Healthy volunteers: The healthy volunteers that were recruited for the control group included people who visited the Health Promotion Center at the First Affiliated Hospital of Anhui Medical University in Hefei, China, for routine health examinations, and through word-of-mouth referrals. The following exclusion criteria were applied for the control group: (1) Presence of chronic liver diseases, neurological or psychiatric diseases, or other diseases that can affect cognitive function; (2) A past history of chronic liver disease, neurologic or psychiatric disorders; (3) Consumption of psychotropic drugs; (4) Alcohol consumption > 50 g/d within the past 3 mo; and (5) Inability to read and write.

Liver cirrhosis group: Consecutive inpatients from the Department of Gastroenterology and Hepatology were recruited. Patients with OHE, which was defined according to the West-Haven criteria[11], were excluded. The diagnosis of liver cirrhosis was based on a combination of physical examination, laboratory tests, medical imaging and endoscopic evidence or on liver histology, if available. The following exclusion criteria were applied for the liver cirrhosis group: (1) A history of OHE, upper gastrointestinal hemorrhage or spontaneous bacterial peritonitis during the past 2 wk; (2) Consumption of lactulose, psychoactive drugs or any antibiotics during the past 2 wk; (3) Presence of neurological or psychiatric diseases, such as Alzheimer’s disease, Parkinson’s disease and nonhepatic metabolic encephalopathy, or a mini-mental status examination (MMSE) score < 25 points; (4) Presence of significant comorbidity, such as heart, respiratory, or renal failure; (5) Presence of hepatocellular carcinoma or other malignancy, previous TIPS or shunt surgery; (6) Alcohol consumption > 50 g/d within the past 3 mo; and (7) Inability to read and write.

All the subjects, both healthy volunteers and patients with liver cirrhosis, were required to have a fair knowledge of numbers and the Chinese alphabet. The research protocol was approved by the ethics committee of the hospital in accordance with the ethical guidelines of the Declaration of Helsinki. Written informed consent to participate was obtained from each subject.

All the five tests of PHES were administered to all the enrolled subjects in the same sequence. The tests were conducted on a one-to-one basis in a quiet room with sufficient light. A specially trained medical doctor assisted the enrolled subjects in finishing these tests.

As some of our enrolled subjects were not familiar with the English alphabet, we replaced the alphabet in NCT-B with the Chinese alphabet in the same order[12]. The results of the NCT-A, NCT-B, and SDT were measured as seconds, including the time needed to correct any errors, and the result of DST was measured as points. The results of the LTT were measured as both the time needed to complete the test (LTTt, seconds) and as the error score (LTTe), LTT = (1 + LTTe/100) × LTTt[13]. Accordingly, a higher result of DST equals better performance, and lower results on the other tests equal better performance. Formulas were constructed to predict the expected results of the five neuropsychological tests. These values were then used as references to which the results from the patients with liver cirrhosis were compared.

The result of DST within ± 1SD from the mean of the control performance was scored as 0 points. Results between -1 and -2SD, between -2 and -3SD and worse than -3SD were scored as -1, -2 and -3, respectively. A result better than mean + 1SD was scored as +1.

The results (NCT-A, NCT-B, SDT and LTT) within ± 1SD from the mean of the control performance were scored as 0 points. Results between +1 and +2SD, between +2 and +3SD, and worse than +3SD were scored as -1, -2 and -3 points, respectively. Those better than mean -1SD were scored as +1 point[10]. The final score of PHES was generated from the sum of the scores of five tests, which ranged between +5 and -15.

On the day of neuropsychological testing, venous blood was taken for routine liver function tests, hematologic parameters and venous ammonia concentration. Venous ammonia was measured within 30 min after blood sampling.

Statistical analyses were performed using the statistical package for the social science (SPSS version 11.0; SPSS, Chicago, IL, United States). Data are expressed as mean ± SD or as proportion. ROC analysis was performed with results of NCT-A, NCT-B and DST comparing to PHES. Continuous and categorical variables were compared using the t test, the one-way ANOVA test, the Mann-Whitney U-test and the χ²-test, respectively. Levene’s test was used in the evaluation of differences in variance. Non-parametric tests were applied if homogeneity of variance assumptions were not met. Multiple liner regression models were used to predict the value of each test for patients with liver cirrhosis. The difference between the expected and observed results for each test was divided by the corresponding SD of the healthy reference population. Kappa statistics were used to study the agreement between the PHES and NCT-A, NCT-B, DST. A two-sided P value < 0.05 was considered significant.

Of 154 healthy volunteers who were recruited, 8 were not able to complete NCT-B and as such, only the remaining 146 volunteers were included. The age and education years of the 146 volunteers were 37.3 ± 10.5 (range 20-67) and 12.0 ± 4.0 (range 2-19) years, respectively, and 99 were men (67.8%). The distribution of subjects according to age was as follows: 20-29 years, 39 (26.7%); 30-39 years, 45 (30.8%); 40-49 years, 40 (27.4%); and ≥ 50 years, 22 (15.1%) (Figure 1). The education years according to age are presented in Figure 2.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Distribution of volunteers according to age.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Comparison of education years between healthy volunteers of various age groups (^aP < 0. 05).

The results of NCT-A, NCT-B, LTT, SDT and DST were 38.289 ± 13.694, 55.846 ± 17.798, 33.287 ± 8.286, 38.035 ± 5.774 and 55.0 ± 14.3, respectively. The results of the five tests were significantly correlated with age and education, and the Pearson’s correlation coefficients are shown in Table 1. In all age categories, the results of all five tests were not significantly correlated with gender (P > 0.05). The variables that affected the results of a neuropsychological test were included in the multiple liner regression models, and the final formulas are shown in Table 2. As shown in Table 2, age and education years were predictors of the results of the five neuropsychological tests in healthy volunteers. As shown in Figure 3, younger age and better education were associated with better DST results.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Distribution of the results from the digit symbol test in healthy volunteers according to age (A) and education years (B).

In the healthy volunteer group, the score of PHES was not correlated with education years (P = 0.992) or age (P = 0.595). Additionally, the PHES did not differ between men and women (P = 0.589).

Of 56 inpatients with liver cirrhosis that were enrolled, 3 were not able to complete NCT-B and thus were not considered further. All tests of the PHES were completed by 53 patients with cirrhosis whose age and education years were 45.6 ± 8.2 years (range 27-62) and 8.2 ± 3.6 years (range 0-15), respectively. The study group comprised 50 (94.3%) men.

The score of PHES in the healthy volunteer group was -0.6 ± 3.7 (median, 0; range -11 to +5). The score of PHES in the liver cirrhosis group was -5.6 ± 4.9 (median, -4; range -13 to +4), significantly lower than that in the volunteer group (Mann-Whitney U = 1476.00, P = 0.000). In the healthy volunteer group, the lower boundary of the 95% range between mean - 2SD and mean + 2SD was -4.0. Using a cutoff for MHE of < -4, 26 of the 53 patients with liver cirrhosis were diagnosed with MHE (49.1%).

The proportion of patients with MHE increased with the increase in the Child-Pugh grade. Specifically, 7 had Child-Pugh grade A (7/24, 29.2%), 14 had Child-Pugh grade B (14/22, 63.6%) and 5 had Child-Pugh grade C (5/7, 71.4%). Compared with compensated cirrhotic patients (Child A), decompensated cirrhotic patients (Child B and C) had a higher proportion of MHE (19/29 vs 7/24; χ² = 6.943, P = 0.008). No differences in age and education years were found between the MHE and non-MHE groups (P > 0.05). Venous ammonia concentration was measured in 26 cirrhotic patients and was found to be similar between the MHE and non-MHE groups (t = 1.086, P = 0.288). In 37 patients with cirrhosis who underwent endoscopic examination, the prevalence of MHE was not associated with esophageal varices (P = 0.584 by Fisher’s exact test). In total, 38 of 53 cirrhotic patients were hepatitis B virus (HBV) positive, while 15 were HBV negative. The prevalence of MHE was similar between the HBV positive and negative groups (χ² = 0.048, P = 0.827) and the prevalence of MHE was not influenced by antiviral therapy. Table 3 shows the characteristics of patients with or without MHE.

International consensus recommends that at least two of the NCT-A, NCT-B, DST and block-design test (BDT) should be used for the diagnosis of MHE[8]. Because the BDT is not easy to use, we compared PHES assessment using NCT-A, NCT-B and DST. Based on the normal range of healthy volunteers, the result of a single test was classified to be abnormal if the score was less than -1 point[10]. Using the NCT-A and DST, we were able to diagnose MHE with a sensitivity of 76.9% and a specificity of 96.3% (AUC = 0.866, K = 0.735), if at least one of the two tests was abnormal (Table 4).

Minimal hepatic encephalopathy (MHE) is widely prevalent in patients with cirrhosis. MHE is associated with impaired health-related quality of life, driving capability and can predict the development of overt hepatic encephalopathy. MHE is not detectable by routine physical or neurological examinations, and a specific neuropsychological/neurophysiological test is needed.

International consensus recommends use of the psychometric hepatic encephalopathy score (PHES) for diagnosing MHE. The PHES has been standardized in Germany, Italy, Spain, India and Korea, but not in China.

This study constructed normal values for the PHES test in healthy Chinese volunteers and evaluated the usefulness of PHES for the diagnosis of MHE in Chinese patients with liver cirrhosis. In the present study, approximately 49% of patients with liver cirrhosis were classified as MHE. Compared to PHES, NCT-A and DST were able to diagnose MHE with a sensitivity of 76.9% and a specificity of 96.3% (AUC = 0.866, K = 0.735).

The results of the five neuropsychological tests of PHES are influenced by age and educational status. Age- and education-corrected nomograms can be used for MHE screening in patients with liver cirrhosis. The proportion of patients with MHE is associated with the severity of liver function. NCT-A and DST are simple and useful tools for the diagnosis of MHE in China.

This is a single-center study from China aiming to validate the use of the PHES for the diagnosis of MHE in cirrhotic patients without overt hepatic encephalopathy. The study, which has created age- and education level-corrected values for the Chinese population, will enable other authors to diagnose MHE in patients with liver cirrhosis and to evaluate interventions.