Risks of Oral Sex

John Tighe
Volume 1, Number 2, March 1991

Research Update

Various levels of risk have been associated with oral sex from the time sexual behaviors were first evaluated for their risk of HIV infection. While early reports were inconclusive, in 1990 three cases of infection were reported -- two cases were published and one case was anecdotally reported -- in which oral sex was the only reported risk behavior. The subjects, men in the San Francisco City Clinic Cohort Study, tested HIV antibody positive to ELISA and Western Blot tests. (1,2)

In the published cases, each subject tested positive after reporting that receptive oral sex with ejaculation was his only high-risk activity. The two subjects indicated that they had not engaged in anal sex during the previous two years. They had participated in episodes of receptive oral sex with ejaculation with many partners.

The subject whose case was reported anecdotally told researchers that in the year since his last negative test result he had engaged in receptive oral sex, and he engaged in a single episode of receptive anal sex in which a condom was used.

Blood samples taken at the time of all three subjects' most recent negative test result showed that the men also had negative results to polymerase chain reaction (PCR) assays. PCR is an advanced laboratory test that can detect HIV when antibodies are absent, such as during the infection "window period," which is the time after an individual is infected, but in which antibodies to the virus have not yet developed.

The cases are the first in the cohort study in which oral sex alone has been identified as the probable route of transmission. The study include about 600 gay and bisexual men in the San Francisco area who are regularly tested for HIV antibody. Most of the men who have tested positive have done so after engaging in anal sex without a condom.

In a separate study initially presented in 1990, researchers reported that 13 of 82 men who tested antibody positive for HIV reported that they engaged in receptive and insertive oral sex since previous negative tests, but no other risk factors, such as anal sex. The individuals from this study were chosen from participants in three San Francisco studies, and included the cases reported by the San Francisco Clinic Cohort Study. The 13 HIV-infected subjects tested antibody positive about one year after their last negative test.

Researchers state that condom use was not consistent in the groupm and it was not known whether subjects had halted their oral sex practices before ejaculation. Researchers have released only preliminary information from their study, and seek to have their findings duplicated elsewhere before they publish their results. (3)

In another study, published in 1988, researchers in a European cohort of gay men reported five cases in which oral sex was the probable route of infection. (4) While subjects from the European study seroconverted in tests performed a mean of 5.4 months after a previous negative test, researchers say that subjects may have been in the infection window period. PCR analysis, which is not subject to such a window period, was not performed for these cases.

Many antibody test counselors report seeing clients who have decribed oral sex as their only risk behavior. The anonymous testing program in San Francisco provides antibody test results to about 200 clients per week. About 8.5% of all clients seen in the program in the first half of 1990 tested antibody positive. A test site supervisor anecdotally reported that of subjects testing antibody positive during 1990, about one male client every other week stated that oral sex was his only risk behavior. Women testing positive have not reported oral sex as an exclusive risk behavior. Clients who have named oral sex as their only risk behavior have stated that for prolonged periods they have not engaged in other risk activities.

Counselors in other parts of the state report seeing a significantly smaller percentage of individuals who state that oral sex has been their only "high-risk" activity.

While most reported cases of HIV infection by oral sex appear to be from the insertive partner to the receptive partner during fellatio, transmission of HIV from receptive partner to insertive partner is also considered a potential risk. A 1988 study reported a case of tranmission from a female prostitute to a 60-year old male client. The man, who had been married for more than 30 years but had not had sex with his wife for several years, reported his only risk activity as insertive fellatio with the prostitute (5). [He also engaged in cunnilingus with the prostitute on several occasions over a number of years, which is also considered a risk factor. --troyer]

Because vaginal secretion, as well as menstrual blood, can contain HIV, researchers consider oral sex with women, cunnilingus, to be a risk behavior.

Some reserchers have disputed the numerous reports of infection through oral sex. They suggest that infected individuals may want to attribute infection to oral sex because they are unwilling to acknowledge that they have participated in unprotected anal sex, a behavior that carries a stigma for some people.

It has also been suggested that individuals may have been engaging in unprotected anal or vaginal intercourse, but were in the infection window period at the time previous tests were conducted.

Subjects' Reported Behaviors in Previous 30 Days (n=327)

     75%   Mutual masturbation
     70%   Oral sex with no semen
     68%   French kissing
     37%   Anal sex with a condom
     22%   Oral-anal contact
     22%   Oral sex with semen                      
      9%   Anal sex--no semen, no condom
      9%   Anal sex--semen, no condom
      2%   Fisting
Source: San Francisco AIDS Foundation -- 1989 survey of gay men in San Francisco. Note: Subjects were not asked if condoms were used during oral sex.

Oral Sex with Male Partners in Previous 30 Days*

                                    oral sex           oral sex
                                   with semen        without semen
Subjects with one primary partner     2.2                 0.5
Subjects with multiple partners       5.6                 3.3
*Mean number of episodes

Source: San Francisco AIDS Foundation -- 1989 survey of gay men in San Francisco [Note: numbers estimated from graph in original HIV Counselor Perspectives printed version by John Troyer]

Assessing the Risk of Oral Sex

Most researchers agree that HIV can be tranmitted during oral sex. However, researchers are hampered in their efforts to determine the level of risk from oral sex for several reasons, including the inability to document cases of transmission beyond a doubt.

It appears that the risk of infection from oral sex with an HIV infected person varies depending on an individual's oral and on the type of oral sex practiced. An individual with gum disease, someone susceptible to ulceration or bruising in the mouth or gums, or someone who vigorously brushes or flosses his teeth prior to or after receptive oral sex is believed to be at increased infection from oral sex.

The American Association of Physicians for Human Rights (AAPHR) issued "refined" guidelines in 1990 on the risk of transmission from sexual activities, including oral sex. All types of oral sex were rated as having "some risk," compated to various forms of anal and vaginal intercourse, which were all ranked as "high risk" behaviors.

The following are AAPHR's rankings of various oral sex practices, in descending order of risk:

  • oral sex with men with ejaculation and without a condom
  • oral sex with women [without a barrier]
  • oral sex with men with preejaculate and without a condom
  • oral sex with men with no ejaculation or preejaculate and without a condom
  • oral sex with men with a condom

Because oral sex with women can put partners in contact with vaginal secretions and blood, AAPHR states this behavior may present a greater risk than oral sex with men who do not ejaculate or secrete preejaculate. In addition, AAPHR's guildelines include concerns that are considered "unresolved." These include the role of preejaculate in transmission and the effectiveness of latex dams or other barriers preventing transmission during oral sex with women.

Researchers attempt to dismiss as incorrect the beliefs that transmission of HIV during oral sex can occur only after ejaculation, or only when an individual swallows another person's semen. In fact, researchers generally believe that the virus can be present before ejaculation, in the form of "pre-ejaculate" or "pre-cum," and that an individual can be infected by pre-ejaculate.

In addition, some epidemiologists state that an insertive partner may have cuts on his penis, or the receptive partner may have cuts in his mouth, and so either partner could be infected from cuts. Also, some men do not always know beforehand when their ejaculate is going to be released and are therefore unable to tell their partners.

Gum disease, which makes an individual susceptible to bruising easily or to developing ulceration, is a common chronic ailment. Some individuals who have experienced signs of gum disease may incorrectly believe that the absence of symptoms means they have recovered and they are free of disease. Dentists reports that the absence of symptoms does not mean an individual is free of gum disease, and that most individuals who have a history of gum disease cntinue to be susceptible to bleeding and open sores. Men who are insertive partners during oral sex may be susceptible to ulcerations and sores on the penis.

Why Reports of Transmission Have Increased

Researchers have suggested several possible reasons for increased reports of transmission attributed to oral sex. Among them are the following:

  • Oral Sex has become much easier to isolate as a risk factor.

    As individuals have reduced the frequency of other risk behaviors, such as unprotected anal sex, oral sex has become easier to identify as a cause of transmission. The actual risk of infection has not necessarily increased, but only recently has the practice of oral sex been considered a possible cause of infection.

  • Frequency of Oral Sex

    Surveys and reports from health educators across the state indicate that gay men are having oral sex with greater frequency now than during the mid-80s or before. In an 1989 survey in San Francisco, 70% of respondents reported having oral sex without the exchange of semen in the previous 30 days, and 22% reported having oral sex with semen. The telephone survey, conducted primarily of gay men, showed an increase in the frequency of oral sex and a decrease in anal sex compared to a similar survey in 1987. (6)

  • Failure to detect throat based gonorrhea, herpes, syphilis, or other sexually transmitted diseases (STD)

    Researchers believe that transmission of HIV may be linked to inflammation of the throat, which is frequently cause by syphilis or herpes. The incidence of several types of STDs has increased in the past three years among gay men in several regions of the country. The prevelence of throat based gonorrhea, for which tests are not routinely performed, has also increased.

  • Intensity of various forms of oral sex

    As individuals have reduced or eliminated other forms of sexual behaviors that can be considered highly physical and penetrative, individuals' oral sex practices may now be more physical and involve more abrasive contact with the mouth.


(1) Lifson AR, O'Malley PM, Hessol NA, et al. HIV seroconversion in homosexual men after receptive oral intercourse with ejaculation: implications for counseling concerning safe sex practices. _American_Journal_of_Public_Health, 1990; 80(12):1509-1511.

(2) Unpublished data. Based on personal conversations with Paul O'Malley, June and July 1990.

(3) Samuel M, Seroconversion for HIV antibody among gay and bisexual men enrolled in three San Francisco cohort studies: risk factors for recent seroconversion. Presentation from the symposium :The Epidemiology of AIDS and HIV Infection in Gay and Bisexual Men: Current Trends and Implications for the Future," 118th Annual Meeting of the American Public Health Association, Sept 30-Oct 4, 1990, New York City.

(4) Rozembaum W, Gharakhanian, Cardon B, et al. HIV tranmission by oral sex. _The_Lancet_ 1988; 1:1395

(5) Spitzer PG, Weiner NJ. Tranmission of HIV infection from a woman to a man by oral sex. _The_New_England_Journal_of_Medicine_, 320(4): 251

(6) San Francisco AIDS Foundation, Communication Technologies. HIV related knowldge, attitudes, and behaviors among San Francisco gay and bisexual men: results from the fifth population-based survey. Unpublished report, 1990.

Truax SR, Ramirez A, Fraziear T. Annual Evaluation of the Anonymous Human Immunodeficiency Virus Testing Program. Sacramento: Office of

HIV Counselor Perspectives, V1N2, 3/91
[email protected] (Thu Dec 7 00:10:24 1995)