Circumcision has been shown to reduce the risk of men contracting several sexually transmitted infections (STIs). Three randomized controlled trials (RCTs) published over the last few years have demonstrated that removing the foreskin of adult men diminished the risk of HIV infection by at least 50%. An article by Tobian and colleagues in last week's issue of the New England Journal of Medicine revealed a weak protective effect of circumcision against two other infections, herpes simplex virus 2 (HSV-2) and human papilloma virus (HPV), which cause genital herpes and penile warts, respectively. The same study showed no effect of circumcision on acquisition of Treponema pallidum, the agent of syphilis. You can find a nice critical analysis of the study here. Because I am interested in diseases caused by spirochetes, I will focus on the syphilis data.
Tobian et al. conducted two RCTs with similar designs. When the data were combined, they found that 50 of 2083 (2.4%) male adolescents and adults in Uganda who underwent circumcision became infected with T. pallidum over the following 24 month period. Similarly, 45 of 2143 (2.1%) control subjects became infected within the same time period, suggesting that circumcision had no effect on contracting T. pallidum. An editorial in the same journal issue points out that the study may have been underpowered to detect a protective effect (i.e., not enough subjects in the study).
Since the Tobian et al. study failed to give a simple answer to the question, I thought it would be illuminating to look at the older observational studies that examined the effects of male circumcision on syphilis transmission. Fortunately, I found a meta-analysis that compiled data from 14 research papers, most of which described cross-sectional studies.
The meta-analysis presented by Weiss and colleagues revealed a slight protective effect of male circumcision. The relative risks (RRs) along with the 95% confidence intervals (CI) are plotted in the graph. The RR in 11 of the 14 studies were adjusted for potential confounding factors such as age. The summary statistics listed at the bottom of the graph indicate a small protective effect (summary RR, 0.67; 95% CI, 0.54-0.83).
Ideally, all studies included in a meta-analysis would have similar RRs. However, if you look carefully at the graph, you will notice a wide variation in the RRs with some of the 95% confidence intervals failing to overlap. Using standard statistical calculations, the investigators determined that it was unlikely that the variation of the RR among the studies was due to chance (P = 0.01). In other words, differences in how the studies were designed and conducted led to significant variation in the outcomes. Consequently, the authors declared that there was significant heterogeneity among the studies and warned that the summary RR "should be interpreted cautiously."
The authors described one potential source of the heterogeneity. Looking at the plots again, you will note that the Cook and Parker studies demonstrated the largest statistically significant protective effect of male circumcision. Those two studies were conducted in the United States and Australia, respectively, where males are circumcised as infants. In contrast, the two largest studies, authored by Gray and Urassa, showed no effect of circumcision on the risk of becoming infected with T. pallidum. Those studies were conducted in Uganda and Tanzania, respectively, where males are not circumcised until they are adolescents or young adults. Many of the circumcised males examined in the two African studies, which were cross-sectional and case-control studies, could have contracted syphilis before being circumcised. Weiss et al. excluded subjects who were circumcised after their first sexual intercourse or after age eleven, but this information was not available for all studies. This would lead to an underestimate of the protective effects of circumcision.
You may look at the large protective effects of infant circumcision observed in the U.S. and Australian studies (RR = 0.25 and 0.19, respectively) and conclude that mass infant circumcision would be beneficial (at least for protection against syphilis). However, both studies involved men visiting STD clinics, and the results may not apply to the general population in those countries.
Tobian, A.A.R., Serwadda, D., Quinn, T.C., Kigozi, G., Gravitt, P.E., Laeyendecker, O., Charvat, B., Ssempijja, V., Riedesel, M., Oliver, A.E., Nowak, R.G., Moulton, L.H., Chen M.Z., Reynolds, S.J., Wawer, M.J., Gray, R.H. (2009). Male circumcision for the prevention of HSV-2 and HPV infections and syphilis. The New England Journal of Medicine 360(13):1298-1309.
Golden M.R. and Wasserheit, J.N. (2009) Prevention of viral sexually transmitted infections--foreskin at the forefront (Editorial). The New England Journal of Medicine 360(13):1349-1350.
Weiss, H.A., Thomas, S.L., Munabi, S.K., and Hayes, R.J. (2006). Male circumcision and risk of syphilis, chancroid, and genital herpes: a systematic review and meta-analysis. Sexually Transmitted Infections 82(2):101-109.