Circumcision and sexually transmitted infections

One important benefit of circumcision, to both HIV-negative and HIV-positive men and their partners, is that it may also prevent infection with certain other STIs. Since many STIs and their symptoms (genital ulcers, urethritis, etc) are associated with higher risk of HIV transmission/acquisition, circumcising men could have a synergistic effect in terms of preventing HIV too.

The 2007 study of HIV-positive men who were circumcised in Rakai¹ found that there was benefit to the circumcised HIV-positive men in terms of STIs. They had a third less genital ulcer disease than those who remained uncircumcised (10.1% versus 15.8%) and this was statistically significant (p = 0.002). However, rates of all STIs and of bacterial vaginosis were the same in the partners of circumcised and uncircumcised HIV-positive men.

In contrast, a further analysis of the Rakai data,² which investigated the effect of circumcision on HIV-negative men and their wives, found reductions in STIs. In the wives of the circumcised men, there was a 25% reduction in Genital Ulcer Disease (GUD), a 20% reduction in bacterial vaginosis, and a 50% reduction in trichomoniasis. Severe bacterial vaginosis fell by 60% (2% in wives of circumcised men versus 6.5% in wives of uncircumcised). All these results were statistically significant.

Amongst the HIV-negative men themselves in the Rakai study, there were 25% fewer herpes simplex virus 2 (HSV-2) infections in circumcised men. Among 62 men who became HIV-positive during the trial, 38 (61%) either had HSV-2 before the trial (47%) or seroconverted simultaneously to HIV and HSV-2 (14%). Men with genital ulcers were 2.89 more likely to get HIV if circumcised but 5.89 more likely to get HIV if they were not circumcised. However, rates of urethral discharge were identical between the two arms of the study: circumcision appeared protective of cutaneous skin lesions but not of internal STIs that attacked the urethral mucosa.

In the Kisumu RCT of circumcision,³ men with HSV-2 were 90% more likely to become HIV-positive.

A 2009 analysis of the Orange Farm trial⁴ found that HIV and HSV-2 independently reinforced each others’ effects. People with one infection in the study were more likely to acquire the other infection.

Circumcision protected against HSV-2 as well as HIV, though the effect was not so strong. In an ‘intent-to-treat’ (ITT) analysis the annual incidence of HSV-2 among men randomised to the control arm was 3.54% and among men in the intervention arm 2.33%, meaning that men in the circumcision arm were 33% less likely to acquire HSV-2. This was not statistically significant (p=0.12).

However, in the ‘as-treated’ (AT) analysis which looked not at which study arm men were in, but whether they had actually been circumcised or not, the risk of HPV was reduced by 45%, which was statistically significant (p = 0.028).

Looking at the increased relative risk of HIV infection among HSV-2 study participants allowed the researchers to calculate that the proportion of HIV infections that were attributable to HSV-2 infection was 26%; in other words a perfect anti-HSV-2 prophylactic might be expected to reduce HIV infection by this amount.

However, it did not find that HSV-2 infection had any effect, positive or negative, on the anti-HIV protective effect of circumcision: men who were circumcised during the study had the same decrease in the risk of HIV infection regardless of HSV-2 status. The protective effect of circumcision was 62, 63 and 55% among men who were HSV-negative throughout, HSV-positive throughout, and who acquired HSV-2 during the study in the ITT analysis. In the AT analysis the protective effect was 76, 80 and 72% respectively.

Another substudy of the Orange Farm RCT⁵ showed that circumcision had a significant protective effect against infection with human papilloma virus (HPV).

The study looked at only three STIs: gonorrhoea, trichomoniasis and HPV. Circumcision offered no protection against gonorrhoea, with identical infection rates between arms (a later analysis⁴ also showed no protective effect against chlamydia). It appeared to offer a degree of protection against trichonomiasis: 3.1% of uncircumcised men versus 1.7% of circumcised men were infected at follow-up, offering a protective effect of 46%, which was of borderline statistical significance.

However, it did offer a significant degree of protection against infection by 13 types of high-risk HPV. At follow-up, 24.8% of uncircumcised and 15.8% of circumcised men had become infected with new types of HPV, and this 47% protective effect was statistically significant (p = 0.0012). After adjusting for confounders the protective effect became smaller (36%) but was still statistically significant.

This study helps to explain why women with circumcised partners are at less risk of high-risk wart virus and trichomoniasis infections.

Non-African studies have also found that circumcision protects against HPV. One US study⁶ involving 463 heterosexual men (84% circumcised) was conducted between 2002 and 2005. Circumcision was strongly associated with a reduced risk of detecting any HPV infection (adjusted odds ratio [AOR]: 0.44, 5% CI: 0.23 to 0.82) or high-risk strains (AOR: 0.47; 95% CI: 0.22 to 0.99) at the top of the penis and in the urethra. Circumcision also seemed to be associated with a reduced risk of any human papilloma virus infection or infection with high-risk strains on the shaft of the penis (p = 0.05). However, although circumcised men had a reduced risk of human papilloma virus infection on the scrotum, anal canal, perianal area and in semen, this reduction was not statistically significant.

Another US study⁷ has found that circumcised men appear more likely to clear human papillomavirus (HPV) infections, including high-risk strains. The HPV Infection in Men study studied 285 men, 88% circumcised, with no history of genital warts, penile or anal cancer, and no current sexually transmitted infections (including HIV and hepatitis C). The study only investigated penile, not anal, HPV infection. Over twelve months, 29.2% of the men acquired a new HPV infection: 19% with oncogenic (cancer-causing) strains and 16% with non-oncogenic strains (some men acquired both).

Although they were no less likely to acquire HPV infection, circumcised men were three times more likely to clear it (AHR, 3.1; 95% CI, 1.2 to 8.2) and six times more likely to clear oncogenic infections (AHR, 6.5; 95% CI, 2.1 to 19.7). The authors comment that "the most important determinant of clearance of any [penile] HPV infection and of clearance of oncogenic HPV infection was circumcision."

Some studies in gay men have failed to find any protective effect of circumcision: for instance, in the 2008 meta-analysis of circumcision in gay men,⁸ the authors found no association between circumcision and reductions in any other STI. Indeed, in post-1996 studies and in higher-quality studies, there was a nearly significant increase in STIs in circumcised men.

On the other hand, a 2009 study from Australia found that circumcision appears to offer gay men some protection from syphilis. Investigators in Sydney performed a prospective study involving 1426 gay men to see if rates of prevalent and incident sexually transmitted infections differed according to circumcision status. All men were HIV-negative at baseline.

After controlling for potentially confounding factors, such as age and number of receptive acts of unprotected intercourse, the circumcised men had a significantly reduced risk of acquiring infection with syphilis (p = 0.019). The protective effect of circumcision against incident infection with syphilis was strengthened when the investigators restricted their analysis to men who expressed a preference for the insertive role during anal sex (p = 0. 033). Circumcision did not reduce the risk of any other infections.

“Circumcised participants in the HIM cohort were at significantly reduced risk of incident syphilis infection”, write the investigators. They do not believe, however, that circumcision is likely to have “a substantial public health impact in reducing acquisition of most sexually transmitted infections in homosexual men.”⁹