Analysis of HIV-1 load in blood, semen and saliva: evidence for different viral compartments in a cross-sectional and longitudinal study.

To quantify the HIV-1 load (measured as copies of viral RNA/ml using competitive reverse transcription-polymerase chain reaction) in blood, semen and saliva and to look for relationships between the viral burden, the clinical and immunological status and antiretroviral therapy. Peripheral blood, semen and whole saliva samples were collected from 26 anti-HIV-1-seropositive patients selected for a cross-sectional study. Nine of the 26 patients provided samples of the three biological fluids for a longitudinal study. HIV-1 RNA was detected in 26 out of 26 samples of plasma, in 25 out of 26 samples of semen and in 24 out of 25 samples of saliva. The median number of HIV-1 copies in plasma was 14 817/ml (range: 167-254 880), in semen was 515/ml (range: 0-196 050) and in saliva was 162/ml (range: 0-72 080). The viral load in semen and in saliva was significantly lower than in plasma (P < 0.0001). The HIV-1 RNA levels in plasma and in saliva were correlated (P < 0.05), but levels in semen were not correlated with either plasma or saliva levels. The HIV-1 copy number in plasma was significantly higher in symptomatic patients than in asymptomatic subjects (P < 0.05). Plasma and saliva HIV-1 RNA levels were higher in subjects with a CD4+ cell count < 200 x 10(6)/l than in subjects with a CD4+ cell count > 200 x 10(6)/l (P < 0.05). The HIV-1 RNA load in either plasma, semen or saliva is not related to antiretroviral therapy. The absence of a correlation between plasma and semen loads suggests that semen and blood are distinct viral compartments. Viral load in semen is not related to the clinical stage of HIV infection or to the CD4+ lymphocyte count. Consequently, HIV-1-infected subjects are potentially infectious at all stages of immuno-deficiency and adequate precautions must always be taken to prevent the sexual transmission of HIV.