Prevalence and Characteristics of Covert/Minimal Hepatic Enc… : Official journal of the American College of Gastroenterology

INTRODUCTION

Hepatic encephalopathy (HE) is a consequential complication of liver cirrhosis that encompasses a range of neurologic and neuropsychiatric abnormalities. Covert/minimal hepatic encephalopathy (C/MHE) is a less severe manifestation of hepatic encephalopathy that is distinct from overt hepatic encephalopathy (OHE). Minimal hepatic encephalopathy (MHE) is characterized by the presence of neuropsychological and neurophysiological abnormalities in patients who exhibit normal mental and neurological examinations (¹). The challenges linked to the clinical diagnosis of grade I HE, which mainly relies on the expertise of the operator and is challenging to standardize across different centers, have led to a suggestion to combine MHE and grade I HE and classify them as covert hepatic encephalopathy (CHE) (²). The available evidence suggests that C/MHE has the potential to not only result in the manifestation of OHE, but also negatively affect an individual’s driving ability, heightening the likelihood of falls and accidents, and contributing to a decline in quality of life (³). Significantly, C/MHE has been identified as a prognostic indicator of unfavorable outcomes in individuals with cirrhosis, thereby posing a substantial public health concern on a global scale.

Owing to its lack of apparent symptoms, C/MHE is typically not expeditiously diagnosed in the context of clinical practice. According to the guidelines (^4–9), the diagnosis of C/MHE should be performed by using available tests that are standardized to the local population norm. The battery of recommended tests for assessment purposes comprises the psychometric hepatic encephalopathy score (PHES), Stroop test, Scan test, inhibitory control test (ICT), critical flicker frequency (CFF) test, repeatable battery for the assessment of neuropsychological status test, continuous reaction time (CRT) test, electroencephalogram, and evoked potential detection. Several guidelines have proposed that the chosen assessments should be widely recognized and possess adequate clinical utility. Nevertheless, the outcomes acquired through the currently accessible examinations exhibit irregularities, thereby rendering the diagnosis of C/MHE a formidable task. Based on the historical research of various diagnostic approaches for C/MHE and the guidelines proposed during different time periods, paper–pencil-based neuropsychological tests are commonly used as the standard diagnostic tests for C/MHE (¹⁰). Among these tests, the PHES is regarded as the most reliable diagnostic tool because of its exceptional sensitivity and specificity. In fact, it is often referred to as the “gold standard” for C/MHE diagnosis (^11,12).

The evaluation of the prevalence of C/MHE within various cohorts is crucial in establishing effective diagnostic and treatment approaches for C/MHE. The global prevalence of C/MHE in individuals with cirrhosis varies between 20% and 80%, owing to differences in sample sizes and diagnostic methodologies used across studies (¹³). Currently, there is a lack of reported data regarding the precise prevalence of C/MHE globally. The objective of this study was to review the latest evidence and assessed the prevalence and characteristics of C/MHE in patients with liver cirrhosis in an evidence-based fashion.

METHODS

This study conducted a systematic review and meta-analysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Statement (¹⁴). The research methodology was duly recorded with PROSPERO (ID: CRD42023425049).

Search strategy

A systematic search of the literature was performed on April 4, 2023. We searched for relevant studies using the PubMed, Embase, and Cochrane library databases without time or language restrictions. The primary items used in the search pertained to “minimal hepatic encephalopathy,” “covert hepatic encephalopathy,” “liver fibrosis,” or “liver cirrhosis” (see Table S1, Supplementary Digital Content, https://links.lww.com/AJG/D98). Possible data from ongoing studies were also identified on ClinicalTrials.gov. We used both MeSH subject words and corresponding free words for literature retrieval. The references from the full text of selected studies were also reviewed for any missed articles.

Selection criteria

The diagnosis of C/MHE was established in accordance with the recommendations of the 1998 Vienna Congress (¹⁵), guidelines from multiple countries in recent years (^4–6), and the diagnostic methods used in prior research. Specifically, C/MHE was diagnosed based on the presence of abnormal results on at least 2 traditional (paper–pencil-based) neuropsychological tests in patients with cirrhosis who did not exhibit OHE. Subsequently, studies that met the following criteria were included: (i) studies should report the prevalence of C/MHE in patients with liver cirrhosis, (ii) studies should provide information about C/MHE diagnostic methods and thresholds, and (iii) full text should be available with no restriction on the language of publication. Studies that met the following criteria were excluded: (i) studies with a sample size of less than 20 patients; (ii) patients with underlying diseases as acute liver failure, portosystemic bypass or shunting rather than liver cirrhosis; and (iii) publication types as reviews, case reports, editorials, letters, and other studies that do not report on primary data. Publications of interest were selected by 2 independent authors. Non-English studies were translated into English using Google Translate. Discussions were conducted regarding disputes until a consensus was reached.

Data analysis

Baseline information gathered for this study comprised the first author, year of publication, country of publication, time period of research, number of study centers, study design, recruitment setting, method for data collection, method for establishing norms, and specific procedures for C/MHE diagnosis. The primary outcome was the pooled prevalence of C/MHE in patients with liver cirrhosis, ascertained through traditional neuropsychological tests. Furthermore, the study conducted prespecified subgroup analyses to examine the impact of various study characteristics (method of diagnosis, method of definition, World Health Organization region, study design, publication year, and sample size) and patient characteristics (gender, Child-Pugh classification, and etiology of cirrhosis) on the overall effect size. Owing to the recommendations of the practice guideline published in 2014 (⁴), the pooled prevalence of C/MHE determined by both traditional neuropsychological tests and computerized neuropsychological tests (or neurophysiological tests) was also summarized. The secondary outcomes encompassed the identification of distinguishing characteristics between patients with C/MHE and those without. When confronted with multiple publications originating from the same source of information, the data were extracted from the most recent and/or most comprehensive report. In the event that supplementary data were required, the authors were contacted.

Quality assessment

The quality assessment of the articles included in our study was conducted using the Joanna Briggs Institute (JBI) Critical Appraisal Tool (¹⁶). The JBI assessment evaluates the potential for bias in research studies through a 9-item checklist, with each item being assigned a categorical grade of “yes,” “no,” “unclear,” or “not applicable.” A uniform quality assessment was used for all incorporated studies, irrespective of their design, as only baseline measurements were used and longitudinal data were not extracted. Studies with no information or negative results in 2 items were considered low risk of bias. Studies with issues in 3 items had a moderate risk of bias, whereas those with issues in more than 3 items had a high risk (¹⁷). The evaluation of quality is conducted by 2 authors, and any potential discrepancies are deliberated on with the involvement of a third author.

Statistical analysis

The DerSimonian and Laird random-effects model was used in all analyses to account for apparent heterogeneities. The prevalence of C/MHE was summarized through the use of meta-analyses of proportions, and the logit transformation of proportions and the variance of the logit were employed (¹⁸). Study-level factors associated with C/MHE prevalence were explored using meta-regression analyses. Possible publication bias was tested qualitatively by visual inspection of the funnel plot and quantitatively by the Egger test. To ascertain potential risk factors for C/MHE, we compared the demographic characteristics, clinical characteristics, and laboratory data of patients with or without C/MHE. The odds ratio (OR) was calculated for dichotomous variables, and the mean difference (MD) or the standardized mean difference (SMD) was calculated for continuous variables, depending on the method used to express the outcomes. The statistical methods proposed by Luo et al (¹⁹) were used to approximate the sample mean and SD for the continuous variable represented as a median with an interquartile range. The study presented pooled estimates along with their corresponding 95% confidence intervals (CIs). The evaluation of statistical heterogeneity was conducted by means of I² and Cochran’s Q test metrics. All statistical analyses were performed using the Comprehensive Meta-Analysis software (version 3.0; Biostat, Englewood, NJ).

RESULTS

The initial search from PubMed, Embase, and Cochrane Library databases yielded 2,858 articles, whereas an additional 7 records were identified from ClinicalTrials.gov. Of the 2,865 records identified, 1,001 studies were removed as duplicates, and 1,641 were removed after reviewing titles and abstracts. After full-text evaluation, 122 studies were excluded for various reasons (Figure 1), and 101 studies (published between 1997 and 2023) finally met our inclusion criteria. Detailed characteristics of the included studies can be seen in Supplementary Digital Content (see Table S2, https://links.lww.com/AJG/D98). In total, 16,786 patients with liver cirrhosis from 25 countries were included in reporting the prevalence of C/MHE. The included studies were performed in the Western Pacific region (n = 28), the South-East Asia region (n = 25), the European region (n = 24), the Americas region (n = 13), the Eastern Mediterranean region (n = 8), and the African region (n = 2). Each study’s risk of bias quality assessment was completed with the JBI’s critical appraisal tool (see Table S3, Supplementary Digital Content, https://links.lww.com/AJG/D98). The risk of bias among the included studies was high in 3 studies, moderate in 20 studies, and low in 78 studies.

Prevalence of C/MHE overall and by study or patient characteristics

The overall pooled prevalence of C/MHE among patients with liver cirrhosis worldwide was estimated to be 40.9% (95% CI 38.3%–43.5%) (Table 1, Figure 2). There was substantial heterogeneity among studies (I² = 90.69%). For the 63 studies that used PHES to diagnose C/MHE, the pooled C/MHE prevalence was 39.9% (95% CI 36.7%–43.1%, I² = 89.49%). For the 32 studies that diagnosed C/MHE with methods other than PHES, the pooled C/MHE prevalence was 43.0% (95% CI 38.3%–47.8%, I² = 92.61%).

On subgroup analysis by study characteristics, the regional prevalence was 36.5% in the Western Pacific region (95% CI 32.2%–41.0%, I² = 89.89%), 48.4% in the South-East Asia region (95% CI 43.4%–58.4%, I² = 90.27%), 37.2% in the European region (95% CI 32.4%–42.3%, I² = 73.95%), 38.2% in the Americas region (95% CI 31.7%–45.0%, I² = 87.67%), 44.9% in the Eastern Mediterranean region (95% CI 36.2%–54.0%, I² = 89.78%), and 58.8% in the African region (95% CI 39.8%–75.5%, I² = 90.05%), respectively. Further analysis found that this rate was 40.8% (95% CI 37.7%–43.9%, I² = 89.98%) for cross-sectional studies, 34.8% (95% CI 28.1%–42.3%, I² = 86.41%) for cohort studies, 44.2% (95% CI 28.9%–60.6%, I² = 13.23%) for case-control studies, and 47.4% (95% CI 39.4%–55.5%, I² = 95.37%) for randomized controlled studies, respectively. Similar pooled prevalence was observed using the definitions of CHE and MHE (41.0% vs 40.8%). A slightly higher pooled prevalence was observed for the publication year of 2014 or earlier (42.3% vs 40.1%) and for the sample size of less than or equal to 200 (42.5% vs 34.8%). Additional analyses based on different countries and PHES thresholds are available in Supplementary Digital Content (see Tables S4-5, https://links.lww.com/AJG/D98).

On subgroup analysis by patient characteristics, the prevalence of C/MHE in male and female patients was 40.5% (95% CI 37.0%–44.2%, I² = 87.95%) and 42.0% (95% CI 38.1%–46.0%, I² = 76.10%), respectively. We found that the prevalence of C/MHE increased with the severity of Child-Pugh classification, which was 29.3% (95% CI 25.9%–32.9%, I² = 81.99%), 45.9% (95% CI 42.4%–49.6%, I² = 69.04%), and 61.6% (95% CI 56.8%–66.2%, I² = 49.28%) in Child-Pugh A, B, and C patients, respectively. Regarding the etiology of cirrhosis, patients diagnosed with alcoholic liver disease exhibited the highest prevalence of C/MHE at 46.2% (95% CI 41.3%–51.1%, I² = 81.59%). By contrast, patients with autoimmune liver disease demonstrated the lowest prevalence at 27.7% (95% CI 17.1%–41.5%, I² = 60.78%). The odds of C/MHE for patients with viral hepatitis and nonalcoholic fatty liver disease were 40.8% (95% CI 36.7%–45.1%, I² = 83.26%) and 34.6% (95% CI 25.5%–45.0%, I² = 69.79%), respectively. Forest plots for the prevalence of C/MHE overall and by study or patient characteristics can be seen in Supplementary Digital Content (see Figures S1-16, https://links.lww.com/AJG/D98).

Prevalence of C/MHE based on the combination of 2 diagnostic methods

A total of 16 studies (17 cohorts) reported the prevalence of C/MHE diagnosed by 2 tests, namely, PHES in combination with Stroop, ICT, CFF, and CRT (Table 2). The pooled prevalence of C/MHE was 34.0% (95% CI 26.4%–42.6%, I² = 74.06%) for the combination with Stroop, 34.4% (95% CI 17.1%–57.2%, I² = 97.32%) for the combination with ICT, 24.1% (95% CI 17.7%–31.9%, I² = 89.91%) for the combination with CFF, and 23.3% (95% CI 16.7%–31.3%) for the combination with CRT (see Figure S17, Supplementary Digital Content, https://links.lww.com/AJG/D98).

Association of baseline characteristics with C/MHE in patients with cirrhosis

A total of 61 studies reported extractable data on the characteristics of patients with or without C/MHE (Table 3). For the association of dichotomous characteristics with the presence of C/MHE, we found individuals with C/MHE were more likely to have a baseline diagnosis of alcoholic liver disease (OR 1.37; 95% CI 1.18–1.59, I² = 30.62%), a Child-Pugh B classification (OR 1.44; 95% CI 1.24–1.67, I² = 35.47%), a Child-Pugh C classification (OR 2.58; 95% CI 2.09–3.13, I² = 21.76%), a history of OHE (OR 4.11; 95% CI 2.49–6.77, I² = 77.50%), and suffer from ascites (OR 2.27; 95% CI 1.60–3.21, I² = 73.89%). These individuals were less likely to have a Child-Pugh A classification (OR 0.45; 95% CI 0.38–0.54, I² = 46.70%) and suffer from hepatocellular carcinoma (OR 0.65; 95% CI 0.47–0.89, I² = 0%). No statistically significant differences were observed between patients with C/MHE and those without C/MHE in relation to gender of male, viral etiology of cirrhosis, or presence of varices.

For the association of continuous characteristics with the presence of C/MHE, we found that older age (MD 2.93; 95% CI 2.00–3.85, I² = 88.55%), lower education years (MD −0.42; 95% CI −0.86 to −0.26, I² = 79.17%), and higher MELD scores (MD 1.54; 95% CI 1.10–1.97, I² = 90.09%) were associated with the presence of C/MHE. For biochemical indicators, higher ammonia level (SMD 1.57; 95% CI 1.02–2.11, I² = 97.99%), higher total bilirubin level (SMD 0.31; 95% CI 0.24–0.39, I² = 28.08%), and higher international normalized ratio (MD 0.08; 95% CI 0.04–0.12, I² = 84.85%) were associated with C/MHE diagnosis, whereas lower albumin level (SMD −0.54; 95% CI −0.65 to −0.42, I² = 69.00%), lower hemoglobin level (SMD −0.65; 95% CI −0.45 to −0.14, I² = 54.34%), and lower platelets level (SMD −0.15; 95% CI −0.22 to −0.07, I² = 0%) were also associated with C/MHE diagnosis. No associations between the C/MHE presence and the levels of alamine aminotransferase, aspartate aminotransferase, creatinine, or white blood cell were observed. Forest plots for the association of patients’ baseline characteristics with C/MHE can be seen in Supplementary Digital Content (see Figures S18-40, https://links.lww.com/AJG/D98).

We further considered only studies that used PHES as a standard diagnostic method for C/MHE, and the results of these studies were again pooled. The factors associated with C/MHE are completely consistent with the results obtained from the overall sample analysis described above (see Table S6, Supplementary Digital Content, https://links.lww.com/AJG/D98).

Heterogeneity and publication bias

The results of the univariable meta-regression analysis indicate that several parameters, namely, geographic region (P = 0.006), sample size (P = 0.016), mean age (P = 0.004), male proportion (P = 0.002), and Child-Pugh A proportion (P = 0.008) exhibited a significant association with the prevalence rate. In the multivariate analysis, the findings indicated that the prevalence of C/MHE was linked with geographic region (P = 0.014), sample size (P = 0.003), male proportion (P = 0.043), and Child-Pugh A proportion (P = 0.033). These variables collectively accounted for 54% of the observed heterogeneity (Table 4). When considering studies that used PHES as a standardized diagnostic tool for C/MHE, it was observed that the geographic region (P = 0.034) was the only significant factor in explaining the observed heterogeneity (see Table S7, Supplementary Digital Content, https://links.lww.com/AJG/D98).

There was no apparent asymmetry in funnel plots (see Figures S41-42, Supplementary Digital Content, https://links.lww.com/AJG/D98) for the primary outcome of this study. No obvious publication bias was found by Egger’ tests for most outcome measures. However, the analysis revealed the existence of publication bias impacting the outcomes of pooled rates in both the non-PHES–based method of diagnosis (P = 0.014) and the randomized controlled study (P = 0.017) subgroups (Table 1). Furthermore, while comparing the characteristics of patients with or without C/MHE (Table 3), we have also seen the presence of publication bias on the outcomes pertaining to gender (P = 0.014), age (P = 0.041), and MELD scores (P = 0.018).

DISCUSSION

An accurate diagnosis of C/MHE is crucial for establishing the appropriate treatment approach and forecasting a dependable patient outcome. To the best of our knowledge, this is the first systematic review and meta-analysis that has assessed the prevalence and characteristics of C/MHE in patients with liver cirrhosis.

Based on data from 101 studies and 16,786 patients, we found that the prevalence of C/MHE is 40.9% in individuals diagnosed with cirrhosis. The prevalence of C/MHE was similar at 39.9% when using the “gold standard” PHES, the diagnostic test advised by the 1998 Vienna Congress (¹⁵). The findings of the subgroup analysis indicate that the prevalence of C/MHE differed based on various study characteristics. Specifically, C/MHE was found to be more prevalent in the Eastern Mediterranean and South-East Asia regions compared with other regions designated by the World Health Organization. In addition, C/MHE was found to be more prevalent in randomized controlled studies compared with other types of studies, more prevalent in the literature published before 2014 compared with more recent publications, and more prevalent in studies with smaller sample sizes (200 participants or fewer) compared with larger sample sizes. Regarding the clinical features, our study revealed that the prevalence of C/MHE was comparable between genders, but notably, more prevalent in individuals diagnosed with Child-Pugh classifications B and C cirrhosis in contrast to those with Child-Pugh classification A cirrhosis. Our findings indicate that the prevalence of C/MHE was comparatively greater among patients diagnosed with alcoholic cirrhosis and viral cirrhosis. These results indicate that, despite the employment of comparable diagnostic techniques across all the studies analyzed, the prevalence of C/MHE varied among different subpopulations. Furthermore, substantial heterogeneity was noted across studies. The heterogeneity of C/MHE prevalence was found to be influenced by several factors, including geographic region, sample size, mean age, sex ratio, and Child-Pugh classification, as determined through the meta-regression analysis. Considering that geographical factors accounted for a large part of the heterogeneity in this study (R² = 0.42), several factors that have not been extensively documented in the research, included race, language, environment, and economy, among others, could potentially contribute to the variations in the prevalence of C/MHE.

In the guideline issued jointly by the European Association for the Study of the Liver and the American Association for the Study of Liver Diseases (⁴), the suggested approach for diagnosing C/MHE is the utilization of PHES in combination with another neuropsychological or neurophysiological test. We, therefore, reported the prevalence of C/MHE ascertained through the utilization of a dual diagnostic approach. Our study revealed a notable decrease in the pooled prevalence of C/MHE when using a combination of several diagnostic approaches. Furthermore, the value was higher in PHES combined with another neuropsychological test than in PHES combined with a neurophysiological test. This may indicate a possible divergence in the emphasis on neuropsychological and neurophysiological assessments in the identification of C/MHE, particularly in light of the C/MHE’s multifaceted nature (²⁰). Notably, in the latest International Society for Hepatic Encephalopathy and Nitrogen Metabolism guideline (²¹), combining 2 or more tests for the diagnosis of C/MHE is not recommended, mainly due to the potential for underestimating the prevalence of C/MHE without enhancing its predictive utility. Therefore, it is expected that the utilization of diverse diagnostic approaches in forthcoming research will continue to be a topic of debate in the immediate future.

This study conducted a comparative analysis of clinical characteristics in patients without OHE, distinguishing between those with C/MHE and those without. Alcoholic cirrhosis was more common in the C/MHE group compared with those without C/MHE. The consumption of alcohol has been found to have a direct or indirect impact on the brain, leading to structural and functional changes that can result in damage. Furthermore, it is plausible that the clinical manifestations of HE related to alcohol consumption may be attributed to the induction of neuroinflammation and the modifications of the gut microbiome (²²). In addition, some multicenter studies have examined potential risk factors that could affect C/MHE, revealing a notable correlation between the severity of liver disease and occurrences of C/MHE (^23–25). Our results were consistent with such findings, showing that C/MHE patients were more likely to have a higher Child-Pugh classification level as well as a higher MELD score. Regarding laboratory findings, it was observed that patients with C/MHE exhibited elevated levels of blood ammonia, total bilirubin, and international normalized ratio, while displaying decreased levels of albumin, hemoglobin, and platelets in comparison with patients without C/MHE. The findings indicate a correlation between the existence of C/MHE and the extent of hepatic dysfunction. Our study also revealed the influence of age and education level on C/MHE presence. The prevailing belief is that advanced age leads to heightened impairments in select areas of the central nervous system, which have a notable impact on cognitive functioning (²⁶). On the other hand, there exists a correlation between elevated levels of education and heightened metabolic activity and enhanced functional connectivity. This phenomenon may contribute to the superior cognitive abilities observed in individuals with higher levels of education (²⁷). Finally, we also discovered a correlation between the manifestation of C/MHE and additional complications stemming from liver cirrhosis. Some studies have reported that patients with a history of OHE are more likely to develop C/MHE (^28,29), and we have obtained the same results. Interestingly, we found that C/MHE was correlated with the presence of ascites but not with the presence of varices. One plausible hypothesis is that ascites is the most commonly occurring decompensated event in patients and is more strongly linked to the development of HE (^30,31). Moreover, the lack of clear definitions may also have contributed to the lack of significance in our results, as several of the included studies did not distinguish between the severity of varices.

This study had certain limitations. First, we only included studies where traditional neuropsychological tests were used for C/MHE diagnosis. Nonetheless, this decision was made after considering the suggestions put forth by prior guidelines and the diagnostic methods used in most studies. Furthermore, due to the relatively broad search method used in this study, we also found that other point-of-care diagnostic methods for C/MHE are not widely used in clinical research. Second, we did not strictly distinguish between the concept of MHE and the concept of CHE. Although the concept of CHE, including MHE and grade 1 HE, has become more widely used in recent years, we found that MHE was still the most widely used concept in most of the included literature. In addition, the inclusion of HE1 was not clearly described in the content of included studies that embraced the CHE concept, and the subgroup analysis and meta-regression analysis indicated that the reliability of our results might not be affected by the choice of term for “covert” or “minimal.” Third, significant heterogeneity was present in prevalence estimates. We, therefore, conducted multiple subgroup analyses and meta-regression models of predetermined factors to identify the source of heterogeneity. However, residual or unrevealed sources of heterogeneity may still exist. For example, not all included studies provided information on how norms were established, and some studies also adopted norms from non-native populations for the diagnosis of C/MHE, which would have caused possible heterogeneity and bias in our results. Finally, although we present some associations of baseline characteristics with C/MHE in patients with cirrhosis, we cannot establish a causal relationship and these factors warrant further validation through prospective longitudinal studies. Some factors that may be related to the presence of C/MHE, such as patient comorbidities, may still need to be analyzed by data from more studies.

In conclusion, C/MHE is a common complication in patients with liver cirrhosis. Our study reports detailed data on the prevalence of C/MHE as well as clinical features associated with C/MHE. These data possess the potential to provide guidance to physicians with respect to the risk of C/MHE in their patients, as well as in the realm of sample size estimation for prospective investigations.

CONFLICTS OF INTEREST

Guarantors of the article: Jin-Lin Yang, MD, and Li Yang, MD.

Specific author contributions: X.-H.L. and Q.L.: contributed equally to this study. X.-H.L., Q.L., and J.-L.Y.: conception and design. X.-H.L., Q.L., and K.D.: analysis and interpretation of the data. X.-H.L. and Q.L.: drafting of the article. J.-L.Y. and L.Y.: critical revision of the article for important intellectual content. All authors: final approval of the article.

Financial support: This study is funded by the Sichuan Science and Technology Program (2022YFH0003) and the National Natural Science Foundation of China (82173253).

Potential competing interests: Nothing to report.

WHAT IS NEW HERE

• ✓ The prevalence of C/MHE was 40.9% among patients with cirrhosis worldwide.
• ✓ The prevalence of C/MHE was 39.9% among patients with cirrhosis when using the psychometric hepatic encephalopathy score as a diagnostic tool.
• ✓ The presence of C/MHE is associated with a variety of demographic and clinical characteristics.

REFERENCES