December 30, 2003
Although the risk for HIV transmission through oral sex is difficult to determine (in large part because the mode of transmission in any given case can be difficult to establish), published studies suggest that the risk associated with fellatio is low, but not zero.
For more information see AIDS 1998; 12:2095-105; 7th Conference on Retroviruses and Opportunistic Infections, Abstract 473; and AIDS 2000; 16:2350-2352).
Investigators in Sydney conducted detailed interviews with 75 HIV acutely infected gay men between 1993 and 1999 to assess all risk behavior in the 6 months preceding seroconversion and to determine the likely mode of transmission.
The researchers established the most likely mode of transmission by sorting exposures according to a risk hierarchy: high risk (sharing needles for intravenous drug use, unprotected receptive and insertive anal intercourse), medium-to-low risk (condom-protected receptive and insertive anal intercourse, blood or semen contact with an open wound, oral sex, and penile-anal external contact without insertion), and no risk (visits to dentists, blood donation in Australia, stepping on a discarded syringe, sex with a verified HIV-negative partner, and mosquito bites).
Each exposure was also adjusted for the likelihood that the partner was infected, the partner's likely viral load, the timing of the contact, relative to seroconversion illness, the presence of skin lesions or breakdown in either partner, duration of the exposure, and occurrence of ejaculation.
HIV transmission was thought to have occurred from high-risk behaviors in 60 men and from low-to-medium risk behaviors in 15 men. Eleven of the men with low-to-medium risk behaviors reported condom-protected anal intercourse, and several noted that they assumed this practice was completely safe. In five cases, the likely source of transmission was concluded to be oral sexual contact. Three of these five men, all of whom had also engaged in protected anal sex, had penile piercings (none of which had been performed recently), and the oral sex involved insertive fellatio. In the fourth case, the infected individual reported no anal intercourse, but receptive oral sex with ejaculation in the setting of Gingivitis and open wounds in the mouth from dental treatment. In the fifth case, an infected individual reported a single instance of condom-protected receptive anal intercourse with a man who was having anal sex for the first time, and a recent history of multiple oral sexual contacts with casual partners.
This is a retrospective, small case series that cannot definitively establish either a route of transmission or the likelihood of transmission per oral sexual act. Questionnaires and interviews are flawed in that recall of past exposures may be inaccurate. However, the methodology employed in this case series was thorough and, in some instances, uncovered risk behaviors that were not initially reported. The link between possible breaches in skin integrity from piercings and the transmission of HIV needs to be further investigated.
Dr. Nagy is Assistant Professor of Medicine at Mount Sinai School of Medicine in New York.
Source: AIDS Clinical Care, December 2003. Published by www.natap.org.