Sexual contact is the most common route of HIV transmission. By December 2001, 51% of all HIV infections among adolescents and adults reported to the U.S. Centers for Disease Control and Prevention (CDC) were sexually transmitted (35% by male homosexual contact, 11% by heterosexual contact in females, 5% by heterosexual contact in males).(1) Worldwide, heterosexual transmission is the most common route of HIV infection. Given the importance of sexual transmission in the HIV epidemic, many HIV prevention strategies have focused on identifying and promoting safer-sex practices. As the name implies, these practices are thought to be "safer" than other sexual practices in that they help reduce (but do not necessarily eliminate) the risk of transmitting HIV from one sexual partner to another.
Clinicians and health educators often have the unique opportunity to discuss topics of an intimate nature in a professional setting. With this privilege comes the responsibility to be respectful and nonjudgmental. In some cases, the goal of safer-sex education may be to help someone minimize risk to him- or herself; in others, it may be to help someone minimize risk to others. The goal of teaching safer sex is to provide not only information, but also counseling to help individuals or groups to make the most appropriate choices for risk reduction.
Not everyone will open a discussion about safer sex with a health care provider. For example, some people may not ask about safer sex because they do not perceive themselves to be at risk. Others may be too embarrassed to open the discussion. It is incumbent on health care providers to perform HIV risk assessment as an integral part of the medical history, and to provide HIV prevention counseling as an integral part of patient education and anticipatory guidance. Risk assessments and appropriate counseling should be performed periodically to facilitate not only initiation, but also ongoing maintenance, of risk-reduction behaviors.
Development of effective antiretroviral therapy (ART) has resulted in optimism for many HIV-infected patients. As efforts to develop even more effective treatments and preventive vaccines continue, it is critical to continue aggressive prevention efforts as a vital component of the battle against HIV. Although ART can result in dramatic reductions in HIV viral load, it is not a cure for HIV disease; thus prevention should still be the first line of defense. In addition, although theoretical models have suggested that ART may combat the HIV epidemic on a population level, models that assumed steady or increased levels of safer-sex practices were more likely to predict reduction in new HIV infections than models that assumed decreased levels of safer sex.(2) ART may reduce, but cannot be expected to eliminate, the potential for an infected individual to transmit HIV to an uninfected individual.(3) Therefore, even individuals receiving effective ART should, at a minimum, initiate and maintain prevention practices with uninfected persons or persons of unknown HIV status. In addition, ART is available only for a small proportion of the world's HIV-infected population. Thus, prevention remains the main line of defense for these individuals.
This chapter reviews the evidence that has led to the development of safer-sex guidelines, and concludes with specific recommendations for safer-sex practices.
| Evidence for Sexual Transmission of HIV|
| Isolation of HIV in Body Fluids|
Researchers can consistently culture or otherwise detect HIV not only in blood, but also in semen (4-6) and cervical secretions (7-9) of infected persons. Infectious HIV exists in saliva,(7-13) tears,(14) and urine (15); however, it has only been recovered from these fluids at extremely low titers. In addition, no report has documented cases of HIV transmission by these fluids. Hence, saliva, tears, and urine are highly unlikely sources of HIV transmission.
Infectious HIV has also been isolated in breast milk, and transmission from HIV-infected mothers to nursing infants has been well documented.(16) Breast milk is not commonly encountered during sexual intercourse. However, should individuals accidentally or intentionally come in contact with HIV-infected breast milk during sex, care should be taken to avoid mucosal contact.
| Epidemiologic Studies and Case Reports|
Epidemiologic evidence in support of male-to-male,(17-28) male-to-female,(29-43) and female-to-male (31,33,36,39,40,42-44) sexual transmission of HIV infection is abundant. The risk of HIV infection among women who have sex with women appears to be largely attributable to other risk factors (sex with men, injection drug use).(45-48) Female-to-female sexual transmission per se is uncommon, with rare case reports of possible HIV transmission by this route.(49-54) A 2003 case report suggested that sexual practices that can expose sex partners to each other's blood, such as the shared use of sex toys or vaginal penetration with hand ("fisting"), are a possible route of female-to-female sexual transmission.(54)
| Number and Selection of Sexual Partners|
Results from early epidemiologic studies of HIV infection in homosexual men revealed that sexual activity with many different partners carries a high risk of HIV infection.(18,21-23) Many of the published reports of heterosexual transmission present no detailed data regarding this risk factor, presumably because the researchers examined transmission from HIV-infected persons to their monogamous sexual partners.(29,30,33,34,37,38,41) Researchers who did specifically analyze this issue were unable to demonstrate an association between number of sexual partners and risk of HIV infection, perhaps because the median number of partners was relatively low in these studies (1-4 partners in 5 years).(32,35,36)
Early research on selection advised that the choice of a partner was the most important determinant of transmission of HIV during a sexual encounter.(55) Current research has shifted emphasis from an explicit concern with absolute numbers of sexual partners to a model that situates an individual's selection of sexual partners in the context of the population seroprevalence, the likelihood that an individual has been tested for HIV, the likelihood that the test result was accurate, the likelihood of infection through insertive or receptive oral, vaginal, or anal sex, and the degree to which condom use reduces the probability of transmission during these acts. The model shows that individuals can reduce their risk by choosing a partner who has tested HIV negative, choosing a safer-sex act, using a condom, or some combination of these factors. For heterosexuals, whose population prevalence was modeled at 1%, choosing one risk-reduction behavior substantially reduced the absolute risk of HIV infection. However, for men who have sex with men (MSM), whose population prevalence was modeled at 10%, the choice of only one risk reduction behavior did not significantly lower the absolute risk of HIV infection.(56) As these models draw their assumptions from the epidemiology of HIV in developed-country settings, the applicability of their conclusions to high-prevalence heterosexual epidemics in the developing world seems limited. It is therefore important in high-prevalence settings to continue to encourage risk reduction behaviors that include both safer sex practices and HIV testing.
When both sexual partners are HIV positive, it is still reasonable to consider safer-sex practices to reduce the likelihood of infection from other sexually transmitted diseases, such as herpes, gonorrhea, chlamydia, syphilis, and hepatitis B and C viruses.
Case reports have confirmed that HIV-positive individuals can acquire different strains of HIV through sexual exposure.(57,58) Acquisition of a new strain of HIV in an individual who is already HIV-infected is known as superinfection. There is not yet a clear understanding of the probability of superinfection at the individual or population level, nor do we fully understand its impact on the long-term health of HIV-infected individuals. Superinfection may cause acute viral syndrome, and transmission of drug-resistant strains may reduce options for future ART,(59,60) suggesting a role for continued vigilance and safer-sex decision making by HIV-infected individuals even when both partners are infected.(56)
| Risk Associated with Specific Sexual Practices|
Epidemiologic investigations of HIV transmission provide substantial evidence that some sexual practices are associated with a high risk of HIV transmission, whereas others are not.
| Penile-Vaginal Sex|
Heterosexual intercourse is presumed to be the most common mode of HIV infection worldwide. Studies of male-to-female and female-to-male transmission provide strong epidemiologic evidence that heterosexual transmission of HIV does occur via penile-vaginal intercourse.(31-41) Vaginal sex during menstruation may increase the risk of transmission from an infected female to an uninfected male,(40) but probably does not increase the risk of transmission from an infected male to an uninfected female.(43,55,56)
The efficiency of heterosexual transmission of HIV and per-act risk of infection are the subjects of debate in the epidemiologic literature. Early epidemiologic studies on heterosexual transmission in Western countries established that male-to-female transmission in the vagina was significantly more likely than female-to-male transmission from the vagina,(39,40,43,61,62) with estimates in three studies ranging from 1.9,(40) 2.3,(61) and 8.0 (62) times greater efficiency of male-to-female transmission. Per-act infectivity in two studies (62,63) was found to be low: 0.0005 and 0.0009 for male-to-female transmission, and 0.0003 and 0.0001 for female-to-male transmission. However, studies conducted in developing countries have estimated that per-act transmission probabilities are greater by a factor of 10 (44,64) for both male-to-female and female-to-male transmission.(65)
Although the greater efficiency of male-to-female versus female-to-male transmission has also been observed in developing countries, a systematic review of the literature found a greatly enhanced efficiency of female-to-male transmission in the high-prevalence epidemics of Asia and sub-Saharan Africa. The ratio of male-to-female summary mean transmission rates in the developing world compared to the rate in Western countries was 2.9, whereas for female-to-male transmission this ratio was 341. Women in some developing countries may be more infectious due to higher prevalence of sexually transmitted infections (STIs) and untreated HIV disease, although the authors state that evidence for the relative importance of these factors is unclear. The greater susceptibility of men in developing countries is also difficult explain, and may include a low prevalence of male circumcision, poor genital hygiene, a high prevalence of genital ulcer disease, and a high prevalence of unprotected sex with women having a high probability of being HIV infected.(65)
| Penile-Anal Sex|
Strong evidence exists that being the receptive partner in unprotected penile-anal intercourse is associated with a high risk of HIV infection. Transmission of HIV to the receptive partner probably occurs as a result of the deposition of HIV-infected semen on traumatized rectal mucosa. More recently, studies have suggested that exposure to infected pre-ejaculate through anal intercourse may also carry a high risk of transmission.(66) Unprotected receptive anal intercourse (URAI) has been consistently described as an independent risk factor for HIV infection among MSM.(18-28) One recent study estimated the per-act risk of HIV infection from URAI with a partner who is HIV-positive at 0.82% (82 in 10,000) and with a partner of unknown serostatus at 0.27% (27 in 10,000).(67)
Several investigators found that receptive penile-anal sex is also a risk factor for male-to-female transmission.(32,35,37,38,40,43) Others failed to find this association among heterosexual couples.(33,34,36,41) Of the latter studies, however, three had small sample sizes, which may have made a relationship between anal sex and HIV infection impossible to detect.(33,34,36) It is probable that unprotected anal sex between serodiscordant heterosexual partners carries a similar per-act risk as it would between MSM, with greater risk incurred by the receptive female partner.
Whether being the inserting partner in unprotected penile-anal sex is an independent risk factor for HIV infection is not well understood. Most early studies did not demonstrate a statistically significant association between this practice and HIV infection among MSM.(18-21,23,25,26,28) This was not taken as evidence that the behavior was free of risk.(27) One recent study has estimated the per-act risk of unprotected insertive anal sex with an HIV-positive or unknown status partner at 0.06% (6 in 10,000).(67) This risk, although 4-14 times less than that estimated for URAI, remains considerable. The lack of more complete information on the risk of unprotected insertive anal sex reflects the research community's attention to the riskier activity URAI, rather than any consensus that unprotected anal sex has been determined to be of low risk to the insertive partner.
| Rectal Douching and Rectal Fisting|
Studies of transmission among MSM have revealed that rectal douching increases the risk of HIV infection.(20,21,23,27) A similar association between fisting (penetration of the anus with the hand) and HIV infection was observed in some studies,(18,19,27) but not others.(21) One presumptive mechanism for transmission via these practices is that they disrupt the mucosal barrier of the rectum and thus facilitate entry of HIV into the bloodstream during subsequent exposure to infected body fluids. In a large multicenter cohort study of MSM, the investigators devised a composite variable called "rectal trauma," composed of enema usage, receptive fisting, report of blood around the rectum, and evidence of scarring, fissure, or fistula on examination.(24) They found that higher rectal trauma scores correlated with increased risk of HIV infection.
| Oral-Penile Sex|
Oral-penile contact (fellatio) is not an efficient route of HIV infection. Estimating precise per-act risk is difficult because so few people practice oral sex to the exclusion of other, higher-risk sexual activities. Nonetheless, the risk of infection from oral sex is believed to be extremely low. Early male-to-male transmission studies consistently failed to demonstrate an increased risk of HIV infection associated with the practice of oral-penile sex.(17-24,25-27) A more recent study of MSM confirmed earlier findings, and further estimated that on a population level, the risk of HIV infection among MSM that is attributable to oral sex is extremely low.(68) Most studies of male-to-female and female-to-male transmission also failed to show any increased risk of HIV infection associated with oral-penile sex.(35,36,38,40) A cohort study among heterosexual serodiscordant couples at an STI clinic in Spain found no seroconversions attributable to oral sex, supporting the conclusion that HIV transmission via oral-penile sex between heterosexuals was extremely low.(69)
Oral-penile contact is not completely risk free, however.(68-70) A study of per-contact risk of infection to the receptive partner found that the probability of infection was 0.06% with a known HIV-positive partner and 0.04% with partners of unknown status. Although these are low probabilities, the authors suggest that oral-penile sex may play a larger role in the epidemic among MSM as more men adopt these behaviors as risk reduction measures.(67) Several case reports have implicated oral-penile contact as a source of male-to-male transmission.(71-73) The most convincing of these reports described a homosexual man who seroconverted despite having had only oral-genital contacts (and no anal-genital contacts) for at least 5 years preceding the estimated date of seroconversion.(73) In one study of heterosexual transmission, repeated oral sex was associated with transmission of HIV from men with AIDS to their spouses, although competing risks also showed significant associations in this study.(31) Because so few people practice oral-penile sex to the exclusion of other sexual practices, it is very difficult to recruit and retain subjects for epidemiologic studies of the HIV risk associated with this practice. Only one study was able to do this (68); another examined monogamous serodiscordant couples whose only unprotected sexual activity was oral sex.(69) No serocoversions were observed in either study. Despite the case reports, the epidemiologic evidence suggests that unprotected oral-penile sex is a low-risk activity.
Oral-genital sex, both oral-penile and oral-vaginal, can transmit STIs other than HIV with varying degrees of efficiency. Receptive oral-penile sex carries the risk of pharyngeal gonorrheal infection for both men and women, and insertive oral-penile sex, although carrying only extremely low, hypothetical risk of HIV infection to the insertive partner, carries a demonstrable risk of urethral gonorrheal infection. Other risks of receptive oral-genital sex include small probabilities of human papillomavirus and hepatitis C transmission. Insertive oral-penile sex is an efficient route for the transmission of herpes simplex virus (HSV).(70) The damage that many STIs cause to mucosa can conceivably increase the likelihood of transmission of HIV through oral sex, although this risk has not been quantified.
| Oral-Vaginal Sex|
The risk of HIV transmission through oral-vaginal sex (cunnilingus) has received less attention than oral-penile sex. There have been case reports of female-to-male (74) and female-to-female (52) transmission of HIV infection via oral-vaginal sex. One study found an association between oral-vaginal sex and female-to-male transmission of HIV, although competing risks also showed significant associations in this study.(31) As with oral-penile sex, conducting an epidemiologic study that can examine oral-vaginal sex in the absence of competing HIV risk behaviors is difficult, and no such studies have been reported. However, all studies that have controlled for competing risk behaviors (35,36,38,40,69,70) have concluded that oral-vaginal sex is extremely low risk.
| Oral-Anal Sex|
Although oral-anal contact is not an independent risk factor for HIV infection based on data from male-to-male transmission studies,(18-20,22,23,25-28) it may be a marker for other high-risk sexual practices. Univariate analysis showed an increased risk of HIV infection associated with oral-anal sex. This association was no longer statistically significant once the authors controlled for other high-risk behaviors.(19,22) Oral-anal sex has been shown to be a route of transmission for hepatitis A and B, and parasitic infections such as giardiasis and amebiasis.(70)
| Methods to Prevent Sexual Transmission of HIV|
| Barrier Methods and Microbicidal Agents|
| Male Condoms: Permeability to Viruses|
Several laboratory experiments were conducted to test the ability of latex condoms to provide an effective physical barrier against HIV. Condoms were tested by placing a solution containing HIV inside the condoms, and culture medium (free of HIV) outside the condoms. No leakage of HIV across latex condoms was demonstrable.(75-78)
Similar studies tested the permeability of latex condoms to other sexually transmitted viruses. Latex condoms are impermeable to HSV (79,80) and hepatitis B virus.(80)
Studies in vitro provided evidence that condoms made of natural membranes do not provide a consistently effective physical barrier against a number of viruses. Leakage of HIV occurred across lambskin condoms in one (75) of two studies.(75,76) Hepatitis B virus (80) and HSV (81) also leak across natural skin condoms.
In the 1990s, polyurethane condoms were introduced in both "male" and "female" varieties (the female condom will be discussed below). Polyurethane condoms, like those made of latex, effectively contain viruses in vitro.(82-86) Additional benefits of the polyurethane condom include use by persons with latex allergies, and anecdotal reports of increased tactile sensitivity through the barrier relative to latex.
| Epidemiologic Evidence for Condom Efficacy|
Several transmission studies demonstrated a statistically significant negative association between condom use and risk of HIV infection,(28,31,34,41,43) whereas others did not.(32,36,38,39) Determining the true extent to which condoms reduce risk is difficult because investigators have used various scales for describing condom use. For example, in one investigation, condom use was reported as "never/not always" versus "always."(41) In another investigation, condom use was reported as "routine use during vaginal intercourse."(31) It is generally accepted by the medical and public health communities that when used properly, latex and polyurethane condoms can significantly reduce the risk of sexual transmission of HIV. Condoms are therefore recommended as an important HIV prevention measure.
| Reasons for Condom Failure|
As the results of HIV transmission studies indicate, being made of material impermeable to HIV in a laboratory (ie, latex or polyurethane) is not sufficient to ensure that condoms will provide complete protection during real-life usage. Condoms can fail to provide complete protection for a variety of reasons, including failure to use them consistently, failure to use them properly, condom breakage, and condom slippage. Studies of latex condom performance during human use reported breakage and slippage rates varying from 1.46% to 18.60%.(87-92) Use of thicker condoms for anal sex (92,93) and having more personal efficacy (technical skill) and experience (number of episodes of prior use) in using condoms (93,94) were associated with lower failure rates. In addition, laboratory evidence suggests that improper use of latex condoms (eg, applying an oil-based lubricant) can make condoms more susceptible to breakage.(95)
When polyurethane condoms were introduced, questions were raised almost immediately as to their safety compared to latex condoms. In general, it was asserted that polyurethane condoms were more prone to breakage and slippage, and this delayed the approval by the U.S. Food and Drug Administration (FDA) of one brand of polyurethane condom. Although the FDA eventually determined that the polyurethane condom was safe for consumer use in 1995,(96) six epidemiologic studies have since addressed this question.(82-97) Three of the six studies found equivalent low rates of breakage and slippage between the two condom types (82,83,97); one study found polyurethane condoms to have higher breakage but equivalent slippage rates compared to latex condoms (86); one study found higher breakage and higher slippage rates with polyurethane condoms (85); and the one study that addressed only breakage found higher rates with polyurethane condoms.(84) It should be emphasized that the breakage and slippage rates of polyurethane condoms are not unacceptably high, and using polyurethane condoms is still considered safer sex practice for those people unable or unwilling to use latex condoms.
All studies on condom efficacy in the United States were conducted with heterosexual couples performing vaginal sex. It is believed that condom failure rates may be higher for anal sex. Condoms designed for specifically for anal intercourse are available in Europe, but there is little data on their performance versus commonly available latex or polyurethane male condoms.(98)
| Female Condoms|
The female condom, made of two flexible polyurethane rings and a loose-fitting polyurethane sheath, is approved for contraception and HIV prevention in heterosexual intercourse. The female condom prevents leakage of HIV in laboratory testing.(99) In studies of acceptability, between 50% and 73% of women respondents liked female condoms as much or better than male condoms.(100-102)Acceptance was somewhat less among their male partners, with only 44% reporting they liked the female condom as much as or better than the male condom.(101,102) Nonetheless, as a female-controlled method, the female condom represents an important advance in HIV prevention.
Researchers have explored whether polyurethane female condoms can be reused. Given their much higher per-unit costs versus male condoms, this question is particularly relevant for women in resource-poor environments. One study has shown that the structural integrity of female condoms is not significantly damaged in up to five uses if disinfected in diluted household bleach and water (1 part bleach to 4 parts water) for not more than 30 minutes, washed in diluted dishwashing liquid or bar soap and water, and air dried or dried carefully by hand. The condoms should always be inspected for holes and tears before reuse, and discarded if any are observed.(103) However, the authors caution that the safest way to use female condoms is to use them only once and then discard them.
Although female condoms are not approved for anal use, some MSM have nonetheless begun to use them for anal sex. The large size and polyurethane composition have been reported anecdotally by MSM to increase sensitivity for the insertive partner compared with latex male condoms. The one study on their use by MSM found that 57% of men reported problems with the condom, including rectal bleeding by the receptive partner.(104) Anecdotal evidence suggests that removing the inner ring, lubricating the inside, placing the condom over the erect penis, lubricating the outside, and then entering the receptive partner eliminates some discomfort and trauma. As with vaginal use, care should be taken that the outer ring of the condom does not enter the rectum. It is important to keep in mind that safety and efficacy have not been demonstrated in anal use, and that female condoms are neither recommended nor approved for this purpose.
| Dental Dams|
Latex dental dams can be placed over the labia and genitalia, or over the anal area, for protection during cunnilingus (oral-vaginal sex) and anilingus (oral-anal sex). These latex squares can be purchased at condom specialty stores and some drugstores (available in different flavors), or similar barriers can be made by cutting a latex condom or a latex glove. The efficacy of these methods has not been studied.
1 summarizes practical instructions for the use of barrier methods.
| Vaginal Microbicides|
Effective topical anti-HIV agents that women could use with or without their sexual partner's knowledge would be of great benefit. Nonoxynol-9 (N-9), a detergentlike molecule once thought to be a leading candidate in the search for vaginal microbicides against HIV,(105-108) is no longer recommended. Studies of the in vivo efficacy of N-9 have shown that N-9 reduces the risk of HIV transmission in some cases but not in others.(109-111) A recent meta-analysis of vaginal N-9 studies found that there is no evidence that N-9 prevents HIV infection in women, and confirmed former research findings that had found a significantly elevated risk of genital lesions and ulcers associated with N-9 use.(112-117)
Several other vaginal microbicides are being studied in animal and clinical trials.(118-121) Available results do not yet support making specific recommendations.(119)
| Rectal Microbicides|
N-9 is no longer recommended for use in anal sex because the chemical has been shown to damage the rectal epithelium.(119,122) Prior to 2000, N-9 had been an ingredient in several water-based lubricants that were marketed specifically to MSM for use with condoms, and there is some evidence that almost half of all MSM in certain locations actively sought out N-9-containing lubricants for use during anal sex.(122) Many manufacturers have since discontinued marketing N-9-containing lubricants to MSM. Public health officials and health care workers should actively discourage MSM from using N-9-containing lubricants for anal sex and promote awareness of the many N-9-free, water-based alternatives available to facilitate condom use.
| Chemoprophylaxis (PEP and PREP)|
People at risk of HIV infection may not always be able to choose safer sexual practices. The reasons may include disempowerment of one sexual partner (particularly women in some heterosexual relationships), sexual assault, depression, or alcohol or drug abuse. In addition, condom breakage or failure may occur.
The use of postexposure prophylaxis (PEP) using antiretroviral medications appears to reduce the risk of HIV infection in health care workers following occupational exposure to HIV (eg, needlesticks or other contact with infected blood). No study has yet quantified efficacy of PEP following sexual exposure to HIV, but nonrandomized studies suggest that PEP may be effective in reducing the risk of HIV infection.(123) PEP is now recommended for sexual exposures, including sexual assault.(123,124) Treatment must be initiated within 72 hours of exposure, and should be followed for 28 days under the supervision of a physician.
The use of antiretroviral medication administered prior to sexual exposure (pre-exposure prophylaxis, PREP) to reduce the risk of HIV infection is under study in high-risk populations.
| Safer-Sex Recommendations|
| General Risk-Reduction Strategies|
On the basis of the evidence just summarized, the following subjects should be incorporated into education and counseling intended to reduce sexual transmission of HIV.
| Decisions About Sexual Activity|
Because sexual contact is the major transmission route for HIV infection, eliminating sexual contact eliminates risk of transmission by this route. Abstinence, however, may be neither desirable nor practicable for many people. Nonetheless, it is an important option to consider, as some persons may feel that at least limited periods of abstinence may be the best choice under certain circumstances.
| Decisions About Partner Selection|
Sexual contact with many persons increases the probability of coming in contact with an HIV-infected partner. Thus, one risk-reducing strategy to consider is a reduction in the number of sexual partners, but this in no way reduces the risk of infection by sexual contact with even a single partner who is HIV positive. Because risk of HIV infection derives only from exposure to HIV-infected partners, avoiding sexual exposure with partners known or likely to be HIV infected would be an appropriate risk-reduction strategy for many people, but in many cases it is not possible to tell whether or not a given partner is HIV positive. One reasonable approach is to choose a partner who is at low risk of being HIV infected and then practice safer-sex techniques with that partner.
| Decisions About Specific Sexual Practices|
Evidence shows that some sexual practices are associated with a greater risk of HIV transmission than others. Proper use of barrier methods can reduce the risk of transmission associated with many of these practices. Thus, decision making about safer sex involves choices about specific sexual practices in addition to choices about partner selection. Based on the scientific evidence discussed in this chapter, Table
2 classifies sexual practices by their level of risk for HIV transmission.
|| ||Centers for Disease Control. HIV/AIDS Surveillance Report 2001: U.S. HIV and AIDS cases reported through December 2001. Year-end edition Vol.13, No.2. http://www.cdc.gov/hiv/stats/hasr1302.htm (accessed November 25, 2003) (see page 15)|
|| || Goedert JJ, Sarngadharan MG, Biggar RJ, Weiss SH, Winn DM, Grossman RJ, Greene MH, Bodner AJ, Mann DL, Strong DM, et al. Determinants of retrovirus (HTLV-III) antibody and immunodeficiency conditions in homosexual men. Lancet. 1984;2(8405):711-716.|
|| || Stevens CE, Taylor PE, Zang EA, Morrison JM, Harley EJ, Rodriguez de Cordoba S, Bacino C, Ting RC, Bodner AJ, Sarngadharan MG, et al. Human T-cell lymphotropic virus type III infection in a cohort of homosexual men in New York City. JAMA. 1986;255(16):2167-2172.|
|| || Kingsley LA, Detels R, Kaslow R, Polk BF, Rinaldo CR Jr, Chmiel J, Detre K, Kelsey SF, Odaka N, Ostrow D, et al. Risk factors for seroconversion to human immunodeficiency virus among male homosexuals. Results from the Multicenter AIDS Cohort Study. Lancet. 1987;1(8529):345-349.|
|| || Detels R, English P, Visscher BR, Jacobson L, Kingsley LA, Chmiel JS, Dudley JP, Eldred LJ, Ginzburg HM. Seroconversion, sexual activity, and condom use among 2915 HIV seronegative men followed for up to 2 years. J Acquir Immune Defic Syndr. 1989;2(1):77-83.|
|| || Johnson AM, Petherick A, Davidson SJ, Brettle R, Hooker M, Howard L, McLean KA, Osborne LE, Robertson R, Sonnex C, et al. Transmission of HIV to heterosexual partners of infected men and women. AIDS. 1989;3(6):367-372.|
|| || Saracco A, Musicco M, Nicolosi A, Angarano G, Arici C, Gavazzeni G, Costigliola P, Gafa S, Gervasoni C, Luzzati R, et al. Man-to-woman sexual transmission of HIV: longitudinal study of 343 steady partners of infected men. J Acquir Immune Defic Syndr. 1993;6(5):497-502.|
|| || Altfeld M, Allen TM, Yu XG, Johnston MN, Agrawal D, Korber BT, Montefiori DC, O'Connor DH, Davis BT, Lee PK, Maier EL, Harlow J, Goulder PJ, Brander C, Rosenberg ES, Walker BD. HIV-1 superinfection despite broad CD8+ T-cell responses containing replication of the primary virus. Nature. 2002;420(6914):434-439.|
|| || Kreiss J, Ngugi E, Holmes K, Ndinya-Achola J, Waiyaki P, Roberts PL, Ruminjo I, Sajabi R, Kimata J, Fleming TR, et al. Efficacy of nonoxynol 9 contraceptive sponge use in preventing heterosexual acquisition of HIV in Nairobi prostitutes. JAMA. 1992;268(4):477-482.|
|| || Veazey RS, Shattock RJ, Pope M, Kirijan JC, Jones J, Hu Q, Ketas T, Marx PA, Klasse PJ, Burton DR, Moore JP. Prevention of virus transmission to macaque monkeys by a vaginally applied monoclonal antibody to HIV-1 gp120. Nat Med. 2003;9(3):343-346.|