Search tips
Search criteria 


Logo of jidLink to Publisher's site
J Infect Dis. 2016 August 15; 214(4): 595–598.
Published online 2016 May 13. doi:  10.1093/infdis/jiw182
PMCID: PMC4957438

Genital Anaerobic Bacterial Overgrowth and the PrePex Male Circumcision Device, Rakai, Uganda


The PrePex circumcision device causes ischemic necrosis of the foreskin, raising concerns of anaerobic overgrowth. We compared the subpreputial microbiome of 2 men 7 days after PrePex device placement to that of 145 uncircumcised men in Rakai, Uganda, using 16S ribosomal (rRNA) RNA gene–based quantitative polymerase chain reaction analysis and sequencing. PrePex users had higher absolute abundance of all bacteria than uncircumcised men (P = .001), largely due to increased numbers of the following anaerobes: Porphyromonas (5.2 × 107 16S rRNA gene copies/swab in the PrePex group and 1.1 × 106 16S rRNA gene copies/swab in uncircumcised men; P = .002), Peptoniphilus (1.0 × 107 and 1.8 × 106 16S rRNA gene copies/swab, respectively; P < .05), Anaerococcus (1.0 × 107 and 1.1 × 106 16S rRNA gene copies/swab, respectively; P < .001), and Campylobacter ureolyticus (1.7 × 105 and 1.6 × 10716S rRNA gene copies/swab, respectively; P < .001). The PrePex-associated increase in anaerobes may account for unpleasant odor and a possible heightened risk of tetanus.

Keywords: male circumcision, PrePex device, anaerobes, penile microbiota, Uganda, tetanus

Randomized trials have shown that medical male circumcision reduces the incidence of male acquisition of human immunodeficiency virus (HIV) and other sexually transmitted viral pathogens [14]. The Joint United Nations Programme on AIDS and the World Health Organization (WHO) recommend that circumcision be an integral component of HIV prevention strategies in 14 priority countries in East and Southern Africa. The goal is to provide circumcision to 20.3 million men to achieve 80% circumcision coverage; thus far, >10 million men have been circumcised through circumcision programs [5].

Lack of basic surgical facilities and skilled personnel are constraints on service provision, and this has motivated the development of simpler circumcision devices [6]. The first device to receive WHO prequalification and the primary one used in Africa is PrePex, which is a nonsurgical method of circumcision that compresses the foreskin, causing ischemic necrosis, and the necrotic foreskin is removed 7 days after placement [7]. Unpleasant odor has been reported prior to PrePex removal, which was thought to be due to anaerobe overgrowth in the subpreputial space or the necrotic foreskin [8, 9]. More recently, 2 men circumcised using PrePex died from tetanus shortly after device removal, among 3800 men who underwent PrePex circumcision in Uganda, with an estimated tetanus-associated death rate of approximately 52.6 cases/100 000 (95% confidence interval, 6.4–190 cases/100 000) [10, 11], which is higher than the estimated tetanus incidence in the general populations of Uganda and elsewhere in sub-Saharan Africa [12]. A review of data on tetanus cases related to circumcision led the WHO to issue a guidance document on tetanus prophylaxis for men receiving the PrePex device and those undergoing surgical methods for circumcision [10].

We previously assessed the subpreputial microbiome in men circumcised using conventional surgery and found a major decrease in anaerobic bacteria following circumcision [13]. We hypothesized that the necrotic foreskin in PrePex users might facilitate anaerobic overgrowth [8, 9], so we conducted a study to assess the microbiome in the subpreputial space at time of PrePex removal and compared this to the microbiome in uncircumcised men.


The men who received the PrePex device were enrolled in a WHO/Ministry of Health PrePex prequalification evaluation, which was classified as nonexperimental/nonresearch by the WHO and the Ministry of Health. All men provided written informed consent to undergo the procedure. As part of the evaluation, we collected a subpreputial swab specimen prior to removal of the PrePex device, 7 days after placement. Permission to import the PrePex device for research was provided by the Uganda National Drug Authority.

We compared the subpreputial microbiome prior to PrePex removal on day 7 to that of a group of 145 uncircumcised men from a prior randomized, controlled trial of circumcision for HIV prevention, conducted in Rakai, Uganda, during 2004–2006 [3]. Although the 2 groups were not contemporaneous, the methods of sample collection, processing, and analysis were identical. Swab samples were collected using sterile cotton-tipped applicators (ThermoFisher Scientific, Waltham, Massachusetts) premoistened with sterile saline, which were rolled over the coronal sulcus twice in a nontraumatic fashion. The swabs were immediately placed in 1 mL of Amplicor specimen transport medium (Roche Diagnostics, Indianapolis, Indiana) and stored at −80°C until assay.

Each sample was lysed in 100 µL of eluted transport medium, using enzyme-free chemical and mechanical lysis as previously described [13]. The lysate was purified using the Qiagen AllPrep DNA/RNA Mini Kit (Qiagen, Valencia, California), and DNA elution was performed using 100 µL of Buffer EB. The bacterial load and microbiome composition were characterized as previously described [13]. In brief, the bacterial load, measured as the number of bacterial 16S ribosomal RNA (rRNA) gene copies per microliter of swab eluent, was determined using a broad-coverage quantitative polymerase chain reaction (qPCR) assay, and microbiome composition was characterized by sequencing barcoded amplicons of the 16S rRNA V3V6 region generated using 50-µL reactions that include 400 nM of each broad range fusion F primer (5′-CCATCTCATCCCTGCGTGTCTCCGACTCAGnnnnnnnnCCTACGGGDGGCWGCA-3′) and fusion R primer (5′-CCTATCCCCTGTGTGCCTTGGCAGTCTCAG-CTGACGACRRCCRTGCA-3′), with the underlined portion denoting FLX Lib-L adapter sequence, the italicized portion denoting the sample-specific 8-nucleotide barcode sequence, and the bold portion denoting the primer sequence. The resultant fusion PCR product were analyzed using 1% E-Gel 96 Agarose (Invitrogen by Life Technologies, Grand Island, New York) to confirm amplification and product band size. The barcoded 16S rRNA gene amplicons were quantified using a broad-coverage qPCR as described earlier and then were pooled in an equimolar fashion and sequenced on the Genome Sequencer FLX instrument (454 Life Sciences, Branford, Connecticut). Resultant sequences were chimera-checked, demultiplexed, and quality checked as previously described [13]. Taxonomic classification was performed using the Ribosomal Database Project Naive Bayesian Classifier (RDP release 10, update 28).

Microbiome composition was characterized using the following metrics: (1) richness, calculated as the number of unique bacterial species found in a sample; (2) prevalence, calculated as the proportion of men with each bacterium of interest; (3) proportional abundance of a bacterium, calculated as a specific bacterium's 16S rRNA gene sequences as a proportion of the total 16S rRNA gene sequences detected; and (4) absolute abundance of a bacterium, defined as the number of 16S rRNA gene sequences attributable to a specific bacterium and calculated as the product of the bacterium's proportional abundance times the total 16S rRNA bacterial load. We also calculated the total anaerobe absolute abundance.

We used χ2 tests to compare bacterial prevalence and the Kolmogorov-Smirnov tests to compare the richness and proportional and absolute abundance of all and specific anaerobes between PrePex users and uncircumcised controls. Adjustment for multiple test comparisons was based on the false-discovery rate.


Men circumcised using the PrePex device (median age, 20.0 years; interquartile range [IQR], 18.0–25.3 years) were younger than the uncircumcised control men (median age, 30.0 years; IQR, 25.0–35.0 years; P < .01). The majority of men in both groups had completed primary school and were engaged in agriculture or fishing. Similar proportions of men in each group reported having either only 1 sex partner in the past year (45.5% in the PrePex group vs 58.6% in the uncircumcised control group; P = .35) or ≥2 partners (40.9% and 41.4%, respectively; P = .99). All uncircumcised controls were married and reported being sexually active in the past year, whereas 50% of men circumcised with the PrePex device were never married, and 13.6% reported no sex partner in the past year.

The subpreputial bacterial load at day 7 was significantly higher in men circumcised using PrePex than in uncircumcised men (1.6 × 108 vs 3.2 × 107 16S rRNA gene copies/swab; P = .001; Figure Figure11A ). There was a significantly lower median number of unique bacterial species in penile specimens (ie, decreased richness) in men circumcised with PrePex (21 species/individual; IQR, 19–24 species/individual), compared with uncircumcised men (26 species/individual; IQR, 20–35 species/individual; P = .03).

Figure 1.
A and B, Comparisons of total penile bacterial absolute abundance (A) and total anaerobe absolute abundance (B). As shown by the box plots in panel A, where the box encompasses the interquartile range (IQR) and the median total penile bacterial absolute ...

A similar core set of anaerobic bacteria—including Peptoniphilus, Anaerococcus, Prevotella, Porphyromonas, and Finegoldia—were found in both study groups; however, compared with uncircumcised men, PrePex users had significantly higher proportional abundance of anaerobes, particularly Porphyromonas (30.2% vs 3.6%; P < .001) and Campylobacter ureolyticus (8.9% vs 0.4%, P < .001; Table Table11).

Table 1.
Summary of Penile Anaerobes With Significantly Different Absolute Abundance in Uncircumcised Men and 7 Days After Placement of the PrePex Device

The absolute abundances of Porphyromonas, C. ureolyticus, and other anaerobes were also significantly higher in men circumcised using PrePex (Porphyromonas, 5.2 × 107 vs 1.1 × 106 16S rRNA copies/swab [P < .01]; C. ureolyticus, 1.7 × 107 vs 1.7 × 105 16S rRNA copies/swab [P < .01]). Other anaerobes that were significantly higher in men circumcised with PrePex, compared with uncircumcised controls, included Peptoniphilus (1.0 × 108 vs 1.8 × 10616S rRNA copies/swab; P < .05) and Anaerococcus (1.0 × 107 vs 1.1 × 10616S rRNA copies/swab; P < .001; Table Table11).

On average, PrePex users had a 6.5-fold higher absolute abundance of total penile anaerobes, compared with uncircumcised controls (1.3 × 108 2.0 × 107 16S rRNA gene copies/swab; P = .001; Figure Figure11B). Adjustment for age or number of sex partners in the past year did not affect the association between total penile anaerobes and PrePex users as compared to findings for uncircumcised controls.


As compared to uncircumcised controls, we found that the penile microbiome of PrePex users at the time of PrePex device removal had a significantly higher total bacterial load and higher absolute abundance of penile anaerobes. This suggests that the necrotic foreskin at day 7 after PrePex device placement is associated with growth of anaerobes, including Porphyromonas, Peptoniphilus, Anaerococcus, and C. ureolyticus. This phenomenon of anaerobiosis is unsurprising, given that the subpreputial space is naturally colonized by anaerobes in uncircumcised men and that the PrePex device causes ischemic necrosis of the foreskin and/or low oxygen tension in the subpreputial space.

This anaerobiosis with the PrePex device might explain the possible increased risk of tetanus in PrePex users with no or incomplete tetanus toxoid immunization. The increase in penile anaerobes may also explain the complaints of unpleasant odor with PrePex use [14, 15].

This study has limitations. We did not sample the subpreputial space prior to placement of the PrePex device, so we could not directly assess the growth of anaerobes over the 7-day period. However, the previously collected swabs from uncircumcised men provided data on the “normal” prevalence and abundance of subpreputial penile bacteria. The number of PrePex users in this study was small, which limited the power to detect statistically significant differences in some anaerobic species. In addition, since our study was conducted prior to the concerns regarding tetanus risk, we did not specifically probe for Clostridium tetani and while we detected sequences from other Clostridium species such as Clostridium septicum, we did not detect sequences from C. tetani in either group. Nevertheless, the significantly higher anaerobe abundance at time of PrePex removal suggests that if the subpreputial space was contaminated by C. tetani spores, the necrotic foreskin may provide a favorable environment for the C. tetani spore germination and growth.

WHO has recommended rigorous skin preparation prior to surgery, avoidance of postoperative wound contamination or application of substances which might contain C. tetani spores, and prophylactic tetanus toxoid immunization for men who may have no or incomplete immunity [10]. As foreskin necrosis is inextricably linked to the mechanism of action for the PrePex device, the phenomenon of anaerobic overgrowth may be unavoidable. Thus, it will be important to evaluate the impact of the pre-surgical skin preparation on the penile microbiome.


Financial support. This work was supported by the National Institutes of Health (R01AI087409-01A1 to L. B. P. and R. H. G.; 1K23AI093152-01A1 to A. A. R. T.).

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.


1. 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; 2:e298. [PubMed]
2. Bailey RC, Moses S, Parker CB et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007; 369:643–56. [PubMed]
3. Gray RH, Kigozi G, Serwadda D et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007; 369:657–66. [PubMed]
4. Tobian AA, Serwadda D, Quinn TC et al. Male circumcision for the prevention of HSV-2 and HPV infections and syphilis. N Engl J Med 2009; 360:1298–309. [PMC free article] [PubMed]
5. WHO. Voluntary medical male circumcision for HIV prevention in 14 priority countries in East and southern Africa, 2015. Accessed 29 August 2015.
6. WHO. WHO technical advisory group on innovations in male circumcision: evaluation of two adult devices, 2015. Accessed 29 August 2015.
7. Tobian AAR, Adamu T, Reed JB, Kiggundu V, Yazdi Y, Njeuhmeli E Voluntary medical male circumcision in resource-constrained settings. Nat Rev Urol 2015; 12:661–70. [PubMed]
8. Kigozi G, Musoke R, Watya S et al. The safety and acceptance of the PrePex device for non-surgical adult male circumcision in Rakai, Uganda. A non-randomized observational study. PLoS One 2014; 9:e100008. [PMC free article] [PubMed]
9. Galukande M, Duffy K, Bitega JP et al. Adverse events profile of PrePex a non-surgical device for adult male circumcision in a Ugandan urban setting. PLoS One 2014; 9:e86631. [PMC free article] [PubMed]
10. WHO. WHO informal consultation on tetanus and voluntary medical male circumcision, 2015. Accessed 29 August 2015.
11. Grund JM, Toledo C, Davis SM et al. Notes from the field: tetanus cases after voluntary medical male circumcision for HIV prevention—Eastern and Southern Africa, 2012–2015. MMWR Morb Mortal Wkly Rep 2016; 65:36–7. [PubMed]
12. WHO. Tetanus (total) reported cases, 2016. Accessed 2 March 2016.
13. Liu CM, Hungate BA, Tobian AA et al. Male circumcision significantly reduces prevalence and load of genital anaerobic bacteria. MBio 2013; 4:e00076. [PMC free article] [PubMed]
14. Vass AA. Odor mortis. Forensic Sci Int 2012; 222:234–41. [PubMed]
15. Vandamme P, Debruyne L, De Brandt E, Falsen E Reclassification of Bacteroides ureolyticus as Campylobacter ureolyticus comb. nov., and emended description of the genus Campylobacter . Int J Syst Evol Microbiol 2009; 60:2016–22. [PubMed]

Articles from The Journal of Infectious Diseases are provided here courtesy of Oxford University Press