Disease progression of HIV-1 infection in symptomatic and asymptomatic seroconverters in Osaka, Japan: a retrospective observational study

BACKGROUND Estimates of the interval from HIV-1 infection to disease progression may be affected by selection bias, and data concerning asymptomatic early seroconverters are limited. We examined the interval until disease progression in HIV-1 seroconverters in whom the timing of infection could be estimated within 1 year before diagnosis. METHODS Subjects included newly diagnosed patients at Osaka National Hospital between 2003 and 2010 who had either (1) symptomatic acute HIV-1 infection with a negative or intermediate reaction on Western blotting and a positive reaction on an HIV RNA test (symptomatic acute group) or (2) a positive reaction on Western blotting at diagnosis and a <1-year interval from the last negative HIV test until the first positive test. The latter was divided into symptomatic recent or asymptomatic recent groups based on the presence or absence, respectively, of any transient fever between the last negative and first positive tests. Disease progression was defined as a fall in the CD4 count to <350 cells/microL on 2 consecutive tests, the start of anti-HIV therapy, or the onset of AIDS-indicator diseases. Information was retrospectively collected from medical records. RESULTS Subjects included 210 patients: 91 in the symptomatic acute group, 72 in the symptomatic recent group, and 47 in the asymptomatic recent group. In the symptomatic acute (0.8 years) and symptomatic recent (2.2 years) groups, the Kaplan-Meier estimate of median interval until disease progression was significantly shorter than that in the asymptomatic recent group (2.9 years). Multivariate analysis by Cox's proportional hazards test showed that the symptomatic acute group (vs. asymptomatic recent group: hazard ratio: 1.93; 95% confidence interval: 1.14-3.36; p = 0.0140) and a baseline CD4 count of <400 cells/microL (hazard ratio: 3.88; 95% confidence interval: 2.57-5.96; p < 0.0001) were independent prognostic factors associated with early disease progression. CONCLUSIONS Symptomatic seroconversion was associated with early disease progression. Furthermore, the estimated median interval until the CD4 count was <350 cells/microL was only 2.9 years even in patients with asymptomatic seroconversion. These results suggest the importance of early diagnosis in early seroconverters.