The Diagnosis and Treatment of Minimal Hepatic Encephalopathy

Hepatic encephalopathy (HE) is a potentially reversible, metabolically caused disturbance of central nervous system function that occurs in patients with acute or chronic liver disease. It encompasses a broad spectrum of neurological symptoms of varying severity and is classified according to clinical symptoms (Table 1) or etiology (Figure 1). Minimal hepatic encephalopathy (MHE), previously known as subclinical or latent hepatic encephalopathy, is at the beginning of this spectrum. It is defined as HE without symptoms on clinical/neurological examination, but with deficits in some cognitive areas that can only be measured by neuropsychometric testing (1). The areas with impairments are attention, visuospatial perception, speed of information processing, especially in the psychomotor area, fine motor skills, and short-term memory (2). MHE has a high prevalence among patients with liver cirrhosis (22% to 74%) (e1) and also occurs in patients with noncirrhotic liver disease such as portal vein thrombosis (e2) or portosystemic shunt (e3). However, the true number of patients with MHE is unknown, firstly because the diagnostic criteria in use around the world are not entirely uniform, and secondly because MHE often remains undiagnosed due to the lack of evident symptoms (e4). However, numerous studies have shown that, although the neurological symptoms are slight, affected patients are markedly impaired in their quality of life and ability to work (3, 4). In addition, in two retrospective studies, patients with liver cirrhosis and MHE had significantly more driving accidents than those without MHE (6, e5). The reasons were more frequent driving errors (speeding, illegal turns) as shown by one study using a driving simulator (e6), a greater tendency to fatigue at the wheel (e7), and subjective overestimation of their own driving skills (e8). Other studies have shown that patients with MHE suffer from falls (5) and from the development of episodic HE more frequently (7, e9). Some studies have even identified MHE as an independent predictor of survival in patients with liver cirrhosis (8, e10). At the same time, current randomized controlled trials (RCTs) indicate that treating MHE leads to an improvement in cognitive abilities (9, 10) and driving performance (11). This review presents the pathogenesis, diagnosis, and options for treatment of MHE.

Figure 1

Nomenclature of hepatic encephalopathy (HE) (1).

Enlarge All figures

Table 1

Semiquantitative grading of mental status in hepatic encephalopathy using the West Haven criteria (modified from Conn et al. [e32]). Grade 0 corresponds to MHE.

Enlarge All figures

Method

PubMed and Medline were searched for original and review articles using a combination of the search terms “minimal hepatic encephalopathy” plus “ammonia,” “lactulose,” “psychometry,” or “rifaximin.” Publications in English and German from the years 1970 to 2011 were evaluated. The reference lists of these articles were also searched for further publications.

Pathogenesis

Ammonia

Ammonia is of central importance in the pathogenesis of HE. Under physiological conditions ammonia is primarily cleared by the synthesis of urea in the liver. If the liver is functionally impaired or a portosystemic shunt is present, this function is compromised and the extrahepatic metabolization of ammonia by the brain and musculature becomes more important (e11). Accumulation of ammonia in the brain of patients with MHE has been shown directly by positron emission tomography (PET) (12). Astrocytes are the only cells in the brain that can fix ammonia, through the formation of glutamine (e11). The intracellular glutamine concentration in the astrocytes rises with ammonia levels in the blood, and causes the cells to swell through the osmosis. This leads overall to the development of low-grade brain edema, which correlates with deterioration in psychometric tests (e12). The close association between brain edema and impaired liver function is also shown by the fact that brain edema and the cognitive impairments are reversible by liver transplantation (e12).

Other factors

Disequilibrium of the gut flora with fecal overgrowth by urease-forming bacteria has been observed in patients with MHE, and therapeutic intervention led to an improvement of this disequilibrium and of psychometric test results (14). In addition, bilateral manganese deposits have been found in the globus pallidus in patients with HE (e13). Both manganese and ammonia are believed to increase the expression of peripheral-type benzodiazepine receptors in the brain (e14). These receptors regulate the production of neurosteroids and are present in increased density in the brain of patients with MHE (e15). Through increased synthesis of neurosteroids, which function as positive regulators of GABA-A receptors, the GABAergic tone in the brain is increased.

Diagnosis

A survey of the American Society for the Study of Liver Diseases revealed that the majority of doctors regard MHE as a significant clinical problem, but only half of those actually tested their patients for MHE (e4). The West Haven criteria for clinical stratification of HE (Table 1), which are in common use, assume manifest neurological symptoms and are therefore of limited suitability in MHE. Although there is international consensus that psychometric tests are the gold standard in the diagnosis of MHE (1), no agreement exists as to what combination of tests should be carried out, and what the threshold value is at which MHE may be reliably diagnosed. This central problem is reflected in the varying reported prevalences for the disease, which range from 22% to 74% depending on which tests are chosen and where the threshold is defined (e1). A general approach to the diagnosis and treatment of MHE based on Ferenci et al. (1) is shown in Figure 2. First, obvious neurological symptoms and cognitive impairment should be ruled out. In addition to the neurological examination, the test that has proved most useful for this purpose is the Mini-Mental State Examination (MMSE). The MMSE is a widely used screening test for the diagnosis of dementia and examines the most important basic cognitive abilities (e16) (Table 2). If both the clinical examination and the MMSE yield normal results, the next step is to quantify any latent cognitive deficits through psychometric testing. Neurophysiological tests and imaging techniques exist to complement these, but are mainly used in experimental settings.

Figure 2

Suggested procedure for diagnosis and treatment of MHE. FIGURE 2 MHE, minimal hepatic encephalopathy, PHES, psychometric hepatic encephalopathy score

Enlarge All figures

Table 2

Mini Mental State Examination (modified from Folstein et al. [e16, e33]). In most studies episodic HE and se - vere cognitive impairment were ruled out by a score above 25 (10, 11).

Enlarge All figures

Ammonia concentration

In episodic HE, the venous ammonia concentration correlates with the severity of neurological impairment (15) and may be used in the differential diagnosis. Ammonia concentrations are less important in the diagnosis of MHE because they do not correlate with the degree of neurological dysfunction (13). In addition, correct measurement of the ammonia concentration requires a venous blood sample obtained without using a tourniquet and immediate laboratory analysis within 20 minutes, which in clinical routine, especially in a doctor’s office, is rarely possible (15). The ammonia concentration is also influenced by factors such as renal function, nicotine consumption, and muscle mass.

Psychometric tests

The Psychometric HE Score (PHES) consists of a series of psychometric tests and was conceived specifically for diagnosing MHE (2). It comprises the number connection test (NCT), the line-tracing test, and the number–symbol test, and takes a total of 20 to 25 minutes (Table 3). Most studies of MHE use the PHES or a selection of its constituent tests. The great advantage of this test is that for some countries, including Germany, comparative data exist from the normal population. The disadvantages of the test are the occurrence of learning effects, which limits repeatability, and the strong emphasis on fine motor skills. There are also differences as to where the border between normal and pathological should be drawn.

Table 3

Psychometric tests recommended for diagnosing minimal hepatic encephalopathy. The lower the overall score, the better the cognitive performance (mod - ified from Ferenci et. al [1])

Enlarge All figures

Neurophysiological tests

To increase objectivity and reproducibility, various neurophysiological tests have been developed. Determination of the critical flicker frequency is based on the assumption that the glial cells of the retina are subject to the same functional impairment as the astrocytes in the brain. A light impulse with an initial frequency of 60 Hz is presented to the patient, who perceives it as a constant light. The frequency is then reduced by 0.1 Hz steps until the patient first perceives the light as flickering. This frequency is the critical flicker frequency. It correlates positively with psychometric test results and is not affected by gender or education, but may be affected by age (16, e17).

Another neurophysiological test is the electroencephalogram (EEG). Changes in the spectral EEG and in the discharge of visual evoked late potentials (P300 wave) are believed to have a higher sensitivity than psychometric tests and to have prognostic significance for progression to episodic HE (e18, e19). Although neurophysiological tests have considerable advantages, their use is limited by the high cost of acquiring the technical apparatus and analyzing the results. For this reason they are mainly used in experimental studies.

Imaging techniques

Various magnetic resonance techniques show pathological changes in patients with MHE. T1-weighted MRI shows a hyperintense signal in the basal ganglia (globus pallidus and substantia nigra), which is interpreted as due to manganese deposits (e20). Although the hyperintensity is not quantitatively correlated to the severity of HE, it does disappear after liver transplantation (e20). Using magnetic resonance spectroscopy, changes can also be demonstrated in the relationship between myoinositol and creatine in patients with MHE (e21). It is assumed that the osmotically active myoinositol is secreted from the cell in order to compensate for the swelling caused by glutamine. Magnetization transfer measurements have shown low-grade brain edema in patients with MHE (e22). Single-photon emission computed tomography (SPECT) and PET show changes in blood flow that correlate with psychometric test results (17, e23).

Treatment

Unlike for episodic HE, there are only a few RCTs with small case numbers on the treatment of MHE (Table 4). The effects of lactulose and rifaximin are the best investigated. So far, RCTs have shown a positive effect of treatment on cognitive abilities, quality of life (9, 10), and driving ability (11); its effect on patients’ ability to work or risk of falling remains unproven. The duration of treatment and choice of medication also remain unclear. Most treatment approaches derive from experience with episodic HE. Since deterioration of cognitive function in patients with liver cirrhosis is primarily triggered by precipitating factors (e24), consistently avoiding these factors is also paramount for patients with MHE (Box).

Box

Precipitating factors for development of MHE and episodic HE

Enlarge All figures

Table 4

Existing studies on the treatment of MHE

Enlarge All figures

Nonabsorbable disaccharides

The nonabsorbable disaccharides lactulose and lactilol are those for which the most comprehensive data is available, because both of these substances have been in clinical use for a long time. In consequence, lactulose is regarded as the first-line therapy for HE (18). Besides their laxative effect, nonabsorbable disaccharides reduce the synthesis and uptake of ammonia by lowering the pH of the colon and also reducing the uptake of glutamine from the gut (e25). It has been shown in several studies that treatment with lactulose significantly improves the performance of patients with MHE in psychometric tests, which is associated with a rise in quality of life (9, 19, 20). Lactulose is also superior to placebo in preventing episodic HE (recurrence in 19.6% of patients in the lactulose group vs. 46.8% in the placebo group, P = 0.001; duration of follow-up: 14 months) (21). The usual oral dose is 15 to 30 mL twice daily, in order to achieve a soft stool several times a day. A course of treatment should continue for at least 3 to 6 months. The adverse effects of the treatment are alteration of taste perception and bloating. Overdosing causes diarrhea which can result in severe dehydration and hyponatremia, which lead to worsening of HE (e26).

Antibiotics

The aim of antibiotic therapy is to reduce ammonia production in the gut. Neomycin was the first antibiotic to be used in the treatment of HE and appears to be as effective as lactulose (e27). Despite their low absorption, the use of macrolides has been reduced in recent years because of their marked oto- and nephrotoxicity, the more so because these adverse effects are particularly serious in patients with reduced liver function. One alternative that is being increasingly used is rifaximin. Rifaximin is an oral antibiotic that is only minimally absorbed in the gut and therefore has a very low adverse effect profile. Although it has been in use since 1987, particularly in the treatment of enteritis, no clinically significant resistance has been observed so far. Rifaximin has been licensed in the USA since 2010 for treatment of HE, and licensing is planned for Germany in 2012. Taking rifaximin improves psychometric test results, quality of life, and driving ability in patients with MHE (10, 11). The exact changes in the effect sizes are presented in Table 4. Bass et al. were able to show in a large study that long-term therapy with rifaximin plus lactulose in patients who had a history of HE gave better protection against renewed episodic HE than did the placebo treatment (hazard ratio with rifaximin: 0.42; 95% confidence interval 0.28 to 0.64; P < 0.001) (22).

Nutritional therapy/nutrition supplementation

The question whether increasing or restricting protein intake is beneficial remains under debate, since under physiological conditions, amino acids are almost fully absorbed in the ileum and consequently contribute little to ammonia production in the colon. It has nevertheless been postulated that excessive protein intake could provoke an increase in blood ammonia levels due to physiological malabsorption. On the other hand, reducing protein intake decreases body muscle mass and hence the ability to absorb ammonia extrahepatically. The European Society for Clinical Nutrition and Metabolism (ESPEN) currently recommends on a purely empirical basis a protein intake of 1 to 1.2 g/kg body weight for patients with HE, with vegetable proteins being preferred to animal proteins (23). In one RCT, oral intake of branched-chain amino acids improved the psychometric test results of patients with HE (e28). In addition, there are indications that oral intake of L-ornithine aspartate, a substrate of the urea cycle, improves the cognitive abilities of patients with HE of varying severity (e29). One further potential candidate for treating MHE is L-acetyl carnitine. Two recent RCTs show improvement of cognitive function and reduction of ammonia concentrations (24, e30). Zinc deficiency is often seen in patients with liver cirrhosis and impairs the metabolization of ammonia (e31). Although the effect of zinc supplementation in HE is not entirely clear, patients with manifest zinc deficiency should receive supplements (18). Probiotics such as yoghurt can have a beneficial effect on the bacterial microflora in terms of lowering ammonia production. In two RCTS, probiotics have led to a significant improvement in MHE (14, 25). The effect of a combination of the above substances has not yet been adequately investigated.

Summary

The borderline between normal or acceptable findings and pathological findings—that is, those that are a threat to health—is fluid in MHE. The problem is that MHE can be a risk to other people, e.g., when it leads to inadequate reactions when driving. Unfortunately this is not easy to measure. In the spectrum of the heterogeneous general population, and given the physiological fluctuations in attention status, patients with MHE are often difficult to identify. For the physician, therefore, the question is whether patients with impaired liver function—e.g., patients suffering from liver cirrhosis and MHE—require a specific treatment. If a patient has never had an episode of HE, we believe that the most important step is to inform the patient about the potential risks. Lifestyle changes, with a balanced diet (not too rich in protein), exercise, enough sleep, and abstinence from alcohol and sedativa, are probably the most appropriate way to proceed. Once episodic HE has been documented, specific medical treatments recommended by the specialist societies for episodic HE should be added to the above changes on a preventative basis. Monotherapy with lactulose or (depending on the individual risk potential) a combination of lactulose with rifaximin should be chosen. Patient compliance is absolutely essential and should be ensured, e.g., by including relatives in the treatment program.

Conflict of interest statement
The authors declare that no conflict of interest exists.

Manuscript received on 20 June 2011, revised version accepted on
1 September 2011.

Translated from the original German by Kersti Wagstaff, MA.

Corresponding author
Prof. Dr. med. Wolfgang Stremmel
Im Neuenheimer Feld 410
69120 Heidelberg, Germany
Wolfgang.Stremmel@med.uni-heidelberg.de

@For eReferences please refer to:
www.aerzteblatt-international.de/ref1012