Improvement of Chronic HCV Infection-Related Depression, Anxiety, and Neurocognitive Performance in Persons Achieving SVR-12

A Real-World Cohort Study

Harmanpreet Kaur; Radha K. Dhiman; Anand V. Kulkarni; Madhumita Premkumar; Virendra Singh; Ajay Kumar Duseja; Sandeep Grover; Gagandeep S. Grover; Akash Roy; Nipun Verma; Arka De; Sunil Taneja; Rohit Mehtani; Saurabh Mishra; Harpreet Kaur


J Viral Hepat. 2022;29(5):395-406. 

In This Article

Patients and Methods

Eligible persons were enrolled from the Liver Clinic of the Postgraduate Institute of Medical Education and Research, Chandigarh, India, during the period between 18 June 2018 and 31 October 2020. We have already described the Punjab Model programme with innovative decentralized care services offered at 25 sites. Briefly, in this model, viraemic persons with or without cirrhosis were provided free of charge DAAs with sofosbuvir plus an NS5A inhibitor, that is daclatasvir (DCV), ledipasvir (LDV) or velpatasvir (VEL), as the standard regimen with the addition of ribavirin for patients with decompensated cirrhosis.[15,16,18] As per the algorithm, the recruited individuals were followed until 12 weeks post-treatment to ascertain the SVR-12. This study was approved by the Institutional Ethical Committee (IEC-D3/2018-866) and was conducted following the Declaration of Helsinki. The trial is accessible at NCT04330508 and is available from All coauthors had access to the final dataset. This prospective cohort study adhered to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.

Patients aged between 18 and 65 years, irrespective of gender, with viraemic CHC infection were recruited prospectively with written informed consent. Exclusion criteria included current or recent overt HE episode, transjugular intrahepatic portosystemic shunt in situ, inability to give informed consent, human immunodeficiency virus (HIV) coinfection, chronic respiratory insufficiency, chronic renal failure, hepatocellular carcinoma (HCC), pre-existing neurological diseases, intake of sedatives, antidepressants, benzodiazepines or benzodiazepine antagonists (flumazenil). Baseline tests for thyroid function, HIV, hepatitis B virus, aspartate, alanine transaminase (AST and ALT), albumin, haemogram, alpha fetoprotein, glycosylated haemoglobin (HbA1c), renal function and ultrasonography of the abdomen were performed in all patients with HCV viral load. AST-platelet ratio (APRI) > 2.0 and Fib-4 > 3.25 were used as cut offs for cirrhosis.

HCV Therapy and Study Design

The Punjab Model algorithm was designed (by RKD) using all-oral DAA-based regimens (Figure S1). Genotype testing was not mandatory. The model uses SOF/DCV in patients without cirrhosis and SOF/VEL in patients with cirrhosis as pan-genotypic regimens and assesses SVR-12 as the outcome. Treatment interruption was defined if drug therapy was interrupted by >7 days or the therapy was discontinued. The diagnoses of cirrhosis and HCV algorithms are reported in Appendix S1.[15] Literacy was defined as the ability to both read and write, with understanding, in any language.[19] Adherence was defined if there was no interruption in therapy, and the patient completed the prescribed regimen on schedule. Residual pill counts were performed at monthly hospital visits for prescription refills to ensure drug compliance.

Study Endpoints

In this study, we assessed the degree of cognitive performance at enrolment and 12 weeks after treatment completion using a standard computerized battery for cognitive domains including visual and verbal memory, attention, visual-motor speed, arithmetic and reaction times, a modification of the Halstead-Reitan Battery. The computer-based tests included reaction times (simple and choice), visual memory, digit span, number connection test (NCT) Stroop test, Best Card Sort test (BCST) and vigilance test for attention. Paper-based tests were conducted with the validated scoring systems for depression (Beck's Depression Inventory-II), anxiety (Generalized anxiety disorder-7) and Montreal Cognitive Assessment (MoCA) at enrolment and 12 weeks after therapy (Table S1).[20] The neurocognitive battery was custom designed using Psychology Experiment Building Language (PEBL) version 2.1.[21,22] Details of the paper-based tests with interpretation (Table S2) and normative data of all the tests for healthy and diseased (nonalcoholic fatty liver disease [NAFLD]) controls (Table S3) are provided in Appendix S1. The HRQoL was performed by vernacular translations of the SF36 ver.2 questionnaire. Our primary objective/endpoints were to ascertain differences in neuropsychological tests and HRQoL associated with achieving SVR-12 in PLHCV. The secondary endpoints were association of scores in the neurocognitive and psychological tests with gender, the presence of cirrhosis, completion of therapy, SVR-12 rate and therapy adherence in a public health setting.

Statistical Analysis and Sample Size Calculation

The neuropsychological variables were expressed as mean values (standard deviation SD) or median (interquartile range [IQR]). Kolmogorov-Smirnov test was used to analyse the hypothesis of normality of the sample. Comparisons between the two time points of cognitive tests (at enrolment and 12 weeks after completion of therapy) were analysed using Student's t-test. Analysis of variance (ANOVA), with post hoc Bonferroni tests, was used for multiple comparisons to detect differences among subgroups. Categorical variables are expressed as frequencies (% of total), and their associations were assessed by Pearson's chi-square or Fisher's exact tests. All tests were two-tailed, and significance was set at p < .05. Correlations between the variables measured were studied by means of Spearman's rho or Pearson's r coefficients where appropriate. Binary logistic regression was performed to identify variables independently associated with treatment failures. All variables that had p < .1 on univariable analysis were used in a model for multivariate logistic regression. The statistical analysis was conducted with the SPSS software, ver. 22 (IBM Corp.).

The primary analysis upon which was the sample size consideration was based on our published results on cure rates in the Punjab Model and prior studies.[15,16] We used a two-factorial design (time course × SVR), a significance level of 5% and a statistical power of at least 80% to detect a medium effect size (d = 0.5) and thus to show a significant group difference. Based on this background, the optimal sample size was estimated as 200 subjects in patients without and 75 patients with cirrhosis. To consider asymmetric subgroups, gender differences and to allow for a moderate dropout rate and additional calculations (secondary study objectives), we aimed to include a total of at least 300 study participants with 25 healthy volunteers as controls for population normative data. In addition, a cohort of 30 persons with NAFLD was used to provide a diseased control group.