Three randomized controlled trials conducted in sub-Saharan Africa have demonstrated that circumcision can dramatically decrease the risk of HIV acquisition among heterosexual men who are potentially exposed to HIV through vaginal sex.1–3 However, the significance of this finding for men who have sex with men (MSM), the group at highest risk for HIV in North and South America, Europe, and Australia, is uncertain. To date, investigators have not conducted randomized trials of circumcision in MSM, and the rationale for conducting such a trial remains controversial. Observational studies evaluating the association of circumcision with HIV4–11 and with sexually transmitted infections (STI) other than HIV have reported mixed results, and most reports focusing on the potential protective effect of the procedure on STI have not included large numbers of MSM.12–20 A recent meta-analysis which used data aggregated from 15 observational studies found no evidence that circumcision was protective against HIV and other STIs among MSM.21
The absence of clear evidence of a protective effect of circumcision against HIV and other STI in MSM may be a consequence of the failure of past studies to stratify analyses by subjects' anal sexual role. MSM can practice exclusively insertive anal sex, exclusively receptive anal sex, both insertive and receptive anal sex, or not engage in anal intercourse at all. Because receptive anal sex poses the greatest per-contact risk for HIV acquisition22 and circumcision would not affect the risk through being exposed to HIV as a receptive partner, we hypothesized that the protective effects of circumcision would vary based on a man's anal sexual behavior. Men who are primarily the insertive partner during anal intercourse would enjoy the greatest protection from circumcision, men who are versatile during anal intercourse would receive some protection, and those who are primarily the receptive partner would receive little or no protection. However, such an effect would likely be diminished if men's anal sexual behavior varied over time. In this study, we tested the hypothesis that the protective impact of circumcision was limited to men who only practice insertive anal sex, using data from MSM tested for HIV and other STIs in Seattle, WA.
MATERIALS AND METHODS
The study population was comprised of men ≥18 years of age who attended the Public Health–Seattle and King County (PHSKC) STD Clinic between October 1, 2001 and May 12, 2006 and reported having anal intercourse with another man in the preceding year. Clinicians asked patients if they had engaged in insertive and/or receptive anal intercourse in the previous 2 and 12 months and entered the responses on a structured medical record, using a check-box system. Clinic staff entered information from the record into an electronic database daily. Patients' circumcision status was based on clinical examination by clinicians and recorded in the STD clinic medical record. We assessed the agreement in how clinicians assessed patients' circumcision status, using data from men seen more than once during the study period, comparing clinician assessments from their first 2 visits and calculating Cohen κ, as described by Armstrong et al.23 If men did not have any circumcision status recorded during any clinic visit during the study period or if clinicians inconsistently recorded a man's circumcision status (e.g., he was assessed as “circumcised” at one visit, and he was assessed as “uncircumcised” at a subsequent visit), we excluded that man's data from further analyses. Because we observed a high level of agreement in how clinicians assessed circumcision (percent agreement = 94%, κ = 0.77), if clinicians neglected to record circumcision status during a visit on a man whose status they had assessed during a prior or subsequent visit, we assumed that status was unchanged and included data from prior visits. Analyses evaluating the association of circumcision with HIV/STI and the stability of anal sexual behavior utilized all HIV/STI testing visits for which data were otherwise complete during the study period. The University of Washington Human Subjects Division approved the study.
We defined men as being newly diagnosed with HIV if they tested positive for HIV during their clinic visit and either reported testing HIV negative in the past year or reported were never having been tested for HIV. Clinicians at the Harborview STD clinic routinely offer HIV testing to all patients who report that they have not previously tested HIV-positive. All patients who agree to have their blood drawn are tested using a second generation HIV enzyme immunoassay (EIA) (Vironostika HIV-1 Microelisa System, bioMerieux, Marcy l'Etoile, France). Clinicians also offer MSM patients at high risk for HIV testing the Oraquick rapid HIV tests (Orasure Technologies, Bethlehem, PA) which were performed on oral fluids or blood specimens obtained by fingerstick. Since 2003, the PHSKC clinic has tested all MSM who agree to a blood draw, using a pooled HIV RNA test to detect early HIV infections. Between 2003 and 2005, our lab used the Procleix HIV-1 Discriminatory Assay (Gen-Probe Inc, San Diego, CA). Thereafter, the lab employed an independently-validated, quantitative, real-time polymerase chain reaction assay. Steckler et al. described pooled HIV RNA testing procedures previously.24 Data from anonymous HIV tests were not included in the current analysis.
We considered men “ever HIV-positive” if they tested positive during a clinic visit or if they reported a previous HIV-positive diagnosis. We defined men as being infected with HSV-2 based on serological testing using the Focus HSV-2 enzyme linked immunosorbent assay (ELISA) (HerpeSelect-2, Cypress, CA). We considered all HSV-2 ELISA test results with an index value of ≥3.5 to be positive. From October 2001 until October 2003, persons with Focus HSV-2 index values of 1.1 to 3.4 were regarded as HSV-2 positive if they were HSV-1 negative by Focus EIA; men who tested Focus HSV-1 positive using the EIA were tested using the University of Washington Western Blot (UW WB) and their HSV-2 status was defined based on the result of the UW WB. We have previously described this procedure which improves the specificity of the Focus HSV-2 ELISA.25 After October 15, 2003, all men with index values between 1.1 and 3.4, regardless of HSV-1 status, were considered to have indeterminate results and were tested using the UW WB to define their HSV-2 status. Urethral gonorrhea and chlamydial infection were diagnosed by culture, ligase chain reaction (Amplicor, Abbott Laboratories, Abbott Park, IL), or Aptima Combo 2 (Gen-Probe, San Diego, CA) during the course of the study period. Serological testing for syphilis was performed using the rapid plasma regain test, with confirmatory testing performed using the Treponema pallidum particle agglutination test (TPPA; Serodia–TP · PA; Fujirebio America, Malvern, PA). We defined rapid plasma regain titers greater than 1:8 to be indicative of prevalent infection of early syphilis.
We used chi-square tests to compare age (<25, 26–35, 36–45, 46–55, 56+ years), race/ethnicity (white, black, Native American, Asian and Pacific Islander, Hispanic, other), and anal sexual repertoire in the previous year (insertive, versatile, receptive) of men based on circumcision status at their first clinic visit, and used generalized estimating equations with robust standard error estimates to assess the association of circumcision with specific STD diagnoses (HSV-2, syphilis, urethral gonorrhea, urethral chlamydia) and HIV infection, using data from all visits to the clinic. We adjusted the associations for race/ethnicity (white/nonwhite) and age (<35 years/≥35 years). We also used generalized estimating equations to examine the association between anal sexual repertoire and race/ethnicity (white, black, Native American, Asian and Pacific Islander, Hispanic, other), age (as previously described), number of sex partners in the previous 12 months (1, 2–9, 10+), a previous HIV diagnosis, and a history of more than 1 visit to the clinic. We used a test of trend to examine how reported anal sexual repertoire changed among men with repeat evaluations during the study period. We performed all analyses using STATA version 10 (Stata Corporation, College Station, TX).
During the study period, 4749 unique MSM who reported anal intercourse in the previous year visited the PHSKC STD Clinic in Seattle for 8337 evaluations. Circumcision status was not assessed or was inconsistently noted for 729 men (15%), and anal sexual repertoire was incomplete for an additional 192 men (4%); we excluded these men from further analysis. Patients whose circumcision status was not noted were similar to those whose status was noted by race/ethnicity (P = 0.72) and reported anal sexual repertoire in the previous 12 months (P = 0.17), but were more likely to be aged ≥56 years (P <0.01). Clinicians evaluated the remaining 3828 men during 6924 visits; 3238 (85%) were circumcised and 590 (15%) were uncircumcised. Characteristics of the men at the first clinic visit are summarized on the basis of circumcision status in Table 1. White men were more likely to be circumcised than men of other races/ethnicities (P <0.01), and men over the age of 35 were more likely to be circumcised than those under age 35 (P <0.01). Circumcised and uncircumcised men were similar in terms of anal sexual behaviors (P = 0.24).
When examining all clinic visits, the proportion of men newly testing HIV-positive or with previously diagnosed HIV did not differ between circumcised and uncircumcised men when stratified by men's anal sexual role in the preceding year and adjusted for race/ethnicity and age (Table 2). Restricting the analysis to the 1595 visits of men who reported only insertive anal intercourse and who reported testing HIV negative in the preceding year, circumcised men were newly diagnosed with HIV at 2.0% of their visits and uncircumcised men tested HIV-positive at 2.8% of their clinic visits (aOR = 0.92; 95% CI: 0.17, 4.97). Similarly, we did not observe a significant association between circumcision status and HIV prevalence (newly diagnosed or previously diagnosed HIV-positive), serologic positivity for HSV-2 or syphilis or testing positive for urethral gonorrhea or chlamydial infections.
Table 3 presents data on the stability of men's anal sexual behavior over time. Evaluating men's cumulative sexual repertoire over multiple clinic visits, the proportion of men who reported being exclusively the insertive partner or exclusively the receptive partner declined with each additional visit, while the proportion classified as versatile increased (Ptrend <0.01). A total of 34.2% of men had a second clinic visit a median of 7 months following their initial clinic visit. Although at the first visit, 12.3% were the receptive partner, 63.4% were versatile, and 24.2% were the insertive partner, using data from the first 2 visits, we classified 6.0% of men as receptive, 79.5% as versatiles, and 14.4% as insertive. The trend towards increased versatility over time persisted when data from additional clinic visits were incorporated.
Older men and men of nonwhite race/ethnicity were significantly less likely to report being versatile in terms of their anal sexual behavior in the previous year (Table 4). A total of 1399 men for whom circumcision status data were available were seen in the clinic >1 time during the study period. Being seen for more than one clinic visit was associated with the type of anal sexual behavior; men who were versatile were more likely to have at least one subsequent visit compared to men who engaged in only receptive or only insertive anal sex (versatile OR: 1.39; 95% CI: 1.23, 1.57; only receptive anal sex OR: 0.73; 95% CI: 0.61, 0.88; only insertive anal sex OR: 0.78; 95% CI: 0.68, 0.90).
Using data collected over nearly 5 years in an urban STD clinic in Seattle, we found no association between lack of circumcision and the risk of infection with HIV, HSV-2, syphilis, and urethral gonorrhea or chlamydia, even when the analysis was restricted to MSM who practiced only insertive anal intercourse. Among MSM evaluated in the clinic multiple times over the study period, a small proportion (11% among those with 3 clinic visits) reported only engaging in insertive anal sex, perhaps providing a partial explanation for why we did not see a protective effect from circumcision.
Our results are consistent with a recent meta-analysis which found no association between circumcision status and HIV or other STIs among MSM.21 Since that study was published, a randomized trial evaluating an adenovirus Type 5 vector-based vaccine in a population comprised primarily of MSM found no association between circumcision and HIV acquisition and, among the trial's placebo recipients, the risk of HIV acquisition was somewhat more common in circumcised men.10 In contrast, Guanira et al. recently reported that among Peruvian MSM who practiced only insertive anal sex, circumcised men were at lower risk for HIV.11 Thus, while the preponderance of evidence available to date does not support the hypothesis that circumcision provides MSM with substantial protection from HIV, there is some heterogeneity across studies, perhaps suggesting the potential for a protective effect in some circumstances.
We also observed no association between lack of circumcision and HSV-2, syphilis, urethral gonorrhea, and urethral chlamydia infections. As with HIV, previous studies examining the association of these STIs with circumcision have yielded mixed results. Our findings are similar to results from one of the male circumcision randomized control trials in South Africa, which found no evidence of circumcision protecting against urethral chlamydia and gonorrhea among heterosexual men.18 A meta-analysis of 10 studies among heterosexual men found that circumcision was associated with a reduced risk of HSV-2 infection, a finding that is further supported by data from recent randomized trials of male circumcision.14,19,20 However, in contrast a population-based household survey found no association of between HSV-2 seropositivity and male circumcision.15 Similarly, population-based studies have not found an association between lack of circumcision and gonorrhea or syphilis infections among heterosexual men,12,13 although other studies have observed such associations or a trend toward a higher risk of gonorrhea and syphilis among uncircumcised men.16,17
Our data may provide some insight into why circumcision may offer relatively little protection to MSM against HIV even as it seems to be highly protective for heterosexual men. At least among MSM serially evaluated in our STD clinic, few consistently practiced only insertive anal sex. Among men evaluated only once, 24% reported having only insertive anal sex, but among men evaluated at least 3 times, only 11% reported being exclusively insertive. Receptive anal intercourse poses a greater per-contact risk for HIV acquisition than insertive anal intercourse,26 and thus most men who engage in both practices probably face a greater risk of becoming HIV infected from their receptive anal sex exposures. Circumcision consequently offers men practicing both insertive and receptive anal sex relatively little overall protection, even if it does somewhat diminish the risk of acquiring HIV through insertive anal sex. However, there seems to be substantial heterogeneity in anal sexual behavior in different culturally defined groups within the United States and between US MSM and MSM in other parts of the world. In our study, nonwhite MSM were less likely to be versatile, and a study conducted among MSM in Peru found that only 16% of MSM were versatile based on men's behavior in the preceding 3 months.27 The variability in MSM' sexual repertoire highlights the need to evaluate the protective efficacy of circumcision in different populations of MSM when attempting to define what role, if any, circumcision may play as an HIV prevention strategy in MSM.
Our analysis is subject to limitations, including missing data on circumcision and sexual behavior and the relatively small number of cases of men diagnosed with early syphilis. While most men in the US are circumcised as infants, a small number of men may have been circumcised in adulthood after acquiring HIV or the other infections. The STD clinic provides treatment on a sliding-scale, and the STD clinic population may not be representative of the MSM population as a whole. Finally, there were several limitations to how we classified men based on their anal sexual behavior. Data are not available for the frequency of insertive and receptive sexual acts and the categories of insertive, versatile, and receptive does not provide information about the relative frequency of these behaviors. Some men may have perceived reporting specific anal sex acts as socially desirable, potentially leading us to misclassify their anal sex role. Also, insofar as versatile men seem to return to the clinic more often than men with more limited sexual repertoires, we may have overestimated the proportion of men who are versatile in our analysis by incorporating multiple visits by the same man. The analysis also has several strengths, including the large size of the population studied, our ability to define circumcision status through clinical examination as opposed to subject self-report, the comprehensive STI testing performed on the study population, and the large number of MSM serologically tested for HIV and HSV-2.
In summary, we did not observe that circumcision was protective against HIV or other STI in this population of MSM presenting to an urban US STD clinic, most of whom were already circumcised, and who reported a versatile anal sexual repertoire. While our findings suggest that male circumcision would not be likely to have a significant impact on HIV or STI acquisition among MSM in Seattle, it remains uncertain whether circumcision could be beneficial in other parts of the world where circumcision is much less common and a higher proportion of MSM may be consistently insertive during anal sex.
1. Auvert B, Taljaard D, Lagarde E, et al. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 Trial. PLoS Med 2005; 2:e298.
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; 369:643–656.
3. Gray RH, Kigozi G, Serwadda D, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: A randomized trial. Lancet 2007; 369:657–666.
4. Grulich AE, Hendry O, Clark E, et al. Circumcision and male-to-male sexual transmission of HIV. AIDS 2001; 15:1188–1189.
5. Templeton DJ, Jin F, Prestage GP, et al. Circumcision status and risk of HIV seroconversion in the HIM cohort of homosexual men in Sydney. In: Fourth International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; 2007; Sydney, Australia. Abstract WEAC103.
6. Mor A, Kent CK, Kohn RP, et al. Declining rates in male circumcision amidst increasing evidence of its public health benefit. PLoS ONE 2007; 2:e861.
7. Buchbinder SP, Vittinghoff E, Heagerty PJ, et al. Sexual risk, nitrite inhalant use, and lack of circumcision associated with HIV seroconversion in men who have sex with men in the United States. J Acquir Immune Defic Syndr 2005; 39:82–89.
8. Millett GA, Ding H, Lauby J, et al. Circumcision status and HIV infection among Black and Latino men who have sex with men in 3 US cities. J Acquir Immune Defic Syndr 2007; 46:643–650.
9. Kreiss JK, Hopkins SG. The association between circumcision status and human immunodeficiency virus infection among homosexual men. J Infect Dis 1993; 168:1404–1408.
10. Buchbinder SP, Mehrotra DV, Duerr A, et al; Step Study Protocol Team. Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): A double-blind, randomized, placebo-controlled, test-of-concept trial. Lancet. 2008; 372:1881–1893.
11. Guanira J, Lama JR, Goicochea P, et al. How willing are gay men to “cut off” the epidemic? Circumcision among MSM in the Andean region. In: Fourth International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; 2007; Sydney. Abstract WEAC102.
12. Laumann EO, Masi CM, Zuckerman EW. Circumcision in the United States. Prevalence, prophylactic effects, and sexual practice. JAMA 1997; 277:1052–1057.
13. Dave SS, Fenton KA, Mercer CH, et al. Male circumcision in Britain: Findings from a national probability sample survey. Sex Transm Infect 2003; 79:499–500.
14. Weiss HA, Thomas SL, Munabi SK, et al. Male circumcision and risk of syphilis, chancroid, and genital herpes: A systematic review and meta-analysis. Sex Transm Infect 2006; 82:101–109; discussion 110.
15. Xu F, Markowitz LE, Sternberg MR, Aral SO. Prevalence of circumcision and herpes simplex virus type 2 infection in men in the United States: The National Health and Nutrition Examination Survey (NHANES), 1999–2004. Sex Transm Dis 2007; 34:479–484.
16. Diseker RA III, Peterman TA, Kamb ML, et al. Circumcision and STD in the United States: Cross sectional and cohort analyses. Sex Transm Infect 2000; 76:474–479.
17. Cook LS, Koutsky LA, Holmes KK. Circumcision and sexually transmitted diseases. Am J Public Health 1994; 84:197–201.
18. Sobngwi-Tambekou J, Taljaard D, Nieuwoudt M, et al. Male circumcision and Neisseria gonorrhoeae
, Chlamydia trachomatis
, and Trichomonas vaginalis
: Observations in the aftermath of a randomized controlled trial for HIV prevention. Sex Transm Infect. 2009; 85:116–120.
19. 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; 360:1298–1309.
20. Sobngwi-Tambekou J, Taljaard D, Lissouba P, et al. Effect of HSV-2 serostatus on acquisition of HIV by young men: Results of a longitudinal study in Orange Farm, South Africa. J Infect Dis 2009; 199:958–964.
21. Millett GA, Flores SA, Marks G, et al. Circumcision status and risk of HIV and sexually transmitted infections among men who have sex with men: A meta-analysis. JAMA. 2008; 300:1674–1684.
22. Vittinghoff E, Douglas J, Judson F, et al. Per-contact risk of human immunodeficiency virus transmission between male sexual partners. Am J Epidemiol 1999; 150:306–311.
23. Armstrong BK, White E, Saracci R. Principles of Exposure Measurement in Epidemiology. Monographs on Epidemiology and Biostatistics, Vol 21. New York: Oxford Press, 1992:104–109.
24. Stekler JD, Swenson PD, Coombs RW, et al. HIV testing in a high-incidence population: Is antibody testing alone good enough? Clin Infect Dis 2009; 49:444–453.
25. Golden MR, Ashley-Morrow R, Swenson P, et al. Herpes simplex virus type 2 (HSV-2) Western blot confirmatory testing among men testing positive for HSV-2 using the focus enzyme-linked immunosorbent assay in a sexually transmitted disease clinic. Sex Transm Dis 2005; 32:771–777.
26. Vittinghoff E, Douglas J, Judson F, et al. Per-contact risk of human immunodeficiency virus transmission between male sexual partners. Am J Epidemiol 1999; 150:306–311.
27. Goodreau SM, Peinado J, Goicochea P, et al. Role versatility among men who have sex with men in urban Peru. J Sex Res 2007; 44:233–239.