Background:  Oral-genital transmission of HIV-1 occurs and HIV-1 RNA can be detected in the saliva; however, culture of HIV-1 from the saliva has been generally unsuccessful (<1% of specimens) due to inactivation of virus by saliva.  Our objective was to describe the predictors and variability of HIV-1 RNA viral load (VL) in the pharynx and the recovery of cultivable HIV-1 from oropharyngeal surfaces.

Methods: Sexually active HIV-1-positive men from Seattle and Lima, Peru, who have sex with men, without bacterial STD’s, were evaluated prospectively at 0, 2, and 4 weeks to assess viral load in blood and swab specimens obtained from the posterior oropharynx.  HIV-1 RNA was quantified using a commercial PCR amplification assay (Roche HIV-1 Monitor).  A subset of 17 men was evaluated for recovery of infectious virus from the saliva, tonsil, and buccal surfaces.

Results: The median CD4 count of 58 subjects was 356 cells/μL; 52% were currently receiving antiretrovirals.  The median baseline VL was higher in blood plasma (4.58 log₁₀ HIV-1 RNA copies/mL) than the pharynx (3.95 log₁₀) (p < 0.001).  The within-subject variability of pharyngeal VL (SD, 0.36 log₁₀) was greater than that of blood VL (SD, 0.22 log₁₀).  Generalized estimating equations were used to show that each 1.0 log₁₀ increase in plasma VL was associated with a 0.4 log₁₀ increase in pharyngeal VL (p < 0.001), and tonsillectomy was associated with a 0.6 log₁₀ reduction in pharyngeal VL (p = 0.007) but antiretroviral therapy was not (p > 0.1).  HIV-1 was cultured from the posterior oropharyngeal surface from 4 (24%) of 17 men.  Median mucosal VL was 6.352 log₁₀ copies/mL in culture positive men vs 4.681 log₁₀ copies/mL in culture negative men (p=0.06).  HIV-1 was cultured from the blood but not the saliva or buccal mucosal surface.

Conclusions: Pharyngeal VL is lower among men with prior tonsillectomy and increased among men with higher plasma viral load.  Our results show for the first time that infectious HIV-1 can be detected from the oropharyngeal mucosal surface.  These data indicate a potential for oral transmission of HIV-1.  Sexual acts that facilitate contact with the posterior oropharyngeal mucosal surface may be associated with an increased risk of oral transmission as has been reported for HHV-8, although additional behavioral data are needed.