The World Health Organization estimates that 3% of the world's population has been infected with hepatitis C virus (HCV), leaving 170 million chronically infected . In the United States, the prevalence of HCV infection is closer to 4 million . In developed countries, HCV co-infects approximately a third of all HIV-infected individuals, 75–90% of HIV-infected intravenous drug users, and nearly 100% of some groups of HIV-infected hemophiliacs . Therefore, of over 1 million HIV-infected individuals in the USA  and Europe , HCV may co-infect between 300 000 and 400 000. The longer life expectancy of HIV-infected individuals since the advent of combination antiretroviral therapy  has highlighted the impact of chronic HCV, which has emerged as a frequent cause of death in HIV-infected people .
HCV belongs to the family Flaviviridae. Neurological disease is associated with many flaviviruses, such as West Nile, Japanese encephalitis, and tick-borne encephalitis viruses. A growing body of literature fuels the notion that HCV infection can lead to neurocognitive impairment. First, neuropsychological testing demonstrated that individuals with chronic HCV disease are impaired on more cognitive tasks than those who clear the infection [8–11]. Second, magnetic resonance spectroscopy demonstrated that HCV-infected individuals have elevated choline/creatine ratios in the basal ganglia and white matter , suggesting inflammation, and reduced N-acetyl aspartate in white matter , suggesting neuronal loss. Third, HCV can replicate in the cells of monocyte lineage , which are resident in (microglia) or can traffic into (macrophages) the central nervous system (CNS) .
The primary objective of this study was to determine the effects of HCV on neuropsychological performance. To accomplish these objectives, we utilized standardized multidisciplinary methods including a neuromedical examination, lumbar puncture, venipuncture, and a comprehensive battery of neuropsychological tests. To account for two of the most prevalent comorbidities, HIV and drug use, we: (i) enrolled both HIV-seropositive and HIV-seronegative subjects; (ii) measured markers of HIV disease activity; (iii) enrolled subjects with and without methamphetamine dependence; (iv) excluded subjects with other substance abuse disorders; and (v) only assessed subjects who had been abstinent for at least 2 weeks. A secondary objective was to determine whether the performance of HCV-infected individuals was associated with measures of viral replication or immune activation.
Participants were 526 volunteers enrolled prospectively in research studies at University of California San Diego (UCSD)'s HIV Neurobehavioral Research Center (HNRC). All subjects provided informed consent before enrollment in the studies, all of which were approved by the UCSD Human Research Protections Program. Almost all (472) were enrolled in a cohort study of the CNS effects of HIV and methamphetamine. In this cohort, subjects were recruited into four groups: HIV-negative/methamphetamine-negative (n = 114), HIV-negative/methamphetamine-positive (n = 119), HIV-positive/methamphetamine-negative (n = 119), and HIV-positive/methamphetamine-positive (n = 120). Methamphetamine-positive subjects met the criteria for dependence, as determined by the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders version IV , within 18 months but were not actively using methamphetamine, heroin, or cocaine. For this analysis of HCV, subjects were excluded if they reported past or current use of IFN-α. The remaining 54 subjects enrolled in other HNRC studies, met criteria for inclusion in the HIV-negative/methamphetamine-negative group and had CSF and plasma specimens in storage at −70°C. These subjects contributed body fluids but not neuropsychological data to our analyses. In total, 312 of the 526 subjects (59.3%) had CSF specimens available for testing.
Each subject underwent a comprehensive neuromedical evaluation that used structured clinical data forms to assess: (i) medical and medication use history; (ii) antiretroviral medications and self-reported adherence; (iii) neurological and general physical examinations; and (iv) routine laboratory studies, including hematology and chemistry panels. The HIV disease stage was assigned according to Centers for Disease Control 1993 classification criteria . Blood was collected by venipuncture. Approximately 15 ml CSF was collected by lumbar puncture using a 22-gauge atraumatic ‘pencil-point’ needle.
Participants underwent standardized neuropsychological assessments of seven ability domains that may be affected by HCV, HIV, and methamphetamine. These seven domains are: verbal fluency; attention and working memory; speed of information processing; learning; delayed recall; abstraction and problem solving; and motor ability. An oral reading test of premorbid intelligence was also administered (Wide Range Achievement Test–3). All neuropsychological tests were administered and scored by trained psychometrists who followed standardized procedures outlined in test-specific manuals. The administered tests and their respective ability domains were: (i) Letter Fluency (FAS) and Category Fluency (animals) (verbal fluency); (ii) Paced Auditory Serial Addition Task and Wechsler Adult Intelligence Scale – III (WAIS-III) letter-number sequencing (attention/working memory); (iii) WAIS-III Digit Symbol, WAIS-III Symbol Search, Trail Making Test A, and Stroop Task (speed of information processing); (iv) Heaton Story Memory Test, Hopkins Verbal Learning Test–R, Heaton Figure Memory Test, and Brief Visuospatial Memory Test–R (learning; delayed recall); (v) Wisconsin Card Sorting Test, Halstead Category Test, Trail Making Test B, and Stroop Task interference ratio (abstraction/problem solving); and (vi) Grooved Pegboard Test (both hands) (motor ability) [18–21].
A global deficit score (GDS) summarized participants’ overall performance on the neurocognitive test batteries. The GDS quantifies the number and degree of impaired test scores across the battery, while attaching relatively less significance to performances that are within normal limits. For each of the test variables above, raw scores were converted to demographically corrected T-scores (M 50; SD 10) using published normative conversions [20,22]. The T-scores were then transformed into deficit scores using the following conversions: ≥ 40T = 0; 39T–35T = 1; 34T–30T = 2; 29T–25T = 3; 24T–20T = 4; ≤ 19T = 5. The deficit scores from each test were then averaged to derive a GDS for each participant. This measure ranges between 0 and 5. One interpretation of this value is the approximate number of standard deviations that performance falls below expected. A score of 0.5 is consistent with definite impairment, and higher scores indicate greater degrees of cognitive impairment [19,23].
HIV and HCV infections were diagnosed by standard clinical antibody detection. HIV RNA in plasma and CSF were measured by reverse transcriptase–polymerase chain reaction (RT–PCR; Amplicor, Roche Diagnostics, Indianapolis, IN, USA). The ultrasensitive assay was used for CSF (nominal detection limit of 50 copies/ml) and the standard assay was used for plasma (nominal detection limit of 400 copies/ml). HCV RNA was also measured in plasma and CSF using RT–PCR (NGI SuperQuant; National Genetics Institute, Los Angeles, CA, USA; nominal detection limit of 100 copies/ml). A fluorescence-activated cell sorter quantified CD4 lymphocytes. At the time of their visit, the urine of each subject was tested with an on-site screening test for common recreational drugs, including amphetamines, cocaine, barbiturates, THC, opiates, benzodiazepines, and phencyclidines (Rapid Response; Biotechnostix, Inc., Markham, Ontario, Canada).
Routine clinical assays, such as hepatic transaminases, were performed by standard methods at the UCSD Medical Center clinical laboratory. Specimens were assayed for selected immune markers [monocyte chemotactic protein (MCP) 1], TNF-α, and soluble TNF receptor (sTNFR) II (p75) using commercial enzyme-linked immunosorbent assay kits (Quantikine, R&D Systems, Minneapolis, MN, USA). The sensitivities of these assays adjusting for dilution were: MCP-1 64 pg/ml, TNF-α 1.0 pg/ml, and sTNFR-II 78 pg/ml. All solid-phase enzyme-linked immunosorbent-based assays have previously been validated for use with human CSF.
Data were analysed using routine statistical methods. The transformation of some variables improved the symmetry of their distributions. For example, log transformation improved the symmetry of HIV-RNA distributions, and square root transformation improved the symmetry of CD4 lymphocyte counts. The distributions of some variables were so highly skewed that we analysed them both as continuous and categorical variables. For example, we analysed TNF-α levels in CSF both as untransformed (continuous) values and as a dichotomous variable encoding detection (i.e. detectable or undetectable). Some analyses refer to ‘target risk conditions’. These three risk conditions were HCV serostatus, HIV serostatus, and methamphetamine dependence. For univariate analyses, we analysed data using both parametric (e.g. analysis of variance or Pearson's correlation) and non-parametric (e.g. Wilcoxon or Spearman's correlation) measures of association, depending on the distributions of the covariates. For multivariate analyses, we used linear regression methods. For example, for immune marker analyses, we constructed a series of regression models, each predicting the concentration of one marker in one body fluid. For this analysis, we limited the independent covariates in each model to HCV serostatus, HIV serostatus, and methamphetamine dependence diagnosis without second or third order interactions. Statistical analyses were performed using JMP (version 5.0 for Mac, SAS Institute, Cary, NC, USA).
Subject characteristics are summarized in Table 1. HCV seropositivity was diagnosed in 112 out of 526 subjects (21.3%). HCV-seropositive subjects were more likely to be older (P < 0.001) and less educated (P < 0.001). In our cohort, HCV-seropositive subjects were not more likely to be HIV infected, but as expected, they were more likely to be methamphetamine-dependent (P < 0.001). HCV and methamphetamine were independently associated with Centers for Disease Control and Prevention (CDC) stage C3 among HIV-seropositive subjects (P = 0.03, t-test comparing HCV and CDC C3; P = 0.003, Cochran–Mantel–Haenszel test for methamphetamine). HCV/HIV-co-infected and HIV-mono-infected subjects did not differ in other disease characteristics, including current CD4 cell counts or the use of antiretroviral agents. HCV-seropositive subjects did not appear to have current substantial hepatocyte injury or advanced liver disease, based on alanine aminotransferase, albumin, and platelet levels.
Hepatitis C virus and neuropsychological performance
Compared with HCV-seronegative subjects, HCV-seropositive subjects had worse neuropsychological performance, whether measured by the GDS (0.58 versus 0.32, P < 0.001, Fig. 1a) or by the blinded clinical rating of the neuropsychological protocols (1.8 times more likely to be diagnosed impaired, 95% confidence interval 1.15–2.82) [21,24].
In a multivariate regression model, worse performance (i.e. higher GDS) was independently associated with HCV (parameter estimate P = 0.004) after controlling for HIV and methamphetamine dependence (model R2 = 0.10, P < 0.001). This was true even after adjusting for the CDC stage and antiretroviral use. To confirm this finding, we performed two additional analyses. First, because HCV-seropositive subjects were more likely to have been recently methamphetamine dependent, we performed a stratified analysis limited to non-methamphetamine users. This analysis identified that HCV seropositivity was still associated with worse neuropsychological performance (median GDS 0.79 versus 0.26, P = 0.002), indicating that the impaired performance associated with HCV was not simply caused by methamphetamine use. Second, we summed the number of target risk conditions present in each individual, and compared this measure with neuropsychological performance. We found that performance worsened as the number of co-existing risk conditions increased (no risks: median GDS 0.21; one risk: 0.32; two risks: 0.42; or all three: 0.74; P < 0.001, Fig. 1b), again favoring the independent effect of HCV on neuropsychological performance.
Neuropsychological performance and viral replication
HCV-RNA levels varied widely in plasma (median 6.4, interquartile range 4.5–6.9 log copies/ml) and were less than 100 copies/ml in all CSF specimens. Overall, plasma HCV-RNA levels were higher in those with memory, but not global, impairment (P = 0.05, Fig. 2a). Among those with impaired memory, higher HCV-RNA levels in plasma correlated with worse memory performance (Spearman's rho 0.31, P = 0.04).
In HIV-infected subjects, HCV seropositivity was associated with higher HIV-RNA levels in the CSF, but not in plasma, after adjusting for antiretroviral use (R2 = 0.26, P < 0.001). This finding is illustrated by a stratified analysis of the 62 HIV-seropositive subjects who were not taking antiretroviral drugs. In this subgroup, HCV-seropositive HIV-seropositive subjects had higher HIV-RNA levels in the CSF, but not in plasma, than HCV-seronegative HIV-seropositive subjects (3.4 versus 2.8 log copies/ml, P = 0.01, Fig. 2b).
Hepatitis C virus and immune activation
HCV infection was associated with higher concentrations of all immune activation markers assayed. In particular, HCV infection was associated with higher MCP-1 levels in plasma (P < 0.001); higher TNF-α levels in plasma (P < 0.001); and higher sTNFR-II levels in plasma (P < 0.001, Fig. 3a) and CSF (P < 0.02). We again confirmed these findings with two additional analyses. First, we constructed a series of multivariate regression models, as described in the Methods section. This analysis confirmed the independent association of HCV seropositivity with higher levels of MCP-1 in plasma, TNF-α in plasma, and sTNFR-II in plasma and CSF. This analysis also extended our initial findings, identifying that HCV infection was independently associated with TNF-α levels in the CSF (FFull − FReduced = 3.8, P = 0.05), after adjusting for HIV and methamphetamine dependence (model R2 = 0.10, P < 0.001). Second, the number of risk conditions present in each individual was associated with concentrations of all immune markers, except TNF-α levels in plasma. Of the markers assayed, HCV infection was most strongly associated with sTNFR-II (P < 0.001). Pairwise comparisons demonstrated that concentrations in plasma increased significantly between each of the four categories (no risks: median 1873 pg/ml; one risk: 2384 pg/ml; two risks: 2912 pg/ml; or all three: 4118 pg/ml; P < 0.001, Fig. 3b).
Based on this analysis of a single-center cohort of over 500 subjects, we concluded that HCV injures the CNS independently of two important comorbidities, HIV and methamphetamine. Our study improved on earlier reports in several areas, including its sample size; its use of comprehensive, standardized assessments; its careful control and characterization of substance use; and its measurement of important biomarkers, such as HCV RNA, HIV RNA, and immune activation markers.
Our primary conclusion is supported by several observations. First, HCV seropositivity was associated with worse neuropsychological performance when analysed by four different approaches (analysis of variance, multivariate regression, stratified analysis, and combined risk conditions). Second, HCV was associated with worse performance even after adjusting for the important HIV disease correlates, CDC stage and antiretroviral use. Third, higher HCV-RNA levels in plasma were associated with neurocognitive impairment, supporting a link between brain injury and HCV replication. Finally, HCV was associated with higher levels of proteins known to be neurotoxic (TNF-α) or strongly associated with HIV-associated dementia (MCP-1).
Our findings suggest that HCV may injure the CNS via several mechanisms. First, proteins encoded by HCV may be directly neurotoxic, similar to the HIV-encoded proteins tat and gp120. The observed association between higher HCV-RNA levels and neurocognitive impairment supports this conclusion. Although we did not detect HCV RNA in the CSF, it has been detected in the CNS using more sensitive assays or assays for negatively stranded RNA intermediates [15,25,26]. Second, HCV may activate monocyte-derived macrophages. Once activated, macrophages may migrate across the blood–brain barrier where they can release inflammatory mediators, including chemokines that can attract additional immune cells. Our finding that markers of macrophage activation were elevated in the plasma and CSF of HCV-infected study participants supports this conclusion. Third, in HCV/HIV-co-infected individuals, trafficking macrophages may also replicate HIV virions. This is supported by the observation that HCV infection was associated with higher HIV-RNA levels in the CSF. Importantly, higher levels of HIV replication in the CSF are strongly associated with the incidence and prevalence of HIV-associated neurocognitive disorders [27,28].
HCV-associated decrements in neuropsychological performance might also be attributed to other causes. For example, hyperammonemia is a well-known complication of advanced liver disease, including that caused by HCV. High levels of ammonia can, in turn, injure the brain via excitotoxic mechanisms. We did not directly test this hypothesis by measuring ammonia levels, but common biochemical markers did not support the theory that our subjects had sufficiently advanced liver disease to be at risk of hepatic encephalopathy. HCV may also adversely affect the CNS by other mechanisms, such as the activation of brain endothelial cells or astrocytes. The current analysis did not directly address these hypotheses by measuring markers associated with these mechanisms.
Other limitations of this analysis include: its cross-sectional design; the imbalance in some subject characteristics (e.g. HCV infection was not balanced across the eight categories defined by HCV, HIV, and methamphetamine), which led to suboptimal variability in some analyses; the reduced power of analyses involving the CSF compared with plasma (of the subjects who had venipuncture, approximately 60% successfully underwent lumbar puncture); and the potentially confounding effects on neuropsychological performance of between-group differences in education, the use of drugs other than methamphetamine, or psychiatric disease. Despite these limitations, our data strongly support the theory that HCV injures the brain, resulting in impairments in neuropsychological performance. Perhaps more importantly, HCV-associated neural injury may be preventable or reversible because HCV infection is potentially curable.
The authors gratefully acknowledge our research volunteers and referring medical providers and grant support from the National Institute on Drug Abuse and the National Institute of Mental Health.
The San Diego HIV Neurobehavioral Research Center (HNRC) Group includes Igor Grant, MD, director; J. Hampton Atkinson and J. Allen McCutchan, MD, co-directors; Thomas D. Marcotte, PhD, center manager; Mark R. Wallace, MD, principal investigator, Naval Medical Center, San Diego; Neuromedical component: Ronald J. Ellis, MD (principal investigator), Scott Letendre, MD, and Rachel Schrier, PhD; Neurobehavioral component: Robert K. Heaton, PhD (principal investigator), Mariana Cherner, PhD, Julie Rippeth, PhD; Imaging component: Terry Jernigan, PhD (principal investigator); Neuropathology component: Eliezer Masliah, MD (principal investigator), T. Dianne Langford, PhD; Clinical trials component: J. Allen McCutchan, MD (principal investigator), Ronald J. Ellis, MD, PhD, Scott Letendre, MD; Data management unit: Daniel R. Masys, MD (principal investigator), F. Michelle Frybarger, BA (data systems manager); Statistics unit: Ian Abramson, PhD (principal investigator), Deborah Lazzaretto, MA.
Sponsorship: This study received support from the National Institute on Drug Abuse (P01 DA12065, R01 DA16015) and the National Institute of Mental Health (K23 MH01779).
2. National Institutes of Health. Concensus statement on management of hepatitis C: 2002
. NIH Consens State Sci Statements
3. Soriano V, Rodriguez-Rosado R, Garcia-Samaniego J. Management of chronic hepatitis C in HIV-infected patients. AIDS 1999; 13:539–546.
4. Centers for Disease Control and Prevention. Cases of HIV infection and AIDS in the United States, 2002
. HIV/AIDS Surveillance Report, Vol. 14. Atlanta, GA: CDC, Department of Health and Human Services; 2003.
5. Euro HIV. HIV/AIDS Surveillance in Europe: mid-year report, 2003
. European Centre for the Epidemiological Monitoring of AIDS and WHO and UNAIDS Collaborating Centre on AIDS. Saint-Maurice: Institut de veille sanitaire, 2004 No. 69.
6. Palella FJ Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338:853–860.
7. Soriano V, Sulkowski M, Bergin C, Hatzakis A, Cacoub P, Katlama C, et al
. Care of patients with chronic hepatitis C and HIV co-infection: recommendations from the HIV-HCV International Panel. AIDS 2002; 16:813–828.
8. Forton DM, Thomas HC, Murphy CA, et al
. Hepatitis C and cognitive impairment in a cohort of patients with mild liver disease. Hepatology 2002; 35:433–439.
9. Hilsabeck RC, Hassanein Tl, Carlson MD, Ziegler EA, Perry W. Cognitive functioning and psychiatric symptomatology in patients with chronic hepatitis C. J Int Neuropsychol Soc 2003; 9:847–854.
10. Hilsabeck RC, Perry W, Hassanein TI. Neuropsychological impairment in patients with chronic hepatitis C. Hepatology 2002; 35:440–446.
11. Ryan EL, Morgello S, Isaacs K, Naseer M, Gerits P. Neuropsychiatric impact of hepatitis C on advanced HIV. Neurology 2004; 62:957–962.
12. Taylor-Robinson SD. Applications of magnetic resonance spectroscopy to chronic liver disease. Clin Med 2001; 1:54–60.
13. Taylor MJ, Letendre SL, Schweinsburg BC, Alhassoon OM, Brown GG, Gongvatana A, et al
. Hepatitis C virus infection is associated with reduced white matter N
-acetylaspartate in abstinent methamphetamine users. J Int Neuropsychol Soc 2004; 10:110–113.
14. Caussin-Schwemling C, Schmitt C, Stoll-Keller F. Study of the infection of human blood derived monocyte/macrophages with hepatitis C virus in vitro. J Med Virol 2001; 65:14–22.
15. Radkowski M, Wilkinson J, Nowicki M, Adair D, Vargas H, Ingui C, et al
. Search for hepatitis C virus negative-strand RNA sequences and analysis of viral sequences in the central nervous system: evidence of replication. J Virol 2002; 76:600–608.
16. Spitzer RL, Williams JBW, Gibbon M, First M. Structured clinical interview for DSM-IV. Washington, DC: American Psychiatric Press; 1995.
17. 1993 Revised Classification System for HIV Infection and Expanded Surveillance. Case definition for AIDS among adolescents and adults
. MMWR Recom Rep
18. Heaton R, Grant I, Matthews C. Comprehensive norms for an expanded Halstead-Reitan Battery: Demographic corrections, research findings, and clinical applications
. Odessa: Psychological Assessment Resources, Inc; 1991.
19. Heaton RK, Grant I, Butters N, White DA, Kirson D, Atkinson JH, et al
. The HNRC 500 – neuropsychology of HIV infection at different disease stages. HIV Neurobehavioral Research Center. J Int Neuropsychol Soc
20. Heaton RK, Kirson D, Velin RA, Grant I. The utility of clinical ratings for detecting early cognitive change in HIV infection.
In: Neuropsychology of HIV infection
. Grant I, Martin A, editors. New York Oxford University Press; 1994. pp. 188–206.
21. Heaton RK, Miller SW, Taylor MJ, Grant I. Revised Comprehensive Norms for an Expanded Halstead-Reitan Battery: Demographically Adjusted Neuropsychological Norms for African American and Caucasian Adults.
Lutz, FL: Psychological Assessment Resources; 2004.
22. Heaton RK, Taylor MJ, Manly J. Demographic effects and use of demographically corrected norms with the WAIS III and the WMS-III.
In: Clinical Interpretation of the WAIS-III and WMS-III
. Edited by Tulsky D, Saklofske D, Heaton RK, et al
. San Diego: Academic Press; 2003. pp.183–210.
23. Carey CL, Woods SP, Gonzalez R, Conover E, Marcotte TD, Grant I, et al
. Predictive validity of global deficit scores in detecting neuropsychological impairment in HIV infection. J Clin Exp Neuropsychol 2004; 26:307–319.
24. Woods SP, Rippeth JD, Frol AB, Levy JK, Ryan E, Soukup VM, et al
. Interrater reliability of clinical ratings and neurocognitive diagnoses in HIV. J Clin Exp Neuropsychol 2004; 26:759–778.
25. Maggi F, Giorgi M, Fornai C, Morrica A, Vatteroni ML, Pistello M, et al
. Detection and quasispecies analysis of hepatitis C virus in the cerebrospinal fluid of infected patients. J Neurovirol 1999; 5:319–323.
26. Morsica G, Bernardi MT, Novati R, Uberti Foppa C, Castagna A, Lazzarin A. Detection of hepatitis C virus genomic sequences in the cerebrospinal fluid of HIV-infected patients. J Med Virol 1997; 53:252–254.
27. Ellis RJ, Hsia K, Spector SA, Nelson JA, Heaton RK, Wallace MR, et al
. Cerebrospinal fluid human immunodeficiency virus type 1 RNA levels are elevated in neurocognitively impaired individuals with acquired immunodeficiency syndrome. HIV Neurobehavioral Research Center Group. Ann Neurol 1997; 42:679–688.
28. Ellis RJ, Moore DJ, Childers ME, Letendre S, McCutchan JA, Wolfson T, et al
. Progression to neuropsychological impairment in human immunodeficiency virus infection predicted by elevated cerebrospinal fluid levels of human immunodeficiency virus RNA. Arch Neurol 2002; 59:923–928.