The objective of this longitudinal study was to analyze the possible influences of clinical and virological parameters on plasma HCV replication evaluated at the beginning and at the end of a time interval of 3 years. Our data indicate that age, gender, a diagnosis of advanced liver fibrosis, CD4+ cell count and successful HIV therapy, leading to undetectable plasma virema, failed to exhibit significant effects on HCV RNA levels.
Furthermore, we focused our interest on the characteristics of HIV-HCV-co-infected patients with persistent serum HCV RNA levels lower than 600,000 IU/ml. Co-infected patients with persistent low HCV RNA levels are typically infected with non-genotype 1 viruses and represent a small proportion (13.4%) of our population of HIV-HCV-co-infected subjects. However, when only baseline data are considered, the proportion of co-infected patients increased to 26.8%, a figure comparable to the 25% reported in the study by Rallon et al.  in a cohort of 196 co-infected patients. The association between HCV genotype 1 and the IL28B C allele with higher HCV RNA values is described in HIV-HCV-co-infected patients [11, 15]; these findings demonstrated that no unintended selection bias occurred.
Barreiro et al.  stated that subjects with HIV-HCV co-infection are more likely than HIV-negative subjects to experience variations in HCV RNA load greater than 0.5 log over a mean period of 43 months. The number of patients surveyed was greater than that in our study (644 versus 82), but the proportion of patients on ART was similar to ours (82% and 86.7%, respectively). Only two HCV RNA controls were available in our study because there was no clinical indication to prompt more frequent measurements. Nevertheless, increases of 0.5 log IU/ml of both the baseline and the final mean HCV RNA values for the patients categorized as having low-level viremia would correspond to values below 800,000 IU/ml, the threshold accepted as “low” ; therefore, our patients maintained their virological classification throughout the study period.
Most included patients (76.8%) were on ART, but a persistent undetectable HIV RNA load with a cut-off of 50 copies/ml was observed in only 58.7% of patients. At a comparable end point (144 weeks of follow-up), the cross-study analysis by Hua et al.  on 279 HIV-HCV-naïve co-infected patients described a rate of virological failure (defined as plasma HIV viremia > 200 copies/ml) of 10.5% (4 patients out of 38 evaluable patients). Of note, HCV status was defined only serologically and not by the presence of detectable HCV RNA, and HBsAg-positive patients were included, albeit the percentage was low (4.3%). ART efficacy was not the purpose of our work; nevertheless, we can report an interesting finding on the long-term efficacy in a small but homogeneous cohort of all HCV patients positive for HCV RNA, with no HBV coinfection and of Caucasian ethnicity who were treated according to the updated HIV guidelines.
Grint et al.  described HCV viral load changes in 1541 patients selected from the EuroSIDA study cohort: the median follow-up was five years, and the only factor able to stabilize HCV viral load was ART. In this study, 1148 subjects (74.5%) were on ART at baseline, with undetectable HIV RNA levels in 657 patients (57.2%); 393 subjects were not taking ART (88.2% with a detectable HIV RNA). Despite this virological picture, the absence or presence of detectable plasma HIV viremia has no significant influence on HCV RNA viral load. Successful ART and undetectable HIV plasma viremia are obviously strictly related if we exclude long-term non-progressing patients. We considered our patients to be responders to therapy when viral replication was lower than 50 copies/ml (400 copies/ml in the study by Grint et al.); however, such a threshold identifies a clinically successful suppression but does not indicate definitively absent plasma HIV replicative activity. There is mounting evidence on the prognostic role of HIV residual viremia (2.5-50 copies/ml) , and we cannot exclude the possibility that residual viremia may influence HCV RNA viral load, as could greater HIV burden. No correlation was observed between CD4+ cell count and HCV RNA values as previously described ; CD4+ cell count increased from T1 to T2 (mean increase of 78 cells/mm3, p = 0.003). The baseline immunocompetence in our enrolled patients was greater than that reported for subjects in the study by Grint et al. 576 cells/mm3 versus 340 cells/mm3 in the subjects on ART .
Interestingly, low HCV RNA viral load at T1 was negatively correlated with CD4+ cell count at T2 (p = 0.01). This finding does not contradict our other data because it was described in a selected population of patients: those with low HCV RNA viral load. This report is the first to describe this correlation, and only speculative hypotheses can be made, including the inverse correlation between CD4 cell apoptosis and serum HCV RNA .
There is a growing evidence of a low rate of prescription of antiviral treatment in co-infected patients. The study by Ioannou et al.  reported that only 994 out of 5999 (18%) HIV-HCV-co-infected patients assisted by the Veterans Affairs healthcare system received an interferon-based regimen. This percentage rises to 25.3% in the EuroSIDA cohort but is accompanied by a reduction in treatment incidence in 2009 . The proper timing at which to begin pegylated interferon and ribavirin treatment, associated with DAA when indicated, must take HIV disease status into account, but such coordination is not an easy task. Concomitant ART may be a risk factor for the development of side effects even in the case of therapy with pegylated interferon and ribavirin without DAA; However, a CD4+ cell number greater than 450/mmc predicts SVR in co-infected patients with HCV genotype 1 [23, 24].