Oral Sex: Risk Factors for HIV Infection

Edited transcription of a talk July 19, 1996. The same talk was presented at Manging Desire.

By Paul O'Malley, Researcher San Francisco Department of Health

 See also

I certainly agree that what is called "safe" or "protected" sex does go against the natural order of things; it is natural for us to want to engage in sex that involves skin-to-skin contact and the exchange of body fluids. Nonetheless, we do have to deal with the reality of HIV statistics in the San Francisco Bay area in the middle of the second decade of the epidemic (in San Francisco, an estimated 42% of men who have sex with men are positive.) Although I will focus in this article on risk factors involved in oral sex, my underlying concern is of course safety in general and so I would first like to say something about the most well-documented and well-known risk factor for HIV transmission among gay men; eight out of ten of the new HIV infections in our study (see sidebar) most probably could have been prevented if HIV-infected precome and semen had been kept out of the men's rectums.

As a way into the discussion of risk factors associated with HIV transmission through oral sex among men who have sex with men, I will try to answer three general questions about oral sex and HIV commonly asked by our study participants.

1. How risky is oral sex as compared to anal sex?

Oral sex is considered less risky for several reasons. First, the mucus membrane lining the mouth is much tougher, thicker, and more resilient than the anal canal, so it is more difficult for HIV to break through. Also, it is easier to remove infectious fluids from the mouth than from the anal canal. To remove precome or semen from the mouth after oral sex, a person can spit out the semen precome and gargle with water, mouthwash, hydrogen peroxide, or alcohol. (We recommend spitting and/or gargling instead of swallowing, just in case the semen, precome, or penile fluids contain blood or other agents such as gonorrhea or chlamydia which can infect the throat.) Finally, recent reports in the scientific literature say that saliva in the mouth contains enzymes which can help neutralize the virus.


2. How risky is insertive oral sex as compared to receptive oral sex?

Receptive oral sex or "giving a blow job" is considered more risky because of the potential that more HIV-infected fluid will enter the body. When someone is giving a blow job, not only is there the potential exposure to semen and precome, there is also the potential exposure to blood from a penile cut, sore, or abrasion, or even from an irritated piercing.

Furthermore, if the man receiving the blow job has another infection in addition to HIV, he may have an unusually high level of HIV present in his body fluids. This is because his immune system may not be as efficient at controlling, HIV since it is also trying to contain the other virus or bacteria. An added problem is that if the additional infection is localized in the penis, then there will probably be urethra inflammation. If so, his penile fluids may contain high levels of 'infection-fighting" white blood cells, which unfortunately also contain HIV. All this could amount to increased infectiousness.

Getting a blow job is considered a very low risk sexual activity. In general, if someone is getting a blow job and has no cuts, scratches or sores on his penis, the possibility of his being exposed to HIV is lower. If the man giving the blow job has good oral hygiene and his mouth is free of infection, then the inserter's penis is primarily being exposed only to saliva. Getting a blow job might also be safer than giving one because ejaculating might flush out the virus. Urinating and washing the penis also reduce the chances of HIV infection and of additional infection.

All this notwithstanding, it is still possible that someone's penis might be exposed to blood or other HIV-infected fluids during a blow job. This might occur if someone has an abrasion or scrape in his mouth which might bleed or if for example, during multiple-partner sex (group sex), he carries in his mouth the remnants of another person's penile fluids or blood from one blow job to the next.

3. Is precum infectious? If so, is it as infectious as semen?

Two studies have isolated HIV from precome but the potential for HIV infection from precome is most likely minimal under normal circumstances. Exposure to precome probably poses less risk of HIV infection than exposure to semen for two reasons: it has fewer infectious particles of HIV per milliliter than semen and even if a lot of precome is discharged into the mouth during oral sex, it is still less than the amount of semen the mouth or throat is exposed to by an ejaculation.

Now, onto the main discussion. There are two sides to the phenomenon of how supposedly "low-risk" oral sex becomes a high-risk activity resulting in infection: the infectiousness of the HIV-positive partner, and the susceptibility of the HIV-negative partner. I would like to go through the important factors involved in each side of the general phenomenon of ncreased-risk oral sex, and then to report our study data specifically.

There are at least five important factors involved in susceptibility to HIV infection through oral sex.

1. A person's genetics, immune system and biological history. Some individuals may have an increased susceptibility due to these factors. Some people, for example, are prone to chronic allergies such as "hay fever." During outbreaks of these allergies, the immune system is activated. The tissue lining of the throat and nasal passage might become inflamed and therefore more permeable to HIV. Also, if someone with an allergy outbreak is constantly blowing his nose, then irritation and bleeding are more likely to occur, which might create a possible route for HIV to enter his system.

I spoke with one man who believed he became infected with HIV by getting a blow job. He told me that he had a history of picking up STDs in the past from insertive oral sex, claiming to have been infected with urethral gonorrhea this way, as well. We can speculate that either due to his genetics, immune system, or some other biological factors, he was not sufficiently resistant to STDs, including HIV, when exposed during insertive oral sex.

2. Oral hygiene. First of all, most Americans across the board have lousy oral hygiene, so this factor may play less of a differentiating role than we might think. Nonetheless, if you have poor oral hygiene or are suffering from gingivitis, receding gums or periodontal disease, or have recently had oral surgery, you may be at increased risk since the virus may more readily find a port of entry into your body. Also, while it is advisable to brush and floss your teeth regularly for excellent oral hygiene, it is probably best not to engage in receptive oral sex for a few hours after flossing or heavy brushing, just in case this leaves abrasions on your gums or mucus membrane.

3. Other infection. You may be at increased susceptibility to HIV if you already have another infection or if you are exposed to HIV and another infection simultaneously. This may be true whether the infection in systemic (such as hepatitis) or localized in your mouth and/or throat (such as gonorrhea or chlamydia.) The reason a systemic infection like hepatitis may increase susceptibility is that if HIV gains entry into your body, your immune system may not be able to fully focus its power against HIV, either because it may already be trying to control another infectious agent or because it may be trying to combat two new infectious agents simultaneously. The reason a localized infection such as gonorrhea or chlamydia might increase your susceptibility to HIV infection is that the tissue lining in your mouth, throat and/or nasal passage might be inflamed, irritated, or abraded by the infectious agent, and so would be an easier target for HIV.

With both systemic and localized infections, another factor which might heighten your susceptibility is the increased amount of lymphocytes in your blood. Lymphocytes are a type of white blood cell activated during infection. Unfortunately, they are also a major target for HIV infection, and it may be easier for the HIV viral particle to locate a susceptible lymphocyte during a period when they are activated due to another infection. The same principle may also apply to other infectious agents such as a cold, flu, strep, staph, as well as to nasal and throat inflammation due to chronic allergies. In general, times when you are feeling run down are times to be more careful.

4. Drug use. The most important factor of drug use is probably the effect the drug has on a person's behavior during sex. Sex on a stimulant such as cocaine, speed, or poppers is likely to be more vigorous, prolonged and intense. Sex may also be prolonged because, on certain drugs, it can take longer to reach orgasm.

Furthermore, during sex on drugs, a person may get caught up in the moment and thus may not notice the wear and tear to his penis and/or throat. He may also be able to ignore any pain that he does feel until after the sex is completed. Study participants do report that, while on speed, the heightened sexual experience and sense of well-being take precedence over everything else--including the concern about being safe.

Study participants also report that popper use makes it easier to engage in rougher or more penetrative sex such as "deep-throating," especially if their sex partner has a large penis. In these situations, the tissue of the throat is more likely to become irritated, especially if the sex is prolonged. Moreover, there has been scientific speculation that popper use causes a dilation of the blood vessels and thus increases the risk of HIV transmission. The biological theory is that if HIV enters the tissue during the "rush" when the mucus membrane lining the mouth and throat is engorged with blood, then the virus more easily locates a white blood cell or lymphocyte to infect. finally, some scientists also believe that certain recreational drugs, such as popper can temporarily suppress the immune system and so increase a person's susceptibility to infection if exposed.

5. Recent immunization against an infection. The immune cells of seven out of ten HIV- uninfected individuals were more easily infected in the test tube after immunization than before immunization, the National Institute of Allergy and Infectious Diseases (NIAID) reported in a recent issue of "The New England Journal of Medicine." The immunization studied was the tetanus booster shot. During the temporary immune system activation which the immunization shot normally causes, the HIV-uninfected individual may be at increased susceptibility to HIV infection via unprotected anal or oral sex (with an infected partner) the study suggests. This enhanced susceptibility may last for several weeks after the immunization.

The other side of increased-risk oral sex--the infectiousness of the HIV- positive sex partner--also involves at least five factors.

1. The stage of HIV infection. Depending on the stage of HIV infection, the HIV-infected person may be a more efficient transmitter of the virus. Shortly after becoming HIV-infected, a person experiences what is called an "acute viral syndrome." During this period, the virus is very active and the person has a high viral load. Theoretically, even though he may not have obvious symptoms of HIV and may be feeling fine, the person who has just become infected with HIV may be very infectious to others during this acute viral syndrome. A person's viral load may be elevated-and he may therefore be more infectious to others--when he has a low T-cell count, a dropping T-cell count, or a diagnosis of AIDS. During these times, the virus is more active in his system, which is the reason for both his symptoms of HIV infection and his increased infectiousness.

An HIV-infected individual may also periodically be more infectious over the course of his HIV infection, due to causes that researchers do not yet completely understand (the explanation may involve a transient infection with a virus or bacteria.) Also, an individual may be less infectious if he is taking anti- viral medications such as AZT, but this has yet to be documented by a research study.

2. Infection with an additional disease. If the HIV-positive person is also infected with another disease such as hepatitis, herpes, gonorrhea, chlamydia, etc. then he might be a more efficient transmitter of HIV. The concurrent infection might activate the immune system of the HIV-positive individual, which would result in more HIV replication, an increased HIV viral load, and possibly greater infectiousness.

It might be useful to return to the scenario I mentioned earlier; if an HIV positive individual also has penile gonorrhea, his urethra will be irritated and inflamed. His immune system will try to control the penile infection by sending lymphocytes to his urethra. These lymphocytes cause the pus-like discharge from the penis when a man has urethral gonorrhea. Since HIV thrives on lymphocytes, a man with urethral gonorrhea probably has more HIV present in his urethra than normal and so discharges more HIV than normal when he is getting a blow job.

3. The infectivity and virulence of HIV. Some strains of HIV are more adept at breaching the immune defenses and entering a cell, and some strains are better at replicating within the cell and destroying the cell in the process.

4. Blood exposure blood is on average 100 times more concentrated with HIV than precome or semen. If a cut or tear occurs during oral sex, the risk of exposure to blood increases. Occasionally, blood is in the precome and semen of individuals with prostate or urethral infections. I have also heard of another possibility if blood exposure oral contact with an irritated or incompletely healed penile piercing. If you or your sex partner have any penile piercings -- especially if they are new -- it is very important to makes sure they have healed completely before you consider engaging in oral sex without a condom. It is also important to always keep the area clean and to keep an eye out for irritation after sex.

5. Recent immunization against an infection. When an HIV-infected individual is immunized, his immune system is activated and his viral load is increased, reported The National Institute of Allergy and Infectious Diseases (NIAID) recently in "The New England Journal of Medicine." The immunization studied was the tetanus booster shot, which was given to 13 HIV-infected volunteers. Following immunization may pose a greater risk of HIV infection (although a comparison of viral load levels in blood and concurrent viral load levels in semen was not conducted.)

While the risk of becoming HIV-infected from receptive oral sex is still considered low under normal conditions, our study group believes these factors played an important role in the infection of five study participants who probably became HIV infected from unprotected receptive oral sex. Four of these men come from the earlier study of 675 men, and one comes from our more recent study (see sidebar.)

As an experienced interviewer, I have no reason to doubt that the information we received from these study participants was reliable. I have personally spoken with four of these men in considerable detail, as well as with the staff member who interviewed the fifth man. Our research group always attempts to re-interview all our newly infected study participants for risk factors associated with HIV infection to make sure the information we have collected is accurate. For these five men, the risk was not low, it was 100%. Oral sex was clearly not safe sex for them.

I. The first man reported 10 episodes of unprotected receptive oral sex with ejaculation during the time he became infected with HIV (HIV sero-converted.) He also remembered noticing that semen was in his nasal passage and sinuses after one occasion of receptive oral sex. One partner with whom he engaged in receptive oral sex on several occasions was known to be HIV- positive. The HIV-positive sex partner had low T-cells, symptoms of HIV infection, and he was taking AZT. The study participant usually deep-throated this partner, who he claimed had a very large penis. The participant reported using poppers routinely in association with oral sex. His mouth was noted to have gum recession on examination both before and after his date of HIV infection. He also reported a history of periodontal disease which had been treated several years previously.

The factors that may have played a role in this new HIV infection are: 1. Sex with HIV+ partner 2. Stage of HIV infection (his sex partner had low T-Cells and probably a high viral load) 3. Drug use (he usually used poppers) 4. Oral hygiene (he had gum recession, periodontal disease). 5. Amount of HIV- infected fluid (his HIV + partner usually ejaculated in his mouth and throat) 6. Deep-throating a large penis.

II. The second man reported having had approximately 400 male sex partners, with whom he reported approximately 900 episodes of receptive oral sex with ejaculation and 20 episodes of rimming during the time frame of his sero-conversion. Although he was unaware of the HIV status of most of his sex partners, one partner with whom he engaged in receptive oral sex was known to be HIV-positive. Regardless, we can assume that approximately 42% of his 400 sex partners may have been HIV-positive, due to the estimated prevalence of HIV infection among men who have sex with men in San Francisco. This man reported the use of poppers, usually in association with oral sex. Gum recession was noted when his mouth was examined, but his mouth was otherwise normal in appearance.

The factors which may have applied in this case are:

1 .Sex with HIV+ partner

2. Drug use (he usually used poppers)

3. Number of partners (he had receptive oral sex with 400 partners)

4. Amount of HIV-infected fluid (he had 900 episodes of receptive oral sex with ejaculation.

III.The third man reported several episodes of unprotected oral sex with an HIV-positive individual during the time he sero-converted. He reported that he was infected with gonorrhea of the throat by his sex partner. The HIV-positive sex partner did not ejaculate into the mouth or throat of the study participant, but gonorrhea-infected fluids from his penis were transmitted to the throat of the study participant and these fluids probably also contained HIV. The study participant continued to engage in receptive oral sex with this partner even after his mouth and throat became inflamed and irritated. He reported use of poppers, usually in association with oral sex. He also reported one episode of insertive anal sex with a condom which did not break or slip during this same time period.

The factors which may have applied in this case are:

1. Sex with HIV+ partner (n.b. the participant did not know this at the time)

2. Drug use (he used poppers)

3. Oral hygiene (inflamed mouth due to gonorrhea infection)

4. Additional infection (his sex partner had urethral gonorrhea and he acquired gonorrhea of the throat; n.b. neither individual realized this at the time)

IV.The fourth man reported no receptive or insertive anal sex during the six months between his last HIV-negative test and his first HIV-positive test. Indeed, he reported no receptive anal sex since 1975. He did report one instance of protected insertive anal sex in which the condom did not break or slip with an HIV-positive sex partner, but this sexual episode -- since it was protected and occurred more than six months before his acute HIV viral syndrome -- is highly unlikely to be related to his new HIV infection. Between his last HIV-negative and first HIV-positive tests, he did report receptive oral sex with four sex partners. Two of the four reported sex partners tested HIV- negative following the study participant's HIV seroconversion and thus were ruled out as possible sources of his HIV infection. With his third sex partner, who was known to be HIV-positive, the study participant had only protected receptive oral sex. His fourth sex partner appears to be the most likely source of his HIV infection.

Our study participant reported 12 episodes of receptive oral sex with this partner, 6 of which involved ejaculation and swallowing. After our study participant tested HIV-positive, his fourth sex partner, who had not known his HIV status, was HIV tested and found out that he was also HIV-positive and, subsequently, that he had a high viral load.

The study participant reported that he routinely used poppers when engaging in receptive deep-throating with this partner. He also reported that his sex partner had a large penis and that he did not use a condom with this man because of the tightness and discomfort it would have caused in his throat. His oral cavity was examined by a physician after his HIV sero- conversion and some gum recession was noted, but otherwise his mouth was normal in appearance. The study participant was also diagnosed with strep throat at the same time as his HIV sero-conversion syndrome. However, since our study participant last had sex with this HIV-positive partner one month before his HIV sero-conversation syndrome, our research group believes this I most likely a coincidence. Unless our participant was a strep carrier, he was probably not infected with strep throat at the same time he was having receptive oral sex with this HIV-infected partner.

The factors that may have played a role in this new HIV infection are:

1. Sex with HIV+ partner (n.b. neither he nor his partner realized this at the time)

2. Stage of HIV infection (his partner had a high viral load)

3. Drug use (he usually used poppers)

4. Additional infection (questionable infection or exposure to strep at a time of exposure to HIV)

5. Amount of HIV-infected fluid (6 episodes of receptive oral sex with ejaculation)

6. Oral hygiene (although gum recession was noted, it was not unusual and his mouth was otherwise normal in appearance)

7. Deep-throating a large penis

V. The fifth man, a participant in our HIV Vaccine Preparedness Study, reported an accident during the time he became infected, which occurred while he was having receptive oral sex with a partner who subsequently came down with AIDS.

The HIV-infected sex partner was not circumcised. When he first thrust his penis into the study participant's mouth, a piece of foreskin was caught between the front teeth. According to the study participant, his sex partner did not realize what had happened to his foreskin because he claimed to have felt only a momentary sting when it occurred and he also claimed to have been "caught up in the act of getting head." The study participant noticed bleeding from the sex partner's torn foreskin when they finished having oral sex and then noticed the piece of foreskin lodged between his front teeth. The study participant also reported 12 episodes of receptive oral sex with ejaculation, protected anal sex with a reported HIV-negative sex partner, and one episode of condom breakage with a reported HIV-negative sex partners.

The factors which may have played a role in this case are:

1. Sex with HIV+ partner

2. Stage of HIV infection (his sex partner was diagnosed with AIDS less than a year after, so he probably had low T-cells and a high viral load when they had sex)

3. Blood exposure (torn foreskin with bleeding)

4. Amount of HIV-infected fluid (12 episodes of receptive oral sex with ejaculation)

Apparently, as with these five men, the factors which I have tried to outline can accelerate the risk of oral sex from low to high either by increasing the susceptibility of the HIV-negative partner, or by increasing the infectiousness of the HIV-positive partner.

The essential thing to remember is that life is unpredictable. We have seen many instances of factors which cannot be anticipated or assessed in advance: the partner who does not know he is HIV-positive because he has only recently become infected and, if he has ever been HIV tested, has tested negative; and the HIV-positive sex partner whose infectiousness is heightened by urethral gonorrhea; and the torn foreskin/blood exposure accident. Clearly, we can not do a complete, accurate, and error-proof risk assessment before we decide to engage in unprotected receptive oral sex.

In short, our decision of what risks to take will necessarily be based on incomplete knowledge. So, once educated and informed, we must base our decision on ourselves as individuals.

Our research groups has conducted two major studies in the past 12 years assessing risk factors for HIV transmission among men who have sex with men. For the first, the "HIV Natural History Study" 675 HIV-negative gay or bisexual men were enrolled between 1984 and 1992 and were followed for an average of 4.5 years. These men consented to testing for HIV antibodies annually and were also interviewed and counseled regarding their sexual practices. For the second and most recent, "The HIV Vaccine Preparedness Study," ongoing since 1993, we have enrolled approximately 1,200 men who have sex with men. The participants are required to be sexually active gay or bisexual males, HIV-negative at entry, consent to HIV antibody testing every 6 months, to be interviewed at each visit about their willingness to participate in HIV vaccine trials, about their sexual behavior, drug usage, and other possible factors that might increase their risk of exposure to HIV.


This more recent study has documented 41 new cases of HIV infection, one (2.4%) of which we attribute to transmission through oral sex. The primary risk factors reported by the 41 are as follows:


1.32 of 41 (78.0%) reported unprotected receptive anal sex as the primary risk factor. 3 out of 32 (approximately 10%) of these who reported unprotected receptive anal sex claimed that it was due to condom breakage and/or slippage.

2. 5 of 41 (12.3%) reported unprotected insertive anal sex as the primary risk factor.

3. 3 of 41 (7.3%) reported IV drug use as the primary risk factor.

4.1 of 41 (2.4%) reported receptive oral sex as the primary risk factor.

The applicability of these data may be somewhat limited for a few reasons. First, the participants in our study may not be representative of all men who have sex with men in San Francisco, because a large percentage were recruited from bars, dance clubs, and the STD clinic, rather than by means of a community-based survey. Also, this study was not designed to look at the oral sex issue, but at the suitability factors for enrollment in an HIV vaccine trial. Our sample size of 41 seroconverters is also fairly small.

In addition, any attempt to ascertain the proportion of HIV infections which result from oral sex faces a fundamental difficulty. Most of our study sero- converters -- and most sero-converters in general -- engage in both anal and oral sex. This multiple-risk activity results in a "masking effect" which makes it impossible to determine the exact percentage of new HIV infections which can be attributed to oral sex. While the ideal study design would be to follow a group of men who report engaging in no form of penetrative sex other than oral sex, the federal government and most public health authorities have never expressed much interest in funding or performing such a study. They believe that unprotected oral sex plays a minor role in the overall spread of HIV in the United States. Though some researchers estimate the proportion of HIV infections due to oral sex is probably 1% or less, other researchers claim the true percentage could be as high as 4% to 5%. And while we attribute 2.4% of the new HIV infections in our study to oral sex, even an estimate as low as 1% would still translate into 6 or 7 new cases of HIV infection due to oral sex each year in San Francisco alone (based on the estimate that 650 men who have sex with men become HIV-infected each year in San Francisco.)

[ Home Page ][ Site Map ][ Links ][ Feedback ][ Submit ][ HIV Prevention Services ]
[ Berkeley Free Clinic ][ Confession Box ][ Counselor Resources ][ Relationships ][ Theorizing Desire ]