Neuropsychiatric Symptoms in Hepatitis C Patients Resemble Those of Patients With Autoimmune Liver Disease but Are Different From Those in Hepatitis B Patients

Meike Dirks; Kim Haag; Henning Pflugrad; Anita B. Tryc; Ramona Schuppner; Heiner Wedemeyer; Andrej Potthoff; Hans L. Tillmann; Kajetan Sandorski; Hans Worthmann; Xiaoqi Ding; Karin Weissenborn

Disclosures

J Viral Hepat. 2019;26(4):422-431. 

In This Article

Results

Questionnaires

Patients and controls did not significantly differ in age and education (Table 1 and Table 3). As expected, however, considering the known sex distributions of the different liver diseases represented in this study, there was a significant difference regarding the sex distribution in our patient groups (P = 0.001).

The median Fatigue Impact Scale Score (FIS) was highest, and thus worst, in the patients with overlap syndrome—followed by those with HCV infection, and lowest in the HBV patients. All patient groups differed significantly from controls (Figure 1, Table S1). Bilateral comparison, in addition, showed a significant difference between HBV and HCV patients (P = 0.01).

Figure 1.

Fatigue Impact Scale Score (FIS) (A), Hospital Anxiety (HADS A) (B) and Depression (HADS D) Score (C), Short-Form-36 (SF36) physical (phys) (D) and mental score (E) in all patient groups compared to controls. Significant differences compared to the control group are marked by * if P < 0.05, ** if P < 0.01. Significant differences between HCV and HBV patients are marked by # if P < 0.05 and ## if P < 0.01

All patient groups demonstrated higher HADS anxiety and depression scores than the controls, and this was significant for all groups but the fairly small group of patients with overlap syndrome. There was no significant HADS score difference between the patient groups. Of note, the median HADS scores of the patients were within or just beyond the normal range.

The SF-36 physical score was significantly worse than in controls in all patient groups except of HBV, while the mental score was significantly worse in all patient groups (Table S1). In a bilateral comparison, HCV patients scored significantly lower than HBV patients in both SF-36 scores while there was no significant difference considering the other groups (Table S1).

Besides the comparison of median scores, we also assessed the frequency of abnormal results for the different questionnaires in the different patient groups studied. Thereby we found that 2/3 of the patients with HCV infection or overlap syndrome achieved FIS scores above 45—the cut-off between normal and abnormal results—compared to about 1/3 of the patients with AIH, HBV or PBC. HCV patients did significantly worse than HBV (P = 0.003) and PBC patients (P = 0.006) in this respect. In contrast, the rate of abnormal HADS scores was quite similar between the patient groups. The physical score of the SF36 was less often below the 25th percentile in the HBV patients than in the other groups, while there was no apparent difference regarding the mental score. For both scores, however, no significant group difference could be shown.

In the HCV and the AIH group, the FIS correlated significantly with the HADS and the SF-36 scores, while in the PBC group, the FIS correlated significantly only with the SF-36 scores. Of note, the results of the HBV patients showed a different pattern: similar to the controls, in who the FIS correlated only with the HADS depression score, they showed no significant correlation between FIS and any of the respective scores (Table 2). In addition, FIS did not correlate with the FIB4 score in any group. The correlation analysis was not performed for the overlap group due to the small number of subjects in this group.

Attention Tests

The attention test results are summarized in Table 3.

In the cancelling d test, the HBV patients showed significantly more errors (in percentage) than controls, while there was no significant difference between the patient groups. In contrast, there was a trend for a decreased test performance regarding the number of correctly processed items in the HCV patients compared to controls (P = 0.005). Again the difference to the other patient groups was not significant. In the TAP battery, the flexible reaction time was significantly prolonged in the PBC and the AIH/PBC overlap group, and in tendency also in the HCV group compared to controls. The reaction time in the HBV group was very close to that in controls (median 787 vs 733 s). Considering Bonferroni correction, the number of errors in the flexible reaction time test did not significantly differ between patients and controls. In the intermodal comparison test, the HCV group and the overlap group tended to prolonged reaction times compared to controls (P = 0.009 and P = 0.003, resp.); the number of errors and misses (not shown) did not differ between groups. Reaction times, number of misses and number of errors in the go/no go test and the divided attention test did neither differ between patient groups and controls, nor between patient groups (data not shown). In contrast, significant differences were observed regarding the (simple and warned) reaction time in the alertness test between HCV patients and controls. Bilateral comparison showed that the HCV patients did also far worse than the HBV patients (simple RT: P = 0.005, warned RT: P = 0.01) (Figure 2). Phasic alertness showed no significant group differences (data not shown). The Kruskal-Wallis test indicated significant group differences for the reaction time (P = 0.004) as well as the number of misses (P = 0.001) in the working memory test but not for the number of errors (not shown). Bilateral comparisons between the patient groups and the control group showed a significantly prolonged reaction time in the HCV patients exclusively combined with a significantly increased number of misses (P = 0.000). Again the HCV patients did significantly worse than the HBV patients (RT: P = 0.002; Figure 2). Applying multiple comparisons, there was no significant difference between the patient groups regarding the percentage of abnormal results in the different attention tests when excluding the small group of patients with overlap syndrome. But there was a remarkable difference in the rate of abnormal working memory test results between HBV and HCV patients especially with regard to reaction time (16.6% vs 29.2%; P = 0.02) and misses (27.7% vs 41.7%).

Figure 2.

Simple (A) and warned reaction time (B), reaction time in the working memory test (C) and number of misses in the working memory test (C) for all patient groups compared to controls. Significant differences compared to the control group are marked by * if P < 0.05, ** if P < 0.01. Significant differences between HCV and HBV patients are marked by # if P < 0.05 and ## if P < 0.01

Memory Tests

All patient groups except of the HBV group did significantly worse than the controls in the Luria list of words test considering the sum score (verbal learning; run 1-5) (Table 3) while there was no significant difference between the patient groups. The percentage of words retrieved from memory after an interval of 10 minutes did neither significantly differ between patients and controls nor between patient groups (data not shown).

AIH, HCV and PBC patients did significantly worse than controls in recognition of words. The HCV patients in addition did significantly worse than controls in the recognition of figures. Of note, HCV patients did significantly worse than HBV patients but achieved results comparable to those of the patients with autoimmune liver disease. Chi square test revealed significant group differences in the percentage of abnormal results with regard to the figure memory (P = 0.02). HCV patients did significantly worse than HBV patients (P = 0.007), AIH (P = 0.03) and PBC (P = 0.007) patients in recognizing figures (Figure 3), but not in recognizing words.

Figure 3.

Percentage of abnormal results with regard to word memory (on the left) and figure memory (on the right) for the different study groups. HCV patients did significantly worse than all patient groups and controls in figure memory, while HCV, AIH and PBC (as well as the overlap group) did worse than HBV patients and controls in word memory

Of note, in both AIH and PBC group patients with immunosuppressive therapy did not differ from those without immunosuppressive therapy regarding the psychometric test results. There was also no significant difference between PCR-positive and PCR-negative HCV and HBV patients (data not shown).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....