E-mail Abstract Author Session Search Abstracts Program


Session 115 Poster Presentations
Determinants of Disease Progression and Mortality
Session Day and Time: Tuesday 1:30 - 3:30 pm
Room: Hall B


908
Determinants of HIV Viral Load in Incident and Prevalent Infections, Rakai, Uganda
R. H. Hray*1, M. J. Wawer2, X. Li1, D. Serwadda3, N. Sewankambo3, T. Lutalo4, N. Kiwanuka5, G. Kigozi5, F. Wabwire-Mangen3, T. Quinn1
1Johns Hopkins Univ, Baltimore, MD; 2Columbia University, New York, NY; 3Makerere Univ, Kamplala, Uganda; 4Uganda Virus Res Inst, Entebbe; and 5Rakai Project, Entebbe, Uganda

Background: The determinants of HIV viremia in sub-Saharan Africa are unknown. However, viral load has been associated with the risk of HIV transmission and disease progression.
Methods: HIV viral loads (VLs) were determined using Roche Amplicor 1.5 RT-PCR in 216 HIV-incident and a sub sample of 1,293 HIV-prevalent participants in a prospective cohort study in Rakai, Uganda, between 1994-2000. Log10 HIV viral load was assessed by gender, age, STD symptoms and HSV-2 serology, and stage of disease. For prevalent subjects with unknown duration of infection, we assumed that 0-1 symptom indicated latency and 2+ symptoms indicated mid-stage disease or AIDS. Deaths were assumed to be due to AIDS. Adjusted viral loads were estimated by multiple linear regression modeling with interactions for age and stage of disease.
Results: Among HIV incident cases, viremia was similar in men and women, but viral load was higher in male than female prevalent cases (4.458 and 4.249 cps/mL, p = 0.01). Viremia was highest among incident cases aged 25-29 (p = 0.01), but increased with age in prevalent cases (p = 0.001). Viremia was higher in incident cases than in asymptomatic prevalent cases (adjusted viral loads 4.393 and 4.234 cps/mL, p = 0.003), and increased with more advanced disease (adjusted viral loads 4.528 cps/mL with 2+ symptoms and 4.885 cps/mL with death). In both incident and prevalent infections, the viral loads were significantly higher in persons reporting genital ulceration (largely due to HSV-2), but HSV-2 seropositivity per se did not significantly affect viremia.
Conclusion: Determinants of HIV viremia differ markedly between incident and prevalent infections (e.g., age, symptomatology). However, in both incident and prevalent cases, genital ulcers mainly due to HSV-2, was associated with higher viral load. Treatment of genital ulceration and HSV-2 suppressive therapy may be of utility in order to reduce HIV viremia, thus potentially reducing HIV transmission and disease progression.