Accelerated Progression of HCV/HIV in Europe
Incidence and Predictors of Severe Liver Fibrosis in HIV-Infected Patients with Chronic Hepatitis C: A European Collaborative Study
Clinical Infectious Diseases 2004;38:128-133
Luz Martín-Carbonero,1 Yves Benhamou,9 Massimo Puoti,10 Juan Berenguer,2 José Mallolas,6 Carmen Quereda,3 Ana Arizcorreta,7 Antonio Gonzalez,4 Jurgen Rockstroh,11 Victor Asensi,8 Pilar Miralles,2 Montse Laguno,6 Leonor Moreno,3 José Antonio Girón,7 Martin Vogel,11 Javier García-Samaniego,1 Marina Nuñez,1 Miriam Romero,1 Santiago Moreno,3 Juan José de la Cruz,5 and Vincent Soriano1
1Hospital Carlos III, 2Hospital Gregorio Marañón, 3Hospital Ramón y Cajal, 4Centros Penitenciarios, and 5Universidad Autónoma, Madrid, 6Hospital Clinic, Barcelona, 7Hospital Puerta del Mar, Cádiz, and 8Hospital General, Oviedo, Spain; 9Hôpital Pitié-Salpêtrière, Paris, France; 10Spedali Civili,
Brescia, Italy; and 11University Hospital, Bonn, Germany
"…we examined the largest series of liver biopsies in HCV-HIVcoinfected patients, that data were always obtained from patients with elevated serum ALT levels and using a standardized protocol, and that multiple health care centers with a large amount of experience with coinfected patients participated in the study. These procedures may have increased the strength of our results, as well as their reliability for generalization to similar patient populations…
…This large study confirms the high prevalence of severe liver fibrosis in HIV-HCV coinfected patients… it is noteworthy that nearly one-half of HCV-HIVcoinfected patients aged >40 years have severe liver fibrosis, which is more than twice the rate seen in subjects infected with HCV alone. Therefore, our findings are in agreement with prior studies that have reported much faster progression of liver damage in HIV-HCVcoinfected patients…
…univariate regression analysis: duration of HCV infection of >15 years were all significantly associated with severe liver fibrosis… In the multivariate analysis, only 3 variables remained independently associated with severe liver fibrosis: age at liver biopsy of >35 years, consumption of >50 g of alcohol per day, and CD4+ T cell count of <500 cells/mm3 at the time of liver biopsy. Duration of HCV infection was removed from the analysis because it strongly correlated with age (r = 0.5; P < .01) and had a weaker association than age…
… In our study, we did not find that HAART had any benefit on liver fibrosis, although prior duration of antiretroviral therapy was not assessed in our study, which might have permitted more accurate assessment of the effect of HAART on liver fibrosis… HAART may have neither a protective effect nor a deleterious impact on HCV-related liver fibrosis in HIV-infected patients…
… anti-HCV treatment must be considered as a priority for coinfected patients, and, in the absence of contraindications, it should be provided at the earliest possible time"
A study was performed in 10 European health care centers in which 914 patients coinfected with hepatitis C virus (HCV) and human immunodeficiency virus (HIV) who had elevated serum alanine aminotransferase (ALT) levels underwent liver biopsy during the period of 1992 through 2002.
Overall, the METAVIR liver fibrosis stage was F0 in 10% of patients, F1 in 33%, F2 in 22%, F3 in 22%, and F4 in 13%. Predictors of severe liver fibrosis (METAVIR stage, F3 or F4) in multivariate analysis were age of >35 years (odds ratio [OR],
2.95; 95% confidence interval [CI], 2.084.18), alcohol consumption of >50 g/day (OR, 1.61; 95% CI, 1.12.35), and CD4+ T cell count of <500 cells/mm3 (OR, 1.43; 95% CI, 1.031.98).
Forty-six percent of patients aged >40 years had severe liver fibrosis, compared
with 15% of subjects aged <30 years. The use of antiretroviral therapy was not associated with the severity of liver fibrosis.
In summary, severe liver fibrosis is frequently found in HCV-HIV coinfected patients with elevated serum ALT levels, and its severity increases significantly with age. The rate of complications due to end-stage liver disease will inevitably
increase in this population, for whom anti-HCV therapy should be considered a priority.
Because there are shared routes of transmission, coinfection with HIV and hepatitis C virus (HCV) is common: it occurs in up to 60%-90% in HIV-positive patients who have history of parenteral blood exposure, who use injection drugs, or who have received blood or blood products. Of note, liver disease progresses faster in persons who are coinfected with both viruses than in persons with HCV monoinfection. The survival of HIV-infected persons began to dramatically increase after the introduction of HAART, and liver disease has become one of the leading causes of morbidity in this population. Furthermore, end-stage liver disease currently represents the leading cause of death among HIV-infected injection drug users (IDUs) and hemophiliacs.
The natural history of HCV infection is quite variable in immunocompetent patients: one-third of patients develop liver cirrhosis 15-20 years after infection, and another one-third develop it 20-30 years after infection. The remaining group of patients, known as "slow fibrotic progressors," develop cirrhosis only after 30 years of HCV infection. Several factors have been associated with an increased risk of progression to cirrhosis, including male sex, older age at the time of HCV infection, and high alcohol consumption. Although it is well established that progression of liver fibrosis is accelerated in patients with HIV infection, factors associated with a more severe stage of liver fibrosis in the HCV-HIV coinfected population have not yet been well characterized. In previous studies, a correlation between higher CD4+ cell counts and less-advanced liver fibrosis was found. However, most of these studies were performed before the widespread use of HAART, and, therefore, the issue needs to be reassessed. The aim of this study was to examine the severity of liver fibrosis in a large HCV-HIV coinfected population and to identify factors associated with more-severe liver fibrosis, with special consideration of the impact of HAART.
PATIENTS AND METHODS
Study population. Ten health care centers from 4 different European countries (Spain, Italy, France, and Germany) where a relatively high proportion of HIV-infected individuals have acquired the infection parenterally (IDUs and hemophiliacs) were invited to participate in a retrospective survey in the year 2002. Eligible patients were those for whom HIV and HCV infections were documented using both serologic and nucleic acid methods, who had elevated serum alanine aminotransferase (ALT) levels, and who had a liver biopsy performed with interpretable findings during the period of 19922002. Patients were excluded if they had normal ALT levels, were positive for hepatitis B surface antigen, had other known causes of liver disease, and/or had received prior anti-HCV therapy.
For each patient, a case report form was completed. It included the patient's basic demographic characteristics and clinical, laboratory, and virologic data. Demographic characteristics included age, sex, risk category, and history of high alcohol intake (defined as the consumption of >50 g of alcohol per day for 12 months). Plasma HIV RNA level, CD4+ T cell count, and antiretroviral therapy received at the time of liver biopsy were recorded for all patients. Moreover, HCV genotype, plasma HCV RNA level, and duration of HCV infection (which was documented for most hemophiliacs and was estimated for IDUs after considering the first year in which needles were unsafely shared) were recorded. Duration of HCV infection was considered to be unknown for subjects infected through sexual contact.
Histologic evaluation. Liver biopsies were performed and evaluated at each health care center. Hepatic fibrosis was quantified using the 5-point score from the METAVIR system, as follows: F0, no fibrosis; F1, portal fibrosis with no septa; F2, portal fibrosis with few septa; F3, bridging fibrosis with many septa; and F4, cirrhosis. This scoring system was chosen because the scores are highly reproducible among pathologists and because it is commonly used throughout Europe. For the purpose of this study, liver fibrosis was stratified into 3 categories: absent or mild (F0 or F1), moderate (F2), and severe (F3 or F4).
Study population. Data from a total of 914 patients were reviewed. Most patients were male (75%) and former IDUs (83%). Up to 25% of patients admitted to prior high alcohol intake. The median age at the time of liver biopsy was 37 years (IQR, 3341 years). The median estimated duration of HCV infection was 16 years (IQR, 1224 years). HCV genotype 1 was the most prevalent genotype (56%), followed by genotype 3 (32%), genotype 4 (9%), and genotype 2 (2%). Plasma HCV RNA levels of >800,000 IU/mL were recorded for 58% of patients. With regard to HIV infection status, the median CD4+ T cell count at the time of liver biopsy was 480 cells/mm3 (IQR, 310672 cells/mm3), and only 9% of patients had a CD4+ T cell count of <200 cells/mm3. HAART was being taken by 55% of patients, whereas 15% of patients were receiving single- or dual-drug therapy. Approximately 30% of patients were not receiving antiretroviral therapy. Among patients receiving HAART, 48% were receiving protease inhibitors (PIs), and 45% were receiving nonnucleoside reverse-transcriptase inhibitors (NNRTIs). Up to 54% of patients had undetectable plasma HIV loads (i.e., HIV RNA level of <200 copies/mL) at the time of liver biopsy.
Stages of liver fibrosis. Absence of or mild fibrosis (METAVIR score, F0 or F1) was noted in 397 patients (43%), and severe fibrosis (METAVIR score, F3 or F4) was recorded in 320 (35%). However, in older patients, the proportion of cases of mild or no fibrosis tended to decrease, whereas the proportion of cases of severe fibrosis tended to increase. Approximately 46% of patients aged >40 years had severe liver fibrosis. 23% had history of alcohol consumption >50 g/day.
Predictors of liver fibrosis. Male sex, high alcohol intake, age of >35 years, and estimated duration of HCV infection of >15 years were all significantly associated with severe liver fibrosis.
HCV-dependent features, such as HCV genotype and HCV load, were not associated with the extent of liver fibrosis. Although severe liver fibrosis was associated with CD4+ T cell counts of <500 cells/mm3, patients receiving HAART had severe liver fibrosis more frequently than did patients who were not receiving any antiretroviral treatment (39% vs. 28%; P < .05). Subjects receiving single- or dual-drug therapy more frequently had intermediate-stage (METAVIR score, F2) fibrosis (32%). When considering the different treatment modalities, there were no significant differences between subjects receiving PIs and those receiving NNRTIs with regard to the severity of liver fibrosis (data not shown).
In the multivariate analysis, only 3 variables remained independently associated with severe liver fibrosis: age at liver biopsy of >35 years, consumption of >50 g of alcohol per day, and CD4+ T cell count of <500 cells/mm3 at the time of liver biopsy. Duration of HCV infection was removed from the analysis because it strongly correlated with age (r = 0.5; P < .01) and had a weaker association than age.
Receipt of HAART was not associated with extent of liver fibrosis in multivariate analysis. Because more subjects receiving HAART were >35 years of age at the time of liver biopsy, compared with untreated patients (72% vs. 42%; P < .01), use of HAART was most likely a confounding factor for age.
This large study confirms the high prevalence of severe liver fibrosis in HIV-HCVcoinfected patients. We found that older age, high alcohol consumption, and low CD4+ T cell count were significantly associated with more-severe liver fibrosis. As in HCV-monoinfected individuals, age seems to be the main factor driving the extent of liver fibrosis in coinfected patients. However, it is noteworthy that nearly one-half of HCV-HIVcoinfected patients aged >40 years have severe liver fibrosis, which is more than twice the rate seen in subjects infected with HCV alone. Therefore, our findings are in agreement with prior studies that have reported much faster progression of liver damage in HIV-HCVcoinfected patients.
As in HCV-monoinfected individuals, high alcohol intake was significantly associated with severe liver fibrosis in our subjects, which supports the current recommendation that alcohol consumption be limited as much as possible in this population. The issue is particularly relevant for these patients, given that many of them are former IDUs who are more prone to abuse of substances, including alcohol. Although only one-fourth of our patients admitted high alcohol consumption, underestimation of alcohol intake is likely in our series.
With respect to HIV-related variables, severe liver fibrosis was only found in subjects with CD4+ T cell counts of <500 cells/mm3. This is in agreement with the results of other studies. In a recent study, nadir CD4+ T cell count was found to be a strong predictor of severe liver fibrosis, an issue that could not be addressed in our study, because this information was not recorded.
Studies of the impact of antiretroviral therapy on HCV-related liver damage have produced conflicting results. If the impairment of cell-mediated immunity due to HIV infection results in a faster progression of liver fibrosis, then HAART should ameliorate this effect and reduce fibrosis progression. This has been supported by 2 studies that examined subjects receiving PI-based regimens. In our study, however, we did not find that HAART had any benefit on liver fibrosis, although prior duration of antiretroviral therapy was not assessed in our study, which might have permitted more accurate assessment of the effect of HAART on liver fibrosis.
It is noteworthy that, in univariate analysis, more-severe liver fibrosis was found among patients receiving HAART, although this association disappeared after adjustment for age. Given that our series included individuals who had undergone biopsy over a 10-year period, patients receiving HAART had undergone biopsy more recently, which indirectly reflects an older population. Consequently, it was age and not HAART use that was the main factor to have influenced liver fibrosis stage in our patients.
Alternatively, more-severe liver damage in patients receiving HAART could be driven by drug hepatotoxicity, which is not an uncommon phenomenon. Although HAART-induced liver toxicity might accelerate HCV-related liver fibrosis, our data do not support this worsening effect. In summary, HAART may have neither a protective effect nor a deleterious impact on HCV-related liver fibrosis in HIV-infected patients.
Finally, we did not find any association between liver fibrosis and specific HCV features, such as HCV genotype or HCV load. Although plasma HCV RNA levels seem to increase after potent antiretroviral therapy is started, thus far, there is no evidence to indicate that this increase results in a worsening of liver lesions.
Our study has some limitations. First, we recognize that any assessment of the impact of immunity on hepatic fibrosis might require repeated liver biopsies in the context of prospective longitudinal studies. However, for ethical and practical reasons, this is not feasible. Second, almost all liver biopsies in our study were performed to assess the indication for anti-HCV therapy. Because patients with decompensated liver disease did not undergo biopsy, our results might ultimately underestimate the real number of patients with cirrhosis among HIV-HCVcoinfected patients. Finally, studies like ours have to admit biases derived from their retrospective design and the lack of uniform interpretation of virologic and histologic data. However, we assessed histologic damage using the METAVIR scoring system, the results of which are highly reproducible among pathologists, and HCV loads were expressed as IU/mL in an attempt to make results for different sites uniform. With these facts in mind, we would like to emphasize that we examined the largest series of liver biopsies in HCV-HIVcoinfected patients, that data were always obtained from patients with elevated serum ALT levels and using a standardized protocol, and that multiple health care centers with a large amount of experience with coinfected patients participated in the study. These procedures may have increased the strength of our results, as well as their reliability for generalization to similar patient populations.
Our findings have important clinical implications. In HIV-negative individuals, it is estimated that the incidence of HCV-related complications will continue to increase during the next decade, despite the decreasing incidence of new HCV infections since 1990, when the first diagnostic tests became available. Our results show that this assumption is particularly true for the HIV-HCVcoinfected population. Because we found that age was the main predictor of severe liver fibrosis, and because the mean age of patients currently attending HIV clinics in developed countries is 4045 years old, the number of patients with end-stage liver disease has not yet peaked and will be on the rise. As long as survival is not compromised by HIV infection, thanks to the efficacy of HAART, more and more, HIV-HCVcoinfected patients will have progression to liver cirrhosis. Indeed, HCV-related disease has already been reported to be a major cause of hospital admission and death among HIV-infected persons in some regions.
Given the more aggressive course of HCV infection in HCV-HIVcoinfected individuals, strategies aimed at prevention of exposure to HCV and at modification of disease progression are urgently needed. Needle-exchange programs for IDUs may be effective for reducing the risk of acquiring HCV infection. For HIV-positive persons who are already infected with HCV, current treatment of hepatitis C with pegylated interferon and ribavirin needs to be considered more openly. Our data argue for providing anti-HCV therapy at the earliest stage possible, because responses diminish as liver fibrosis worsens and the CD4+ T cell count decreases. In addition, there is growing concern about the risk of interactions between medications for HCV infection and antiretroviral drugs. Up to one-third of coinfected patients permanently cleared HCV after receiving a standard course of pegylated interferon plus ribavirin, with relatively good tolerance.
In conclusion, HCV-HIVcoinfected patients develop severe liver fibrosis more frequently than do HCV-monoinfected persons. Nearly one-half of patients aged >40 years with elevated serum ALT levels have severe liver fibrosis (METAVIR score, F3 or F4). Because liver fibrosis worsens rapidly with age, anti-HCV treatment must be considered as a priority for coinfected patients, and, in the absence of contraindications, it should be provided at the earliest possible time. Given the magnitude of the problem worldwide, strategies aimed at prevention of HCV exposure, at reducing alcohol consumption, and at treating HCV infection in early stages should be warranted in HIV-infected populations at risk.