Randomized Trials Needed to Ascertain Association of Male Circumcision With Decreased HIV Risk

Although there appears to be a significant association between male circumcision and decreased risk for HIV infection, randomized trials of male circumcision to prevent HIV infection are needed, researchers from the School of Public Health at the University of Illinois-Chicago and the University of Manitoba-Winnipeg conclude in an analysis of multiple epidemiological studies examining HIV transmission and the procedure published in the November issue of the Lancet Infectious Diseases. About 80% of people with HIV worldwide are infected through sexual intercourse, and in Africa, about 90% of infections occur through vaginal intercourse. Two ecological studies found that countries where fewer than 20% of the men are circumcised have HIV prevalence rates "several times higher" than countries where approximately 80% of the men have undergone the procedure. In sub-Saharan Africa, about 30% of societies do not traditionally practice male circumcision, with most of these communities found in the eastern and southern regions where HIV prevalence is highest. Similar associations between circumcision and HIV transmission have not been found in Europe and North America where most of the men are circumcised and the majority of male HIV acquisition is attributed to intravenous drug use or sex between men. The researchers reviewed additional cross-sectional and case-control studies and found that 28 reported that circumcision had a significant protective effect against HIV infection in men. Five additional studies indicated a trend toward a protective effect. When researchers adjusted 17 of the studies for demographic variables and sexual behavior, they found that the adjusted odds of HIV infection for circumcised men ranged between 0.19 and 0.69. Eight studies found no association between circumcision and HIV acquisition. One study found that circumcision was associated with an increased risk of HIV infection, but the data relied on information gained from HIV-positive pregnant women in Rwanda who were asked about the circumcision status of their partners.

Shortcomings
Several factors -- such as the prevalence of other STDs, age at circumcision, religious associations and other biological factors -- could have played roles in the outcomes of the studies analyzed, the researchers note. The following are summaries of possible confounding factors:

• Sexually transmitted diseases: Uncircumcised men are known to be at 50% or more greater risk for certain STDs such as syphilis and genital herpes than circumcised men. The presence of such STDs also places them at greater risk of HIV infection. The authors note, "[STDs] and lack of circumcision create a cycle of amplification in some populations," which facilitates HIV transmission and can in turn increase the frequency of STDs.
• Age at circumcision: The age of circumcision may also play a role in the procedure's effectiveness against HIV. In most developed countries and areas where Judaism and Islam are practiced, circumcision occurs shortly after birth. However, in most African countries, the procedure is performed as part of a "rite of passage into manhood" and is conducted between the ages of 10 and 20. As long as the procedure occurs before the onset of sexual activity, it should have the same protective effect against HIV acquisition, the authors note. However, if the men have already become sexually active they may have already been exposed to HIV.
• Religion and culture: The procedure is also "highly related to ethnicity and religious beliefs," which may also play a role in HIV transmission. Such ethnic or religious groups may have different sexual behaviors than groups that do not circumcise, thereby lowering or increasing their chances of HIV infection. Men who have been circumcised also tend to come from "more educated or more urban" groups.
• Hygiene and degree of circumcision: Different genital hygiene practices, such as the strict cleansing practices of Muslims, and the degree of circumcision may also play a role. The presence of foreskin can lead to more scratches, tears and abrasions during intercourse, providing "portals of entry for HIV." The presence of a "warm, moist environment under the foreskin" is also conducive to viral survival. The authors recommend further study of the "immunohistochemistry" of the foreskin to determine the influence of these factors on HIV transmission.
• Over- and under-reporting: Misreporting may also confound the data in the studies analyzed. One study in Uganda found that 20% of the men who said they were circumcised were in fact uncircumcised.

Conclusions
The authors conclude that although "[m]uch has been learned over the past decade" about HIV acquisition and circumcision, more information is needed before public health officials can make recommendations in favor of the procedure as a means of reducing the HIV transmission rate. They note that there may be risks associated with the procedure, depending upon the environment in which the procedure is performed. Although the risk of complications in the developed world is between 0.2% and 0.6%, anecdotal evidence suggests the risk in "traditional settings" may be serious, warranting further study. The authors also note that those who perform the procedure in developing areas may be lacking the tools to safely circumcise. Only a randomized trial that is able to control for confounding factors will be able to "estimate with confidence the effect of circumcision on HIV risk," the authors conclude, noting that such a trial is set to begin in 2002 and others are in the proposal stage (Bailey et al., Lancet Infectious Diseases, November 2001)