Assessment and Management of Sleep Disturbance in Cirrhosis

Purpose of Review This review presents an in-depth overview of the sleep–wake phenotype of patients with cirrhosis, together with available pharmacological and non-pharmacological treatment strategies. A set of simple, practical recommendations is also provided. Recent Findings The understanding of the pathophysiology of sleep disorders in this patient population has improved over the past decade, especially in relation to the interplay between homeostatic and circadian sleep regulation. In addition, new tools have been utilised for both screening and in-depth investigation of the sleep–wake profile of these patients. Finally, a number of studies have evaluated the efficacy of novel treatment strategies, often with encouraging results. Summary Since sleep disturbances are common in patients with cirrhosis, more so than in patients with other chronic diseases of similar severity, their assessment should become routine hepatological practice, along with the initiation of adequate treatment.


Introduction
Sleep disturbances are common in patients suffering from liver cirrhosis [1][2][3] and they impinge on their health-related quality of life (H-RQoL) [4], thus representing a challenging topic of clinical relevance.
Difficulties sleeping can be due to several factors: pain and discomfort associated with the disease, poor sleep hygiene, medications that interfere with sleep, pruritus and fatigue (especially in primary biliary cirrhosis) [5], but at the same time they are pathophysiologically related to liver disease, and they represent, at least to a certain extent, a direct consequence of liver dysfunction. Notably, insomnia is also reported in wellcompensated cirrhotics, with no evident reasons for disturbed sleep such as itching, tense ascites, or treatment with diuretics [1,4].
The recognition of sleep disturbance in cirrhosis and the understanding of its underlying pathophysiological mechanisms are of crucial importance in the management of the disease, since they may translate into potential improvement of these patients' quality of life.

Phenotype
Sleep disturbances in cirrhosis were first characterized in 1954, when Sherlock et al. described sleep-wake inversion (i.e. the combination of restless nights and excessive daytime sleepiness) in patients with severe overt hepatic encephalopathy (HE) [3]. These features represent an extreme of the wide range of sleep disturbances exhibited by cirrhotic patients, even by those without signs of overt HE.
B. Sleep-wake signs/symptoms secondary to the aetiology of liver disease: unusual events associated with sleep, such as heavy snoring, apnoea, or periodic limb movements (metabolic cirrhosis); delayed sleep habits (delayed sleep onset, delayed wake-up time), preference for evening activities and insomnia (alcohol-related cirrhosis) [17][18][19].
C. Sleep-wake signs/symptoms related to HE: hypersomnia/somnolence/excessive and inappropriately timed sleepiness/numerous and long daytime naps (up to sleep-wake inversion in severe overt HE).
A summary of the available studies, where phenotypic characteristics of sleep disturbances in cirrhotic patients are described, is provided in Table 1.

Pathophysiology and Relation to Hepatic Encephalopathy
The origin of sleep-wake disturbances in cirrhosis is likely to be multifactorial.
It has been observed that patients with cirrhosis present delayed sleep habits/evening preference [1]. These features are, in turn, associated with impaired sleep quality. Evening type cirrhotic patients present an increased sleep latency and worse sleep quality than their counterparts with earlier sleep times [4].
Based on the assumption that sleep is regulated by two primary components, a homeostatic process, that builds up during wakefulness and declines during sleep [20], and a circadian process, with near 24-h periodicity, Montagnese et al. [9•] have proposed a model to explain the abnormal interaction between homeostatic and circadian components of sleep in cirrhotic patients.
With regard to the homeostatic sleep-wake regulation in patients with cirrhosis, it has subsequently been shown that hyperammonaemia/HE correlates with daytime sleepiness and disruption of sleep architecture. In 2012, Bersagliere et al. [21] demonstrated that the administration of a so-called amino acid challenge (AAC), a mixture of 54 g of amino acids which is used to simulate hyperammonaemia/mild HE, leads to a significant increase in daytime subjective sleepiness and changes in the sleep EEG architecture in cirrhotic patients, with a consequent impairment in their ability to generate restful sleep. These findings support the interpretation of HE as a vigilance defect [4]. Indeed, the absence of daytime sleepiness has been shown to have a high negative predictive value in relation to the occurrence of HE-related hospitalizations, thus suggesting that patients who do not report daytime sleepiness may not need formal neuropsychiatric assessment or particularly close monitoring in relation to their risk of HE [22].
Based on the hypothesis that the sleepiness-inducing effect of ammonium is mediated by adenosine, a known regulator of sleep/wake homeostasis, a recent study performed on hyperammonaemic rats (i.e. fed with an ammoniumenriched diet for 4 weeks) showed, after sleep deprivation, a larger increase in adenosine levels in hyperammonaemic animals in comparison to controls, highlighting the role of adenosine in mediating the sleepiness/sleep-inducing effects of hyperammonaemia [23••]. Interestingly, the increase in putrescine, a molecule that is presumably correlated with neuroinflammation [24] and/or blood brain barrier alterations [25], was higher in the hyperammonaemic group, pointing to a more than additive effect underlying the association between hyperammonaemia/neurocognitive impairment and insomnia.
However, there are multiple, mostly unexplored mechanisms through which cirrhosis and its complications might affect sleep-wake patterns. Further pathophysiological factors may contribute to sleep impairment, as sleep abnormalities have also been reported in patients with little or no evidence of neuropsychiatric disturbance related to HE [2] and in a number of studies no significant relationships have been observed between night sleep disturbances and the presence/ degree of HE [4,22].
Abnormalities in the circadian rhythm of melatonin of both central (reduced cerebral sensitivity to dark/light cues as a consequence of a dysfunction of the central suprachiasmatic nucleus circadian clock) and peripheral origin (reduced melatonin clearance, high daytime melatonin levels and low urinary 6-sulfatoxymelatonin concentration) may play a role, although without offering a comprehensive explanation for the observed sleep-wake disorders [11,26].
Disturbances in the 24-h rhythm of skin temperature have been recently reported in patients with cirrhosis [27]. In physiological conditions, skin temperature starts to decrease in the evening and a redistribution of heat from core to peripheral regions, driven by vasodilatation of distal skin sites, occurs [28][29][30]. The peripheral heat loss is measured by the distal-proximal gradient (DPG), which is an excellent predictor of sleep latency [28,29]. In patients with cirrhosis, an impaired thermoregulation, in terms of absolute values and time-course of DPG, has been reported [27]. These alterations may be explained by the hyperdynamic circulatory syndrome [31], caused by the generalized state of vasodilatation, which may hamper heat dissipation in these patients. Moreover, such abnormalities were shown to parallel the severity of the disease and the associated sleep-wake abnormalities [27].
A model for the interaction and the effects of circadian and homeostatic dysfunction in cirrhosis, based on the current available evidence, is presented in Fig. 1 [20].

Comorbidities
The presence of comorbid conditions should be considered in the assessment of sleep impairment in cirrhotic patients and it represents a non-negligible issue in the evaluation and proper treatment.

Obstructive Sleep Apnoea
An exemplifying condition is represented by obstructive sleep apnoea (OSA) in metabolic cirrhosis. The prevalence of this well-recognized cause of sleep fragmentation and excessive daytime somnolence [32] is increased in obese patients [33]. Emerging data support the hypothesis that OSA-related oxidative stress and hepatic ischemia-reperfusion injury may contribute to the progression from steatosis to nonalcoholic steatohepatitis (NASH) and, ultimately, metabolic cirrhosis [34][35][36][37][38].
Bajaj et al. [39] analysed the complex relationship between OSA and metabolic cirrhosis in terms of sleep quality, daytime sleepiness, cognition and driving simulation, highlighting the fact that OSA is a cause of daytime sleepiness in cirrhosis and concluding that OSA should be considered as a modulator of cognitive function and sleep quality in chronic liver disease. Moreover, Continuous Positive Airway Pressure (CPAP) has been shown to be effective in the treatment of OSA both in patients with and without cirrhosis, significantly improving executive function and sleep quality in both groups [39].

Alcohol Misuse
The negative influence of alcohol consumption on the sleep-wake cycle represents another example of the complex relationship between liver disease and its pathogenetic factors.
In recent years, a better understanding of the relationship between sleep homeostasis and genetic substrates of circadian physiology on one hand, and psychiatric disorders, including alcohol misuse, on the other, has lent support to the idea that alcohol consumption and chronobiological disruption reciprocally interact, thus resulting in a vicious cycle [40,41]. Indeed, reciprocal interactions occur between the circadian system and ethanol consumption at both physiological and genetic levels [42]. Studies on animal models suggest that mutations of circadian clock genes (e.g. Per2 and CLOCK), are associated with increased alcohol intake, which in turn is promoted by perturbed circadian behaviour [43].
In addition, the adverse effects of alcohol consumption on sleep are well characterised both in healthy individuals and alcohol misusers. In social drinkers, acute alcohol intake determines a reduced sleep onset latency and an increased quality (delta power) and quantity of NREM sleep [44,45], whereas severe insomnia, excessive daytime sleepiness, and altered sleep architecture are the consequences of prolonged alcohol misuse [46,47]. Abstinence from alcohol in alcohol misusers results in severe and protracted sleep disruption, with insomnia, sleep fragmentation, and alterations in sleep architecture, that may persist for several years after withdrawal. Furthermore, sleep impairment is a predictive factor of relapse [45].
Therefore, it is reasonable to hypothesize that circadian deregulation (i.e. delayed sleep habits, irregular life-style, lack of sleep-wake, food intake and physical activity routines) may promote alcohol misuse, which in turn disrupts sleep and alters circadian rhythms.

Primary Biliary Cirrhosis
Sleep disorders complained by patients with primary biliary cirrhosis (PBC), a chronic autoimmune cholestatic disorder, may be partially explained by symptoms of liver disease, notably fatigue and pruritus [48,49].
A significant correlation has been observed between the severity of pruritus and sleep impairment. PBC patients who show the greatest perception of pruritus and significantly complain about itching present longer sleep latency and earlier wake-up times, when compared to patients who are not considerably troubled by this symptom [5].
Similarly, fatigue negatively influences sleep-wake habits in PBC patients as it is significantly correlated with sleepiness in the morning and early afternoon hours [5,48].

Hepatitis C Virus Infection
It has been observed that over half of patients with chronic hepatitis C complain of chronic fatigue, depression, reduced quality of life and sleep disturbance, in particular daytime sleepiness and poor sleep quality [50,51]. The combination of such signs and symptoms has been termed "hepatitis C virus (HCV) syndrome" [52,53] and it includes both hepatic and extra-hepatic manifestations of the infection, which are largely independent of the stage of liver fibrosis and the genotype [54].
In a study conducted on HCV-infected patients, with the aim of characterizing sleep disturbances in this population by means of actigraphy, questionnaires and sleep diaries, patients achieved lower quality of life scores and higher scores for depression, fatigue and sleep disturbances, in particular higher nocturnal activity and worse sleep efficiency, than healthy controls. Fatigue and quality of life scores correlated with bad sleep quality and daytime sleepiness [55].
While the above serve as examples of how the aetiology of liver disease and related comorbidities may impinge on the ultimate sleep-wake profile of a patient with cirrhosis, the consideration of aetiology/comorbidities and the formal diagnosis/exclusion of concomitant sleep-wake disorders are recommended (vide infra).

Tools
Sleep health can be assessed by measuring the following parameters: night sleep quality, sleep-wake timing, daytime sleepiness.
The evaluation of sleep-wake behaviour in patients with cirrhosis comprises a heterogeneous mix of methodologies which can be distinguished into subjective and objective/ semi-quantitative.

Subjective Sleep Assessment
This is based on daily sleep diaries and questionnaires. Sleep diaries have been widely utilized for collecting patient's habitual daily routines over time in sleep-wake research. These diaries have been commonly adapted to the population/ features of the study but a comparison between studies is difficult because of the absence of standardization. Recently, a 'Consensus Sleep Diary' with standard indices to evaluate sleep-wake disturbances and particularly insomnia has been approved [56]. There is agreement that such tool should provide information on a set of parameters including night sleep onset latency (SOL), wakefulness after initial sleep onset (WASO), total sleep time (TST), total time spent in bed (TIB), sleep efficiency, and sleep quality or satisfaction, which reflects a subjective global appraisal of each night's sleep [56]. Based on this approach, some authors have utilized this standardized diary for evaluating self-perception of sleep quality in patients with cirrhosis, confirming delayed sleep-wake timing and frequent night awakenings [27].
The Pittsburgh Sleep Quality Index (PSQI) is the gold standard test among self-administered tools utilized for evaluating the subjective sleep quality and sleep disturbances over the preceding month, as well as for distinguishing between 'good' and 'poor' sleepers [57]. It consists of 19 individual items grouped in seven components: sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, sleep medication, and daytime dysfunction, each component being weighted equally on a 0-3 scale. The sum of all components provides a total PSQI score (ranged between 0 and 21), where scores ≥ 5 identifies poor sleepers. The PSQI takes around 10 min to be filled in and 5 min to be scored. Depending on the severity of cirrhosis, time required to complete the test and answering properly all the items can become complicated. In these settings, using simplified and quick the questionnaires, such as the validated Sleep Timing and Sleep Quality Screening Questionnaire (STSQS) [58], becomes a more suitable option for the diagnosis of subjective sleep disturbances in this population. This questionnaire takes approximately 2 min to be completed, without extra time for scoring. STSQS collects information of sleep quality rated on a 1-9 analogue scale (1: 'best sleep ever', 9: 'worst sleep ever') providing also information about habitual sleep timing: bedtime, sleep latency, night awakenings, and wake-up/get-up time. Recently, Gencdal et al. [15] have confirmed a significant correlation between STSQS and PSQI for the diagnosis of sleep disturbances in patients with cirrhosis.
Excessive daytime somnolence is one of the manifestations of the abnormal sleep-wake rhythm in patients with cirrhosis [9•] and it is commonly evaluated by means of The Epworth Sleepiness Scale (ESS) [59]. Subjects rate their likelihood of 'dozing off' in eight different daytime situations: sitting and reading, watching TV, sitting inactive in a public place like a theatre or meeting, as a passenger in a car for an hour without a break, lying down to rest in the afternoon when circumstances permit, sitting and talking to someone, sitting quietly after lunch without alcohol, in a car, while stopped for a few minutes in traffic. The likelihood of 'dozing off' is rated from 0 (unlikely) to 3 (very likely). The higher the ESS score (range 0-24), the sleepier the subject. A score ≥ 11 is considered abnormal. The detection of a strong correlation between poor cognition and excessive daytime sleepiness [6•] highlights the importance of sleepiness evaluation in the analysis of sleep disturbances in this population.

Objective Sleep Assessment
Although the assessment of subjective sleep quality provides valuable information and it is recommended for screening purposes, these tools do not offer quantitative information about sleep architecture or sleep stages.
For in-depth studies, polysomnography (PSG) represents the gold standard diagnostic tool. This technique monitors brain electrogenesis, eye movements, skeletal muscle activity, blood oxygen levels, and heart/breathing rhythms during sleep.
A small number of PSG studies of patients with cirrhosis are available [60][61][62].
A correlation between the severity of HE on one hand and PSG patterns, the duration of sleep and its organization on the other was first identified in 1972 by Kurtz et al. [63] who analysed 18 polygraphic recordings from 15 cirrhotics in different stages of encephalopathy [63].
In more recent years, polysomnographic features and sleep aspects in cirrhotic patients have been extensively characterized, as well as the role of liver dysfunction severity [60][61][62]. Teodoro et al. [60] have showed that cirrhosis is associated with shorter sleep time, reduced sleep efficiency, increased sleep latency, increased rapid eye movement (REM) latency and reduced REM sleep [60], confirming the existence of a disturbed sleep structure in this population.
Nevertheless, PSG is an expensive and labour-intensive tool. In addition, since it initially disrupts sleep, its application normally requires an adaptation night (i.e. first night polysomnography on for the subject to get used to it, second night actual recording) to avoid the well-known 'firstnight effect' [64]. Due to the aforementioned limitations, this technique is generally employed in the research field and currently not included in the routine assessment of sleep quality.
Actigraphy is a semi-quantitative technique that records temporal rest-activity patterns and allows the analysis of micro/macrostructure of nocturnal sleep [65]. Moreover, it is cost-effective compared to PSG, unobtrusive and can be utilized in free-living conditions. The actigraph is a wristwatch like device that includes an accelerometer to monitor the subject's movements. The simple assumption underlying the technique is wake = movement; sleep = lack of movement. Recorded data are analyzed with a software package to estimate sleep parameters such as total time in bed, sleep latency, real sleep time, assumed sleep, wake time after sleep onset, or number of awakenings per night. In liver units, actigraphy has been used in combination with complementary structured interviews or validated questionnaires with the idea of obtaining more detailed information about sleep-wake cycle beyond data provided by body movements, confirming poor sleep quality, prolonged sleep latency and reduced sleep efficiency in cirrhotic patients [66•, 67].
A list of the most important steps that should not be neglected in the routine assessment and management of sleep disorders in cirrhotic patients, including practical recommendations and diagnostic advice provided by the literature of the last few years, is reported below and summarized in the algorithm shown in Fig. 2 [22]. & Detailed and complete pharmacological history should be collected, in order to identify medications that could be responsible for/contribute to sleep disturbance [68]. & Specific enquiry about sleep behaviour, with detailed assessment of sleep quality, sleep timing, and daytime sleepiness should be performed with appropriate tools, i.e. daily sleep diaries and sleep quality questionnaires [58]. These include the PSQI, to assess sleep quality and sleep disturbances over the preceding month, and to differentiate 'good' from 'poor sleepers'; the simplified STSQS, which provides information on both sleep quality and sleep timing, and the ESS, to evaluate excessive daytime somnolence.

Treatment
Since sleep-wake assessment is not part of routine hepatological practice and the pathophysiology of sleep disorders is not completely elucidated [9•], limited data are available on how sleep-wake disturbances should be treated in patients with cirrhosis. Here follows a summary of both non-pharmacological and pharmacological approaches that have been administered and evaluated over the last few years.

Non-pharmacological Approaches
Since cirrhotic patients often suffer from 'pill burden', nonpharmacological treatments have often been preferred to medication [69]. For example, exposure to bright light in the early hours of the morning and avoidance of bright light exposure in the evening are sleep and light hygiene practices that should be encouraged. Drawing parallels between delayed sleep phase syndrome (DSPS), a circadian rhythm sleep disorder characterized by considerable delays in sleep onset/wake times, effectively treated with exposure to bright light in the morning [70], and the circadian deregulation and delayed melatonin response observed in cirrhosis, appropriately timed bright light therapy has been considered as a potentially beneficial non-pharmacological approach to sleep disturbances in cirrhosis [16]. One encouraging case report described in 2011 suggests that treatment with morning light might be effective [71]: an 82-year-old cirrhotic woman with a tendency to sleep-wake inversion underwent controlled lighting administration with a wallmounted lamp, with variable light intensity/spectrum, in order to advance her sleep-wake cycle. Light was more intense and blue enriched in the morning, whereas it became less intense and red enriched during the afternoon/ evening hours. As a result, a progressive improvement of sleep-wake rhythms was recorded, with reduced daytime sleepiness and fewer night awakenings. Nevertheless, a subsequent small randomized control trial of 12 cirrhotic patients did not confirm these preliminary findings, showing no obvious beneficial effect after administration of bright light therapy in terms of sleep onset, quality, and daytime sleepiness. These results are most likely in relation to the severity of disturbance at baseline, since sleep and circadian rhythms in hospitalized, decompensated patients with cirrhosis are extremely compromised [66•].
Mindfulness-based stress reduction (MBSR) is an attractive approach, the beneficial effects of which on sleep disturbances in cirrhotic patients have recently been evaluated [72]: a 4-week dedicated mindfulness and supportive group therapy approach significantly improved depression, sleep quality and H-RQoL in a group of 20 cirrhotic patients. In particular, besides a significant reduction in Beck Depression Inventory (BDI), also PSQI and, consequently, overall H-RQoL scores were significantly improved after treatment. Moreover, this approach significantly reduced the perceived burden, improved depression, and enhanced sleep quality of the caregivers of patients with cirrhosis.
Progressive neuromuscular relaxation training, lavender warm sponge bath and footbath were shown to improve sleep-related symptoms in patients with chronic liver disease, producing a significant reduction in self-rating scales of sleep [73].

Pharmacological Treatments
Sleep disorders are often inadequately treated in cirrhosis because of the role of the liver in metabolising psychoactive medication and the fact that the therapeutic-toxic threshold of these medications is very narrow in cirrhosis [72,74]. These patients are fairly sensitive to psychoactive medication, and the risk of precipitating severe HE when administering psychoactive drugs for sleep impairment is not negligible [9•]. For example, when an "aetiological" treatment was attempted by Spahr et al. [75] who administered the histamine H1 blocker hydroxyzine to patients with minimal HE and sleep impairment, despite an increase in sleep efficiency in treatment-group vs placebo-group, one patient developed an acute episode of HE reversible upon cessation of treatment, highlighting the need of caution when prescribing this type of drugs [75].     The impairment of melatonin metabolism observed in cirrhosis, as a result of the dysfunction of the central SCN circadian clock [26,76], may represent an interesting field of research for future therapeutic strategies. Nevertheless, several variables need to be considered, such as overnight melatonin clearance, delays in the nocturnal rise of melatonin and in its time to peak [11,26,77]. Therefore, the benefits and risks of melatonin administration are worthy of specific, formal studies.
Lactulose is an effective treatment for HE. Singh [79].
Recently, rifaximin has been shown to improve objective sleep architecture parameters on 24-h PSG, with increased REM sleep after a 28-day course of treatment, although no changes were detected in the subjective quality of sleep and sleepiness [80].
Another interesting subject of study is represented by the association of ammonia-lowering [L-ornithine-L-aspartate (LOLA)] and vigilance-enhancing medication (caffeine). A pilot study [81] conducted on six healthy volunteers and six cirrhotic patients sought to assess the effects of the administration of an amino acid challenge (AAC-see above), alone and in combination with either LOLA and caffeine. Results showed that both the administration of LOLA and caffeine could contain the post-AAC increase in capillary ammonia levels in healthy volunteers. In this group, the administration of caffeine also resulted in a reduction in subjective sleepiness, in line with previous studies [82], and in the amplitude of the EEG on several frontal/temporal-occipital sites. This was taken as evidence that the association of ammonia-lowering and vigilance-enhancing medication is worthy of further study. The timing of caffeine administration is also worthy of study in patients with cirrhosis.
Hypnotics should be used with caution, especially in decompensated patients, and chosen amongst those with negligible hepatic metabolism, short half-life, no active metabolites, and limited lipophilia, to avoid prolonged intracerebral action [9•].
A list of the available studies, evaluating specific nonpharmacological approaches or pharmacological treatments, and collective reviews, is provided in Table 2.

Future Perspectives
In conclusion, over the last few years, progress has been made in terms of both the description of the sleep-wake abnormalities associated with cirrhosis, and the understanding of their pathophysiology. However, this has not yet translated into well-defined therapeutic strategies or well-designed treatment trials.
In parallel, a phenomenal amount of progress has been made by basic chronobiologists in the understanding of the functions of the liver clock and its role, dependence and ability to dissociate from the main, cerebral circadian clock [83,84]. Meanwhile, the circadian and sleep-wake consequences of the timing of meals [85] and physical activity [86] are being better defined.
The 2017 Nobel prize in Physiology/Medicine went to the discovery of the molecular mechanisms controlling circadian rhythms. Evidence of the health consequences of both circadian deregulation and sleep loss [87,88] is emerging at an impressive rate. Such exciting discoveries have remarkable translational, clinical and therapeutic potential, and hepatology may represent a favourable environment in this respect. We should attempt to rise to the challenge.

Compliance with Ethical Standards
Conflict of Interest Chiara Formentin, Maria Garrido, and Sara Montagnese each declare no conflicts of interest.
Human and Animal Rights and Informed Consent This article does not contain any studies with human or animal subjects performed by any of the authors.
Open Access This article is distributed under the terms of the Creative Comm ons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.