Systematic Review With Meta-analysis

Neuroimaging in Hepatitis C Chronic Infection

G. Oriolo; E. Egmond; Z. Mariño; M. Cavero; R. Navines; L. Zamarrenho; R. Solà; J. Pujol; N. Bargallo; X. Forns; R. Martin-Santos


Aliment Pharmacol Ther. 2018;47(9):1238-1252. 

In This Article

Abstract and Introduction


Background Chronic hepatitis C is considered a systemic disease because of extra–hepatic manifestations. Neuroimaging has been employed in hepatitis C virus–infected patients to find in vivo evidence of central nervous system alterations.

Aims Systematic review and meta–analysis of neuroimaging research in chronic hepatitis C treatment naive patients, or patients previously treated without sustained viral response, to study structural and functional brain impact of hepatitis C.

Methods Using PRISMA guidelines a database search was conducted from inception up until 1 May 2017 for peer–reviewed studies on structural or functional neuroimaging assessment of chronic hepatitis C patients without cirrhosis or encephalopathy, with control group. Meta–analyses were performed when possible.

Results The final sample comprised 25 studies (magnetic resonance spectroscopy [N = 12], perfusion weighted imaging [N = 1], positron emission tomography [N = 3], single–photon emission computed tomography [N = 4], functional connectivity in resting state [N = 1], diffusion tensor imaging [N = 2] and structural magnetic resonance imaging [N = 2]). The whole sample was of 509 chronic hepatitis C patients, with an average age of 41.5 years old and mild liver disease. A meta–analysis of magnetic resonance spectroscopy studies showed increased levels of choline/creatine ratio (mean difference [MD] 0.12, 95% confidence interval [CI] 0.06–0.18), creatine (MD 0.85, 95% CI 0.42–1.27) and glutamate plus glutamine (MD 1.67, 95% CI 0.39–2.96) in basal ganglia and increased levels of choline/creatine ratio in centrum semiovale white matter (MD 0.13, 95% CI 0.07–0.19) in chronic hepatitis C patients compared with healthy controls. Photon emission tomography studies meta–analyses did not find significant differences in PK11195 binding potential in cortical and subcortical regions of chronic hepatitis C patients compared with controls. Correlations were observed between various neuroimaging alterations and neurocognitive impairment, fatigue and depressive symptoms in some studies.

Conclusions Patients with chronic hepatitis C exhibit cerebral metabolite alterations and structural or functional neuroimaging abnormalities, which sustain the hypothesis of hepatitis C virus involvement in brain disturbances.


Hepatitis C virus (HCV) infection is a major public health problem, with an estimated annual incidence of between 3 and 4 million cases and a worldwide prevalence of 2.8%. Between 50% and 80% of infected individuals develop chronic hepatitis C (CHC), which may progress to cirrhosis and hepatocellular carcinoma.[1] Today, because of its extra–hepatic manifestations, CHC is considered a systemic disease.[2] As far as the central nervous system is concerned, around 50% of infected individuals may develop neurological or psychiatric disorders, which are independent of the severity of the liver disease.[3] Among psychiatric and somatic symptoms, up to 80% of patients can suffer from fatigue,[4] and up to 50% present depressive symptoms, anxiety and weakness,[5] causing impairment in social and occupational functioning as well as reduction in quality of life.[6,7] Moreover, HCV–infected individuals often complain of "brain fog," which specifically includes forgetfulness and difficulty concentrating, in addition to fatigue, malaise and anhedonia.[8] In fact, mild cognitive impairment in CHC patients has been reported in many studies,[9–11] with slower psychomotor speed, alteration in working memory and deficits in attention and concentration being the most commonly observed features.[9,10] It has been estimated that around one third of infected patients may show neurocognitive disability, even in the absence of cirrhosis and encephalopathy.[12–14] However, such estimation could had been exaggerated, to the extent that no clear and direct attribution to HCV infection can be made, as other factors may contribute to cognitive impairment in such patients. Several reviews[3,15,16] pointed out that confounding factors such as advanced liver disease or history of neuropsychiatric disorder or substance use disorder were not systematically assessed in studies addressing cognitive impairments in CHC patients. Furthermore, most of these studies were cross–sectional, with small samples, and HCV–infected patients were often selected from hospital populations, which may account for selection bias.[16]

The high prevalence of neuropsychiatric symptoms reported in CHC is supported by several lines of converging evidence pointing to a neuropathogenic effect of the virus.[13] Several studies suggest that HCV can replicate in monocytes/macrophages, which are able to cross the blood–brain barrier and access to the central nervous system in a process known as "Trojan horse" mechanism.[17–20] Thus, the brain may serve as an important reservoir for subsequent viral replication, as indicated by the detection of specific HCV–RNA strands in post–mortem brains of CHC patients.[21] Furthermore, the evidence of viral quasi–species diversity between the central nervous system and liver, supports the notion of independent viral evolution,[4] rendering the central nervous system a potential source of relapse after anti–viral therapy.[16] Considering the questionable blood–brain barrier penetration ability of the new direct acting anti–virals (DAAs),[22] such reservoir should be taken into account in long–term rebound or outcomes. Late viral relapses are extremely rare after successful treatment with DAAs.[23,24] However, a recent exhaustive systematic review underlined the absence of evidence to determine the effect of sustained viral response on long–term outcome.[25] Regarding neurobiological consequences, it has been suggested that the infected microglial cells may increase secretion of pro–inflammatory cytokines such as tumour necrosis factor–α (TNF–α) or interleukins (IL),[26] which have been associated with the physiopathology of depressive and cognitive symptoms.[27]

Beside some evidence of direct neuroinvasion, the systemic chronic immune system activation that characterises CHC may also account for the pathogenesis of neuropsychiatric symptoms. It is well known that pro–inflammatory cytokines, such as IL–1, IL–6 and TNF–α, may interact with several neurobiological pathways, interfering with neurotransmission and neurotrophic mechanisms.[28] For example, the activation of the indoleamine–2,3–dioxygenase enzyme secondary to neuroinflammation and the consequent induction of the tryptophan catabolites pathways have been related to the depletion of serotonin levels and the augmentation of glutamate neurotoxicity through an increase in quinolinic acid.[29] In the same way, immune activation in central nervous system has been related to a reduced activity of tetra–hydro biopterin, an enzyme involved in dopamine synthesis and whose disruption has been associated with decreased dopamine levels.[30]

In the last 2 decades, neuroimaging has been used in HCV–infected patients in the search for in vivo evidence of these central nervous system alterations.[16] Structural and functional techniques have been used to define anatomical alterations, metabolic or neurotransmission abnormalities, and connectivity disruption in CHC. Nevertheless, the majority of these studies centre on demonstrating the neuropsychiatric effects of therapy with interferon–α (IFN–α), which was the main molecule used to treat CHC[31–33] before the development of the new DAAs, and had been associated with major depressive disorder, malaise and fatigue.[34] Few neuroimaging studies have focused on chronic HCV infection in patients without treatment, and even more less those with a healthy control group. Moreover, the samples selected are heterogeneous, especially with regard to the clinical severity of the CHC, the presence of neuropsychiatric comorbidities and the diagnosis of active substance use disorder. Finally, the attempts to find correlation between neuroimaging findings and cognitive or psychiatric symptoms have yielded inconclusive results.[35]

Thus, in order to examine the in vivo evidence of central nervous system alterations in CHC, we performed a systematic review (and meta–analysis if sufficient data were available) of studies investigating HCV chronic–infected patients (treatment naive or previously treated without achieving a sustained viral response), using neuroimaging techniques, and assessed the correlation of these alterations with neuropsychiatric symptoms.