Background. HIV viral load set points may vary between virus subtypes and host characteristics. The HIV epidemic in French Guiana entails a mix of viruses and populations of cosmopolitan origins. In this epidemiological context, we aimed to determine whether at the scale of our territory, we could identify differences in HIV-1 viral load setpoints in our hospital cohort. MethodsThe French Guiana hospital cohort was retrospectively analyzed between 1992 and 2023. Bootstrapped (100 replications) quantile (median) regression modelled the initial HIV-1 viral load with age, sex, nationality, and CD4 count at diagnosis as independent variables. ResultsAfter adjusting for period of analysis and CD4 count at diagnosis, bootstrapped quantile regression (median) showed that Haitians (minus 20 088 copies, P<0.0001) and Guyanese (minus 14087, P=0.039) had a significantly lower viral load at HIV diagnosis than French, males had a greater viral load at HIV diagnosis than females (plus 16 430 copies, P<0.0001), each additional year of age was associated with more copies (+345 per year, P=0.002). When compared to those with more than 500 CD4 per ml, persons with CD4 counts at diagnosis below 200 per ml had a greater initial viral load (plus 94 932 copies, P<0.0001), as did those with 200-350 CD4 per ml (plus 13 088 copies, P<0.0001), and those with [350-500 CD4 per ml] (plus 5 643 copies, P<0.0001). Non B viruses seemed to have an independently greater viral load at diagnosis than B viruses (plus 58402, P=0.029). Conclusions: We show some significant differences in the viral load set point of different groups. For nationalities we are inclined to attribute this to differences in the frequency of infecting subtypes. This suggests that models of incidence or date of infection based on CD4 decline may be distorted by this situation.
Read full abstract