The PHES battery does not detect all cirrhotic patients with early neurological deficits, which are different in different patients

Abstract

Background and aims: The psychometric hepatic encephalopathy score (PHES) is the "gold standard" for minimal hepatic encephalopathy (MHE) diagnosis. Some reports suggest that some cirrhotic patients "without" MHE according to PHES show neurological deficits and other reports that neurological alterations are not homogeneous in all cirrhotic patients. This work aimed to assess whether: 1) a relevant proportion of cirrhotic patients show neurological deficits not detected by PHES; 2) cirrhotic patients with mild neurological deficits are a homogeneous population or may be classified in sub-groups according to specific deficits.

Methods: Cirrhotic patients "without" (n = 56) or "with" MHE (n = 41) according to PHES and controls (n = 52) performed psychometric tests assessing attention, concentration, mental processing speed, working memory and bimanual and visuomotor coordination. Heterogeneity of neurological alterations was analysed using Hierarchical Clustering Analysis.

Results: PHES classified as "with" MHE 42% of patients. Around 40% of patients "without" MHE according to PHES fail two psychometric tests. Oral SDMT, d2, bimanual and visuo-motor coordination tests are failed by 54, 51, 51 and 43% of patients, respectively. The earliest neurological alterations are different for different patients. Hierarchical clustering analysis shows that patients "without" MHE according to PHES may be classified in clusters according to the tests failed. In some patients coordination impairment appear before cognitive impairment while in others concentration and attention deficits appear before.

Conclusions: PHES is not sensitive enough to detect early neurological alterations in a relevant proportion of cirrhotic patients. Oral SDMT, d2 and bimanual and visuo-motor coordination tests are more sensitive. The earliest neurological alterations are different in different cirrhotic patients. These data also have relevant clinical implications. Patients classified as "without MHE" by PHES belonging to clusters 3 and 4 in our study have a high risk of suffering clinical complications, including overt HE and must be diagnosed and clinically followed.

Fig 1. Patients “without” MHE according to PHES show impaired performance in some psychometric tests.

The tests were performed by patients classified as “without” or “with” MHE according to PHES and by controls. (A) Percentage of individuals of each group that fail the indicated number of tests or more. (B) Percentage of tests failed by each group (mean±SD). (C) Each bar represents the percentage of tests failed by each individual subject. (D) Percentage of total cirrhotic patients (“without” or “with” MHE according to PHES) failing the indicated tests: Oral SDMT test, total items; d2 test, TOT: total correctly processed; bimanual coordination; concentration, measured with the d2 test and in visuo-motor coordination.

Fig 2. Percentage of individuals of each group showing impaired performance in each test.

The tests were performed by patients classified as “without” or “with” MHE according to PHES and by controls. The percentage of individuals of each group showing impaired performance in Oral SDMT test, total items (A); d2 test, TOT: total correctly processed (B); bimanual coordination (C); concentration, measured with the d2 test (D) and in visuo-motor coordination (E), are given. The failure criterion was to obtain a score equal or lower than the mean minus 2 times the standard deviation of the control group in A, B and D and equal or higher than the mean plus 2 times the standard deviation of the control group in C and E.

Fig 3. Patients “without” MHE according to PHES show impaired performance in Oral SDMT test, d2 test and bimanual and visuo-motor coordination.

The mean scores of each group in Oral SDMT test, total items (A); d2 test, TOT: total correctly processed (B); bimanual coordination (C); concentration, measured with the d2 test (D) and in visuo-motor coordination (E), are given. Values are the mean±SEM of 56 patients classified as “without”, 41 “with” MHE according to PHES and 52 controls. Values significantly different from controls are indicates by asterisks: *p<0.05; ***p<0.001.

Fig 4. Dendrogram showing the clustering of all individuals in main groups and sub-groups.

All subjects included in the study have been arranged according to the similarity-dissimilarity of their performance in the combination of tests indicated using hierarchical clustering analysis. This analysis provides clusters including subjects more closely related between them than subjects assigned to a different cluster. These clusters are visualized in the dendrogram shown. The first column shows the number identifying each subject, which have been coloured according to their classification by the PHES: control (black), without MHE (red) or with MHE (green). These colours are repeated in the second column to facilitate its identification. The tests failed by each individual are indicated by blue colour in the corresponding box. The following tests have been included: PHES, critical flicker frequency (CFF), the congruent, neutral and incongruent tasks of the Stroop test, Bimanual and visuo-motor coordination, d2 test, TOT: total correctly processed and d2-CON: concentration, measured with the d2 test; Oral SDMT test, Digit Span and Letter-number test.

Fig 5. Follow-up of clinically relevant outcomes in the different clusters of patients.

(A) Percentage of patients with complications in each group. Different clinical complications are expressed as: EX-1, exitus by causes related to liver disease; EX-2, exitus by causes non-related to liver disease; HE, overt hepatic encephalopathy; D, other complications (ascites, variceal bleed, hepatocellular carcinoma, spontaneous bacterial peritonitis…). (B) Percentage of patients without complications in each group. In clusters 3 and 4 only patients without MHE (according to PHES score) are shown. Tests failed by patients in each group are shown in parentheses. Differences of all clusters were performed by Fisher’s exact test by comparing with cluster 5 (patients without MHE with no failed tests). *p<0.05; **p<0.01; ***p<0.001.