Education and debate

How does male circumcision protect against HIV infection?

Robert Szabo, medical resident ^a,  Roger V Short, professor ^b. 

^a Faculty of Medicine, Monash University, Wellington Road, Melbourne 3168, Australia, ^b Department of Obstetrics and Gynaecology, University of Melbourne, Royal Women's Hospital, 132 Grattan Street, Melbourne 3053, Australia

Correspondence to: R V Short

In his otherwise excellent review of the AIDS epidemic in the 21st century, Fauci presented no new strategies for preventing the spread of the disease.¹ He made no mention of male [or female] circumcision, yet there is now compelling epidemiological evidence from over 40 studies which shows that male circumcision provides significant protection against HIV infection; circumcised males are two to eight times less likely to become infected with HIV.²

[The evidence is far from "compelling". Each of the studies has its own flaws. See the relevant page of this site. The extraordinary history of circumcision as a panacea, and before that as a rite, strongly suggest that latter-day claims of prophylaxis should be regarded with a sceptical, if not jaundiced, eye. Few if any men can be truly neutral about circumcision. The temptation to justify what was done to oneself seems almost irresistible.]

Furthermore, circumcision also protects against other sexually transmitted infections, such as syphilis and gonorrhoea, ^{3 4} [This is contradicted by Laumann] and since people who have a sexually transmitted infection are two to five times more likely to become infected with HIV,⁵ circumcision may be even more protective. The most dramatic evidence of the protective effect of circumcision comes from a new study of couples in Uganda who had discordant HIV status; in this study the woman was HIV positive and her male partner was not.⁶ No new infections occurred among any of the 50 circumcised men over 30 months, whereas 40 of 137 uncircumcised men became infected during this time. [But as Peiperl points out, over one-third (29 out of 79) of the circumcised men in this study were HIV-positive before the study began.] Both groups had been given free access to HIV testing, intensive instruction about preventing infection, and free condoms (which were continuously available), but 89% of the men never used condoms, and condom use did not seem to influence the rate of transmission of HIV. [Why not? What's going on? Did circumcision status influence the rate of usage of condoms?] These findings should focus the spotlight of scientific attention onto the foreskin. [No, on the condoms or their use] Why does its removal reduce a man's susceptibility to HIV infection? [This is the fallacy of post hoc ergo propter hoc, after this therefore because of this. Circumcision does not take place in a social vacuum. What is associated with circumcision and intactness in Ugandan society? It very commonly goes with religion, for example, and religion in turn influences sexual practice in a variety of ways:

• choice of partner
• degree of monogamy
• use of alcohol and hence unprotected sex
• attitudes towards homosexuality]

	Methods

To compile the information for this review a Medline search was done using the terms circumcision, HIV, Langerhans' cells, penis, foreskin, and prepuce, and extensive email correspondence with other researchers was also undertaken. Histological observations were carried out on samples of penile tissue obtained from 13 perfusion fixed cadavers of men aged 60-96 years, seven of whom had been circumcised. [The advanced age of the men from whom the samples were taken, in considering a process that is age-related, throws suspicion on this work.]

	The pathogenesis of sexually acquired HIV infection

Between 75% and 85% of cases of HIV infection worldwide have probably occurred during sexual activity.⁷ Most cases of primary HIV infection are thought to involve HIV binding initially to the CD4 and CCR5 receptors found on antigen presenting cells - which include macrophages, Langerhans' cells, and dendritic cells - in the genital and rectal mucosa.

The most widely accepted model for the sexual transmission of HIV is based on infection of the genital tract of rhesus macaques with simian immunodeficiency virus. ^{8 9} After female macaques are inoculated intravaginally with simian immunodeficiency virus, the virus targets the Langerhans' cells located in the vaginal mucosa. [And do Short and Szabo consider advocating the amputation of vaginal mucosa? The fatal sexism of a policy that protects men but not women, compared with one that protects both, such as condom use, is not considered.] Once infected, these cells fuse with adjacent CD4 lymphocytes and migrate to deeper tissues. Within two days of infection, the virus can be detected in the internal iliac lymph nodes and shortly thereafter in systemic lymph nodes. This ultimately leads to a fatal infection.

Similarly, infection in male macaques occurs when simian immunodeficiency virus is inoculated into the penile urethra or onto the foreskin; the same sequence of cellular events involving the infection of Langerhans' cells is then likely to occur.⁹ Infected Langerhans' cells have also been detected in the penile mucosa of male rhesus macaques that have chronic simian immunodeficiency virus infection.⁹ In humans, histological studies have identified antigen presenting cells in the mucosa of the inner foreskin and urethra.¹⁰ Therefore it seems likely that antigen presenting cells at these mucosal sites are the primary target for HIV in men.

In vitro studies have shown that the CD4 receptor is generally necessary, although insufficient on its own, to permit HIV-1 to enter host cells.¹¹ The entry of HIV-1 into cells requires an additional chemokine receptor, usually CCR5, although CXCR4 is used by cells that become infected during the later stages of the disease.¹² After primary infection occurs, the virus mutates, which allows it to utilise other chemokine receptors, such as CXCR4, and thus spread to a variety of cell types. However, more than 99% of HIV-1 isolates from acutely infected patients are homologous, indicating that one specific variant is likely to be responsible for most cases of primary HIV infection.¹³ HIV variants that are transmitted to other individuals almost invariably use CCR5 as a coreceptor and are therefore named R5 viruses, to reflect their specific requirement for a coreceptor.¹⁴

(Credit: WELLCOME TRUST) Circumcision in ancient Egypt shown on a relief from Saqqara (c 2200 BC). Used with the permission of the Wellcome Institute for the History of Medicine, London [What is the relevance of this picture? The Egyptians did not circumcise for medical reasons but to sanctify priests and mark slaves. It is a general rule that articles advocating circumcision cite its great antiquity, as if that justified it.]

	How HIV enters the penis

About 70% of men infected with HIV have acquired the virus through vaginal sex, and a smaller number have acquired it from insertive anal intercourse.⁷ Thus, on a global scale most men who are HIV positive have acquired the virus via the penis. This raises questions of how HIV enters the penis and why men who are uncircumcised are potentially more susceptible to becoming infected with HIV.

The uncircumcised penis consists of the penile shaft, glans, urethral meatus, inner and outer surface of the foreskin, and the frenulum, the thin band connecting the inner foreskin to the ventral aspect of the glans. A keratinised, stratified squamous epithelium covers the penile shaft and outer surface of the foreskin. This provides a protective barrier against HIV infection. [There is no evidence for this claim.] In contrast, the inner mucosal surface of the foreskin is not keratinised¹⁵ and is rich in Langerhans' cells,¹⁰ making it particularly susceptible to the virus. [Again, this is conjecture.] This is particularly important because during heterosexual intercourse the foreskin is pulled back down the shaft of the penis, and the whole inner surface of the foreskin is exposed to vaginal secretions, providing a large area where HIV transmission could take place. [In that case, why are women, with the much greater area of their vaginal surfaces, not vastly more susceptible to HIV infection from men?]

There is controversy about whether the epithelium of the glans in uncircumcised men is keratinised; some authors claim that it is not,¹⁵ but we have examined the glans of seven circumcised and six uncircumcised men [...aged 60 to 96...], and found the epithelia to be equally keratinised. In circumcised males only the distal penile urethra is lined with a mucosal epithelium. However, this is unlikely to be a common site of infection because it contains comparatively few Langerhans' cells.¹⁰ [This is an example of begging the question, assuming what you have to prove.]

Ulcerative or inflammatory lesions of the penile urethra, foreskin, frenulum, or glans that are caused by other sexually transmitted infections may provide additional potential routes for HIV transmission. In uncircumcised males, the highly vascular frenulum is particularly susceptible to trauma during intercourse, [It is equally so in circumcised males - if it has not been ablated - and in addition traumas and tearing of the scarred ring are only possible in circumcised men] and lesions produced by other sexually transmitted infections commonly occur there. Thus, male circumcision further reduces the risk of infection by reducing the synergy that normally exists between HIV and other sexually transmitted infections.⁵

	Conclusions

Of the estimated 50 million people infected with HIV worldwide, about half are men, most of whom have become infected through their penises [... and a vast number of whom were circumcised]. The inner surface of the foreskin, which is rich in HIV receptors, and the frenulum, a common site for trauma and other sexually transmitted infections, must be regarded as the most probable sites for viral entry in primary HIV infection in men. ["Must"? Yet a vast number of them don't have one.] Although condoms must remain the first choice for preventing the sexual transmission of HIV, they are often not used consistently or correctly, they may break during use, and there may be strong cultural and aesthetic objections to using them. Cultural and religious attitudes towards male circumcision are even more deeply held, [and the perfectly rational view that it reduces sexual pleasure, and the human rights issue of consent] but in the light of the evidence presented here circumcising males seems highly desirable, ["Desirable"? A strange choice of word, suggesting a sexual rather than rational motivation] especially in countries with a high prevalence of HIV infection. [Yet Ethiopia and the US already have high prevalences of both HIV infection and circumcision.] Although neonatal circumcision is easy to perform, and has a low incidence of complications,¹⁶ [Other references cite much higher rates] it would be 15-20 years before a programme of circumcision had any effect on HIV transmission rates. Circumcision at puberty, as practised by many Muslim communities, would be the most immediately effective intervention for reducing HIV transmission since it would be done before young men are likely to become sexually active [... and knew what they had been deprived of].

It may also be time to re-think the definition of "safe sex." Since the penis is the probable site of viral entry, neither infected semen nor vaginal secretions should be allowed to come in contact with the penis, particularly in uncircumcised males. Thus, mutual male masturbation during which a penis is exposed to the potentially infected semen of another male should be regarded as risky sexual behaviour. [Szabo and Short present no evidence for this claim. They have reasoned backward from their analysis of the literature of heterosexual transmission.]

New preventive strategies are needed that could be used by men or women before the onset of intercourse. The disadvantage of topical virucides, such as nonoxinol 9, is that they may cause local irritation and thus increase susceptibility to HIV infection. The development of topically active agents that could block HIV binding sites, such as CCR5, and which could be applied to the penis or vagina to create a "chemical condom," might be more effective and acceptable than any mechanical barrier or surgical intervention.

	Acknowledgments

We thank Professor John Mills for helpful comments on an early draft of the manuscript and Professor Daine Alcorn and the staff of the Department of Anatomy, University of Melbourne, for supplying and processing the specimens from human cadavers.

Contributors: RS reviewed all the relevant literature, carried out the histological examination of the specimens, and wrote the first draft of the manuscript. RVS initiated the study and participated in redrafting of the paper. Both authors will act as guarantors.

	Footnotes

Funding: None.

Competing interests: None declared.

	References

1.	Fauci AS. The AIDS epidemic: considerations for the 21st century. N Engl J Med 1999; 341: 1046-1050[Medline].
2.	Halperin DT, Bailey RC. Male circumcision and HIV infection: 10 years and counting. Lancet 1999; 354: 1813-1815 [An unreferreed opinion-piece.] [Medline].
3.	Cook LS, Koutsky LA, Holmes KK. Circumcision and sexually transmitted diseases. Am J Public Health 1994; 84: 197-201[Medline].
4.	Moses S, Bailey RC, Ronald AR. Male circumcision: assessment of health benefits and risks. Sex Transm Infect 1998; 74: 368-373[Medline].
5.	Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Transm Infect 1999; 75: 3-17[Medline].
6.	Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med 2000; 342: 921-929 [See below for a commentary on the relevant part of this study.] [Medline].
7.	Joint United Nations Programme on HIV/AIDS. The HIV/AIDS situation in mid 1996: global and regional highlights. Geneva: United Nations, 1996. (UNAIDS fact sheet 1 July 1996.)
8.	Spira AI, Marx PA, Patterson BK, Mahoney J, Koup RA, Wolinsky SM, et al. Cellular targets of infection and route of viral dissemination after an intravaginal inoculation of simian immunodeficiency virus into rhesus macaques. J Exp Med 1996; 183: 215-225[Abstract].
9.	Miller CJ. Localization of simian immunodeficiency virus-infected cells in the genital tract of male and female rhesus macaques. J Reprod Immunol 1998; 41: 331-339[Medline].
10.	Hussain LA, Lehner T. Comparative investigation of Langerhans cells and potential receptors for HIV in oral, genitourinary and rectal epithelia. Immunology 1995; 85: 475-484[Medline].
11.	Zaitseva M, Blauvelt A, Lee S, Lapham CK, Klaus-Kovtun V, Mostowski H, et al. Expression and function of CCR5 and CXCR4 on human Langerhans cells and macrophages: implications for HIV primary infection. Nature Med 1997; 3: 1369-1375[Medline].
12.	Dragic T, Litwin V, Allaway GP, Martin SR, Huang Y, Nagashima KA, et al. HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. Nature 1996; 381: 667-673[Medline].
13.	Zhu T, Mo H, Wang N, Nam DS, Cao Y, Koup RA, et al. Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 1993; 261: 1179-1181[Medline].
14.	Kahn JO, Walker BD. Acute human immunodeficiency virus type 1 infection. N Engl J Med 1998; 339: 33-39[Medline].
15.	Barreto J, Caballero C, Cubilla A. Penis. In: Sternberg SS, ed. Histology for pathologists. 2nd ed. Philadelphia: Lippincott-Raven, 1997.
16.	Morris B. In favour of circumcision. Sydney: University of New South Wales Press, 1999. [Damned by the critics. The level of Morris's scholarship may be seen on this site.]

(Accepted 11 May 2000)

from a Commentary by Laurence Peiperl, MD on The Rakai Study: Risk Factors for Heterosexual Transmission:

Caution must be exercised in applying the results of this study to individual risk-reduction strategies. For example, the result that no circumcised men seroconverted during the course of the study might be casually interpreted to mean that a circumcised man cannot be infected with HIV through heterosexual sex. This conclusion is illogical, as the relatively small sample size (50 circumcised seronegative men) and short duration of follow-up (median less than 2 years) allow for a significant possibility that infection of circumcised men would be observed in a larger population or over a longer period of time. The fact that over one-third (29/79) of the circumcised men in this study were HIV-positive on entry demonstrates that transmission to circumcised men occurs in this population, and it seems extremely unlikely that heterosexual transmission did not contribute.

HIV InSite Journal Club, April 14, 2000
Discussion of study by Quinn et al., in the New England Journal of Medicine, March 2000.