Objective Previous randomised trial data have demonstrated that male circumcision reduces Mycoplasma genitalium prevalence in men. We assessed whether male circumcision also reduces M genitalium infection in female partners of circumcised men.
Methods HIV-negative men were enrolled and randomised to either male circumcision or control. Female partners of male trial participants from the intervention (n=437) and control (n=394) arms provided interview information and self-collected vaginal swabs that were tested for M genitalium by APTIMA transcription-mediated amplification-based assay. Prevalence risk ratios (PRR) and 95% CI of M genitalium prevalence in intervention versus control group were estimated using Poisson regression. Analysis was by intention-to-treat. An as-treated analysis was conducted to account for study-group crossovers.
Results Male and female partner enrolment sociodemographic characteristics, sexual behaviours, and symptoms of sexually transmitted infections were similar between study arms. Female M genitalium prevalence at year 2 was 3.2% (14/437) in the intervention arm and 3.6% (14/394) in the control arm (PRR=0.90, 95% CI 0.43 to 1.89, p=0.78). In an as-treated analysis, the prevalence of M genitalium was 3.4% in female partners of circumcised men and 3.3% in female partners of uncircumcised men (PRR=1.01, 95% CI 0.48 to 2.12, p=0.97).
Conclusions Contrary to findings in men, male circumcision did not affect M genitalium infection in female partners.
- Epidemiology (Clinical)
Statistics from Altmetric.com
Mycoplasma genitalium is a sexually transmitted infection (STI), and growing evidence is demonstrating that it is associated with urethritis, cervicitis, salpingitis and pelvic inflammatory disease.1 The prevalence of M genitalium among women in the general population is approximately 1–5%.1–4 However, M genitalium infection is substantially higher among HIV-positive women and female sex workers with the prevalence ranging from 10% to 26%.5–8 M genitalium infection has also been associated with an increased risk of acquiring HIV.9
Three randomised trials in South Africa, Kenya and Uganda, demonstrated that male circumcision (MC) significantly decreases HIV acquisition in men.10–12 Additionally, these trials have shown that MC reduces herpes simplex virus type 2 (HSV-2) and human papillomavirus (HPV) in men, and the Kenyan trial also demonstrated that MC decreases M genitalium infection.13–16 The Ugandan trial also showed that female partners of circumcised men have decreased prevalence of genital ulcer disease (GUD), Trichomonas vaginalis, bacterial vaginosis and HPV.17 ,18
We used data from a randomised controlled trial of MC in HIV-negative men in Rakai, Uganda, to assess the efficacy of MC for reducing prevalence of M genitalium in the female partner.
Materials and methods
Participants, study design and randomisation
The Rakai Health Sciences Program (RHSP) in Rakai, Uganda, enrolled 5596 HIV-negative men in MC trials for HIV/STI prevention.10 ,17 Men were eligible for enrolment if they were uncircumcised, aged 15–49 years, had no medical indications or contraindications for MC, and provided written informed consent. Men were randomly assigned to receive immediate MC (intervention arm, n=2786) or MC delayed for 24 months (control arm, n=2810).
Consenting female partners of male trial participants who were married or in long-term consensual relationships were invited to participate in a separate follow-up study.17 All female participants provided written informed consent. The effects of MC on female STIs were secondary trial outcomes. As previously described, there were 648 women in the intervention arm and 597 women in the control arm, who were persistently HIV-negative, married, concurrently enrolled with their husband who participated in the trial and had a swab collected at enrolment.17 Of these women, 549 women in the intervention arm and 502 women in the control arm had swabs collected at year 2.17 However, swab samples from 220 randomly selected women (112 (20.4%) from the intervention arm and 108 (21.5%) from the control arm) were exhausted after being used for previous studies. Thus, the current study included the remaining 437 women from the intervention arm and 394 women from the control arm at year 2. There were no differences in terms of age, marital status, religion, education, sexual partners, non-marital relationships, condom use, alcohol use, transactional sex, receipt of voluntary counselling and testing, HPV prevalence or self-reported symptoms of GUD, vaginal discharge or dysuria between this population and the primary trial population.17 The primary objective of this analysis was to assess the efficacy of MC of HIV-negative men on prevalence of M genitalium in the female partner.
At each annual study visit, women were interviewed to ascertain sociodemographic characteristics, sexual risk behaviours, and health status. During the mid-point trial study visit (year 1), women presenting with either discharge (n=148, 17.8%) or dysuria (n=46, 5.5%) were treated with metronidazole and azithromycin to cover vaginal and cervical infections. Women presenting with genital ulcers (n=16, 1.9%) were treated with azithromycin and acyclovir. Women were also asked to provide blood samples and self-administered vaginal swabs. They were instructed to squat, insert a 20-cm Dacron or cotton-tipped swab and to rotate the swab high in the vaginal vault. After collection, the women handed the swab to a field worker who placed the swab in 1 mL of AMPLICOR specimen transport medium (Roche Diagnostics, Indianapolis, Indiana, USA). This approach to specimen collection was well accepted, with compliance rates of over 90% at study visits. The specimens were maintained at 4–10°C for less than 6 h until they were frozen at −80°C.
The trials were approved by the Uganda National Council for Science and Technology, the Science and Ethics Committee of the Uganda Virus Research Institute (Entebbe, Uganda), the Committee for Human Research at Johns Hopkins University Bloomberg School of Public Health (Baltimore, Maryland, USA), and the Western Institutional Review Board (Olympia, Washington, USA).10 ,17 ,18 The trials were overseen by independent Data and Safety Monitoring Boards. The trials were registered with ClinicalTrials.gov, NCT00425984 and NCT00124878.
M genitalium, HPV and HIV detection
M genitalium infection was detected using the APTIMA transcription-mediated amplification-based Research Use Only (RUO) assay (Gen-Probe, San Diego, California, USA), as previously described.19
HPV Linear Array (Roche Diagnostics, Indianapolis, Indiana, USA) was used to detect 37 HPV genotypes.14 HIV status was determined using two separate ELISAs, and discordant results were confirmed by HIV-1 Western Blot.10
Enrolment and follow-up characteristics, sexual risk behaviours and STI symptoms in men and their female partners were tabulated by study arm, and differences assessed by χ2 tests. All p values were two-sided.
The primary assessment of MC efficacy for reduced female M genitalium prevalence used an intention-to-treat analysis. An as-treated analysis was also carried out, in which intervention arm women were classified as crossover exposures if their male partner remained uncircumcised at the annual follow-up visit, and partners of control arm men were classified as crossover exposures if the male partner underwent MC from other sources during the follow-up interval in which the procedure was performed.
Prevalence risk ratios (PRR) and 95% CI of M genitalium prevalence in intervention versus control group were estimated using Poisson regression. Multivariate Poisson regression was used to estimate adjusted PRRs (adjPRRs) after adjusting for enrolment covariates that differed significantly between the arms at p<0.15 and known risk factors for M genitalium infection, which included enrolment age, treatment for vaginal discharge, dysuria, or genital ulcers at the mid-point trial visit, and the year 2 follow-up number of sex partners (1 vs >1), condom use, non-marital relationships and transactional sexual intercourse.
Analyses were performed using STATA V.11.2 (StataCorp LP, College Station, Texas, USA).
Male baseline sociodemographic characteristics, sexual behaviours and symptoms of STIs were similar between study arms, except more men in the control arm reported drinking alcohol prior to sexual intercourse (p=0.01) (table 1). The female enrolment characteristics were similar between trial arms (table 1). Additionally, HPV prevalence was similar at enrolment for the female partners between the two trial arms.
The overall prevalence of M genitalium at year 2 was 3.4% (28 cases of 831 women). In the intention-to-treat analysis, M genitalium infection was detected in 14 female partners of men in the intervention group, and in 14 female partners of men in the control group at the 2-year follow-up visit. Female partner prevalence at year 2 was 3.2% in the intervention arm and 3.6% in the control arm (PRR=0.90, 95% CI 0.43 to 1.89, p=0.78) (table 2). After adjustment, the PRR (adjPRR) of M genitalium infection in female partners of intervention relative to control arm men was 0.93 (95% CI 0.43 to 2.03, p=0.86).
In an as-treated analysis, the prevalence of M genitalium was 3.4% in female partners of circumcised men and 3.3% in female partners of uncircumcised men (PRR=1.01, 95% CI 0.48 to 2.12, p=0.97).
Self-reported rates of female partners’ sexual behaviours and STI symptoms were assessed by the male partner's trial arm and follow-up interval. There were no differences at year 2 among the female partners between study arms in self-reported number of sexual partners, GUD, vaginal discharge, or dysuria (table 3). There were also no differences in the report of non-marital relationships (2.3% (10/435) of women in the intervention arm and 3.4% (13/388) of women in the control arm, p=0.36). However, there were more women in the control arm who reported differences in condom use (22.7% (88/388)) compared with women in the intervention arm (13.1% (57/435), p=0.001).
We found that MC of HIV-negative men did not affect prevalence of M genitalium in the female partner. The lack of MC efficacy for prevention of M genitalium infection in female partners of HIV-negative men is likely due to multiple factors. M genitalium infection from outside relationships could have diluted the potential efficacy of MC. The study may have lacked the power to detect an effect of MC on female partners’ M genitalium infections. The three MC trials have shown consistently that MC reduces viral STIs among men.20 ,21 More recent evidence has shown that MC modifies and reduces the penile microbiome,22 but has limited to no impact on bacterial STIs for men, specifically syphilis, Neisseria gonorrhoeae, Chlamydia trachomatis and Trichomonas vaginalis.14 ,23–25 These pathogens are usually found in urethral cells and not in the foreskin. While M genitalium is a urethral pathogen, the foreskin may also be a reservoir for M genitalium.26 Consequently, it may be biologically plausible that MC has no effect on M genitalium infection in female partners since the impact of MC on bacterial STIs among men is limited. Further research is needed in this area to understand the role of the foreskin, bacterial pathogens and HIV risk.
There are limitations with this study. The women were all in stable partnerships with HIV-negative men, and may represent a self-selected population of more compliant lower-risk participants in both arms. Although the MC trial in Kenya showed the circumcised men had lower prevalence of M genitalium,15 we unfortunately did not have urine or urethral swabs to evaluate the men in this trial. The risk behaviours and symptoms of STIs are self-reported, and the data were potentially vulnerable to recall and reporting bias. We do not know the prevalence of M genitalium at enrolment. However, female partner participants of the two trial arms were likely similar since there was no difference in the demographics, sexual behaviours or HPV prevalence at enrolment. We were unable to assess the impact of MC on M genitalium acquisition, and a portion of the year 2 M genitalium prevalence could represent chronic infection from enrolment.
Despite the benefits for female partners of circumcised men, such as reduced rates of high-risk HPV transmission, GUD, trichomoniasis and bacterial vaginosis,17 ,18 MC did not affect the prevalence of M genitalium infection in their female partners in this study.
Previous randomised trial data have demonstrated that male circumcision reduces Mycoplasma genitalium prevalence in men and there are derivate benefits for female partners.
It is not known whether male circumcision also reduces M genitalium infection in female partners of circumcised men.
This study demonstrated that contrary to findings in men, male circumcision does not affect M genitalium infection in female partners.
We are most grateful to the study participants and the Rakai Community Advisory Board whose commitment and cooperation made this study possible. The authors also thank Gen-Probe Hologic for the donation of the RUO Mycoplasma genitalium TMA reagents.
Handling editor Jackie Cassell
Contributors All authors contributed to the study design, data collection, data analysis, writing and reviewing the paper.
Funding The trials were funded by the Bill and Melinda Gates Foundation (#22006.02) and the National Institutes of Health (#U1AI51171). The Doris Duke Charitable Foundation (grant #2011036) along with the Division of Intramural Research of the National Institute of Allergy and Infectious Diseases, NIH provided laboratory support. The Fogarty International Center (#5D43TW001508 and #D43TW00015) contributed to training. AART was supported by the Doris Duke Charitable Foundation and NIH 1K23AI093152-01A1. CG was supported by NIAID NIH A1U0168613 and NIBIB NIH U54EB007958.
Competing interests There are no potential conflicts of interest relevant to this article by all authors, except Charlotte Gaydos reports having previously received research funding from Gen-Probe Hologic.
Ethics approval The trials were approved by the Uganda National Council for Science and Technology, the Science and Ethics Committee of the Uganda Virus Research Institute (Entebbe, Uganda), the Committee for Human Research at Johns Hopkins University Bloomberg School of Public Health (Baltimore, Maryland, USA), and the Western Institutional Review Board (Olympia, WA, USA). The trials were overseen by independent Data and Safety Monitoring Boards.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.