Our meta-analysis quantitatively assessed the influence of HIV infection on the course of HCV infection. The overall OR for histological cirrhosis or decompensated liver disease or liver cancer or death was 3.40 (95% CI = 2.45 and 4.73) by the random effect model. However, these studies had a significant heterogeneity. Some factors could explain the heterogeneity, including different outcomes in these studies, methodological differences in study design, different number of patients in cohort, bias in selection of patients for biopsies, effect of duration of HCV infection on progression to severe liver disease, and publication bias.
We also performed subgroup analyses to determine the priority according to the different outcomes and duration of HCV infection. Striking differences were found in studies examining different end points of histological cirrhosis, decompensated liver disease, hepatocellular carcinoma or death. The combined adjusted OR for histological cirrhosis, decompensated liver disease, hepatocellular carcinoma and death was 1.47 (95% CI = 1.27 and 1.70), 5.45 (95% CI = 2.54 and 11.71), 0.76 (95% CI = 0.50 and 1.14), and 3.60 (95% CI = 3.12 and 4.15), respectively. There was a smaller difference between HIV-HCV coinfected and HCV-infected patients with regard to the development of cirrhosis or liver cancer, but there was a substantial increased risk of developing decompensated liver disease or death.
In the pre-HAART era, many patients coinfected with HIV and HCV died of opportunistic infection, lymphoma or wasting syndrome due to severe immunodeficiency, which is the main risk factor for death of HIV-HCV coinfection patients. The declined HIV-related mortality after widespread use of HAART parallels the emergence of HCV-related liver disease as an important cause of mortality in coinfected patients. Studies indicate that HAART has a protective effect on fibrosis progression in patients with HIV-HCV coinfection. On the other hand, HAART may enhance liver damage in some HIV-HCV coinfected individuals through drug-related hepatotoxicity. Only 6 studies[28,32,35,38,39,42
] in our analysis introduced the state of HAART in patients with HIV-HCV co-infection. The proportion of patients who were receiving HAART in cohort ranged 83%-95%. However, we could not acquire the primary data comparing progression of HCV infection in pre-HAART and HAART era, and the other studies were performed before the widespread use of HAART. We, therefore, did not examine the impact of HAART on the progression of HCV infection. This is an important limitation in our study and its impact on the progression of HCV-induced liver disease needs to be explored.
Other factors may have influenced the level of liver damage in HIV-HCV coinfected patients. Important fields for further study include the effects of HAART on HCV-related liver disease progression, duration of HCV infection, and alcohol consumption.
The results of our study suggest that HIV infection can significantly change the natural history of HCV infection, especially in the development of death or decompensated liver disease. Because most cohorts in our analysis were composed of patients with hemophilia or injection drugs, and most patients studied were males, caution should be taken in generalizing the results of this meta-analysis of women and racial or ethnic populations not represented in these studies. All these factors may potentially impact the natural history of HCV infection.