CROI: Circumcision could particularly benefit higher-risk men

This article is more than 15 years old. Click here for more recent articles on this topic

Clinical trials may have understated the HIV prevention benefit of circumcision, according to the lead investigator on a recently reported study. The benefit appears to grow over time and may be highest in men with multiple partners, the Fourteenth Conference on Retroviruses and Opportunistic Infections heard this week in Los Angeles.

As already reported, two trials of circumcision as an HIV prevention measure for men in Rakai, Uganda and Kisumu, Kenya were halted early last December when it became apparent that in both trials circumcision had approximately halved the risk of acquiring HIV.

Ronald Gray, lead investigator of the Rakai trial, gave more details to the Fourteenth Conference on Retroviruses and Opportunistic Infections in Los Angeles last week. He said that the benefit of circumcision was probably greater than the preliminary efficacy of 51% would indicate. This is both because the benefit, for reasons as yet unclear, appears to grow over time and because the highest-risk men, namely those with multiple partners and/or with genital ulcer disease, appeared to particularly benefit.


voluntary male medical circumcision (VMMC)

The surgical removal of the foreskin of the penis (the retractable fold of tissue that covers the head of the penis) to reduce the risk of HIV infection in men.


The surgical removal of the foreskin of the penis (the retractable fold of tissue that covers the head of the penis) to reduce the risk of HIV infection in men.

genital ulcer disease

Any of several diseases that are characterised by genital sores, blisters or lesions. Genital ulcer diseases (including genital herpes, syphilis and chancroid) are usually sexually transmitted.

statistical significance

Statistical tests are used to judge whether the results of a study could be due to chance and would not be confirmed if the study was repeated. If result is probably not due to chance, the results are ‘statistically significant’. 

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.

Almost 5,000 men aged between 15 and 49 in Rakai, a rural district of Uganda, were randomised either to immediate circumcision or to be offered circumcision at the end of the two-year study. Fifty per cent of the men reported extramarital partners and 40% reported (inconsistent) condom use. Men who turned out to be HIV-positive on screening were referred to a parallel and ongoing study which is studying the effect of circumcision on HIV transmission by positive men.

The study was stopped early when interim analysis showed that circumcised men had a significantly reduced risk of HIV acquisition.

The risk of HIV infection was compared between the circumcised and uncircumcised men at follow-up visits at six, twelve and 24 months after circumcision. Data were also gathered on rates of genital ulcer disease and urethral infections in circumcised and uncircumcised men. The men were also asked if they had had symptoms suggestive of a sexually transmitted infection, including general ulcers, discharge from the penis, or urethral pain. They were also tested for HIV and for syphilis, herpes, HPV, gonorrhoea, chlamydia and trichonomiasis at every visit, though so far the lab results for STIs other than HIV have not been analysed.

In an ‘intent-to-treat’ analysis, the incidence of HIV infection in the circumcision group was reduced by 51%. It was 0.66 per 100 person years in the circumcised men and 1.33 per 100 person years among uncircumcised men. This difference was statistically significant (p = 0.007).

Gray told the conference that the protective effect of circumcision appeared to increase over time. HIV incidence for circumcised men was 1.19% a year from 0-6 months after circumcision, 0.42% from 6-12 months and 0.40% from 12-24 months. This reduction over time was statistically significant too (p=0.0014). The corresponding incidence rates in uncircumcised men for the same time periods were 1.58%, 1.19% and 1.19%.

Gray said he “had no idea” why the protective effect of circumcision appeared to increase over time, but speculated that it was due to increased keratinisation of the glans of the penis.

However he added that because of the premature stoppage of the trial, 73% of person-time had accrued, but only 44% of the men were in the second year of follow-up. If the study had continued as planned, the protective effect of circumcision over time might have been stronger still.

There were also ‘crossovers’ between the intervention and control arms of the trial. There were 146 crossovers in the intervention arm, i.e. men who were randomised for circumcision but then did not go ahead with it, and 33 in the control arm, i.e. men who decided not to wait and got circumcised elsewhere. This means the ‘as treated’ efficacy of circumcision was greater that the ‘intent to treat’ efficacy.

Taking both these factors together and taking the ‘as-treated’ figures as if all men had stayed in the trial for 24 months, Gray calculated that the true efficacy of circumcision was 60% rather than 51%.

Gray added that circumcision appeared to offer greater benefit to higher-risk men.

The efficacy was 45% in men with one partner but 70% in men with two or more; it was 36% in men whose only sex was with their wives but 66% in men who had extramarital partners.

“It may be more efficacious in higher-risk people,” he said. “This is possibly due to induced mucosal immune response in regular partners” – in other words, because men acquire a degree of immunity to the HIV of regular partners, as other studies have demonstrated.

Gray said that circumcision appeared to protect against some, but not all, other sexually transmitted infections.

Three per cent of circumcised men experienced genital ulcers compared to 6% of uncircumcised men, a 47% difference which was statistically significant (p < 0.0001). However, rates of urethral discharge were identical between the two arms of the study at 2%, as were rates of urethral pain at 3%. Gray commented that circumcision appeared protective of cutaneous skin lesions but not of internal STIs that attacked the urethral mucosa.

Genital ulcers were associated with a considerably increased risk of HIV acquisition. In men without genital ulcers, HIV incidence was 0.63 a year in circumcised men and 1.1% a year in uncircumcised men – an efficacy for circumcision of 34%. But HIV incidence in men with genital ulcers was 1.8% a year in circumcised men and 6.3% a year in uncircumcised men – a circumcision efficacy of 71%. 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 – thus circumcision again offers more protection to men at higher risk of HIV.

Circumcision did not appear to affect the sexual activity as there were no consistent or substantial differences in reported sexual risk behaviour between the two arms. In fact condom use, though inconsistent, was slightly higher in circumcised men and alcohol use was slightly higher in uncircumcised men. This suggested that circumcision was not “disinhibiting” men, though Gray stressed that the men received much more intensive monitoring and support than they would if circumcision was rolled out as a national programme.

The circumcision operation takes 20-25 minutes. Moderate or severe side-effects following circumcision were reported by 4% of men who remained HIV-negative and 3% of the circumcised men who became infected with HIV.

Gray said that adverse events reported in the Gates-funded trial of HIV-positive men were similar.

Eighty-one per cent of men reported that the operation wound was completely healed within 30 days, and 89% of men did not resume sex until the wound was certified healed.

After the trial closure, no less than 80% of the men in the control arm elected to be circumcised.

Just five (0.2%) of the men developed ‘severe’ side-effects, meaning they had to be recalled to hospital. When questioned, Gray listed these. Two developed uncontrolled bleeding that required further stitching; one developed a serious bacterial infection; one ripped his stitches open “lifting a big bag of coffee”; and one developed a post-operative attack of herpes severe enough to require hospitalisation, a side-effect also seen in one of the men in the HIV-positive trial. All side-effects resolved with treatment. Gray said in answer to a question that if circumcision was rolled out, he would not expect ‘replicability’ of quite such good post-operative results, as surgeons would inevitably not be as highly trained.


Gray R et al. Randomized trial of male circumcision for HIV prevention in Rakai, Uganda. Fourteenth Conference on Retroviruses and Opportunistic Infection, Los Angeles, abstract 155aLB, 2007.

Wawer M et al. Effects of male circumcision on genital ulcer disease and urethral symptoms, and on HIV acquisition: an RCT in Rakai, Uganda. Fourteenth Conference on Retroviruses and Opportunistic Infections, Los Angeles, abstract 155bLB, 2007.