psychometric and neurophysiological tests, although available, are copyrighted and require a psychologist for administration and interpretation. Hence these tests become difficult to apply in a regular clinic setting, although they are ideal for research settings. Therefore recent studies have been performed for tests that can be applied in the clinical settings by personnel other than psychologists. These tests are the critical flicker frequency (neurophysiological test) and two computerized psychometric test systems, the Cognitive Drug Research (CDR) and the inhibitory control tests.

Tests applicable in the clinical setting that do not require psychological expertise

Critical flicker frequency (CFF) is a test of the functional efficacy of the cortex, which has a direct correlation with psychometric abnormalities.15 In this test, the patient indicates the minimum frequency at which a flickering light is still perceived as flickering and not fused, using a portable device. Stepwise increase of the light frequency from 25 Hz allows the determination of the fusion-frequency threshold, which is the frequency at which the flickering light is perceived by the patient as fused light. Conversely, a stepwise decrease in the frequency of the light pulse from 60 Hz downwards is also performed and the CFF threshold is determined as the frequency when the impression of fused light switched to a flickering one. Studies have shown that although a CFF threshold of 38–39 Hz could differentiate between manifest HE (i.e. early stages of OHE) and no HE, it was less sensitive in differentiating MHE from manifest HE. This test has been tested in Spain and India, with good results.4, 28 Encouragingly, it can be performed by clinic personnel without the need for a psychologist within a short period of time and, apart from the equipment, has minimal costs. However, CFF has not been validated for the US population.

The CDR consists of five psychometric subsets that test attention power, attention continuity, speed of memory and quality of episodic and working memory. This battery has been developed by Cognitive Drug Research (CDR) Ltd (Goring-on-Thames, United Kingdom).29 These tests have 50 parallel forms and have population norms for the UK. A recent study compared the CDR with the PSE-syndrome test and showed improvement after liver transplantation and worsening after a nitrogen challenge. In this study, MHE patients were impaired in all subsets and there was worsening of the quality of working and episodic memory after a nitrogen challenge. CDR will be available from the UK at an assessment cost of 30 pounds sterling.29

Inhibitory control test (ICT) is a computerized variant of the continuous performance test that assesses sustained attention and response inhibition.30 It has been used in the description of traumatic brain injury, schizophrenia and attention deficit disorder.4 ICT consists of 1728 stimuli, 40 lures and 212 targets that are presented within 13 min after a training run. A higher lure and lower target rate represent worse psychometric performance. ICT has been validated in the US population with 88% sensitivity for the diagnosis of MHE when a patient had >5 lures using a standard psychometric battery as the gold standard.31 ICT also predicted the development of OHE and changed appropriately with the clinical state of the patients, i.e. improved after therapy and worsened after shunting procedures. ICT performance was also significantly associated with driving simulator performance and traffic accidents in the US.4, 11 Clinic personnel were able to administer ICT with minimal training and it was found to be cheaper than psychometric test administration.4 It may be appropriate for MHE testing at the clinic level in a US population and is currently available for free download at

Challenges for the diagnostic testing of MHE in the United States

The diagnosis of HE using psychometric tests is difficult in the US because of a dearth of tests with norms that have been used in HE. The DST is the only component of the PHES that has norms for the US, whereas the ICT has norms for two states at this time.32 RBANS has only age-based, and not age- and education-based, norms for the US, and has not been validated for HE systematically. In the US, certain psychometric batteries, such as those containing DST and RBANS, require a psychologist for procurement and administration, adding another barrier to the routine use of these tests in the clinic. In addition to all these travails, the US has a high proportion of car drivers who can potentially be affected by minimal HE and these diagnostic methods need to be adapted to the US population before interventions can be planned.

The next section will review the current therapeutic strategies for hepatic encephalopathy

Treatment options and practical issues in the management of hepatic encephalopathy

Treatment goals and options are dependent on the stage and acuity of HE (Table 3). Several studies have been performed in MHE using lactulose, probiotics and other agents.4, 33–36 Although these studies demonstrate improvement in the psychometric and neurophysiological variables, these studies are short term in duration. In addition, apart from Prasad et al. who showed improved quality of life and a minor decrease in OHE development in a smaller study, clinically relevant outcomes in MHE have not been studied to date.4, 35 A large, multi-centre MHE trial with clinically relevant outcomes is needed. Until such a trial is available, there is no current standard of care for this stage of MHE. Unlike MHE, the management of OHE is better defined and includes both the therapy of the OHE episode and the prevention of recurrences.

Table 3.   Goals for the treatment of overt hepatic encephalopathy
Acute episode of HE
 Treatment of precipitating factors
 Improvement in mental status
 Evaluation for liver transplant
Out-patient management after an episode of HE
 Prevention of recurrent episodes of HE
 Improvement of daily functioning
 Evaluation for liver transplant

Management of the acute OHE episode

The overall management consists of properly identifying OHE, gauging its severity, treating potential precipitating factors and using treatments specifically directed towards OHE (Figure 2). The differential diagnosis of OHE has already been discussed in the clinical diagnosis section.

Figure 2.

 Proposed Algorithm for in-patient HE Management*. Simple algorithm for in-patient management of HE with focus on differential diagnosis, investigation of precipitating factors and admission and triage strategies. *There is no standard of care for MHE therapy at this time. HE, hepatic encephalopathy; ICU, intensive care unit.

Precipitating factors

Once the diagnosis of OHE is confirmed, an extensive search for potential precipitating factors should be instituted along with treatment of OHE. Precipitants work by increasing the underlying inflammatory milieu, increasing ammonia production or reducing the threshold for mental status decline, or a combination of the above.

The leading causes are gastrointestinal bleeding, sepsis and dehydration resulting from diuretics, diarrhoea or vomiting. Also important are hyponatraemia, surgical intervention, transjugular intra-hepatic porto-systemic shunting (TIPS), constipation and the use of sedative and narcotic drugs. Another important cause in patients with prior OHE is non-adherence to medications either because of adverse effects or other reasons.8 A careful exclusion of these factors is essential as the OHE is probably a marker of these underlying issues and their treatment can reverse OHE in most cases. In cases where no specific precipitating factor is identified despite an exhaustive search, treatment specific for OHE should be instituted.8

Clinical approach to management of a patient admitted with OHE

There are a series of questions that need to be answered when a clinician is presented with a patient with possible OHE:

  • (i) Does the patient really have OHE?
  • (ii) If yes, is there a precipitating factor?
  • (iii) Should the patient be admitted to the hospital?
  • (iv) If so, should the patient go to the floor or the ICU?
  • (v) When should the patient be intubated?
  • (vi) What route should be used to administer lactulose?
  • (vii) How often should ammonia levels be drawn, if at all?

After confirmation of the diagnosis of OHE and identification of precipitating factors, patients with OHE should be assessed for consciousness, orientation and ability to protect the airway. The inability to protect airways is the main reason why patients who have stage 3 or higher OHE should be transferred to the intensive care or a step-down unit. Patients with stage 2, i.e. disorientation, should be admitted to the hospital in most cases to evaluate potentially serious precipitating factors. There should be a low threshold for airway intubation in patients in the intensive care unit, especially those with concurrent GI bleeding, to prevent aspiration. A complete infectious work-up, including diagnostic paracentesis, pan-culturing of urine and blood, chest X-ray and skin examination for possible cellulitis, should be carried out. Patients with diarrhoea could also benefit from stool toxin analysis for Clostridium difficile. Unless there is evidence of GI bleeding or sepsis, there is currently no role for prophylactic antibiotics in patients admitted with OHE. Metabolic abnormalities should be corrected as noticed and hydration should be gentle, with care not to cause pulmonary congestion.

Ammonia levels do not provide any additional information beyond the observation of change of mental status in OHE in the individual patient and, therefore, may not predict or correlate with actual clinical outcomes.8, 37

An analysis of consciousness and orientation should be performed at least 2–3 times per day, especially in patients admitted to the floor.

Specific treatment interventions in HE

Therapeutic options for HE are directed towards the reduction of ammonia production, increase in fixation and/or excretion of ammonia, direct neurological action and modifying porto-systemic collaterals.8, 37, 38 The direct neurological therapies such as bromocriptine and flumazenil have fallen out of favour because of their potential for decreasing seizure threshold and efficacy concerns. The ammonia fixation therapies such as l-ornithine-l-aspartate and phenyl-butyrate have limitations regarding their availability in the US and inconsistent results in HE studies.39 Modifications of porto-systemic collaterals are only necessary in cases of persistent HE in which patients do not respond to adequate therapy, or in recalcitrant OHE after TIPS.40

Therefore, the majority of therapeutic options currently in use are directed towards the gut. There are several therapies that have been used in research studies but lactulose and rifaximin form the bulk of the current therapeutic options in HE.

Lactulose is a non-absorbable disaccharide that is fermented in the colon. The exact mechanism of action remains unclear; however, acidification of colonic contents and mass evacuation of bacteria have been proposed.8 It has been used for several decades, with anecdotal and clinical trial experience. As the use of lactulose pre-dated randomized controlled trials, the Cochrane review did not find any significant difference in outcomes in patients treated with and without lactulose.41 However, the large sample sizes needed to treat HE have not been achieved in prior lactulose trials; therefore, it is not entirely accurate to dismiss the use of lactulose. The administration of lactulose while the patient is admitted with HE is associated with improvement in mental status; but as precipitating factors are simultaneously being corrected, it is difficult to pinpoint the correct reason for improvement.42

Lactulose should be given in enema form in patients with stage 3 or higher OHE, and orally if the patient is able to tolerate it through that route. Lactulose enemas are mixed with water and specifically aid in patients with poor stool output. Adverse effects of lactulose range from minor gastroenterological issues with nausea, vomiting and diarrhoea to severe dehydration, hypernatraemia and ileus.43 The use of lactulose, therefore, should be tempered with education of patients and their family members regarding the possible adverse effects. However, there is a lack of scientific investigation into the mechanism of action of and the optimum dosage of lactulose, both in the in-patient and the out-patient setting.14

Rifaximin is a non-absorbable antibiotic that has been used to treat HE in several European countries.44, 45 It has a favourable impact and the Cochrane review recommends the use of non-absorbable antibiotics.41, 44, 46 It is currently available in 200 mg form, which is given up to 1200 mg/day. Use of rifaximin is slowly becoming mainstream and is well tolerated. The drug expense remains a concern, although a recent study noted the reduced hospitalization rates after rifaximin therapy compared with that of lactulose.47

There is no significant role of neomycin, flumazenil, metronidazole and zinc as stand-alone therapies for HE. There are several other drugs in the pipeline for HE that are undergoing trials in the US.39 Other drugs that have been used outside the US are l-ornithine-l-aspartate infusion and oral forms, acetylcarnitine and acarbose.39

Out-patient treatment after an episode of HE

The goals for therapy in this situation are very different from those in an acute HE episode (Table 3). Prevention of recurrent episodes is key to normalization of daily functioning. A recent study demonstrated that lactulose therapy significantly reduced recurrent HE episodes after the first one compared with placebo in a randomized trial. This trial showed good adherence with lactulose, although this adherence may be hard to achieve outside clinical trials.48 A recent trial showed that rifaximin 550 mg BID along with lactulose was significantly more effective than lactulose alone in the prevention of HE episodes in patients who had had two or more HE episodes in the past 6 months. The safety profile was good in this large trial.49

Therefore at this time, based on the current prescribing guidelines, both lactulose and rifaximin can be used to prevent recurrent episodes of OHE.

Management of persistent OHE

The role of porto-systemic shunting in patients whose mental status does not improve even after maximal therapy with first- and second-line agents is important to investigate. A study by Riggio et al. found that 71% of patients with persistent OHE had patent, large porto-systemic shunts compared to only 14% without persistent OHE.40 This can improve after embolization or coiling using interventional radiological expertise. A minority of patients may develop persistent, disabling HE after TIPS, which does not respond to medical therapy. This may also require radiological interventions, such as ballooning, to occlude the TIPS shunt.38 Other causes of persistence may be an undiscovered source of sepsis, i.e. abscesses, or an inability to tolerate medications prescribed for OHE.

Liver transplant

The ultimate management goal for OHE is the replacement of the diseased liver. This is especially important to recognize in the MELD era when OHE is not given enough significance.50 Stewart et al. showed that OHE worsens prognosis over and above MELD score and results in worsened survival depending on the initial stage on admission.50 Therefore, liver transplant work-up is crucial for the management of OHE after correction of the acute insult and prevention of recurrences.

Summary and challenges for the future

Clinical scales for the diagnosis of HE are fraught with subjectivity until there is evidence of disorientation. Mental status changes and neuropsychological changes are two different planes that need to be tested before deciding what strategy is needed to diagnose HE. Mental status changes beyond disorientation (stage 2 or higher) have good inter-rater reliability; however, the simple clinical assessment of patients with normal mental status or cognitive complaints without disorientation is not objective or adequate. There are several psychometric and neurophysiological tests that can be used to study these patients, which should be performed according to locally available expertise, norms and population.

Therapeutic challenges for OHE depend on the acuity and severity of the clinical condition. At this time, therapy for MHE needs to await long-term trials that focus on clinically relevant outcomes. The overall management consists of properly identifying OHE, gauging its severity, treating potential precipitating factors and using treatments specifically directed towards OHE. Out-patient management goals for OHE are prevention of recurrence and improvement in daily functioning. At this time, lactulose and rifaximin have emerged as leading therapeutic options for in-patient and out-patient OHE therapy in the US. Liver transplant work-up in these individuals needs to be initiated to ensure a lasting improvement in mental status.

The challenge for the future is to evaluate cognitive function as a continuum with clinically important outcomes to increase their relevance and to also develop methods that can be performed in the clinic. Therapeutic strategies for MHE need to be developed in the context of clinically relevant outcomes and those for OHE need to be refined to fit the individual patient while minimizing cost and adherence issues.


Declaration of personal interests: Jasmohan Bajaj has served as a consultant and an advisory board member and has received research funding from Salix and Ocera Pharmaceuticals. Declaration of funding interests: No specific funding was involved in this study.