Incomplete improvement of visuo-motor deficits in patients with minimal hepatic encephalopathy after liver transplantation



Previous studies have suggested reversibility of minimal hepatic encephalopathy in patients with liver cirrhosis after liver transplantation (LT), however, this topic is controversially discussed. We investigated this issue in a prospective study on liver cirrhotic patients listed for LT. Patients were investigated before and after liver transplantation (on average 21 months later) using a neuropsychological test battery which measured visuo-constructive and visuo-motor ability, verbal fluency, and memory function. To assess visuo-motor and visuo-constructive functions, we performed 4 tests: Rey Complex Figure Test copy, trail making tests A and B, and digital symbol test. The average percentile score of the tests, arbitrarily named the visuo-motor and visuo-constructive performance score (VMCP), was calculated. After LT, the patients did not demonstrate a significant increase of VMCP (P = .29) and additionally showed significantly lower VMCP score (P = .041) compared to control group. Analysis of individual responses showed that only 7 of 14 patients improved their VMCP values after LT. These data indicate that the cirrhosis-associated visuo-motor deficits subside or disappear only in some of the patients after LT, whereas a significant number of patients show no improvement of the visuo-motor and visuo-constructive function. We concluded that monitoring of cognitive and visuo-motor functioning is important for the post-transplant rehabilitation of patients with liver cirrhosis. (Liver Transpl 2004;10:77–83.)

Mild cognitive and psychomotor deficits in patients with liver cirrhosis constitute a neuropsychiatric syndrome known as minimal hepatic encephalopathy.1–5 Recent data have suggested that minimal hepatic encephalopathy may be a mild clinical form of metabolic hepatic encephalopathy and is distinguished from other clinical forms—episodic and persistent encephalopathy—primarily by the absence of overt clinical neurological symptoms, such as asterixis, flapping tremor, and pronounced fluctuations of vigilance.

The mechanisms and prognosis of minimal hepatic encephalopathy are not completely established and the data on this issue are controversial.3, 5 It has been recently found that patients with minimal hepatic encephalopathy demonstrate cognitive deficits, affective symptoms, changes of behavioral patterns, and bioregulation disorder.5–8 Comprehensive neuropsychological testings in these patients have shown pronounced deficits in attention, visuo-constructive ability, and psychomotor function, and less prominent deficits in other neuropsychological domains, such as memory and verbal ability.9–11

The most substantial deterioration has been found in the psychomotor domain and in particular in visuo-motor tasks.1, 2, 12–16 Executive tasks involving visual search, motor sequencing, and visuo-motor coordination are frequently impaired. This impairment has been suggested to be characteristic for minimal hepatic encephalopathy.9, 17, 18 Magnetic resonance-spectroscopy and positron emission tomography studies have shown metabolic alterations in different brain areas, such as occipital and cingulate regions which are supposed to be involved in the control of attention and in visuo-motor coordination.19

Several studies have recently investigated the effect of liver transplantation (LT) on cirrhosis-associated neuropsychiatric deficits. Patients with liver cirrhosis after LT have shown substantial improvement of symptoms of minimal hepatic encephalopathy.20, 21 Other studies have also shown improvement of quality of life after LT that coincided with a decrease of the deficits in visuo-constructive and verbal ability.22 In a longitudinal study of neuropsychiatric outcome in patients after LT, Tarter and coworkers have reported a slow-rate post-transplant improvement and a near-normal post-transplant functioning.23, 24

On the other hand, a considerable number of earlier and recent studies have observed a significant rate of persistence of different cirrhosis-associated neuropsychiatric deficits. Lazeyras et al. have recently demonstrated the persistence of the hepatic parkinsonism18 associated with basal ganglia hyperintensity after LT. The findings of other investigators have also suggested an incomplete improvement of neuropsychological and neurological deficits specific for hepatic encephalopathy after LT.24

The emerging evidence of possible persistence of some cirrhosis-associated neuropsychiatric deficits within months after LT raises the question whether these deficits are reversible. The aim of this prospective study was to investigate the long-term changes in visuo-motor function in patients with liver cirrhosis after liver transplantation.


LT, liver transplantation; CI, calcineurin inhibitors; SD, standard deviation; TMT, trail making test; DST, digit symbol test; RCFT, Rey complex figure test; VMCP, visuo-motor and visuo-constructive performance; CYA, cyclosporin A; AZA, azathioprine; MMF, mycophenolate mofetil; TAC, tacrolimus.

Patients and Methods


Fourteen subsequent patients with biopsy-proven liver cirrhosis treated at the outpatient clinic (mean age 55.6 [SD = 7.92]; all right-handed; 7 females, 7 males) and listed for liver transplantation (LT) were included in this study. Patients underwent a comprehensive hepatologic work-up before the entry in the waiting list for LT and gave their informed consent. The main clinical parameters of the patients are summarized in Table 1. At the time of pre-transplant testing patients had no clinical or laboratory signs of acute inflammation, severe gastrointestinal bleeding, anemia, electrolyte abnormalities, or renal insufficiency. The following exclusion criteria were applied in this study: 1) overt hepatic encephalopathy (persistent or episodic, clinical grade II-IV) as revealed by a standard clinical neurological investigation;25 2) significant cortical atrophy or other structural brain changes as revealed by conventional neuroimaging; 3) regular use of psychotropic substances, such as benzodiazepines; 4) known major psychiatric disorder, as defined by DSM-IV classification; and 5) less than 6 months of complete alcohol abstinence. At the time of post-transplant testing all the patients were in stable condition and, in particular, showed no clinical or laboratory signs of acute inflammation, anemia, electrolyte abnormalities, or other current severe clinical problems. They presented with therapeutic serum levels of immunosuppressive drugs regarding the calcineurin inhibitors, cyclosporin A, and tacrolimus, and had no acute transplant rejection episodes. The immunosuppressive therapy after LT consisted of a calcineurin inhibitor-based regimen in association with a rapid-taper steroid protocol (within three months) and azathioprine or mycophenolate mofetil (see Table 1).

Table 1. Demographic and Clinical Characteristics of the Patients
 Clinical ParametersNumber of Patients
  1. Abbreviation: SD, standard deviation.

Child-Pugh stageA/B/C3/6/5
Duration of cirrhosisLess than 5 years10
 Between 5 and 10 years2
 More than 10 years2
Post-transplant medication  
 CYA and AZA5
 CYA and MMF5
 TAC and MMF4
Post-transplant neuropsychological testsMonths between pre- and post-transplant testsMonths, Mean (SD/Range)
  25.4 (7.0/14–36)
 Months after transplantation21.0 (7.7/11–33)

Twenty two age-matched healthy persons (mean age 52.4 [SD = 6.49]) constituted the control group which was not different from the patient group in respect to the number of education years, sex, and handedness.

Psychometric Tests

The neuropsychological test battery applied in this study was based on the recent data on sensitivity and specificity of psychometric tests in hepatic encephalopathy.9 This test battery included the following neuropsychological tests: trail making tests A and B (TMT A and TMT B26), digit symbol tests27, Rey Complex Figure test (RCFT copy, RCFT recall, RCFT delayed recall28), tests for short- and long-term memory;29 semantic verbal fluency test (VFT), as well as a verbal intelligence test.30

To minimize the age-dependent differences in the subject's performance the age-adjusted percentile scores based on the large population data27–30 were assigned for each test parameter. One score point was assigned for every 10 percentiles producing the maximum of 10 score points per test. For example, the performance ranged by 30% was scored with 3 points and the performance ranged by 60% was scored with 6 points.

To obtain the measure of overall visuo-motor and visuo-constructive performance, we calculated the average percentile score of the 4 selected tests: RCFT copy, TMT A, TMT B, and digit symbol test. The average score of these tests was arbitrarily named the visuo-motor and visuo-constructive performance (VMCP) score. The complete psychometric battery was repeated in the post-transplant phase of treatment on average 21 months after liver transplantation (see Table 1).

Statistical Analysis

The comparison between groups was performed using non-parametric statistics as follows: Mann-Whitney U test for comparison between controls and patients and Wilcoxon matched pairs test for comparison between pre- and post-transplant scores within the group of patients. The analysis of association between neuropsychological and clinical parameters as well as between the improvement rates of psychometric parameters and clinical parameters was done using the Spearman correlation coefficient. Data processing and statistical analysis were performed using the SPSS 10.0 software package.


The group analysis of the pre-transplant data showed significant deficits in visuo-motor function, visuo-constructive ability, and verbal fluency in patients with cirrhosis as compared with the age-matched control group (Table 2). The difference between group mean values of the score points ranged between 1.59 score points (Rey Complex Figure Test/copy) and 3.27 score points (verbal fluency test) and showed statistically significant differences in single visuo-motor tests, in the overall VMCP score, and in the verbal fluency test (see Table 2). Other neuropsychological functions such as short- and long-term memory as well as verbal intelligence did not significantly differ between patients and controls.

Table 2. Mean Values of Percentile Scores of Different Neuropsychological Parameters in Controls and Patients Prior to and After Liver Transplantation
  • *

    Mann-Whitney U test.

  • **

    Wilcoxon matched-pairs test.

  • Abbreviations: VMCP, visuo-motor and visuo-constructive performance; RCFT-copy, Rey Complex Figure Test/copy; TMT A and TMT B, trail making tests part A and B; DST, digit symbol test; VFT, verbal fluency test; SD, standard deviation.

Controls vs. pre-transplantP-value*0.0040.0610.0060.0180.0480.003
Pre-transplant vs. post-transplantP-value**0.2860.9520.0910.4180.8320.344
Controls vs. post-transplantP-value*0.0410.1080.1800.2520.0760.936

The analysis of the effects of LT on the neuropsychological parameters showed a tendency toward a post-transplant increase of the mean scores of patients' group concerning all parameters that were decreased pre-transplant. The improvement of these scores ranged between 0.14 score points (digit symbol test) and 3.0 (verbal fluency test). Statistically significant improvement of performance, however, was observed only in two recall parameters of the Rey Complex Figure Test (recall and delayed recall; P = .005 and P = .01). Table 2 summarizes the group means and standard deviations of the control group, pre-transplant and post-transplant group, as well as the corresponding P-values.

The analysis of the correlation between improvement rates of single tests showed that post-transplant changes were highly consistent between different visuo-motor tests. The improvement rates of the visuo-motor tests significantly correlated as follows: TMT A vs. TMT B (r = 0.658, P = .010), digit symbol test vs. TMT B (r = 0.545, P = 0.044), and TMT B vs. RCFT copy (r = 0.655, P = 0.011).

The average visuo-motor performance (VMCP) score also demonstrated a tendency toward improvement after LT. The group mean value of the VMCP score increased in patients from 3.96 score points pre-transplant to 4.65 score points post-transplant. The mean group improvement of the VMCP corresponded to the average increase of the age-adjusted percentile by 6.9 %, but this change did not achieve statistical significance (Fig. 1 and Table 2).

Figure 1.

Visuo-motor and visuo-constructive performance (VMCP). Average percentile score of four tests: RCFT copy, TMT A, TMT B, and digit symbol test. Means and confidence intervals of 95%. P-values of Mann-Whitney U test (controls vs. patients pre-transplant and controls vs. patients post-transplant) and Wilcoxon matched-pairs test (patients pre-transplant vs. patients post-transplant).

The analysis of the individual long-term responses to LT showed no improvement in 7 of 14 patients in respect to their VMCP scores (Fig. 2). The increase of the average VMCP score in those patients who demonstrated post-transplant improvement ranged between 1.38 and 4.13 score points and was on average 2.52 score points. In those patients who showed no post-transplant improvement, the average decrease was −1.13 score points and ranged between −0.13 and −2.13 score points. In Fig. 2, the pre-transplant VMCP scores are plotted against the post-transplant scores. This figure visualizes the distribution of the improvement rates within the group of patients and shows that there are two opposite tendencies in the post-transplant course of treatment: toward improvement and toward deterioration of the visuo-motor performance. No differences in the extent of post-transplant improvement were found in patients with both high and low pre-transplant VMCP scores (Fig. 2).

Figure 2.

Comparison of the visuo-motor and visuo-constructive performance (VMCP) score pre- and post-transplant. Average pre-transplant VMCP score of individual patients are plotted versus post-transplant VMCP scores. The dotted line (- - -) represents the imaginary position of the points if no post-transplant change would have taken place. This line divides patients into two subgroups – those who improved (above the dotted line) and those who did not improve (below the dotted line) their VMCP scores after liver transplantation.

The analysis of the clinical factors in the pre-transplant and post-transplant course of the disease with regard to their influence on the rate of post-transplant improvement was performed using Spearman correlation statistics. The duration of liver cirrhosis and its severity (as determined by Child-Pugh classification), presence and severity of esophageal varices, history of gastrointestinal bleedings, and previous episodes of hepatic encephalopathy did not show any significant impact on the rate of post-transplant improvement. Furthermore, no influence of the etiology of cirrhosis on pre-transplant level and post-transplant improvement of visuo-motor functioning was found. Additionally, a comparison between 2 subgroups of patients—those who improved and those who did not improve their post-transplant scores—was performed. These 2 groups did not differ in duration and severity of liver cirrhosis, presence and severity of esophageal varices, history of gastrointestinal bleedings, previous episodes of hepatic encephalopathy, time between pre- and post-transplant neuropsychological testing, education, and age. Concerning the clinical parameters of the post-transplant course, statistical analysis showed no correlation of the post-transplant improvement rate with different immunosuppressive treatment regimes, re-transplantation, other major clinical events and duration of the time period between first and second neuropsychological tests.


Recent studies have suggested high clinical relevance of minimal hepatic encephalopathy as this condition has a substantial impact on the level of every day functioning and quality of life in patients with liver cirrhosis.31, 32 Changes of neuropsychological deficits in response to liver transplantation (LT) should be taken into account because residual neuropsychological deficits, such as visuo-motor slowing may significantly influence the everyday functioning of post-transplant patients and reduce the general clinical and psycho-social benefit of liver transplantation. The importance of this issue grows due to the constantly increasing waiting time for liver transplantation and because a neuroprotective treatment for patients with cirrhosis-associated cerebral deficits is not currently available.

Recent studies on the effects of LT on minimal hepatic encephalopathy have shown substantial improvement of neuropsychological parameters,12, 22, 33, 34 reduction of the encephalopathy-associated changes of EEG10, 35, and improvement of cirrhosis-associated brain metabolic alterations detected with the proton magnetic resonance spectroscopy.12, 20 On the other hand, several previous reports have shown incomplete reversibility of the crucial cirrhosis-associated deficits.18, 24

In this long-term study liver cirrhotic patients showed a tendency toward post-transplant improvement in different tests measuring visuo-motor function, visuo-constructive ability, and verbal fluency tests, as well as in the average visuo-motor performance score. This finding is consistent with previous reports that have shown post-transplant improvement of different neuropsychological and clinical parameters.12, 21, 36 According to the hypothesis of complete reversibility of the minimal hepatic encephalopathy, significant improvement of the visuo-motor deficits after liver transplantation could be expected. Our data showed the opposite effect: the patients showed no significant group effect of LT in improvement of their visuo-motor function and failed to reach the functional level of the control group after LT. The incomplete improvement of the patients group in our study obviously results from the inhomogeneous response of the patients to LT. Seven of 14 patients did not improve their visuo-motor performance and even demonstrated lower VMCP scores in the post-transplant tests. Those patients who improved their post-transplant scores showed this effect to a large extent, which is sufficient to be considered as complete reversibility. On the other hand, half of the patients showed reduction of their VMCP scores and, therefore, further deterioration of the neuropsychological functioning after liver transplantation.

Due to the small number of patients investigated in this study, definitive conclusions about the rate of non-improvement of neuropsychological functioning after liver transplantation are not admissible. Nevertheless, the following statistical evaluation can be made. Using standard distributional assumptions (binomial distribution) a 95% confidence interval of the non-improvement rate (7 of 14 patients) can be evaluated as having a range from 23% to 77%. This means that in a larger study of similar design a substantial part of at least 23% of patients would not show any improvement of neuropsychological functioning after LT. Although these findings seem at first sight surprising, they are consistent with those previous reports, which have shown incomplete reversibility of cirrhosis-associated motor deficits18 and incomplete improvement of functional psychological capacity after liver transplantation.24

Why did some patients show no improvement after transplantation? It could be suggested that cirrhosis-associated factors influencing post-transplant improvement may originate from both pre-transplant and post-transplant phases of treatment. Etiology and severity of cirrhosis did not show any significant influence on the rate of post-transplant improvement in this study. Correlations between the rate of post-transplant improvement and other clinical factors, such as different immunosuppressive regimes, re-transplantation, and other major clinical events in the course of post-transplant treatment were also not found in this study.

It could not be definitely ruled out that changes detected in this study may be partly caused by the natural inter-subject variability of neuropsychological functioning. On the other hand, the large extent of the observed post-transplant changes (25 percentile in average for improvement and 11 percentile for deterioration) points toward the involvement of important pathophysiological mechanisms. It is well established that in the long-term course of cirrhosis, metabolic factors, such as hyperammonemia, deteriorate brain functions. Liver transplantation leads to significant reduction of neurotoxic factors and to improvement of the functional ability of the brain. Some patients appear to be more susceptible for cirrhosis-associated neurotoxic factors. The underlying mechanisms of this susceptibility are still unknown. The recently published study by Kril and Butterworth37 has shown diencephalic lesions and cerebellar degeneration in a significant number of patients with liver cirrhosis. Our data are consistent with these findings and suggest that although complex brain functions, such as visuo-motor coordination, may obviously be restored after liver transplantation, in some cases, deterioration may become irreversible.

Due to some limitations, such as the relatively small group size, this study cannot provide definitive evidence concerning the effects of LT on neuropsychological functioning. Nevertheless, several important conclusions can be drawn. This study shows that the cirrhosis-associated visuo-motor deficits are not completely reversible and may even worsen in some patients during the course of the post-transplant treatment. On the other hand, even the patients with a relatively high pre-transplant level of functioning may substantially improve their performance after LT, which suggests that potentially all patients suffer from cirrhosis-associated deterioration of the brain function of different severity and may need neuro-protective treatment in the pre-transplant phase. Additionally, this study points toward neuro-degenerative processes underlying the cerebral pathophysiology of minimal hepatic encephalopathy. These processes may cause irreversible neuropsychological impairment in patients with liver cirrhosis. Further studies should be performed to search for predictors of the irreversibility of neuro-psychiatric deficits in patients with liver cirrhosis.


The authors thank Dr. G. Kemmler for support in the statistical analysis and Simone Auffinger for her technical assistance.