Detection and quantification of HIV-1 in semen: identification of a subpopulation of men at high potential risk of viral sexual transmission

Background: To assess HIV burden in both acellular and cellular fractions o f semen in men with different levels of blood plasma HIV RNA by a cross-sec tional study. Patients: Fifty-two HIV-1-seropositive men (21 receiving antiretroviral the rapy) with CD4 cell counts ranging from 1 to 1170 x 10(6)/l. Methods: Semen was separated into seminal plasma and fractions enriched in motile spermatozoa or non-spermatozoal cells. HIV RNA was quantified by the HIV-Monitor technique (Roche) in blood plasma, seminal plasma and spermato zoa fractions. HIV DNA or infectious virions in cellular fractions were det ected by either PCR or qualitative viral culture. Results: HIV RNA was detected in 86.5% of seminal plasma specimens and in 1 4.6% of spermatozoa fractions; HIV DNA was detected in 57.1% of nonspermato zoal cell fractions. HIV RNA levels in blood plasma and seminal plasma were correlated (r(s) = 0.56, P < 0.0001, Spearman's rank test). A majority of men had lower levels in seminal plasma than in blood plasma: one-third had HIV-positive seminal cell fractions. However, 20 men (38.5%) with HIV RNA l evels in seminal plasma (median: 4.65 log(10) copies/ml) comparable to or h igher than those in blood plasma had all HIV-positive non-spermatozoal cell s or spermatozoa fractions with a high frequency of positive cultures. Conclusion: A high frequency of men had detectable HIV in semen. We identif ied a subpopulation demonstrating high levels of HIV RNA in seminal plasma, comparable to or higher than those in blood plasma, frequently associated with a substantial viral shedding in seminal cells, raising the possibility of viral production within the genital tract and suggesting heterogeneity in the potential of HIV sexual transmission among infected men. (C) 1999 Li ppincott Williams & Wilkins.