From
Journal of Viral Hepatitis
Neuropsychological Tools in HepatologyA Survival Guide for the Clinician
S. Montagnese; S. Schiff; M. De Rui; M. M. E. Crossey; P. Amodio; S. D. Taylor-Robinson
Abstract and Introduction
Abstract
Neuropsychological assessment has three main applications in clinical hepatology: (i) to detect, grade and monitor liver failure-related cognitive alterations in end-stage liver disease (hepatic encephalopathy), (ii) to substantiate complaints of attention or concentration difficulties in patients with non-cirrhotic chronic hepatitis C viral infection, and (iii) to screen patients who are being considered for liver transplantation for early signs of dementia. However, there is limited agreement on how cognitive assessment should be conducted in these patients, and how results should be interpreted and used to implement clinical decisions. In this review, we summarize the available literature on neuropsychological dysfunction in patients with cirrhosis and with chronic hepatitis C viral infection and provide some guidance on how to utilize neuropsychological assessment in practice.
Introduction
Patients with end-stage liver disease have long been known to exhibit cognitive deficits in a variety of forms, which have been termed hepatic encephalopathy (HE). This spectrum ranges from minimal impairment, only detected on neuropsychological assessment (so called 'minimal hepatic encephalopathy' [MHE] – formerly labelled 'subclinical'),
[1,2] through to overt hepatic encephalopathy (OHE), which can manifest as mild confusion and behavioural change through to deep coma.
[3,4] More recently, it has become apparent that patients with chronic hepatitis C virus (HCV) infection complain of a variety of neuropsychiatric sequela including confusion ('brain fog'), anxiety and depression, in the absence of significant liver disease.
[5] Most studies on these issues have been performed by hepatologists who have acted as amateur psychologists. In this review, we discuss some of the pitfalls of neuropsychological assessment and put the published studies on HE and non-cirrhotic chronic HCV infection into context.
Neuropsychological Assessment
The human mind produces an enormous set of complex behaviours. These events represent the net result of what cognitive psychologists refer to as
mental operations and, as a whole, make up an individual's
cognitive ability. This can be separated into domains, such as language, memory and attention (
Appendix 1), and each domain can be assessed by specific
neuropsychological tests.
Many conditions, including liver disease, impinge on the brain and on the patients' capacity to perform mental operations, thus on their cognitive ability.
Neuropsychological assessment is the diagnostic process aimed at measuring an individual's cognitive ability by means of specific neuropsychological tests.
[6] Neuropsychological evaluation is not based on the patient's history or complaints, or on his/her carer's observations, but relies on direct assessment of the patient's performance.
Neuropsychological diagnosis is based on the systematic exploration of cognitive performance, with a view to defining mental functioning in patients with potential neurological or psychiatric illnesses.
In general, neuropsychological evaluation aims at defining impairment in aspects of cognition (i.e. memory, attention, language, executive function, etc.), which are thought to be sensitive to certain pathologies, and thus to be reliable indicators of their presence or absence. In this context, the scores obtained by a patient in a set of neuropsychological tests define the
neuropsychological profile of the patient. The concept of neuropsychological profiling is an important one because, despite inter-individual differences (
vide infra), a neuropsychological profile can, to some extent, 'define' a disease. For example, in individuals with suspected cerebro-vascular or Alzheimer-type mild cognitive impairment, isolated memory deficits on neuropsychological testing would support the diagnostic hypothesis.
Neuropsychological assessment requires meticulous attention to confounding variables. For example, physical pain may easily distract the patient from test instructions and test performance. As a consequence, the patient may be wrongly qualified as having attention or concentration deficits. Other variables that often confound neuropsychological assessment are mood (such as anxiety or depression) and sleep-wake disturbances. In addition, the evaluation of cognitive function needs to take into account socio-biological variables, such as age, sex, level of education and occupation, which are all known to affect performance. In recent years, considerable interest has arisen in the novel concept of
cognitive reserve, or the degree of neuroprotection that derives from chronic enhancement of mental, social and physical activity.
[7] Cognitive reserve can heavily impinge on performance in one or more cognitive domains. For example, professional sportspeople have been shown to have above average skills in certain cognitive domains, in relation to the type of sport practiced and the level of expertise.
[8] Along the same lines, a London taxi-driver, who has learnt the geography of a big city almost by heart, may perform well on tests involving spatial attention abilities, despite a progressive disease that is impinging on his cognitive ability, including spatial attention. The same applies to accountants being tested on calculus and so on and so forth. This is because when an individual is tested, his/her baseline, premorbid performance is generally unknown. In addition, neuropsychological tests are scored in relation to reference, normative data obtained from large, supposedly healthy groups, stratified by country of origin or ethnicity, sex, age and level of education. Variables which make up the cognitive reserve, such as occupation and hobbies, are virtually impossible to account for in any meaningful way. Of importance to the hepatologist, the impact of cognitive reserve on both specific neuropsychological functions
[9] and activities of daily living (i.e. driving)
[10] has been documented also in patients with cirrhosis.
It is therefore obvious that to obtain clinically meaningful results, a neuropsychological assessment needs to be performed in a systematic fashion by adequately trained, experienced personnel.
Neuropsychological Assessment in Hepatology
In routine hepatological practice, neuropsychological assessment has different aims:
- To detect, grade and monitor liver failure-related cognitive alterations (HE).
- To substantiate complaints of 'brain fog' or 'concentration difficulties' in patients with HCV infection. In these patients, it has also been used to monitor treatment-related cognitive alterations (i.e. the side effects of interferon).
- To screen individuals undergoing a transplant work-up for early signs of neuropsychiatric impairment (i.e. early dementia).
Hepatic Encephalopathy
HE is a reversible neuropsychiatric syndrome, which occurs in patients with cirrhosis and/or significant portal-systemic shunting.
[3] HE manifests as a spectrum of change, which may or may not be clinically apparent. The neuropsychiatric changes detected in patients with cirrhosis by clinical examination are collectively termed OHE. The patient typically appears 'slow', somnolent, and sometimes euphoric, while in other instances obviously confused and disorientated. An involuntary jerking of the outstretched hands (
flapping tremor) can be detected. The diagnosis can be difficult, especially in mild forms, thus a history obtained from a family member is valuable.
[11] The detection of an obvious precipitating factor (i.e. infection or dehydration) can also help. It is important to notice how patients with cirrhosis are prone to the development of metabolic encephalopathies other than HE, such as those related to hyponatremia, hypoglycaemia and nutritional deficits, which should be ruled out.
The term MHE is used to describe the occurrence of neuropsychological and/or neurophysiological abnormalities (i.e. slowing of the electroencephalogram) in patients with cirrhosis who appear neuropsychiatrically normal on clinical examination.
[1,12,13] The prevalence of MHE varies considerably, depending on the patient population studied and the tools utilized for the diagnosis. MHE predicts the subsequent occurrence of OHE,
[14] impinging on the patient's ability to perform complex tasks, such as driving,
[10,15,16] and usually impairing quality of life.
[17] Therefore, it is important to screen for, recognize and adequately treat this syndrome.
Neuropsychological Dysfunction in Hepatic Encephalopathy: The Underlying Theory
Over the years, several different neuropsychological tests have been utilized to describe the effect of HE on various cognitive domains, to 'quantify' OHE and to diagnose MHE.
[1,18] A review of the literature between 1970 and 2004 retrieved over 80 different tests, often exploring more than one cognitive domain.
[19] Still, the large majority explored attention, motor speed and executive function (
Appendix 1); a smaller number explored memory and the remainder other cognitive domains.
Attention has a crucial role within the cognitive system,
[20] and it impinges on other cognitive functions. It is common experience that if a person does not pay attention to something, he or she may not be able to remember it later on, which does not necessarily imply that memory is not functioning properly. In addition, attention is the net result of three separate sub-functions: vigilance, spatial attention and selective attention (
Appendix 1). Vigilance has not been directly measured in patients with HE, but there are both clinical
[21] and electrophysiological data
[22] to suggest that it is affected.
[23] Selective attention seems to be more compromised than spatial attention in these patients, possibly explaining why they may be easily distracted
[24] and why there is an inability to cope with conflicting tasks, or tasks which require to 'switch' between different sets of information.
[25] The simplest, paper-and-pencil switching test is Trail-making test B: the patient is asked to connect numbers and letters in alternating order (1-A-2-B-3-C-4-D…), thus continuously switching between the alphabetical and the numerical sequence. Trail-making test B has been widely used in patients with cirrhosis and has proven more sensitive in detecting HE-related cognitive changes than Trail-making test A,
[26] where the patient is only asked to connect numbers from 1 to 25. This is in line with results obtained by computerized neuropsychological tests, which can be helpful in dissecting a complex task in its components, and measure the time taken to perform each of them.
[23] However, the delay in reaction time also depends on the complexity of the task, being progressively more marked in
simple reaction times (i.e. reaction to a visual stimulus by pressing a key), compared with
choice reaction times (i.e. pressing different keys in response to different visual stimuli), compared with
reverse selection choice reaction times (i.e. reaction/inhibition to different visual stimuli).
[27] In agreement with this concept, and based on the profound delay observed in patients with cirrhosis on reverse selection choice tasks,
[28] a computerized inhibition test was recently proposed for MHE screening.
[29,30] Along similar lines, a motor delay which was proportional to the cognitive load/difficulty of the task was observed in a working memory test.
[31]
Limited consensus or solid data exist on which neuropsychological tests should be used to diagnose MHE and/or to quantify mild OHE. The approaches utilized so far to address this problem can be criticized as they have been 'circular' (i.e. choosing neuropsychological tests to measure functions that have been shown to be abnormal in patients with HE by neuropsychological tests), 'statistical' (i.e. choosing tests that are more often abnormal in these patients) or 'comparative' (i.e. choosing tests that often agree/overlap with other, non-neuropsychological measures of HE, regardless of the absence of a gold standard). Despite these problems, a set of reasonable suggestions were first provided by an expert panel at the 11th World Congress of Gastroenterology in 1998, and published in 2001.
[12] The experts suggested that the presence of MHE should be defined by either: (i) two abnormal tests of a set of four (Trail-making Tests A and B, Symbol Digit, Block Design), or (ii) an abnormal Psychometric Hepatic Encephalopathy Score (PHES), based on a battery including Trail-making Tests A and B, Digit Symbol, Serial Dotting and Line Tracing.
[32,33] The first criterion was derived from a previous, substantially arbitrary definition of MHE as a condition in which two psychometric tests are abnormal;
[34] no information was provided on the total number/type of tests to administer. The second criterion was more solid, as the PHES battery was derived using discriminant analysis, in a formal attempt to maximize the separation of patients with cirrhosis with no electroencephalographic abnormalities from normal subjects and from alcohol misusers with no liver disease.
[32,33]
The experts of the 11th World Congress of Gastroenterology panel recommended the use of neuropsychology as an alternative to that of neurophysiology, and, where possible, the combination of both techniques.
[12]
More recent guidelines were produced by the International Society for Hepatic Encephalopathy and Nitrogen Metabolism (ISHEN), separately for neuropsychology
[35] and neurophysiology.
[36] The former endorsed either the use of PHES or that of a more comprehensive battery, the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS),
[37] which had been used in studies of patients with end-stage liver disease on a transplant list.
[38] However, there is limited experience with RBANS in hepatology; the battery is time-consuming and it seems to exceed the requirements for standard, routine HE assessment.
The ISHEN guidelines did not formally endorse any of the computerized tests utilized so far in patients with cirrhosis, which include the Posner Test,
[24] the Scan Test,
[31] the Inhibitory Control Test
[29] and the Cognitive Drug Research battery.
[39] These have generally been chosen based on known cognitive features of HE and cover psychomotor speed, selective attention, inhibition and working memory. However, their use has not yet become widespread, and available normative data are limited.
Neuropsychological Dysfunction in Hepatic Encephalopathy: Actual Practice
Two of the authors of this review, SM (hepatologist) and SS (neuropsychologist), run a joint clinic for the assessment of neuropsychiatric dysfunction in Internal Medicine at the University of Padova, Italy. Our experience is that real patients are generally more complex than those enrolled in trials and observational studies and that attribution of reported or detected neuropsychological abnormalities to liver disease may not necessarily be straightforward. A standard HE evaluation at the Padova clinic includes: (i) a doctor's review, clinical and neurological examination, (ii) a structured interview with the neuropsychologist, (iii) a comprehensive neuropsychological evaluation, including both paper-and-pencil and computerized tests, (iv) an electroencephalogram, (v) a full blood count, liver function tests, renal function and electrolytes, venous ammonia, vitamin B12 and thyroid function tests. The case reports in Figs 1 & 2 illustrate how even this may not be enough and a definitive diagnosis may require further laboratory tests, cerebral imaging and re-evaluation after treatment. While it is difficult to imagine that such a comprehensive system may be routinely implemented in busy clinics and district hospital wards, it is equally important to remember that inaccurate neuropsychological diagnoses may impinge on quality of life,
[17] post-transplant neurological outcome
[40] and even survival.
Click Image To Enlarge
Figure 1.
Example of neuropsychiatric evaluation in a patient with alcohol-related cirrhosis. PHES, Psychometric Hepatic Encephalopathy Score, CRP, C-reactive protein, EEG, electroencephalogram, MRI, magnetic resonance imaging.
Click Image To Enlarge
Figure 2.
Example of neuropsychiatric evaluation in a patient with hepatitis C virus-related cirrhosis. PHES, Psychometric Hepatic Encephalopathy Score, EEG, electroencephalogram, TIPS, transjugual intrahepatic portal-systemic shunt.
Hepatitis C Virus and the Brain
In recent years, there has been growing evidence that neuropsychological changes in patients with chronic HCV infection may occur long before the development of significant liver fibrosis/cirrhosis.
[41] A considerable proportion of patients with HCV infection complain of 'brain fog', weakness, fatigue and difficulties concentrating.
[5] These problems do not seem to relate to HCV genotype or HCV replication, and their exact pathophysiology remains unknown, although a number of potential mechanisms have been proposed.
[42]
Neuropsychological Dysfunction in Hepatitis C Virus: The Underlying Theory
In 2001, Forton
et al.[43] described the presence of cerebral metabolic abnormalities – elevated choline/creatine ratio – in the frontal white matter and basal ganglia of HCV-infected patients by use of proton magnetic resonance spectroscopy. These alterations were not detected in either healthy age-matched controls or in patients with hepatitis B.
[43] The following year, the same group of authors showed significant impairment in concentration and working memory in 27 patients with HCV infection and active viral replication, compared with 20 controls and 16 anti-HCV antibody positive, but HCV RNA-negative patients.
[44]
Since these seminal observations, a number of studies have been published on neuropsychological performance in patients with hepatitis C.
[45–51] Unfortunately, these have often included (i) patients with cirrhosis, (ii) patients who had acquired the infection by previous intravenous drug misuse, (iii) patients who had co-infections, (iv) patients who had had treatment with interferon not long before the study, (v) patients who were on psychoactive medication, or (vi) patients who complained of significant fatigue, which may all impinge on cognitive performance. More importantly, neuropsychological examination was not necessarily conducted in a systematic way, with different studies testing different, often isolated cognitive functions, making the results even more heterogeneous. Finally, very stringent cut-off values for abnormality were utilized (i.e. 1.5 SD from the reference values), at least in some instances.
[48] Nevertheless, impairment was reported in sustained attention, executive function, verbal learning, verbal recall, memory and working memory.
[5] To date, only two studies have been completely negative. In the first, HCV-positive patients were selected amongst supposedly healthy individuals, screened for purposes of blood donation, although the medical provenance of all subjects in the study was not fully elicited;
[51] in the second, they were children/teenagers, probably with a shorter interval between infection and cognitive assessment, compared with all other studies.
[52]
Interestingly, limited research has been performed to compare the neuropsychological burden of HCV-related cirrhosis and that of cirrhosis of other aetiologies, although clinical experience does not suggest there are significant differences.
Therefore, while patients with HCV infection have fairly definite magnetic resonance spectroscopy alterations,
[43,44,47,48] which are unlikely to be explained by previous drug misuse or HE, their neuropsychological profile remains poorly defined. This is certainly worthy of further investigation, and so are the complex interactions between neuropsychological impairment, disease perception, mood, fatigue, daily functioning and quality of life.
[53] These aspects are likely to become key to treatment choices for HCV-infected patients in the near future.
Neuropsychological Dysfunction in Hepatitis C Virus: Actual Practice
The practical difficulties of neuropsychological evaluation in patients with hepatitis C are even greater than those described for HE, for two main reasons: (i) the neuropsychological profile of these patients is not defined, and there is no consensus on which tests should be used, (ii) these patients often have additional reasons for cognitive dysfunction, such as psychoactive treatment, co-infections, previous drug/alcohol misuse and even the innate impulsiveness that characterizes addictive personalities.
[54,55] This trait can affect both the approach to and the performance on several tests.
Within this very complex setting, also attention to common confounders needs to be meticulous, as illustrated in Fig. 3.
Click Image To Enlarge
Figure 3.
Example of neuropsychiatric evaluation in a patient with chronic hepatitis C infection. PHES, Psychometric Hepatic Encephalopathy Score, Lp(a), lipoprotein(a), HDL/LDL, high/low-density lipoprotein, MRI, magnetic resonance imaging.
Conclusions
When patients with cirrhosis or with non-cirrhotic chronic hepatitis C infection present with cognitive disturbance, it is often difficult to establish a causal relationship between their liver disease and their neuropsychological problems. Alcohol-related cerebral damage, bacterial infection, malnutrition, anaemia, vitamin deficiencies and renal failure can all be present in these patients, together with unrelated diseases of high prevalence, such as hypothyroidism and vascular/degenerative dementia.
For these reasons, neuropsychological batteries should be chosen, which include: (i) specific tests for the liver disease-related cognitive alterations of interest, and (ii) tests aimed at excluding gross, unrelated cognitive damage. The latter are even more important for patients who are being considered for transplantation, as subsequent treatment with immunosuppressants, especially calcineurin inhibitors, may unmask or accelerate cerebro-vascular disease.
[40] Thus, when we evaluate a patient with cirrhosis, the battery will need to cover attention, visual-constructive, visual-motor and motor abilities (like PHES). If the patient is older than 55 years or is on a transplant list, long-term memory and executive function should also be tested, as an abnormal performance on Digit Symbol or Trail-Making test B, which are part of PHES, is also observed in early dementia. Similarly, visual-constructive impairment is common to both HE and Alzheimer's type dementia. However, the latter is also characterized by disturbance in long-term memory and praxis (
Appendix 1). The presence of risk factors for cerebro-vascular disease, such as systemic hypertension, diabetes and hypercholesterolaemia, should prompt even more accurate differential diagnosis, as cerebro-vascular disease is associated, like HE, with extra-pyramidal motor disturbance. Previous substance or alcohol misuse is associated with working memory, planning, inhibition and decision-making issues, which can be picked up not only on test results but also on the patient's approach to tests performance.
[55] Therefore, while hepatologists require simple tools for routine neuropsychiatric screening of their patients,
[56] accurate neuropsychiatric diagnosis often requires the interaction between hepatologists and neuropsychologists, meticulous history taking, laboratory tests, neurophysiology, brain imaging and re-evaluation over time. A consensus is required as to a practical approach for neuropsychological testing in both HE and in chronic HCV infection. While no consensus exists on simple tests to use in the clinic, the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL) have commissioned a joint working party on this subject. The results of the working party's report are awaited in the next year.
http://www.medscape.com/viewarticle/762582
h of hospitalization in a patient with a recent contact with the Health Care system (i.e. when the patient has attended a hospital or a hemodialysis clinic, or has received intravenous chemotherapy during the 30
