Terrence Higgins Trust Sceptical of Circumcision-HIV Link


The original of this article is in the July 2006 Newsletter of the UK's HIV prevention and treatment organisation, the Terrence Higgins Trust.


HIV and Circumcision

A new study from Uganda appears to show women are less likely to get HIV from circumcised men, reigniting the debate about whether circumcision should be advocated to reduce HIV transmission. But is the debate really so clear cut?

For decades male circumcision has been justified on preventative health grounds: sexually transmitted infections (after World War One), cervical cancer (from the 50s) and urinary tract infections (from the 80s). Now HIV is cited as a justification.

This latest research into women’s vulnerability to infection(1) joins work that for the most part has previously only considered men. Some show circumcision reduces the risk, others that it increases it, while some demonstrate no effect. Yet how research is carried out and reported can mislead.

Circumcision polarises opinion, with few men neutral about it (and most researchers in this area are men). When researchers come from cultures with a strong consensus around circumcision, how impartial will their findings be? Any discussion about the merits of circumcision as an HIV prevention strategy is necessarily coloured by social attitudes towards the practice. In Europe circumcision of young males is rare except on medical or religious grounds; interestingly, there are few studies carried out by Europeans concluding circumcision is a valid HIV prevention tool. However, circumcision is still common in the United States.

Lower HIV prevalence rates in some parts of Africa appear to be linked to higher levels of circumcision and some previous studies looking at the risk of men getting HIV from women suggested circumcision reduced the risk by 50-75%.(2)(3) But such studies need to be examined for biases, such as small or skewed samples or confounding factors not being taken into account, such as:

  • ethnicity or religion – linked to both circumcision state and sexual behaviour
  • sexual practice – including the common African practice of ‘dry sex’, heavily implicated in the spread of HIV(4)
  • female genital cutting/ circumcision – frequently practised alongside male circumcision and a possible factor in the spread of HIV.
In the recently published Ugandan study HIV acquisition was recorded in women with HIV positive circumcised or uncircumcised male partners. Circumcision appeared to reduce HIV transmission to females by about 30%. However, this was not statistically significant as the sample was too small, a fact the report did not make clear.

How circumcision might reduce HIV transmission:

Removing the foreskin takes away mucosal surfaces that can harbour HIV

The foreskin also carries Langerhans cells that HIV seeks out to infect (5)

Circumcision may mean lower incidence of ulcerative STIs, a known risk factor for HIV infection

The foreskin is thin and liable to minute tears, facilitating transmission of the virus both ways.

Research suggesting circumcision protects against HIV transmission has not been carried out over a particularly long time. It may be that circumcision only delays infection and cannot prevent it. Also, very many ‘cut’ men become HIV positive and some nations that routinely circumcise such as the USA and Ethiopia have high rates of HIV infection.(6) Mathematical modelling has shown that if circumcised men increase their number of partners any protective effect disappears and HIV incidence rises. There is also the question of the effect on sexual behaviour and condom use if circumcised men believe they cannot get or pass on HIV.

One study showed that circumcision has much less protective value with higher viral load and showed circumcision after puberty failed to protect (this may have indicated that Muslims in the study, circumcised very young, exhibited other factors that explained their lower infection rate).(7) In addition, the practice has been shown to have no or only limited effect in protecting against STIs, a major co-factor in the spread of HIV, especially in the developing world.(8)

Sex between men
The 2001 Gay Men’s Sex Survey(9) found circumcised men were slightly more likely to have contracted HIV than ‘uncut’ men (6.1% compared to 5%), a small but statistically significant difference seen across ages and ethnic groups. However, these men were much more likely to have become infected through the lining of the anus than through the penis.

Australian researchers found no association between circumcision status and infection through insertive unprotected anal intercourse (UAI) - and men who had become infected without reporting UAI were also no more likely to be ‘cut’. This report concluded that the foreskin is not the main source of HIV infection in gay men who become infected by insertive UAI and that circumcision is not strongly protective against HIV infection in gay men.(10)

Should circumcision be promoted?
If further randomised control trials eventually demonstrate beyond reasonable doubt a protective effect, should health promoters encourage boys and men to be circumcised?

The procedure requires hygiene and anaesthetic, the availability of which cannot be assumed in the developing world. In such conditions, complications (including permanent injury to the penis) are not uncommon. Crucially, it would be unwise to assume that findings from the developing world can be transposed onto populations in the industrialised world with better health care and much lower burdens of STIs.

Yet in countries where access to HIV treatment is poor or nonexistent, might anything that delays or prevents infection be welcome? Although not culturally acceptable among many populations, research in Botswana suggests that 81% of men would undergo circumcision if it could prevent HIV transmission.(11) In addition, in cultures where women find negotiating condom use difficult, circumcision may save many women and girls from infection.

Ethical considerations
Childhood circumcision is controversial as it involves the removal, without consent, of a healthy part of the body with a definite function (to lubricate and increase sensation during sex or masturbation, and provide a protective cover for the glans). In addition, many see circumcision as mutilation. Ethical reservations can be overcome if circumcision is left until after childhood, when consent can be obtained. Might a health promotion consensus may be possible around the following principles?

  • The procedure should not be carried out on minors or promoted to adults.
  • It should be made available on request to consenting young adult males before they become sexually active.
However, if the conclusions of researchers around circumcision are applied to other cultures there may be accusations of health promotion ‘colonialism’. There is also a danger of coercion being used in any State-sponsored procircumcision drives, as happened during family planning campaigns in India and China.

Care should be taken when evidence is so contradictory or unreliable. An association has not been clearly established. More randomised clinical trials (that take into account confounding factors) are needed before circumcision can be confirmed as a potential prevention strategy.

Reducing viral load or treating ulcerative STIs would almost certainly make more impact on transmission rates than circumcising. Circumcision at birth would take 15-20 years to start to impact on HIV transmission rates, during which time other HIV prevention technologies will hopefully become available. Until then this is certain: a condom offers more protection than circumcision and circumcised men will still be told they need condoms to protect themselves and their partners.

Richard Scholey,
Programme Development Officer


1. Male circumcision and the risks of female HIV and STI acquisition in Rakai, Gray R et al. Uganda. Thirteenth Conference on Retroviruses and Opportunistic Infections, Denver, abstract 128, 2006

2. Auvert B et al. Impact of male circumcision on the female-to-male transmission of HIV. IAS Conference on HIV pathogenesis and treatment, Rio de Janeiro, abstract Tu0a0402, 2005 (found a 75% protective effect)

3. Baeten JM et al. Female-to-male infectivity of HIV-1 among circumcised and uncircumcised Kenyan men. Journal of Infectious diseases 191:546-553, 2005 (found infections reduced by 50%)

4. ‘Dry sex’ is a common practice in many African populations of rubbing drying herbal preparations and other substances into the vagina to make it drier and tighter, in order to increase male pleasure. It is believed to facilitate HIV transmission through lacerations and inflammation, increased condom failure and lack of vaginal secretions to combat infection.

5. However, Langerhans cells are found in all genital skin tissue. One study found an excess of such cells in the foreskin, another found very few. Nevertheless both concluded the foreskin led to vulnerability to infection and recommended circumcision.

6. Around 1 million Americans are estimated to be living with HIV as of 2003, with prevalence estimated at 0.6%, the second highest among Western nations after Spain (0.7%). Glynn M, Rhodes P. Estimated HIV prevalence in the United States at the end of 2003. National HIV Prevention Conference. June 2005, Atlanta. Abstract T1-B1101. Ethiopia’s HIV prevalence is estimated to be 4.4% (www.unaids.org)

7. Gray RH et al. Male Circumcision and HIV acquisition and transmission: cohort studies in Rakai, Uganda. AIDS, Volume 14, Number 15:2371-2381, October 20, 2000.

8. As footnote 1.

9. Reid et al (2002) Know the score: Findings from the National Gay Men’s Sex Survey 2001.www.sigmaresearch.org.uk /downloads/report02d.pdf (the 46th page of the pdf)

10. Gruhlich AE et al. Circumcision and male-to-male sexual transmission of HIV. AIDS, 2001 Jun 15; 15(9): 1188-1189

11. Kebaabetswe P et al. Male circumcision: an acceptable strategy for HIV prevention in Botswana. Sexually Transmitted Infections 79:214-219, 2003. (81% said they would be circumcised themselves and 89% said they would circumcise a male child).

Related pages:

Back to the Intactivism index page.