Lack of male circumcision has long been linked to the higher HIV prevalence in Africa. First identified in the eighties, the associations between circumcision, sexually transmitted diseases (STDs) and HIV transmission have been extensively studied. At the XIII International AIDS Conference in Durban this July, more evidence was presented supporting circumcision as a risk-lowering factor for HIV transmission. What was surprising, however, were studies suggesting that adult male circumcision, if available, would be an acceptable – even desirable – option for many African men at risk.
An intact (uncircumcised) penis is biologically different from a circumcised one. The foreskin (prepuce) is a folded-over flap of skin and mucosa that covers the head (glans) of the penis. In intact boys, the prepuce adheres to the glans until about the age of eight, when it loosens and becomes retractable. The glans and the part of the prepuce protecting the glans are mucosal tissue – similar to tissue found in the vagina or mouth. During circumcision, the foreskin is pulled back, a section of the prepuce is cut away and the glans is exposed. After exposure, the mucosa of the glans and remaining prepuce begins to keratinize and toughen, eventually becoming like the more durable form of skin found on other exposed parts of the body.
When mucosal tissue is irritated or inflammed, immune system scavengers and other white blood cells are attracted to the area. These cells, including macrophages and dendritic and Langerhans cells, are the primary entry points for HIV infection. One theory of the protective value of circumcision proposes that the reduced area of exposed mucosa affords less opportunity for HIV to enter the immune system. There is also an observed relation between HIV transmission and the prevalence of sexually transmitted diseases such as herpes simplex 2 (HSV-2), a cause of penile lesions. If uncircumcised men are more easily infected with HSV-2 and other STDs, as many reports suggest, then they may also become more susceptible to HIV infection. On the other hand, some physicians have proposed that circumcised penises, lacking the 'gliding' mechanism of the intact foreskin, may possibly cause more irritation to vaginal tissue during intercourse. This would increase the chances for HIV transmission to women.
Although the biological mechanisms for a protective effect of circumcision await more research, the observations of epidemiology studies that find protective associations must be considered seriously. But are these effects due to circumcision or to other behavioral factors? Certain cultural or religious practices, such as washing the penis after coitus, may themselves contribute significant protective benefits incidental to circumcision. On the other hand, critics of circumcision have argued that daily washing of the uncircumcised penis is unnecessary and may cause mucosal inflammation. Without evidence from randomized controlled trials, blanket recommendations for changes in cultural practices are risky. In the meantime, well-established prevention messages about consistent condom use and avoiding high-risk encounters continue to be valid advice for men whether they have foreskins or not.
When mucosal tissue is irritated or inflammed, immune system scavengers and other white blood cells are attracted to the area. These cells, including macrophages and dendritic and Langerhans cells, are the primary entry points for HIV infection. One theory of the protective value of circumcision proposes that the reduced area of exposed mucosa affords less opportunity for HIV to enter the immune system. There is also an observed relation between HIV transmission and the prevalence of sexually transmitted diseases such as herpes simplex 2 (HSV-2), a cause of penile lesions. If uncircumcised men are more easily infected with HSV-2 and other STDs, as many reports suggest, then they may also become more susceptible to HIV infection. On the other hand, some physicians have proposed that circumcised penises, lacking the 'gliding' mechanism of the intact foreskin, may possibly cause more irritation to vaginal tissue during intercourse. This would increase the chances for HIV transmission to women.
Although the biological mechanisms for a protective effect of circumcision await more research, the observations of epidemiology studies that find protective associations must be considered seriously. But are these effects due to circumcision or to other behavioral factors? Certain cultural or religious practices, such as washing the penis after coitus, may themselves contribute significant protective benefits incidental to circumcision. On the other hand, critics of circumcision have argued that daily washing of the uncircumcised penis is unnecessary and may cause mucosal inflammation. Without evidence from randomized controlled trials, blanket recommendations for changes in cultural practices are risky. In the meantime, well-established prevention messages about consistent condom use and avoiding high-risk encounters continue to be valid advice for men whether they have foreskins or not.
In the 1980s, Dr. Francis Plummer of the University of Nairobi observed that uncircumcised men were eight times more likely than circumcised men to have had genital ulcers. Additionally, uncircumcised men with genital ulcers in his studies had a 50% chance of becoming infected with HIV after only a single sexual encounter with an infected prostitute.
Twelve years later, Anne Buvé of the Institute of Tropical Medicine in Belgium continues to contribute evidence that lack of circumcision and presence of other STDs are linked with increased HIV prevalence. She performed a large cross-sectional analysis of factors associated with having HIV in four African towns, two of known high HIV prevalence and two with a low, stable HIV prevalence. In each town, 1,000 men and 1,000 women were interviewed and examined. In Yaounde, Cameroon and Cotonou, Benin, two towns with an adult male HIV prevalence of about 4%, nearly all men studied (>99%) were circumcised. In Kisumu, Kenya, with an adult male HIV prevalence near 20%, just 28% of men were circumcised. In Ndola, Zambia, where the adult male prevalence of HIV infection is around 23%, the circumcision rate was only 9%.
In Kisumu, 10% of HIV infections occurred in circumcised men and lack of circumcision emerged as a strong independent predictive factor for acquiring HIV infection. Fully 25% of Ndola's circumcised men were HIV-positive. The protective effect of circumcision could not be ascertained in Ndola because too few circumcised men were available to achieve statistical precision. The overall survey also found positive associations with HIV infection for alcohol use, genital herpes and a history of sexually transmitted disease. Interestingly, no cases of syphilis were observed in any circumcised men.
Buvé concluded that male circumcision is protective against HIV infection, although the magnitude of the effect may differ between populations. One risk to this intervention, she warns, is that newly circumcised men may believe they are now HIV-proof and discontinue their use of condoms.
Ronald Gray of Johns Hopkins University in Baltimore presented findings from observational studies of HIV incidence in a cohort of 5,507 HIV-negative men and studies of smaller numbers of sero-discordant couples in Rakai, Uganda. The rate of new infections observed in these cohorts between 1994 and 1998 was 1.1 per 100 person years (py) among circumcised men and 1.8 per 100 py among uncircumcised men. Circumcision before puberty was associated with reduced HIV incidence. The rate of infection for men who had been circumcised before puberty was 0.9 per 100 py compared to 1.5 per 100 py for those circumcised when older than 12 years. The benefit of postpubertal circumcision in this study was not statistically significant.
Nearly all (over 99%) of the 737 Muslim men in this cohort were circumcised; only 3.7% of non-Muslims were circumcised. Additionally, all Muslims were circumcised before puberty whereas only 48% of non-Muslims were. Overall, HIV incidence was reduced by -0.9 per 100 py among Muslims compared to uncircumcised non-Muslims in Rakai. Gray observed that certain Muslim behaviors such as non-use of alcohol might confer additional protection. Alcohol use has been highly associated with paying for sex, non-use and misuse of condoms and increased risk-taking. Another possibly protective Muslim practice is polygamous marriage, which creates closed sexual networks for men with multiple wives. An understudied Muslim behavior that may also help reduce transmission is postcoital genital washing, routinely performed prior to prayer. Gray believes that the cumulative effect of these additional protective factors may have contributed to an observed reduction of HIV infection by -0.5 per 100 py among Muslim men compared to circumcised non-Muslims.
Among the discordant couples with HIV-negative males in Rakai, no HIV infections occurred in 50 circumcised men, whereas new infections occurred in 16.7 per 100 py among uncircumcised men. For couples with HIV-positive men, there was no difference in the rate of transmission between circumcised and uncircumcised men for those with viral loads over 50,000 copies/mL (25 per 100 py). However, among men with viral loads less than 50,000 copies/mL, there were no observed transmissions of HIV from circumcised men to their partners compared to 9.6 per 100 py transmissions from uncircumcised men.
David Serwadda, of Makerere University in Kampala, Uganda analyzed the Rakai incidence data to estimate the potential impact of prophylactic circumcision of HIV-negative men on a population-wide basis. This issue gains urgency because of the principle that any intervention resulting in a reduction of HIV incidence among men will lead to a lower prevalence among women and ultimately have a damping effect on the pace of the epidemic. Based on the Rakai incidence data, Serwadda proposed a potential reduction of HIV acquisition ranging between 11% and 22%.
Unfortunately, observational studies such as these do not allow conclusions about the protective value of circumcision to be generalized to other populations. This raises the question of whether a randomized trial of circumcision is warranted. While prepubertal circumcision may afford better protection, it would require two decades to investigate that intervention. Serwadda thinks that trials of adult male circumcision may be feasible in highly exposed populations where smaller protective effects might be observed.
If male circumcision is a potentially effective intervention that deserves a controlled trial, then current attitudes and practices need to be studied and described. It is quite reasonable to ask, as the next two presenters did, how many men would be willing to undergo an invasive and irreversible operation on what respondents often called "the site of their manhood."
Insight into this question was offered by reports on knowledge and beliefs about circumcision in two populations: the residents of Carletonville, South Africa, where circumcision is fairly common among various ethnic and cultural groups, and the Luo people of Kenya, who do not routinely practice circumcision.
South African researcher Reathe Taljaard explored attitudes toward male circumcision in the heavily HIV-affected region of Carletonville, where he found widely varying prevalence and ages at circumcision. Initiation into the traditional culture, whether Zulu, Pedi or Xhosa, is a rite of passage to manhood for many boys during their teenage years. Young men at this transitional stage may spend several weeks sequestered at an initiation school learning about traditional life and values. In some cultures, this ritual passage has included circumcision, performed by the traditional circumciser, to give evidence that the boy has become a man. In recent years though, young men have sought circumcision by local medical practitioners as an alternative to the traditional cutting. This is true both for boys who go on to initiation school as well as for those who prefer to attain manhood without benefit of ritual. Fear of pain and stories of "cutting short" have driven this trend. Indeed, on the second day of the AIDS Conference, a news brief appeared in the Durban Daily News about a traditional circumciser charged with culpable homicide in the death of a young initiate after a botched Xhosa ceremony in the Eastern Cape.
Taljaard examined general beliefs about circumcision among women as well as men, since wives often made the clinic appointments for their husbands to have the procedure. Among various common beliefs were ideas that a circumcised penis was associated with cleanliness, did not gather dirt, did not suffer damage when entering a virgin, was immune to HIV (but not STDs), and was easier to aim. Other beliefs were that circumcision enhanced sexual performance, increased the size of the penis and brought respect. Some thought that it was bad luck not to be circumcised; some said that Westerners learned circumcision from Africans.
Robert Bailey of the University of Chicago interviewed adult Luo men and women from non-urban areas of the Nyanza province of Kenya concerning the acceptability of male circumcision. This is an area where circumcision is not traditionally practiced, and 90% of men are not circumcised. The prevalence of HIV is estimated at 27 to 35% among women in the area. Despite the low traditional prevalence of circumcision, there was widespread belief that circumcised men were cleaner and were less likely to contract HIV or STDs. The belief that circumcised men and their partners derived greater enjoyment from sex was widespread. Given the choice, 60% of men said they would prefer to be circumcised, and 74% of men and 88% of women would have their son circumcised.
Bailey and colleagues also interviewed regional medical practitioners about their knowledge and experience with circumcision. They also inventoried clinic supplies and instruments necessary for performing the operation. Only 39% of clinicians were circumcised themselves, and 40% had never performed a circumcision. Knowledge of the risks, benefits and proper procedures were low, and only one of eight clinics inventoried had the proper supplies and instruments. Bailey also reported that knowledge of the principle of informed consent was lacking.
These results suggest that the obstacle to a controlled trial of prophylactic male circumcision may not be patient acceptance so much as a lack of training, experience and ethical guidance for performing the research in affected areas.
Bailey R, et al. "Trial Interventions Introducing Male Circumcision to Reduce HIV/STD Infections in Nyanza Province, Kenya: Baseline Results." XIII International AIDS Conference, Durban, South Africa. July 9-14 2000;
Buvé A, et al. "Male Circumcision and HIV Spread in Sub-Saharan Africa." XIII International AIDS Conference. Durban, South Africa. July 9-14 2000
Circumcision Information and Resource Pages.
Gray R, et al. "HIV Incidence Associated with Male Circumcision in a Population-Based Cohort, and HIV Acquisition/Transmission Associated with Circumcision and Viral Load in Discordant Couples: Rakai, Uganda." XIII International AIDS Conference. Durban, South Africa. July 9-14 2000;
Serwadda D, et al. "Potential Efficacy of Male Circumcision for HIV Prevention in Rakai, Uganda." XIII International AIDS Conference. Durban, South Africa. July 9-14 2000;
Taljaard R, et al. "Cutting It Fine: Male Circumcision Practices and the Transmission of STDs in Carletonville." XIII International AIDS Conference, Durban, South Africa. July 9-14 2000;