Liver Fibrosis Progression in Human Immunodeficiency Virus and Hepatitis C Virus Coinfected Patients 

Jules Levin, NATAP

Hepatology, October 1999, p. 1054-1058, Vol. 30, No. 4
Authors conclude HIV accelerates HCV Fibrosis Rate. Severe immunosuppression (<200 CD4s) also associated with accelerated fibrosis. CD4 increases from HAART may slow fibrosis (HCV) progress

Yves Benhamou1, Marie Bochet2, Vincent Di Martino1, Frederic Charlotte3,Felipe Azria1, Anne Coutellier4, Michel Vidaud1, Franois Bricaire2,Pierre Opolon1, Christine Katlama2, and Thierry Poynard1 for the for the MULTIVIRC Group 

>From the 1Service d'HŽpato-GastroentŽrologie, Groupe Hospitalier PitiŽ Salptrire and UPRES-A 8067, Paris, France; 2Service de Maladies Infectieuses, Groupe Hospitalier PitiŽ-Salptrire, Paris, France; 3Service d'Anatomie Pathologique, Groupe Hospitalier PitiŽ-Salptrire, Paris, France; and 4Service de MŽdecine Interne, Groupe Hospitalier PitiŽ-Salptrire, Paris, France. 

Abstract-
The natural history of hepatitis C virus (HCV) infection in human immunodeficiency virus (HIV)-infected patients has never been studied according to the concept of liver fibrosis progression. The aim of this work was to assess the fibrosis progression rate in HIV-HCV coinfected patients and in patients infected by HCV only. 

A cohort of 122 HIV-HCV coinfected patients was compared with a control group of 122 HIV-negative HCV-infected patients. Groups were matched according to age, sex, daily alcohol consumption, age at HCV infection, and duration and route of HCV infection. The fibrosis progression rate was defined as the ratio between fibrosis stage (METAVIR scoring system) and the HCV duration. The prevalence of extensive liver fibrosis (METAVIR fibrosis scores 2, 3, and 4) and moderate or severe activity were higher in HIV-infected patients (60% and 54%, respectively) than in control patients (47% and 30%, respectively; P < .05 and P< .001, respectively). 

Results-
The median fibrosis progression rate in 
coinfected patients and in control patients was 0.153 (95% confidence interval [CI], 0.117-0.181) and 0.106 (95% CI, 0.084-0.125) fibrosis units per year, respectively (P < .0001). HIV seropositivity (P< .0001), alcohol consumption (>50 g/d, P = .0002), age at HCV infection (<25 years old, P < .0001), and severe immunosuppression (CD4 count 200 cells/µL, P < .0001) were associated with an increase in the fibrosis progression rate.

In coinfected patients, alcohol consumption (>50 g/d), CD4 count (200 cells/µL), and age at HCV infection (<25 years old) (P < .0001, respectively) were associated with a higher fibrosis progression rate. HIV seropositivity accelerates HCV-related liver fibrosis progression. In coinfected patients, a low CD4 count, alcohol consumption rate, and age at HCV infection are associated with a higher liver fibrosis progression rate. (HEPATOLOGY 1999;30:1054-1058.) 

Discussion-
We have conducted a study of the natural history of HCV infection in HIV-infected patients according to the concept of liver fibrosis progression. This concept has been validated in a large series of HCV-positive untreated patients with paired biopsies.12 Our main finding was that liver fibrosis progressed faster in HIV-HCV coinfected patients than in patients infected only by HCV. According to the median fibrosis progression rate in untreated patients, the expected time to cirrhosis was 26 years in HIV-infected patients and 34 years in non-HIV-infected patients. Multivariate analysis showed that HIV infection, alcohol consumption (>50 g/d), age at HCV infection (>25 years old), and CD4 cell count (200 cells/µL) were independent factors associated with higher liver fibrosis progression rates. The fibrosis progression rate was lower in the simulated group than in the HIV-positive group. This difference also suggests the influence of HIV infection because these 2 groups differed only according to HIV seropositivity. Gender did not influence fibrosis progression of HIV-HCV coinfected patients; however, a large proportion of males were included in our study. Previous studies have shown more severe liver damage in HIV-positive patients than in HIV-negative patients.6,7,9,10,11,17-19 However, these studies were mostly retrospective, included a small number of patients, and did not distinguish fibrosis from necroinflammatory activity. To characterize liver fibrosis and the activity grades, we used a highly reproducible method (METAVIR scoring system).13,14 HIV-positive patients had never been matched to HIV-negative patients according to the factors involved in the natural history of liver fibrosis progression in chronic hepatitis C. Correct matching should be an important point because the HIV-HCV coinfected population and the population of patients infected only by HCV may be different. As shown in the current study, the HIV-HCV coinfected group has a younger age at HCV infection, included more alcohol drinkers (more than 50 g/d of alcohol), and more men than previously reported in the whole population of patients infected only by HCV.12 Bias of selection was reduced by the careful match of HIV-negative control patients with HIV-positive patients regarding sex, alcohol consumption, risk factors for HCV infection, age at HCV infection, and age at liver biopsy. 

Among HIV-HCV coinfected patients, age at HCV-infection (>25 years old) and alcohol consumption (>50 g/d), together with the CD4 cell count (200 cells/µL), were independently associated with higher liver fibrosis progression rates. Therefore, an HIV-infected patient with 200 CD4 cells/µL or less and drinking more than 50 g/d of alcohol may develop cirrhosis in 16 years whereas an HIV-infected patient with more than 200 CD4 cells/µL and drinking 50 g or less of alcohol daily may have an expected time to cirrhosis of 36 years. As observed in experimental studies, this finding is consistent with the role of immunity in the fibrosis progression process.20 In HIV-infected patients, the CD4 lymphocyte counts are negatively correlated with HCV viral load.21 A high serum HCV-RNA load related to impaired immune functions could be involved in the severity of liver disease.22 

Our study presents some limitations. The estimated fibrosis progression rate per year was defined as the ratio between the fibrosis stage and the estimated duration of HCV infection. In this model it is assumed that the patient has no liver fibrosis the day of infection and that fibrosis progression is constant. However, duration of HCV infection remains an estimate. To accurately estimate the duration of HCV infection, the studied population included only IV drug users and transfused patients. In this population, contamination occurs at the date of the first transfusion or during the first year of IV drug injection in 90% of patients.23 Furthermore, it is also possible that some patients already had fibrosis (i.e., owing to alcohol) the day of infection. Thus, an ideal assessment would have been to prospectively follow-up a large representative sample of patients from HCV infection to death, with repeated liver biopsy and without treatment. Obviously, such a study is both ethically and pragmatically impossible. However, our observation of paired biopsies performed in a large number of untreated HCV-infected patients12 and in a small number of HIV-HCV coinfected patients showed that estimated and observed fibrosis progression rates were not different. Thus, although not perfect, the estimation of liver fibrosis progression rates using a single liver biopsy remains an interesting tool for the study of the natural history of chronic hepatitis C. 

Other factors were not investigated in our study. The influence of HCV genotypes could not be assessed because it was only available in a small number of patients. However, the HCV genotype is not associated with the fibrosis progression rate in patients infected only by HCV12,24 and its influence on the severity of liver disease is controversial in liver transplant recipients.25 

Combination antiretroviral therapies may also play a role in the fibrosis progression rate. The influence of HIV-1 protease inhibitor therapies on the natural history of liver fibrosis progression rate of HIV-HCV coinfected patients is unstudied. Although the liver fibrosis progression rate was lower in patients who received HAART, the difference did not reach significance in our study; however, the number of HAART-treated patients was small. Studies including more patients are needed to reevaluate the impact of HAART on liver fibrosis progression. An increase of the CD4 cell count related to such therapies is expected to reduce the fibrosis progression rate of HCV-HIV coinfected patients, although long-term use of combination antiretroviral therapy that includes an HIV protease inhibitor does not influence HCV viral load.26 On the other hand, antiretroviral therapy may induce liver damage.27 The control of HIV with effective combination antiretroviral therapy together with a major reduction of alcohol consumption may delay the expected time to cirrhosis. The recommendations of limiting alcohol consumption and maintaining a high CD4 count are mandatory especially in HIV-coinfected patients, who are poor responders to interferon therapy.28 

We conclude that HIV seropositivity accelerates fibrosis progression related to chronic hepatitis C. HIV infection, alcohol consumption of more than 50 g/d, CD4 lymphocyte count of less than 200 cells/µL, and age of more than 25 years at HCV infection are associated with an increased rate of liver fibrosis progression in HCV-coinfected patients. These factors should be considered in anti-HCV therapeutic strategies.