EASL Clinical Practice Guidelines


General principles

At this time, only OHE is routinely treated [10]. Minimal hepatic encephalopathy and CHE, as its title implies, is not obvious on routine clinical examination and is predominantly diagnosed by techniques outlined in the previous section. Despite its subtle nature, MHE and CHE can have a significant effect on a patient's daily living. Special circumstances can prevail where there may be an indication to treat such a patient (e.g., impairment in driving skills, work performance, quality of life, or cognitive complaints). Liver transplantation is mentioned under the treatment recommendations.


Comments on management strategy

Patients with higher grades of HE who are at risk or unable to protect their airway need more intensive monitoring and are ideally managed in an intensive care setting. Alternative causes of encephalopathy are not infrequent in patients with advanced cirrhosis. Technically, if other causes of encephalopathy are present, then the episode of encephalopathy may not be termed HE. In the clinical setting, what transpires is treatment of both HE and non-HE.

Controlling precipitating factors in the management of OHE is of paramount importance, because nearly 90% of patients can be treated with just correction of the precipitating factor [89]. Careful attention to this issue is still the cornerstone of HE management.

Therapy for episodes of OHE

In addition to the other elements of the four-pronged approach to treatment of HE, specific drug treatment is part of the management. Most drugs have not been tested by rigorous randomized, controlled studies and are utilized based on circumstantial observations. These agents include nonabsorbable disaccharides, such as lactulose, and antibiotics, such as rifaximin. Other therapies, such as oral branched-chain amino acids (BCAAs), intravenous (IV) L-ornithine L-aspartate (LOLA), probiotics, and other antibiotics, have also been used. In the hospital, a nasogastric tube can be used to administer oral therapies in patients who are unable to swallow or have an aspiration risk.

Nonabsorbable disaccharides

Lactulose is generally used as initial treatment for OHE. A large meta-analysis of trial data did not completely support lactulose as a therapeutic agent for treatment of OHE, but for technical reasons, it did not include the largest trials, and these agents continue to be used widely [90]. Lack of effect of lactulose should prompt a clinical search for unrecognized precipitating factors and competing causes for the brain impairment. Though it is assumed that the prebiotic effects (the drug being a nondigestible substance that promotes the growth of beneficial microorganisms in the intestines) and acidifying nature of lactulose have an additional benefit beyond the laxative effect, culture-independent studies have not borne those out [[75], [91]]. In addition, most recent trials on lactulose have been open label in nature. Cost considerations alone add to the argument in support of lactulose [92]. In some centers, lactitol is preferred to lactulose, based on small meta-analyses of even smaller trials [[93], [94]].

In populations with a high prevalence of lactose intolerance, the use of lactose has been suggested [95]. However, the only trial to show that stool-acidifying enemas (lactose and lactulose) were superior to tap-water enemas was underpowered [96]. The use of poly ethylene glycol preparation [97] needs further validation.

The dosing of lactulose should be initiated [98] when the three first elements of the four-pronged approach are completed, with 25 mL of lactulose syrup every 12 h until at least two soft or loose bowel movements per day are produced. Subsequently, the dosing is titrated to maintain two to three bowel movements per day. This dose reduction should be implemented. It is a misconception that lack of effect of smaller amounts of lactulose is remedied by much larger doses. There is a danger for overuse of lactulose leading to complications, such as aspiration, dehydration, hypernatremia, and severe perianal skin irritation, and overuse can even precipitate HE [99].


Rifaximin has been used for the therapy of HE in a number of trials [100] comparing it with placebo, other antibiotics, nonabsorbable disaccharides, and in dose-ranging studies. These trials showed effect of rifaximin that was equivalent or superior to the compared agents with good tolerability. Long-term cyclical therapy over 3–6 months with rifaximin for patients with OHE has also been studied in three trials (two compared to nonabsorbable disaccharides and one against neomycin) showing equivalence in cognitive improvement and ammonia lowering. A multinational study [101] with patients having two earlier OHE bouts to maintain remission showed the superiority of rifaximin vs. placebo (in the background of 91% lactulose use). No solid data support the use of rifaximin alone.

Other therapies

Many drugs have been used for treatment of HE, but data to support their use are limited, preliminary, or lacking. However, most of these drugs can safely be used despite their limited proven efficacy.


An updated meta-analysis of eight randomized, controlled trials (RCTs) indicated that oral BCAA-enriched formulations improve the manifestations of episodic HE whether OHE or MHE [[102], [130]]. There is no effect of IV BCAA on the episodic bout of HE [127].

Metabolic ammonia scavengers

These agents, through their metabolism, act as urea surrogates excreted in urine. Such drugs have been used for treatment of inborn errors of the urea cycle for many years. Different forms are available and currently present as promising investigational agents. Ornithine phenylacetate has been studied for HE, but further clinical reports are awaited [103]. Glyceryl phenylbutyrate (GPB) was tested in a recent RCT [104] on patients who had experienced two or more episodes of HE in the last 6 months and who were maintained on standard therapy (lactulose ± rifaximin). The GPB arm experienced fewer episodes of HE and hospitalizations as well as longer time to first event. More clinical studies on the same principle are under way and, if confirmed, may lead to clinical recommendations.

L-ornithine L-aspartate (LOLA)

An RCT on patients with persistent HE demonstrated improvement by IV LOLA in psychometric testing and postprandial venous ammonia levels [105]. Oral supplementation with LOLA is ineffective.


A recent, open-label study of either lactulose, probiotics, or no therapy in patients with cirrhosis who recovered from HE found fewer episodes of HE in the lactulose or probiotic arms, compared to placebo, but were not different between either interventions. There was no difference in rates of readmission in any of the arms of the study [106].

Glutaminase inhibitors

Portosystemic shunting up-regulates the intestinal glutaminase gene so that intestinal glutaminase inhibitors may be useful by reducing the amounts of ammonia produced by the gut.


This antibiotic still has its advocates and was widely used in the past for HE treatment; it is a known glutaminase inhibitor [107].


As short-term therapy [108], metronidazole also has advocates for its use. However, long-term ototoxicity, nephrotoxicity, and neurotoxicity make these agents unattractive for continuous long-term use.


This drug is not frequently used. It transiently improves mental status in OHE without improvement on recovery or survival. The effect may be of importance in marginal situations to avoid assisted ventilation. Likewise, the effect may be helpful in difficult differential diagnostic situations by confirming reversibility (e.g., when standard therapy unexpectedly fails or when benzodiazepine toxicity is suspected).


Simple laxatives alone do not have the prebiotic properties of disaccharides, and no publications have been forthcoming on this issue.


A recent RCT on OHE patients on rifaximin given daily IV albumin or saline showed no effect on resolution of HE, but was related to better postdischarge survival [109].


Prevention of overt hepatic encephalopathy

After an episode of OHE

There are no randomized, placebo-controlled trials of lactulose for maintenance of remission from OHE. However, it is still widely recommended and practiced. A single-center, open-label RCT of lactulose demonstrated less recurrence of HE in patients with cirrhosis [33]. A recent RCT supports lactulose as prevention of HE subsequent to upper gastrointestinal (GI) bleeding [110].

Rifaximin added to lactulose is the best-documented agent to maintain remission in patients who have already experienced one or more bouts of OHE while on lactulose treatment after their initial episode of OHE [101].

Hepatic encephalopathy after TIPS

Once TIPS was popularized to treat complications of PH, its tendency to cause the appearance of HE, or less commonly, intractable persistent HE, was noted. Faced with severe HE as a complication of a TIPS procedure, physicians had a major dilemma. Initially, it was routine to use standard HE treatment to prevent post-TIPS HE. However, one study illustrated that neither rifaximin nor lactulose prevented post-TIPS HE any better than placebo [111]. Careful case selection has reduced the incidence of severe HE post-TIPS. If it occurs, shunt diameter reduction can reverse HE [112]. However, the original cause for placing TIPS may reappear.

Another important issue with TIPS relates to the desired portal pressure (PP) attained after placement of stents. Too low a pressure because of large stent diameter can lead to intractable HE, as noted above. There is a lack of consensus on whether to aim to reduce PP by 50% or below 12 mmHg. The latter is associated with more bouts of encephalopathy [113]. It is widely used to treat post-TIPS recurrent HE as with other cases of recurrent HE, including the cases that cannot be managed by reduction of shunt diameter.

Hepatic encephalopathy secondary to portosystemic shunts (PSSs)

Recurrent bouts of overt HE in patients with preserved liver function consideration should lead to a search for large spontaneous PSSs. Certain types of shunts, such as splenorenal shunts, can be successfully embolized with rapid clearance of overt HE in a fraction of patients in a good liver function status, despite the risk for subsequent VB [114].


Discontinuation of prophylactic therapy

There is a nearly uniform policy to continue treatment indefinitely after it has successfully reversed a bout of OHE. The concept may be that once the thresholds for OHE is reached, then patients are at high risk for recurrent episodes. This risk appears to worsen as liver function deteriorates. However, what often occurs are recurrent bouts of OHE from a well-known list of precipitating factors. If a recurrent precipitating factor can be controlled, such as recurrent infections or variceal hemorrhages, then HE recurrence may not be a risk and HE therapy can be discontinued. Even more influential on the risk for further bouts of OHE is overall liver function and body habitus. If patients recover a significant amount of liver function and muscle mass from the time they had bouts of OHE, they may well be able to stop standard HE therapy. There are very little data on this issue, but tests positive for MHE or CHE before stopping HE drug therapy will predict patients at risk for recurrent HE.


Treatment of minimal HE and covert HE

Although it is not standard to offer therapy for MHE and CHE, studies have been performed using several modes of therapy. The majority of studies have been for less than 6 months and do not reflect the overall course of the condition. Trials span the gamut from small open-label trials to larger, randomized, controlled studies using treatments varying from probiotics, lactulose, and rifaximin. Most studies have shown an improvement in the underlying cognitive status, but the mode of diagnosis has varied considerably among studies. A minority of studies used clinically relevant endpoints. It was shown, in an open-label study [115], that lactulose can prevent development of the first episode of OHE, but the study needs to be replicated in a larger study in a blinded fashion before firm recommendations can be made. Studies using lactulose and rifaximin have shown improvement in quality of life [[34], [116]] and also in driving simulator performance [117]. Probiotics have also been used, but the open-label nature, varying amounts and types of organisms, and different outcomes make them difficult to recommend as therapeutic options at this time [[118], [119], [120], [121]].

Because of the multiple methods used to define MHE and CHE, varying endpoints, short-term treatment trials, and differing agents used in trials to date, routine treatment for MHE is not recommended at this stage. Exceptions could be made on a case-by-case basis using treatments that are approved for OHE, particularly for patients with CHE and West Haven Grade I HE.



Modulation of nitrogen metabolism is crucial to the management of all grades of HE, and nutritional options are relevant. Detailed recent guidelines for nutrition of patients with HE are given elsewhere [122]. Malnutrition is often underdiagnosed, and approximately 75% of patients with HE suffer from moderate-to-severe protein-calorie malnutrition with loss of muscle mass and energy depots. Chronic protein restriction is detrimental because patient's protein requirements are relatively greater than that of healthy patients and they are at risk of accelerated fasting metabolism. Malnutrition and loss of muscle bulk is a risk factor for development of HE and other cirrhosis complications. Sarcopenia has been proven to be an important negative prognostic indicator in patients with cirrhosis [[123], [124]]. All HE patients should undergo an assessment of nutritional status by taking a good dietary history, with anthropometric data and muscle strength measurement as practical, useful measures of nutritional status. In the undressed patient, particular attention is paid to the muscle structures around the shoulders and gluteal muscles. Pitfalls are water retention and obesity. Although body mass index is rarely helpful, the height-creatinine ratio may be useful, as well as the bioimpedance technique. More advanced techniques, such as dual-energy X-ray absorptiometry/CT/MR, are rarely useful for clinical purposes. The patient should undergo a structured dietary assessment, preferably by a dietician, or other specially trained staff. The majority of HE patients will fulfil criteria for nutritional therapy. The therapy is refeeding by moderate hyperalimentation, as indicated below. Small meals evenly distributed throughout the day and a late-night snack [125] should be encouraged, with avoidance of fasting. Glucose may be the most readily available calorie source, but should not be utilized as the only nutrition. Hyperalimentation should be given orally to patients that can cooperate, by gastric tube to patients who cannot take the required amount, and parenterally to other patients. The nutrition therapy should be initiated without delay and monitored during maintenance visits. The use of a multivitamin is generally recommended, although there are no firm data on the benefits of vitamin and mineral supplementation. Specific micronutrient replacement is given if there are confirmed measured losses, and zinc supplementation is considered when treating HE. If Wernicke's is suspected, large doses of thiamine should be given parenterally and before any glucose administration. Administration of large amounts of nonsaline fluids should be adjusted so as to avoid induction of hyponatremia, particularly in patients with advanced cirrhosis. If severe hyponatremia is corrected, this should be done slowly.

There is consensus that low-protein nutrition should be avoided for patients with HE. Some degree of protein restriction may be inevitable in the first few days of OHE treatment, but should not be prolonged. Substitution of milk-based or vegetable protein or supplementing with BCAAs is preferable to reduction of total protein intake. Oral BCAA-enriched nutritional formulation may be used to treat HE and generally improves the nutritional status of patients with cirrhosis [126], but IV BCAA for an episode of HE has no effect [127]. The studies on the effect of oral BCAA have been more encouraging [[128], [129]] and confirmed by a recent meta-analysis of 11 trials [130]. Ultimately, the effects of these amino acids may turn out to have more important effects on promotion of maintenance of lean body mass than a direct effect on HE.


Liver transplantation (LT)

Liver Transplantation remains the only treatment option for HE that does not improve on any other treatment, but is not without its risks. The management of these potential transplant candidates as practiced in the United States has been published elsewhere [[131], [132]], and European guidelines are under way. Hepatic encephalopathy by itself is not considered an indication for LT unless associated with poor liver function. However, cases do occur where HE severely compromises the patient's quality of life and cannot be improved despite maximal medical therapy and who may be LT candidates despite otherwise good liver status. Large PSSs may cause neurological disturbances and persistent HE, even after LT. Therefore, shunts should be identified and embolization considered before or during transplantation [133]. Also, during the transplant workup, severe hyponatremia should be corrected slowly.

Hepatic encephalopathy should improve after transplant, whereas neurodegenerative disorders will worsen. Therefore, it is important to distinguish HE from other causes of mental impairment, such as Alzheimer's disease and small-vessel cerebrovascular disease. Magnetic resonance imaging and spectroscopy of the brain should be conducted, and the patient should be evaluated by an expert in neuropsychology and neurodegenerative diseases [134]. The patient, caregivers, and health professionals should be aware that transplantation may induce brain function impairment and that not all manifestations of HE are fully reversible by transplantation [135].

One difficult and not uncommon problem is the development of a confusional syndrome in the postoperative period. The search of the cause is often difficult, and the problem may have multiple origins. Patients with alcoholic liver disease (ALD) and those with recurrent HE before transplantation are at higher risk. Toxic effects of immunosuppressant drugs are a frequent cause, usually associated with tremor and elevated levels in blood. Other adverse cerebral effects of drugs may be difficult to diagnose. Confusion associated with fever requires a diligent, systematic search for bacterial or viral causes (e.g., cytomegalovirus). Multiple causative factors are not unusual, and the patient's problem should be approached from a broad clinical view [136].

Economic/cost implications

As outlined under epidemiology, the burden of HE is rapidly increasing and more cases of HE will be encountered, with substantial direct costs being attributed to hospitalizations for HE and to indirect costs. The patients with HE hospitalized in the United States in 2003 generated charges of approximately US$ 1 billion [[40], [137]]. Resource utilization for this group of patients is also increasing as a result of longer lengths of stay and more complex and expensive hospital efforts, as well as a reported in-patient mortality of 15%. There are no directly comparable EU cost data, but by inference from epidemiological data, the event rate should be approximately the same and the costs comparable, differences between U.S. and EU hospital financing notwithstanding. These costs are an underestimate, because out-patient care, disability and lost productivity, and the negative effect on the patient's family or support network were not quantified [138].

The cost of medications is very variable to include in analyses because it varies widely from country to country and are usually determined by what the pharmaceutical companies believe the market can sustain. Regarding the beneficial effects of rifaximin, cost-effective analyses based on current drug prices favor treatments that are lactulose based [[92], [139]], as do analyses of accidents, deaths/morbidity, and time off from work [73] in patients with MHE or CHE. Therefore, until the costs of other medications fall, lactulose continues to be the least expensive, most cost-effective treatment.