University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Circumcision in HIV-infected men and its effect on HIV transmission to female partners in Rakai, Uganda: a randomised controlled trial
Global Health Sciences Literature Digest
Published January 14, 2010
Journal Article

Wawer MJ, Makumbi F, Kigozi G, et al. Circumcision in HIV-infected men and its effect on HIV transmission to female partners in Rakai, Uganda: a randomised controlled trial. Lancet. 2009 July; 374(9685):229-37.

In Context

Studies in Africa have shown that among HIV-uninfected men, circumcision can decrease the risk of HIV acquisition by 50%-60%.(1,2) For asymptomatic HIV-positive men with high CD4 cell counts, circumcision also can reduce the rates of genital ulcer disease.(3) As a result, it is now a recommended strategy for HIV prevention in many countries. It is not known whether male circumcision in HIV-infected men might also benefit female partners, although data from several observational studies suggest that circumcision may decrease male-to-female HIV-transmission.(4)


To determine whether male circumcision can reduce the transmission of HIV and sexually transmitted infections (STIs) from HIV-infected men to their uninfected female partners


Rakai District, southwestern Uganda

Study design

Randomized, non-blinded, controlled trial


HIV-infected, uncircumcised men aged 15-49 years, without evidence of immunosuppression by World Health Organization (WHO) stage (stage 1 or 2), or CD4 count (>350 cells/mL), and their confirmed HIV-uninfected steady female sexual partners or wives. In the intervention group, 87 men and 93 of their sexual partners were enrolled; in the control group, 68 men and 70 partners were enrolled. The number of female partners was greater because some men were polygamous.


HIV acquisition in women


Enrolled men were circumcised using the sleeve procedure. They and their partners were asked to refrain from sexual intercourse until wound healing was confirmed at a clinic visit. At baseline, men and women participants were screened for syphilis using RPR and TPHA, questioned about current and past STI-associated symptoms and were interviewed about sexual behavior. Among men, viral load was measured. Men in both groups were then followed regularly at 6, 12, and 24 months post-enrollment, with sexual behavior questionnaires and blood samples. At baseline, women were evaluated for Trichomonas vaginalis and bacterial vaginosis then followed at 6, 12 and 24 months with HIV testing, STI evaluation, and behavioral questionnaires. Data about sexual behavior between couples for use in analyses were taken from women�s interview responses. Men from the control group were offered surgery following trial completion. All couples received risk reduction counseling and free condoms at all visits. The effect of circumcision on HIV transmission was evaluated using Kaplan-Meier estimates, and adjustment for covariates were made using Cox proportional hazards modeling.


Retention rates of the women partners were 90%-95% at 6 months and 82%-86% at 24 months. Overall, 17 (18%) women in the intervention, and 8 (12%) in the control group acquired HIV; the cumulative probability of transmission was 21.7% (95% confidence interval (CI): 12.7%-33.4%) and 13.4% (95% CI: 6.7%-25.8%), respectively (P=0.29); the adjusted HR was 1.49 (95% CI: 0.62-3.57; P=0.37). Using only six-month data, a sub-analysis of the association of early resumption of sex (before complete healing) and HIV acquisition revealed that in couples who resumed sex early, the female partner was significantly more likely to acquire HIV than the control group (rate ratio (RR)=3.5; 95% CI: 1.1-10.8; P=0.04), and slightly more likely than women of intervention couples who delayed sex (RR=2.9; 95% CI: 1.0-8.5; P=0.06). There were no significant associations with HIV acquisition and reported sexual behavior, condom use, or baseline viral load. At 24 months, women in the control group were slightly more likely to have a Trichomonas infection than those whose male partners were circumcised (15% vs 7%, P=0.056). An evaluation of viral load before and four weeks after circumcision among control men who later opted for surgery revealed a mean increase in intra-individual viral load of 0.20 log10 copies/mL (P<0.01).


This trial did not demonstrate that circumcision of HIV-infected men reduces transmission to women. Transmission may be increased among men who undergo circumcision and resume sexual intercourse before complete wound healing. Data on STIs was sparsely reported upon. It is possible that the trial was underpowered to detect a protective effect of circumcision; based on observational data, the authors estimated that 220 couples would have been required in the study for analysis to detect a more than 50% reduction in HIV incidence.

Quality Rating

This was a high-quality randomized trial conducted with appropriate methodology, oversight, and statistical analysis. From the article, it is not possible to judge the representativeness of the sample. Although one of the secondary outcomes was effect of circumcision on STI transmission, information on whether and how STIs were treated at baseline and follow-up, or how they were taken into consideration in analysis, were not described in sufficient detail (7% of all participants were RPR/TPHA-positive at enrollment).

Programmatic Implications

Although this study did not demonstrate that circumcision in HIV-infected men reduces transmission to women, arguments still can be made for offering the procedure. Excluding HIV-positive men could result in stigmatization and indirect identification of those who are seropositive. Circumcised HIV-uninfected men might use their circumcision status to negotiate unsafe sex. Other data supports circumcision in reducing STD acquisition among those who have had the procedure. It will be important for programs that provide circumcision to emphasize the importance of refraining from sex before wound healing, as this may increase the probability of infecting a partner.


  1. Puren A. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Med 2005;2:1112-22. Abstract not available.
  2. Bailey RC, Moses S, Parker CB, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomized controlled trial. Lancet 2007;369:643-56.
  3. Kigozi G, Gray RH, Wawer MJ, et al. The safety of adult male circumcision in HIV-infected and uninfected men in Rakai, Uganda. PLoS Med 2008;5:e116.
  4. Gray RH, Kiwanuka N, Quinn TC, et al. Male circumcision and HIV acquisition and transmission: cohort studies in Rakai, Uganda. Rakai Project Team. AIDS 2000;14:2371-81.