The relation between symptoms, viral load, and viral load set point in primary HIV infection

Objectives: To examine the relation between symptoms, initial viral load, and viral load set point in primary HIV infection (PHI). Design: Prospective cohort of patients with preseroconversion or recent seroconversion HIV infection (typically < 60 days) in San Francisco. Methods: Subjects were questioned about 21 potential PHI symptoms at enrollment and were subsequently followed with viral load measures. Results: The analysis included 57 subjects with preseroconversion HIV infection and 120 with recent seroconversion. In univariate analysis, most symptoms and the total number of symptoms were each associated with a significantly higher initial viral load. In stepwise multiple linear regression, however, only the number of symptoms was independently associated with a higher initial viral load, with an increase in the initial viral load of 0.08 log(10) per additional symptom (P < 0.001). In univariate analysis, more PHI symptoms were associated with a higher viral load set point, but in a multivariable mixed-effects model, this association was accounted for by the initial viral load, which was strongly correlated with viral load set point (R = 0.44, P < 0.001). Conclusions: A high initial viral load was associated with more symptoms during PHI. The strong correlation between initial HIV-1 RNA viral load levels and viral load set point suggests that early interactions between the HIV-1 virus and a new host, even before fully developed adaptive immune responses, are important in establishing viral load set point.