Archives of Biochemistry and Biophysics 536 (2013) 197–203

Contents lists available at SciVerse ScienceDirect

Archives of Biochemistry and Biophysics

journal homepage: www.elsevier .com/ locate/yabbi

Review

The slowed brain: Cortical oscillatory activity in hepatic encephalopathy

Markus Butz a,b,⇑,1, Elisabeth S. May a,1, Dieter Häussinger c, Alfons Schnitzler a

a Heinrich-Heine-University Düsseldorf, Medical Faculty, Institute of Clinical Neuroscience and Medical Psychology, Universitätsstraße 1, D-40225 Düsseldorf, Germanyb Sobell Department of Motor Neuroscience and Movement Disorders, UCL Institute of Neurology, Queen Square, London WC1N 38G, United Kingdomc Heinrich-Heine-University Düsseldorf, Medical Faculty, Department of Gastroenterology, Hepatology and Infectious Diseases, Universitätsstraße 1, D-40225 Düsseldorf, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Available online 16 April 2013

Keywords:Critical flicker frequency (CFF)Electroencephalography (EEG)Magnetoencephalography (MEG)OscillationsSlowingSynchronization

0003-9861/$ - see front matter � 2013 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.abb.2013.04.004

⇑ Corresponding author at: Sobell Department oMovement Disorders, UCL Institute of Neurology, Q38G, United Kingdom. Fax: +44 20 72789836.

E-mail addresses: m.butz@ucl.ac.uk, markus.butz@1 These authors contributed equally to this work.2 Abbreviations used: CFF, critical flicker frequency;

GABA, gamma-aminobutyric acid; HE, hepatic encephshowing any signs of HE; MEG, magnetoencephalograsomatosensory evoked potentials.

Oscillatory activity of the human brain has received growing interest as a key mechanism of large-scaleintegration across different brain regions. Besides a crucial role of oscillatory activity in the emergence ofother neurological and psychiatric diseases, recent evidence indicates a key role in the pathophysiology ofhepatic encephalopathy (HE).

This review summarizes the current knowledge on pathological alterations of oscillatory brain activityin association with liver dysfunction and HE in the context of spontaneous brain activity, motor symp-toms, sensory processing, and attention. The existing literature demonstrates a prominent slowing ofthe frequency of oscillatory activity as shown for spontaneous brain activity at rest, with respect to def-icits of motor behavior and motor symptoms, and in the context of visual attention processes. Theobserved slowing extends across different subsystems of the brain and has been confirmed across differ-ent frequency bands, providing evidence for ubiquitous changes of oscillatory activity in HE. For example,the frequency of cortico-muscular coherence in HE patients appears at the frequency of the mini-asterixis(612 Hz), while cirrhotics without overt signs of HE show coherence similar to healthy subjects, i.e. at13–30 Hz. Interestingly, the so-called critical flicker frequency (CFF) as a measure of the processing ofan oscillating visual stimulus has emerged as a useful tool to quantify HE disease severity, correlatingwith behavioral and neurophysiological alterations. Moreover, the CFF reliably distinguishes patientswith manifest HE from cirrhotics without any signs of HE and healthy controls using a cut-off frequencyof 39 Hz.

In conclusion, oscillatory activity is globally slowed in HE in close association with HE symptoms anddisease severity. Although the underlying causal mechanisms are not yet understood, these results indi-cate that pathological changes of oscillatory activity play an important role in the pathophysiology of HE.

� 2013 Elsevier Inc. All rights reserved.

Introduction

In recent decades, oscillatory activity of the human brain hasenjoyed a renewed and ever-growing interest in the scientific com-munity [1,2]. Neuronal oscillations are periodic fluctuations in theamplitude of recorded neuronal activity, which occur at differentfrequencies. When the cortical area engaging in oscillatory activityis large enough, this synchronized neuronal activity is recordableat the surface of the skull using electroencephalography (EEG)2

[3] or magnetoencephalography (MEG) [4]. EEG allows recording

ll rights reserved.

f Motor Neuroscience andueen Square, London WC1N

uni-duesseldorf.de (M. Butz).

EEG, electroencephalography;alopathy; HE 0, cirrhotics notphy; mHE, minimal HE; SEP,

the electric activity of the brain, while MEG measures the magneticfield created by neuronal currents. Both measure brain activity atmultiple locations covering the entire scalp. Thereby, it is for exam-ple possible to study spontaneous, i.e. resting state, activity of thebrain. In addition, M/EEG are often used to study so-called sensoryevoked potentials in response to different sensory stimuli and so-called cognitive evoked potentials in the context of cognitive tasks[3].

The most prominent and well-known oscillatory signal re-corded from the surface of the awake human brain is the so-calledalpha band activity in the frequency range between 8 and 12 Hz,which is also known as the Berger-rhythm [5]. Other frequencybands, which have been associated with specific sensory, motor,and/or cognitive functions are the delta (1–3 Hz), theta (3–7 Hz),beta (13–30 Hz), and gamma frequency bands (30–100 Hz). Iftwo or more separated brain areas show oscillatory activity witha stable phase and amplitude relation over time at the same fre-quency, a functional coupling between these areas can be assumed[6]. This coupling can be mathematically expressed in terms of

198 M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203

coherence, which quantifies the degree of linear dependency be-tween two signals and can also be computed between the brainand external signals, e.g. muscle activity.

Oscillatory activity is believed to be a key mechanism of large-scale functional integration across different brain regions and thebasis of cortical network interactions [6]. Apart from this crucialrole in healthy subjects, changes of oscillatory activity are knownto play a critical role in the pathophysiology of various neurologicaland psychiatric diseases such as Parkinson’s disease, schizophre-nia, and epilepsy [6,7].

In the recent past, alterations of oscillatory brain activity havealso been implicated in the emergence of hepatic encephalopathy(HE) [8,9]. A number of studies shed further light on the mecha-nisms of impaired brain function in HE by highlighting the impor-tance of changes in oscillatory brain activity, which parallel diseaseseverity. Interestingly, there is no specific subsystem of the brainprimarily affected in HE, like e.g. the motor system in Parkinson’sdisease. Instead, HE is associated with a variety of symptomsencompassing different subsystems. Thus, HE offers the uniqueopportunity to study the progressive, gradual, and fine-scaleddeterioration of oscillatory activity across the brain and diseasestates.

Fig. 1. Increasing severity of hepatic encephalopathy (HE) is paralleled by slowingof cortico-muscular coupling. In patients with liver cirrhosis and HE of varyingseverity, cortico-muscular coupling, i.e. coherence, was determined during forearmelevation. Both the CFF (A) and the frequency of coherence between arm muscleactivity and the contralateral primary motor cortex (B) were slowed with increasingdisease severity as determined by detailed neuropsychometric and clinical testing.In addition, the frequency of coherence correlated with disease severity asquantified by the CFF (C). These findings imply a parallel slowing in the motorand visual system in HE. CFF = critical flicker frequency, EMG = electromyography,HE 0 = patients with liver cirrhosis not showing any symptoms of HE, HE 1 = HEgrade 1, HE 2 = HE grade 2 or higher according to the West-Haven Criteria,SHE = subclinical HE. Figure adapted from [55].

The critical flicker frequency (CFF)

The capacity of the visual system to process flickering, i.e. oscil-lating, visual stimuli has been extensively used as measure for theeffects of psychotropic drugs as well as the characterization of dif-ferent neurological diseases such as Alzheimer’s disease and multi-ple sclerosis [10–12]. The CFF is the frequency threshold at which afast flickering light is no longer perceived as continuous but as aflickering light and it is a parameter increasingly used to quantifyand monitor the severity of HE (e.g. [13–17]).

Kircheis and colleagues were the first to systematically evalu-ate the application of the CFF as a diagnostic tool for the quanti-fication of low-grade HE [14]. The authors demonstrated aninverse correlation between the CFF and the clinical grade ascharacterized by detailed computerpsychometric and clinical test-ing. By using a cut-off value of 39 Hz, the CFF reliably distin-guishes patients with manifest HE from healthy control subjectsand HE 0 cirrhotic patients, i.e. cirrhotic patients not showingany HE symptoms. However, an overlap between CFF values ofpatients without HE, with minimal HE (mHE) , with HE of grade1, as well as healthy control subjects has been reported andyielded mixed results regarding the value of the CFF for the diag-nosis of mHE [14,18–20]. Nevertheless, a recent meta-analysis re-vealed a high specificity and a moderate sensitivity and it wasthus recommended to use the CFF as an addition to psychometrictesting for the diagnosis of mHE [21].

Importantly, the CFF offers the advantage to be simple and fastto administer and interpret [22,23]. It provides a fine-graded andcontinuous measure, meeting recent demands to approach theneurocognitive changes in HE as a continuum rather than defininga limited number of distinct stages of the disease [24]. Using theCFF as a behavioral oscillatory parameter of HE disease severity,close correlations with pathophysiological changes in oscillatorybrain activity and behavior could be shown (see sections ‘‘Motorsystem’’ and ‘‘Attention’’ as well as Figs. 1 and 2).

It is important to note, however, that the indicative role of theCFF for changes of oscillatory brain activity in HE, which is sug-gested in this review, is so far solely based on correlation analyses,which do not prove causality per se. As long as a causal prove ismissing, spurious correlations or correlations due to a commonrelation to a third factor like HE severity cannot be entirely ruledout.

Spontaneous brain activity

Since the early description by Foley and others [25,26], it hasbeen known that HE patients present significant changes of electricbrain activity, of which a general slowing is the main characteristic.Resting state EEG measurements are used for the detection of min-imal/subclinical HE, the objective quantification of overt HE, andthe management of acute liver failure [27].

Spontaneous oscillatory activity in HE patients is characterizedby a slowing and disorganization of alpha band activity (8–12 Hz)and increased activity in the lower theta band (3–7 Hz) [28–29].With progressing disease severity, theta band activity is becoming

Fig. 2. Increasing severity of hepatic encephalopathy (HE) is paralleled by slowingof oscillatory gamma band activity in early visual cortices. Healthy subjects andpatients with liver cirrhosis and HE up to grade 1 underwent an audiovisualattention paradigm inducing oscillatory gamma band activity in early visualcortices. Subjects with a pathological critical flicker frequency (CFF), i.e. below39 Hz, and subjects with a CFF in the physiological range, i.e. above 39 Hz, showedstimulus-induced gamma band activity (A), which could be localized in early visualareas (B). Across all subjects, increasing disease severity as quantified by the CFFwas associated with a slowing of the peak frequency of stimulus-induced gammaband activity (C). These findings imply a parallel slowing of oscillatory gamma bandbrain activity and the CFF in low grade HE. High CFF = subjects with a CFF above39 Hz, low CFF = subjects with a CFF below 39 Hz. Figure adapted from [63].

M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203 199

stronger, resulting in triphasic waves, composed of two electroneg-ative waves followed by a positive wave [30,31]. They are observa-ble in all metabolic encephalopathies and believed to originatefrom alterations of thalamo-cortical-coupling [31,32]. In more se-vere states of HE, EEG activity is dominated by even slower oscilla-tory activity in the delta band (1–3 Hz), before the EEG finallyappears flat in most severely affected patients, i.e. shows isoelec-tric activity [27,28,31]. In accordance with these results, a lowermean dominant peak frequency has been observed (e.g. [33,34])and significant EEG alterations could even been shown in cirrhoticswithout signs of overt HE [29,33]. Thus, a slowing of spontaneousbrain activity appears as a prominent phenomenon in HE, even inearly, i.e. subclinical, stages of the disease.

There is evidence for a link between EEG activity and behavioralparameters [35,36] as measured by the psychometric hepatic

encephalopathy score (PHES) [37]. Moreover, correlations betweenlaboratory values relevant in the pathophysiology of HE, i.e. ammo-nia and indole, and the mean dominant peak frequency could bedemonstrated [38]. In addition, a correlation between an increaseof oscillatory activity in the theta band and ammonia blood levelswas reported for mHE patients [29].

Quantitative spectral EEG analysis provides objective parame-ters of brain activity changes [39] and has prognostic value forthe occurrence of overt HE and mortality in cirrhotics [40,41]. Fur-thermore, there was recovery of EEG activity after liver transplan-tation [42]. Attempts to optimize the analytical approach toautomatically interpret EEG activity in HE include spatio-temporaldecomposition [43], autoregressive modelling [44], coherencemeasures [45], and artificial neural network-expert system soft-ware [46]. However, the observed changes of oscillatory activityvary significantly between individuals, hindering the developmentof standard software for the automatic analysis and interpretationof EEG activity in HE.

Motor system

Motor symptoms are key symptoms in HE. They include sym-metric akinetic-rigid syndromes, postural tremor, and early pos-tural and gait impairment [47,48]. Apart from ataxia, the mostevident motor symptom in manifest HE is mini-asterixis [49]. Thisalmost rhythmical, tremor-like phenomenon, which occurs pri-marily at 6–12 Hz in the upper limbs, is present during maintainedposture and starts after a latent period of about 2–30 s. In more se-vere stages of HE, an additional, more striking motor symptom oc-curs, commonly referred to as asterixis. First described by Adamsand Foley in 1949 [50], asterixis is characterized by sudden lapsesof posture at the wrist, metacarpophalangeal, or hip joints [51,52]and commonly termed flapping tremor.

A previous MEG study suggested that mini-asterixis arises froma pathologically slowed drive of the primary motor cortex towardsthe muscles [53]. In this work, coupling between muscle and brainwas measured by cortico-muscular coherence. Coupling was stron-gest in the primary motor cortex contralateral to the affected handand appeared in the individual frequency of the mini-asterixis, i.e.between 5 and 12 Hz. Importantly, such a pathologically slowedcoupling was only present in cirrhotics suffering from mini-aster-ixis; cirrhotics without this symptom showed cortico-muscularcoherence at much higher frequencies similar to healthy subjects,i.e. in the beta frequency range (13–30 Hz). From this finding[53], the authors concluded that the primary motor cortex is in-volved in the generation of mini-asterixis and that the observedcoupling may reflect a pathologically slowed and synchronizedmotor-cortical drive. The same group demonstrated that mini-asterixis is also accompanied by slowed thalamo-motor-corticalcoupling in the same slow frequency range [54]. Hence, the authorssuggested that a slowing of oscillatory coupling in a network ofbrain areas is involved in the generation of mini-asterixis.

Subsequently, these results were replicated in a larger cohort ofpatients [55]. In addition, a correlation between the frequency ofcortico-muscular coherence and the parallel slowing of the CFFcould be demonstrated (see Fig. 1) [55], indicating a close associa-tion between oscillatory coupling patterns in the motor systemand the capacity to process an oscillating visual stimulus. Finally,a recent study showed that the slowing of the CFF correlates witha slowing and impairment of behavioral motor performance asmeasured by fastest finger movements and movement ratingscales [56].

First reports on single patients using EEG also revealed associa-tions between asterixis and oscillatory activity at 16–20 Hz [57] aswell as a biphasic wave [58] in central areas likely to involve the

200 M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203

primary motor cortex. However, more work is needed in thiscontext.

Sensory processing

With the primary goal to develop new tools for the diagnosis ofboth overt and minimal/subclinical HE, numerous studies mea-sured basic auditory, visual, and somatosensory processing interms of sensory evoked potentials in patients with liver dysfunc-tion and HE (for reviews see [59–60]). As described in detail in areview by Montagnese et al. [60], these studies show a large vari-ation in the types of patients studied, the means to diagnose HE,and the methodological approaches to acquire and analyze thedata, resulting in a large variability of results. Altogether, there isevidence for a delay, a deformation, and eventual loss of singlecomponents of visual evoked potentials, auditory brain stemevoked potentials, and somatosensory evoked potentials in HE.Brainstem auditory evoked potentials show the least, somatosen-sory evoked components the most consistent results across stud-ies, so that the latter are seen as the most promising tool for thediagnosis of HE [60].

As an example, somatosensory processing in HE has many timesbeen investigated by measuring somatosensory evoked potentials(SEP) in response to median nerve stimulation. These studies con-sistently reported a prolongation of the peak and inter peak laten-cies in combination with a later deformation or loss of the corticalSEP components, thus providing evidence for altered and delayedprocessing of simple somatosensory stimuli in HE (e.g. [61,62]).Across studies, these alterations consistently correlate with HE dis-ease severity and mirror clinical improvement/worsening andtherapeutic effects [60].

Although these studies indicate changes of the processing ofsimple sensory stimuli across different modalities in HE, actualalterations in oscillatory brain activity have only been analyzedand reported by Kahlbrock et al. [63], who demonstrated a slowingof oscillatory activity in the gamma band elicited by a complex vi-sual grating stimulus (see section ‘‘Attention’’ for details). How-ever, besides eliciting stimulus-evoked potentials, sensorystimulus processing is known to be associated with distinct mod-ulations of oscillatory brain activity. Simple somatosensory stimuli,for example, are associated with a characteristic modulation ofoscillatory alpha band activity, which is initially suppressed indi-cating cortical activation and then rebounds above baseline levelsbefore stimulus processing is completed (e.g. [64]). Like the laten-cies of stimulus-evoked potentials, the time course of this modula-tion might be delayed in HE, possibly in combination with ageneral slowing of the alpha peak frequency. While results fromprevious SEP studies make corresponding changes of oscillatoryactivity plausible, such pathophysiological changes associated withsensory processing in HE still need to be examined and mightpotentially show a diagnostic usability superior to conventionalevoked component analysis.

Attention

Attentional deficits are one of the key symptoms of HE. They arepresent even in the mildest forms, show a continuous decline withincreasing HE severity and hepatic dysfunction, and are accompa-nied by alterations of arousal in higher stages of the disease, whichin the extreme manifest as stupor and hepatic coma [65–66]. Inthose patients who seem clinically unaffected, i.e. those with min-imal HE, attention deficits have even been assumed to be the mainsymptom [66].

In a series of studies, Amodio and co-workers for example dem-onstrated early attentional deficits in a visual attention orienting

paradigm in cirrhotic patients without overt symptoms of HE[65,67,68]. Results from these studies showed increased reactiontimes and error rates as well as a reduced basal arousal in combi-nation with an impaired capacity to disengage attention previouslyfocused on a cued location in cirrhotic patients compared tohealthy controls. Weissenborn et al. also studied cirrhotic patientswithout overt symptoms of HE and performed an extensive batteryof attention tests [69], showing dysfunctions of mHE patients in al-most every attentional function studied. Even cirrhotic patientswithout any indication of mHE differed from healthy controls insome tests. Taken as a whole, these behavioral results indicatecomplex attentional dysfunctions very early in the developmentof cognitive impairments associated with liver dysfunction.

On the neurophysiological level, several studies using EEG andthe measurement of cognitive evoked potentials have been per-formed in HE. An often-examined measure of cognitive deficits inHE is the auditory evoked potential P300 (e.g. [32,41,70,71]). Inthe classic paradigm, the P300 is elicited by rare auditory stimuliembedded in a series of irrelevant stimuli and is thought to bean indicator of stimulus evaluation and attentional target process-ing [72]. The P300 latency has been shown to be a good marker ofovert HE grade 1 and 2 [32,70] and has also been confirmed to bedelayed in a proportion of cirrhotic patients not showing manifestclinical symptoms [41,70,71,73]. Other recent neurophysiologicalstudies of attentional deficits demonstrated altered mismatchevoked activity in response to auditory stimuli in patients withmHE [74] and diminished visual evoked potentials in the atten-tional blink paradigm in cirrhotic patients without signs of HE[75]. Thus, early deficits of attention associated with liver dysfunc-tion and HE are accompanied by corresponding neurophysiologicalchanges as measured by cognitive evoked potentials. Again, thesereported changes in evoked potentials make correspondingchanges in oscillatory activity likely.

Recently, attentional deficits in HE were for the first time di-rectly related to changes in oscillatory brain activity during anattention task by Kahlbrock and co-workers [63]. In this study,an audiovisual attention paradigm was employed, which requiredshifts of attention between visual and auditory stimuli. In healthysubjects, these attentional shifts are known to be associated withcorresponding strong and gradual modulations of the power ofstimulus-induced oscillatory gamma band activity in early visualbrain areas [76]. Twenty-six cirrhotic patients with HE up to grade1 and eight healthy, age-matched control subjects underwent aclinical assessment including the CFF and completed the attentionparadigm while brain activity was measured using MEG. Duringthe MEG recording, visual and auditory stimuli were presentedsimultaneously. The subjects’ task was to attend to and react tothe visual while ignoring the auditory stimulus or vice versa.Behavioral data as well as oscillatory brain activity were then ana-lyzed, compared between the different attentional conditions, andexamined as a function of HE disease severity. On the behaviorallevel, attentional performance was closely related to HE diseaseseverity as quantified by the CFF. With decreasing CFF, reactiontimes to the auditory stimulus increased, while the percentage ofcorrect responses to the visual stimulus decreased. With respectto oscillatory brain activity, the visual stimulus consistently in-duced a well-known response in terms of strong oscillatory activityin the gamma band over early visual cortices [63,77] in both cir-rhotic patients and healthy controls (see Fig. 2A and B). Impor-tantly, only subjects with a CFF value in the healthy range, i.e.above 39 Hz [14], showed a significant modulation of the powerof visual oscillatory activity in the gamma band with attention,i.e. displayed stronger gamma band activity when the visual stim-ulus was attended compared to when the auditory stimulus wasattended as shown in healthy subjects [63]. Subjects with a CFF be-low 39 Hz did not show this modulation associated with the shift

M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203 201

of attention. In addition, the peak frequency of oscillatory gammaband activity elicited by the visual stimulus showed a strong neg-ative correlation with the CFF, indicating that a slowing of stimu-lus-induced activity in the visual system is closely related toincreasing HE disease severity as assessed by the CFF (seeFig. 2C). Thus, HE patients showed a slowing of attention-relatedoscillatory brain activity and were not able to modulate the powerof activity in the gamma band by top-down attention in this task,which is deemed crucial to enhance the processing of an attendedstimulus [78]. These results thereby demonstrate for the first timethat attentional deficits in HE are related to corresponding patho-logical changes of the associated oscillatory brain activity, in linewith the hypothesis that a slowing of oscillatory brain activityunderlies the different symptoms of HE [9].

Potential mechanisms of altered oscillatory brain activity

According to the current pathogenetic concept of HE (see [79]as well as [80] in this issue), the emergence of disturbances ofoscillatory brain activity precedes the development of HE symp-toms after a cascade of complex events. These involve, amongsta range of other factors like inflammatory processes [81] and ge-netic alterations [82], the accumulation of neurotoxins such asammonia, resulting astrocytal dysfunction due to astrocyte swell-ing and a low-grade cerebral edema associated with generation ofoxidative/nitrosative stress (see [80,83] in this issue), and distur-bances of glio-neuronal communication and neuronal function[84]. These pathophysiological processes are believed to ulti-mately lead to changes of synaptic plasticity and alterations ofdifferent oscillatory networks in the brain, which then cause thelarge variety of symptoms found in patients with HE. However,it is not yet understood how exactly these different pathophysio-logical mechanisms relate to the observed disturbances of oscilla-tory brain activity.

With respect to neurotransmission in the brain, alterations inmultiple neurotransmitters systems have been reported in HE,including the gamma-aminobutyric acid (GABA), glutamatergic,dopaminergic, and serotonergic neurotransmitter systems [85].Changes in both neurotransmitter systems and oscillatory brainactivity patterns are known to be closely linked in neurologicaland psychiatric diseases such as Parkinson’s disease [86–88] orschizophrenia [7,89]. Accordingly, an interrelation between altera-tions in neurotransmission and oscillatory brain activity in HE canbe hypothesized.

As an example, disturbances in inhibitory GABA-mediated neu-rotransmission and the GABAergic tone in the brain have been dis-cussed as a factor in the pathogenesis of HE [90–92], although theunderlying mechanisms are not yet fully understood (see [93] inthis issue). Notably, GABAergic neurotransmission has also beenstrongly implicated in the generation of oscillatory brain activity.GABAergic feedback from interneurons, for example, is believedto be an important factor in the generation of the alpha band activ-ity (e.g. [94]). Regarding oscillatory activity in the gamma band, itwas recently shown that the frequency of neuronal gamma bandactivity as measured using MEG correlates positively with the con-centration of GABA in visual [95] and motor cortices [96] in humansubjects. This provides evidence that the generation of oscillatorygamma band activity, in particular its frequency, depend on theGABAergic neurotransmitter system. It is thus conceivable that al-tered GABAergic neurotransmission is at least part of the mecha-nism of the recently described progressive frequency slowing ofstimulus-induced gamma band activity in early visual cortices inHE [63].

A further important pathophysiological factor in the develop-ment of HE is a low-grade cerebral edema, which is closely

related to the elevated ammonia levels associated with liver dys-function [97]. Based on a recently developed method for themeasurement of localized water content in the brain using MRI[98], it became possible to directly and quantitatively demon-strate this low-grade cerebral edema in patients with HE byrevealing a significant global increase in cerebral water content[99]. Like the reviewed alterations of oscillatory brain activity,spatial redistributions of water content furthermore showed arelation to HE disease severity as quantified by a clinical gradingand the CFF [99]. However, it is again not yet known if and towhat extent the low-grade cerebral edema is causally relatedto the global pathological slowing of oscillatory brain activityin HE.

Interestingly, changes of oscillatory activity in HE appear tobe different from other pathologies as they are specifically char-acterized by a gradual slowing of oscillatory activity in parallelto disease severity. This slowing occurs in different frequencybands and different brain areas, suggesting a global phenomenoninvolving the entire brain. In contrast, changes of oscillatoryactivity in Parkinson’s disease for example are primarily charac-terized by excessive synchrony in specific and stable frequencybands and brain areas, i.e. changes in the beta band (13–30 Hz) and the individual tremor frequency in the motor system[6,88]. Also for schizophrenia, impairments of neural synchronyhave been described for the beta, but specifically and mostlyfor the gamma frequency band (30–80 Hz) [7]. While the patho-physiological mechanisms of the global slowing in HE are notyet understood, they are clearly different from other neurologicaland psychiatric diseases. Similar phenomena can only be ex-pected in other conditions of brain impairment of metabolic ori-gin [31].

Conclusions

Research addressing oscillatory brain activity in HE over the lastdecades has revealed a close association between HE and a slowingof oscillatory activity across different frequency bands and subsys-tems of the brain. Given the large variety of different symptomspresent in HE, global alterations of oscillatory brain activity helpto understand the pathophysiological mechanisms of HE and cannow be seen as a key phenomenon of this disease. In simple words,HE seems to be characterized by a globally ‘slowed brain’. Futureresearch needs to establish the extent to which the reportedchanges of oscillatory activity in HE generalize across different fre-quency bands, brain areas, and experimental settings and if theyare causally linked to the observed symptoms and other patho-physiological markers of HE.

Role of the funding source

A.S. and D.H. acknowledge funding from the German ResearchFoundation (SFB575 and SFB974), M.B. was supported by a MarieCurie Intra European Fellowship (FP7-PEOPLE-2009-IEF-253965).E.S.M. received travel allowances by the Boehringer IngelheimFonds, the Integrated Graduate School 575, and the German Aca-demic Exchange Service (DAAD). The funding sources had noinvolvement in the study design, in the collection, analysis andinterpretation of data, in writing the report or in the decision tosubmit the article for publication.

Conflict of interest

G.K. and D.H. belong to a group of patent holders for a bedsidedevice for determination of critical flicker frequency.

202 M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203

Acknowledgments

The authors would like to thank Lucinda Crook and Louise Mar-shall for skilful language support.

References

[1] G. Buzsaki, Rhythms of the Brain, Oxford University Press, Oxford, 2006.[2] W. Singer, Neuron 24 (1999) 49–65. 111–125.[3] D.L. Schomer, F.H. Lopes da Silva (Eds.), Niedermeyer’s

Electroencephalography: Basic Principles, Clinical Applications, and RelatedFields, Lippincott Williams & Wilkins, Philadelphia, PA, USA, 2011.

[4] P.C. Hansen, M.L. Kringelbach, R. Salmelin (Eds.), MEG: An Introduction toMethods, Oxford University Press, USA, New York, 2010.

[5] H. Berger, Arch. Psychiatr. Nervenkr. 87 (1929) 527–570.[6] A. Schnitzler, J. Gross, Nat. Rev. Neurosci. 6 (2005) 285–296.[7] P.J. Uhlhaas, W. Singer, Neuron 52 (2006) 155–168.[8] A. Schnitzler, M. Butz, J. Gross, G. Kircheis, D. Häussinger, L. Timmermann, in:

D. Häussinger, G. Kircheis, F. Schliess (Eds.), Hepatic Encephalopathy andNitrogen Metabolism, Springer, Dordrecht, The Netherlands, 2006, pp. 245–254.

[9] L. Timmermann, M. Butz, J. Gross, G. Kircheis, D. Häussinger, A. Schnitzler,Metab. Brain Dis. 20 (2005) 337–346.

[10] S. Curran, S. Wilson, S. Musa, J. Wattis, Int. J. Geriatr. Psychiatry 19 (2004) 575–581.

[11] M.L. Daley, R.L. Swank, C.M. Ellison, Arch. Neurol. 36 (1979) 292–295.[12] J.M. Smith, H. Misiak, Psychopharmacologia 47 (1976) 175–182.[13] E. Biecker, I. Hausdorfer, F. Grunhage, H. Strunk, T. Sauerbruch, Digestion 83

(2011) 24–31.[14] G. Kircheis, M. Wettstein, L. Timmermann, A. Schnitzler, D. Häussinger,

Hepatology 35 (2002) 357–366.[15] M.M. Lauridsen, P. Jepsen, H. Vilstrup, Metab. Brain Dis. 26 (2011) 135–139.[16] C. Montoliu, G. Gonzalez-Escamilla, M. Atienza, A. Urios, O. Gonzalez, A.

Wassel, R. Aliaga, R. Giner-Duran, M.A. Serra, J.M. Rodrigo, V. Belloch, V. Felipo,J.L. Cantero, Neuroimage 61 (2012) 1165–1175.

[17] N.J. Shah, H. Neeb, G. Kircheis, P. Engels, D. Häussinger, K. Zilles, Neuroimage41 (2008) 706–717.

[18] R.K. Dhiman, R. Kurmi, K.K. Thumburu, S.H. Venkataramarao, R. Agarwal, A.Duseja, Y. Chawla, Dig. Dis. Sci. 55 (2010) 2381–2390.

[19] M. Romero-Gomez, J. Cordoba, R. Jover, J.A. del Olmo, M. Ramirez, R. Rey, E. deMadaria, C. Montoliu, D. Nunez, M. Flavia, L. Company, J.M. Rodrigo, V. Felipo,Hepatology 45 (2007) 879–885.

[20] P. Sharma, B.C. Sharma, V. Puri, S.K. Sarin, J. Hepatol. 47 (2007) 67–73.[21] F.J. Torlot, M.J. McPhail, S.D. Taylor-Robinson, Aliment. Pharmacol. Ther. 37

(2013) 527–536.[22] G. Eberhardt, Flimmerfrequenz-Analysator Automatische Messmethode

Version 3.00., Dr. G. Schuhfried GmbH, Mödling, Austria, 1994.[23] F. Gundling, E. Zelihic, H. Seidl, B. Haller, A. Umgelter, W. Schepp, C. Dodt, Ann.

Hepatol. 12 (2013) 108–114.[24] J.S. Bajaj, J.B. Wade, A.J. Sanyal, Hepatology 50 (2009) 2014–2021.[25] J.M. Foley, C.W. Watson, R.D. Adams, Trans. Am. Neurol. Assoc. 51 (1950) 161–

165.[26] B.G. Parsons-Smith, W.H. Summerskill, A.M. Dawson, S. Sherlock, Lancet 273

(1957) 867–871.[27] J.M. Guerit, A. Amantini, C. Fischer, P.W. Kaplan, O. Mecarelli, A. Schnitzler, E.

Ubiali, P. Amodio, Liver Int. 29 (2009) 789–796.[28] M.G. Davies, M.J. Rowan, J. Feely, Metab. Brain Dis. 6 (1991) 175–186.[29] M. Senzolo, P. Amodio, M.C. D’Aloiso, S. Fagiuoli, F. Del Piccolo, D. Canova, A.

Masier, M. Bassanello, G. Zanus, P. Burra, Transplant Proc. 37 (2005) 1104–1107.

[30] R.G. Bickford, H.R. Butt, J. Clin. Invest. 34 (1955) 790–799.[31] T. Pang, M. Selvitelli, D.L. Schomer, F.H. Lopes da Silva, Metabolic disorders and

EEG, in: D.L. Schomer, F.H. Lopes da Silva (Eds.), Lippincott Williams &Wilkins, Philadelphia, PA, USA, 2011, pp. 395–411.

[32] M.G. Davies, M.J. Rowan, J. Feely, Scand. J. Gastroenterol. 25 (1990) 1205–1214.

[33] F. Kullmann, S. Hollerbach, G. Lock, A. Holstege, T. Dierks, J. Scholmerich, Eur. J.Gastroenterol. Hepatol. 13 (2001) 513–522.

[34] C.C. van der Rijt, S.W. Schalm, G.H. De Groot, M. De Vlieger,Electroencephalogr. Clin. Neurophysiol. 57 (1984) 423–426.

[35] S.S. Olesen, C. Graversen, T.M. Hansen, R.A. Blauenfeldt, J.B. Hansen, K. Steimle,A.M. Drewes, Scand. J. Gastroenterol. 46 (2011) 988–996.

[36] P. Amodio, F. Campagna, S. Olianas, P. Iannizzi, D. Mapelli, M. Penzo, P. Angeli,A. Gatta, J. Hepatol. 49 (2008) 346–353.

[37] K. Weissenborn, J. Hepatol. 49 (2008) 308–312.[38] S. Montagnese, A. Biancardi, S. Schiff, P. Carraro, V. Carla, G. Mannaioni, F.

Moroni, N. Tono, P. Angeli, A. Gatta, P. Amodio, Hepatology 53 (2011) 558–566.[39] P. Amodio, P. Marchetti, F. Del Piccolo, M. de Tourtchaninoff, P. Varghese, C.

Zuliani, G. Campo, A. Gatta, J.M. Guerit, Clin. Neurophysiol. 110 (1999) 1334–1344.

[40] P. Amodio, F. Del Piccolo, E. Petteno, D. Mapelli, P. Angeli, R. Iemmolo, M.Muraca, C. Musto, G. Gerunda, C. Rizzo, C. Merkel, A. Gatta, J. Hepatol. 35(2001) 37–45.

[41] N. Saxena, M. Bhatia, Y.K. Joshi, P.K. Garg, S.N. Dwivedi, R.K. Tandon, Liver 22(2002) 190–197.

[42] R.R. Reeves, F.A. Struve, R.S. Burke, Clin. EEG Neurosci. 37 (2006) 34–40.[43] S. Montagnese, C. Jackson, M.Y. Morgan, J. Hepatol. 46 (2007) 447–458.[44] P. Amodio, R. Orsato, P. Marchetti, S. Schiff, C. Poci, P. Angeli, A. Gatta, G.

Sparacino, G.M. Toffolo, Neurophysiol. Clin. 39 (2009) 107–115.[45] P. Marchetti, C. D’Avanzo, R. Orsato, S. Montagnese, S. Schiff, P.W. Kaplan, F.

Piccione, C. Merkel, A. Gatta, G. Sparacino, G.M. Toffolo, P. Amodio,Gastroenterology 141 (2011) 1680–1689. e1681–1682..

[46] P. Amodio, A. Pellegrini, E. Ubiali, I. Mathy, F.D. Piccolo, R. Orsato, A. Gatta, J.M.Guerit, Clin. Neurophysiol. 117 (2006) 2243–2251.

[47] P.R. Burkhard, J. Delavelle, R. Du Pasquier, L. Spahr, Arch. Neurol. 60 (2003)521–528.

[48] L. Spahr, F. Vingerhoets, F. Lazeyras, J. Delavelle, R. DuPasquier, E. Giostra, G.Mentha, F. Terrier, A. Hadengue, Gastroenterology 119 (2000) 774–781.

[49] B.T. Shahani, R.R. Young, in: M. Shahani (Ed.), The Motor System:Neurophyiology and Musle Mechanisms, Elsevier, Amsterdam, 1976, pp.301–306.

[50] R.D. Adams, J.M. Foley, Trans. Am. Neurol. Assoc. 74 (1949) 217–219.[51] S. Leavitt, H.R. Tyler, Arch. Neurol. 10 (1964) 360–368.[52] H.R. Tyler, S. Leavitt, J. Chronic Dis. 18 (1965) 409–411.[53] L. Timmermann, J. Gross, G. Kircheis, D. Häussinger, A. Schnitzler, Neurology

58 (2002) 295–298.[54] L. Timmermann, J. Gross, M. Butz, G. Kircheis, D. Häussinger, A. Schnitzler,

Neurology 61 (2003) 689–692.[55] L. Timmermann, M. Butz, J. Gross, M. Ploner, M. Südmeyer, G. Kircheis, D.

Häussinger, A. Schnitzler, Clin. Neurophysiol. 119 (2008) 265–272.[56] M. Butz, L. Timmermann, M. Braun, S.J. Groiss, L. Wojtecki, S. Ostrowski, H.

Krause, B. Pollok, J. Gross, M. Sudmeyer, G. Kircheis, D. Haussinger, A.Schnitzler, Acta Neurol. Scand. 122 (2010) 27–35.

[57] Y. Ugawa, R. Hanajima, Y. Terao, I. Kanazawa, Clin. Neurophysiol. 114 (2003)1278–1284.

[58] J. Artieda, J. Muruzabal, R. Larumbe, C. Garcia de Casasola, J.A. Obeso, Mov.Disord. 7 (1992) 209–216.

[59] P. Amodio, A. Gatta, Metab. Brain Dis. 20 (2005) 369–379.[60] S. Montagnese, P. Amodio, M.Y. Morgan, Metab. Brain Dis. 19 (2004) 281–312.[61] R.A. Blauenfeldt, S.S. Olesen, J.B. Hansen, C. Graversen, A.M. Drewes, Eur. J.

Gastroenterol. Hepatol. 22 (2010) 1323–1330.[62] S.S. Yang, N.S. Chu, Y.F. Liaw, Gastroenterology 89 (1985) 625–630.[63] N. Kahlbrock, M. Butz, E.S. May, M. Brenner, G. Kircheis, D. Haussinger, A.

Schnitzler, Neuroimage 61 (2012) 216–227.[64] S. Della Penna, K. Torquati, V. Pizzella, C. Babiloni, R. Franciotti, P.M. Rossini,

G.L. Romani, Neuroimage 22 (2004) 1438–1446.[65] P. Amodio, S. Schiff, F. Del Piccolo, D. Mapelli, A. Gatta, C. Umilta, Metab. Brain

Dis. 20 (2005) 115–127.[66] K. Weissenborn, K. Giewekemeyer, S. Heidenreich, M. Bokemeyer, G. Berding,

B. Ahl, Metab. Brain Dis. 20 (2005) 359–367.[67] P. Amodio, P. Marchetti, F. Del Piccolo, G. Campo, C. Rizzo, R.M. Iemmolo, G.

Gerunda, L. Caregaro, C. Merkel, A. Gatta, Hepatology 27 (1998) 1517–1523.[68] P. Amodio, P. Marchetti, F. Del Piccolo, G. Sartori, M. Prior, C. Merkel, A. Gatta,

Metab. Brain Dis. 10 (1995) 335–345.[69] K. Weissenborn, S. Heidenreich, J. Ennen, N. Ruckert, H. Hecker, Metab. Brain

Dis. 16 (2001) 13–19.[70] V. Gallai, A. Alberti, S. Balo, G. Mazzotta, C. Clerici, G. Gentili, C. Firenze, A.

Morelli, Acta Neurol. Scand. 91 (1995) 358–361.[71] N. Saxena, M. Bhatia, Y.K. Joshi, P.K. Garg, R.K. Tandon, J. Gastroenterol.

Hepatol. 16 (2001) 322–327.[72] J. Polich, Clin. Neurophysiol. 118 (2007) 2128–2148.[73] I. Ciecko-Michalska, M. Senderecka, J. Szewczyk, A. Panasiuk, A. Slowik, M.

Wyczesany, M. Binder, T. Mach, Adv. Med. Sci. 51 (2006) 273–277.[74] V. Felipo, J.F. Ordono, A. Urios, N. El Mlili, C. Gimenez-Garzo, C. Aguado, O.

Gonzalez-Lopez, R. Giner-Duran, M.A. Serra, A. Wassel, J.M. Rodrigo, J. Salazar,C. Montoliu, Hepatology 55 (2012) 530–539.

[75] I. Ciecko-Michalska, M. Binder, M. Wyczesany, J. Szewczyk, M. Senderecka, J.Wojcik, T. Dziedzic, A. Slowik, T. Mach, J. Physiol. Pharmacol. 63 (2012) 339–346.

[76] N. Kahlbrock, M. Butz, E.S. May, A. Schnitzler, Neuroimage 59 (2012) 673–681.[77] N. Hoogenboom, J.M. Schoffelen, R. Oostenveld, L.M. Parkes, P. Fries,

Neuroimage 29 (2006) 764–773.[78] P. Fries, Annu. Rev. Neurosci. 32 (2009) 209–224.[79] D. Häussinger, H. Sies, Arch. Biochem. Biophys. 536 (2013) 97–100.[80] B. Görg, F. Schliess, D. Häussinger, Arch. Biochem. Biophys. 536 (2013) 158–163.[81] D.L. Shawcross, G. Wright, S.W. Olde Damink, R. Jalan, Metab. Brain Dis. 22

(2007) 125–138.[82] M. Romero-Gomez, M. Jover, J.A. Del Campo, J.L. Royo, E. Hoyas, J.J. Galan, C.

Montoliu, E. Baccaro, M. Guevara, J. Cordoba, G. Soriano, J.M. Navarro, C.Martinez-Sierra, L. Grande, A. Galindo, E. Mira, S. Manes, A. Ruiz, Ann. Intern.Med. 153 (2010) 281–288.

[83] V. Lachmann, B. Görg, D. Häussinger, Arch. Biochem. Biophys. 536 (2013) 143–151.

[84] D. Häussinger, B. Görg, Curr. Opin. Clin. Nutr. Metab. Care 13 (2010) 87–92.[85] D. Häussinger, A.T. Blei, in: J. Rodes, J.-P. Benhamou, A.T. Blei, J. Reichen, M.

Rizzetto (Eds.), The Textbook of Hepatology: From Basic Science to ClinicalPractise, Wiley-Blackwell, Oxford, 2007.

[86] M. Weinberger, J.O. Dostrovsky, NeuroReport 22 (2011) 151–156.[87] P. Brown, Mov. Disord. 18 (2003) 357–363.

M. Butz et al. / Archives of Biochemistry and Biophysics 536 (2013) 197–203 203

[88] C. Hammond, H. Bergman, P. Brown, Trends Neurosci. 30 (2007) 357–364.[89] G. Gonzalez-Burgos, T. Hashimoto, D.A. Lewis, Curr. Psychiatry Rep. 12 (2010)

335–344.[90] S. Ahboucha, Curr. Mol. Pharmacol. 4 (2011) 1–13.[91] O. Cauli, R. Rodrigo, M. Llansola, C. Montoliu, P. Monfort, B. Piedrafita, N. El

Mlili, J. Boix, A. Agusti, V. Felipo, Metab. Brain Dis. 24 (2009) 69–80.[92] E.A. Jones, D.F. Schafer, P. Ferenci, S.C. Pappas, Yale J. Biol. Med. 57 (1984) 301–

316.[93] O.A. Sergeeva, Arch. Biochem. Biophys. 536 (2013) 122–130.

[94] M.L. Lorincz, K.A. Kekesi, G. Juhasz, V. Crunelli, S.W. Hughes, Neuron 63 (2009)683–696.

[95] S.D. Muthukumaraswamy, R.A. Edden, D.K. Jones, J.B. Swettenham, K.D. Singh,Proc. Nat. Acad. Sci. USA 106 (2009) 8356–8361.

[96] W. Gaetz, J.C. Edgar, D.J. Wang, T.P. Roberts, Neuroimage 55 (2011) 616–621.[97] D. Häussinger, F. Schliess, Gut 57 (2008) 1156–1165.[98] H. Neeb, V. Ermer, T. Stocker, N.J. Shah, Neuroimage 42 (2008) 1094–1109.[99] N.J. Shah, H. Neeb, G. Kircheis, P. Engels, D. Haussinger, K. Zilles, Neuroimage

41 (2008) 706–717.