Gray RH, Kigozi G, Serwadda D, Makumbi F, Watya S, Nalugoda F, Kiwanuka N, Moulton LH, Chaudhary MA, Chen MZ, Sewankambo NK, Wabwire-Mangen F, Bacon MC, Williams CF, Opendi P, Reynolds SJ, Laeyendecker O, Quinn TC, Wawer MJ. Male Circumcision for HIV Prevention in Men in Rakai, Uganda: A Randomised Trial. Lancet. 2007 Feb 24;369(9562):657-66.
To investigate the effect of male circumcision on HIV incidence in men.
A randomized controlled trial in which men were randomly assigned to receive immediate circumcision (intervention group) or to have circumcision delayed for 24 months (control group).
Rural Rakai district, Uganda.
A total of 6,461 uncircumcised men aged 15-49 years were screened at a central study facility and in mobile facilities in the rural communities. Of these, 1,046 were excluded due to refusal to undergo voluntary counseling and testing or a positive HIV test. Another 314 were incompletely enrolled and 101 had medical contraindications. Of the 5,000 men enrolled, four were excluded due to double enrollment. Ultimately, 4,996 men participated in this study; 2,474 in the intervention group and 2,522 in the control group. Baseline characteristics between the intervention and control groups were very similar at the time of enrollment. Approximately 48% of the participants were married, 68% were Catholic, and 66% had primary education. Roughly 20% reported no sexual partners in the past year, while 47% had one partner, 23% had two partners, and 12% had three or more partners. Additionally, 51% reported having a non-marital partner in the past year, 38% reported not using condoms in the past year, 29% had inconsistent condom use, and 14% had consistent condom use. Transactional sexual intercourse was low, at 2%.
Treatment assignment was randomly generated in blocks of 20, stratified on community, with each community receiving four blocks of 20 assignment envelopes. Because enrollment occurred concurrently at more than one site, this procedure ensured balance within the sites. In the intervention group, surgery was provided to 91% of men within two weeks of enrollment. Trained and certified physicians performed the circumcisions in well-equipped operating rooms. Post-operative follow-up visits were scheduled at 24-48 hours, 5-9 days, and 4-6 weeks. HIV testing, physical examination, and interviews were repeated at 4-6 weeks, 6-, 12-, and 24-month follow-up visits.
The primary outcome was HIV incidence. For incidence rate and Poisson regression calculations, HIV sero-conversion was estimated assuming that infection occurred at the mid-point between the last negative and the first positive serological tests, or at the time of the first positive RT-PCR for those participants seen during the period before HIV antibody sero-conversion. For participants who were positive by PCR but who were negative for HIV antibody, the date of the positive PCR was used as the date of infection. Time from enrollment was accumulated up to the 24-month follow-up visit and HIV incidence was estimated per 100 person-years. A modified intention-to-treat approach was used for analysis, which included all participants who were serologically or PCR negative at enrollment. The primary modified intention-to-treat population included crossovers and participants who reported periods of sexual abstinence during the 24 months of follow-up.
Retention rates at 12 months were equivalent in both groups (90-92%). NIAID terminated the trial for efficacy on December 12, 2006, by which time 44% of men in both groups had reached their 24-month follow-up time point. In the modified intention-to-treat analysis, HIV incidence over 24 months was 0.66 cases per 100 person-years in the intervention group and 1.33 cases per 100 person-years in the control group. After adjusting for age, marital status, and sexual risk behaviors at enrollment, the incidence rate ratio (IRR) was 0.49 (95% confidence interval [CI] 0.29-0.81; p=0.003). In a Kaplan-Meier survival curve for time-to-detection of HIV infection, the difference between the cumulative probabilities of HIV detection was significant (p=0.003) and the risk ratio was 0.43 (0.24-0.75). The as-treated Poisson analysis, which assigned person-time according to the actual circumcision status of participants, showed an incidence of 0.61 cases per 100 person-years in the intervention group and 1.35 cases per 100 person-years in the control group with an IRR of 0.45 (95% CI 0.25-0.78; p=0.002). The as-treated Kaplan-Meier risk ratio was 0.40 (0.23-0.70, p=0.003).
The rates of HIV acquisition were lower among circumcised men in all strata of characteristics, risk behaviors, and symptoms of sexually transmitted infections examined, with the exception of men who reported no sexual activity within the follow-up interval of sero-conversion. HIV incidence was lower in circumcised than in uncircumcised men in all age groups and categories of marital status and education. Circumcision reduced HIV acquisition irrespective of the number of partners, non-marital relationships, condom use, consumption of alcohol before sexual intercourse, and transactional sexual intercourse. The efficacy of circumcision for prevention of incident HIV was 51% in the Poisson intention-to-treat analysis. In the as-treated Poisson analysis, efficacy was 55% and the Kaplan-Meier estimate of efficacy was 60%. The prevalence of self-reported genital ulcers during the preceding interval was lower in the intervention group than in the control group (3.1% vs 5.8%, PRR 0.53, 95% CI 0.43-0.64; p<0.0001). There was no consistent or substantial evidence of behavioral disinhibition after circumcision in the study population. The rate of all adverse events related to the surgery in the intervention group was about 8%, all of which were successfully managed and resolved.
The authors conclude that male circumcision reduced HIV incidence in men without behavioral disinhibition and they state that circumcision can be recommended for HIV prevention in men.
Based on the Jadad criteria for randomized controlled trials, this study was of good quality. Medical workers could not be blinded to treatment. However, there was no discussion as to whether non-medical staff who gave HIV tests, administered questionnaires, and counseled participants about risk reduction were blinded to treatment. Measurements of sexual risk behaviors and symptoms of sexually transmitted diseases were based on self-report, which could be under- or over-reported. Generalizability of study results to other populations could be limited.
This study's efficacy findings are compatible with observational data, as well as data from a randomized trial in South Africa(1) (60% intention-to-treat efficacy and 76% as-treated efficacy in a semi-urban population aged 18-24 years) and a trial in Kenya(2) (53% intention-to-treat efficacy and 60% as-treated efficacy in an urban population aged 18-24 years), which suggests similar efficacy in divergent populations. Additionally, circumcision also reduced the rate of self-reported symptoms of genital ulcer disease (GUD), which is comparable with the protective effects of circumcision on GUD in observational studies.(3)
The results of this study show that circumcision can be used to reduce the risk of heterosexually acquired HIV infection in adult men. The public health implications are significant with the potential to avert large numbers of infections, both in the circumcised men and in their sexual partners. However, the use of surgery for a prevention program must only be used and scaled up under the appropriate circumstances. Careful attention must be paid to training of personnel, provision of facilities, equipment and supplies, postoperative care to minimize and manage complications, and monitoring of the quality of services and surgical outcomes. Circumcision of younger boys may be a safer, simpler, and less expensive option, although the HIV benefits will be delayed. Additionally, circumcision programs must emphasize that circumcision provides only partial protection, and that there is a critical need to practice safe sex after circumcision (limiting the number of sex partners and consistently using condoms). Further, trials that are stopped early could overestimate efficacy and long-term post-circumcision trial surveillance will be essential. Perceptions of circumcision efficacy may lead to an exaggerated belief in the protective effects of the procedure, thus engendering increases in HIV risk behaviors.
- Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, Puren A. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med 2005 Nov;2(11):e298.
- Bailey RC, Moses S, Parker CB, Agot K, Maclean I, Krieger JN, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007 Feb 24;369(9562):643-56.
- Weiss HA, Thomas SL, Munabi SK, Hayes RJ. Male circumcision and risk of syphilis, chancroid, and genital herpes: a systematic review and meta-analysis. Sex Transm Infect 2006 Apr;82(2):101-9; discussion 110. Review.