Discordance in HIV-1 viral loads and antiretroviral drug concentrations comparing semen and blood plasma


Dr Paul A MacPherson, Division of Infectious Diseases, Module G, Box 223, Ottawa Hospital, General Campus, 501 Smyth Road, Ottawa, ON, Canada K1H 8L6. Tel: +1 613 737 8899 ext. 73917; fax: 1 613 737 8164; e-mail: pmacpherson@ottawahospital.on.ca



For individuals not on antiretroviral therapy, the risk of heterosexual transmission of HIV appears negligible when blood plasma (BP) viral loads are <1500 HIV-1 RNA copies/mL. It is not clear whether this observation can be extrapolated to individuals on highly active antiretroviral therapy (HAART). Because of differential tissue penetration, antiretroviral drug concentrations may be sufficient to maintain an undetectable viral load in the BP yet not achieve adequate levels to suppress HIV in the genital tract. Therefore, we wanted to correlate HIV viral loads and drug concentrations in semen plasma (SP) and BP.


Thirty-three men were included. All were on combination antiretroviral therapy with an undetectable BP viral load for at least 1 year. Blood and semen samples were collected within 2 h of each other and tested for HIV RNA by the NucliSens QT (bioMerieux, St Laurent, QC, Canada) method; drug concentrations were determined by liquid chromatography tandem mass spectrometry.


Two of the 33 patients (6.1%) with BP viral loads below detection had time-matched HIV viral loads in SP ≥700 copies/mL. Both patients were on efavirenz, the SP concentrations of which were ≤10% of the levels in BP and well below the minimal therapeutic drug monitoring target concentration required to suppress HIV.


Because, at least in part, of poor drug penetration into the genital tract, an undetectable HIV viral load in the BP does not guarantee an undetectable viral load in semen. In view of this, caution should be taken in concluding that patients on HAART with suppressed viraemia are sexually non-infectious.