Effect of hepatitis C virus genotype on CD4+ T cell count in a cohort of antiretroviral-naive HIV-1-infected individuals.

To the Editor—Over the past few years, there has been substantial controversy regarding whether HIV-1/hepatitis C virus (HCV)–coinfected individuals experience more-rapid HIV-1 disease progression than do HIV-1–monoinfected individuals [1–3]. Yoo et al., in a cohort of largely untreated young HIV-1/HCV–coinfected participants with hemophilia who were followed for 7 years, observed lower CD4 + T cell counts and a higher risk of AIDS-related mortality in the participants infected with HCV genotype 1 than in the participants infected with other genotypes [4]. However, as was suggested by the authors, the peculiar characteristics of the participants make the results of the study not directly applicable to other coinfected groups. We repeated the analysis of Yoo et al. using the CD4 + T cell counts and HIV-1 RNA loads recorded in the Italian Co-hort Naive for Antiretrovirals (I.Co.N.A.) database over the period during which the patients remained untreated. The I.Co.N.A. is a multicenter, prospective, observational study cohort that includes HIV-1– infected adults who were antiretroviral naive at the time of enrollment [5]. Specifically , we created an age-adjusted random coefficient linear regression model with CD4 + T cell count as the response variable [6]. We then further adjusted for HCV RNA loads measured in plasma that had been obtained and stored before the initiation of therapy. Mean CD4 + T cell counts were compared overall across the HCV genotype groups as well as between groups by the construction of prespeci-fied contrasts. Our analysis focused on 302 adult patients , 72 (23.8%) of whom were female, who were tested for HCV RNA load and genotype. Of these, 153 (50.7%) were infected with genotype 1, 7 (2.3%) were infected with genotype 2, 95 (31.5%) were infected with genotype 3, and 47 (15.6%) were infected with genotype 4. The mean (‫ע‬SD) numbers of CD4 + T cell count measurements taken during the follow-up period (stratified by genotype) were 1.80 (1.75) for genotype 1, 1.71 (0.76) for genotype 2, 2.55 (2.56) for genotype 3, and 2.64 (2.72) for genotype 4 (, Krus-P p .02 kal-Wallis test); the mean (‫ע‬SD) follow-up period for all patients was 4.57 (8.33) months. Risk categories were injection drug use (255 [84.4%] patients), homosexual contact (11 [3.6%] patients), and heterosexual contact (27 [8.9%] patients). In our basic model that adjusted for age alone, absolute CD4 + T cell counts were significantly lower in the HCV genotype 1 group than in the HCV genotype 3 …