Dear Dr. Potterat and colleagues,

Thank you for responding to our manuscript. We have carefully reviewed your comments. Below, please find our responses to the questions raised.

The first comment raised concerns the fact that “sexual factors may have played a lesser role in observed HIV and syphilis prevalence’s than nonsexual factors.” The sexual transmission of sexually transmitted infections including HIV-1 is a significant risk factor of HIV-1 acquisition in female sex workers and men who have sex with men (MSM), and has been related to HIV-1 acquisition in preliminary analyses from our incidence cohort [1,2]. You note that the discrepancy between the relationship we found between syphilis and receptive anal intercourse (RAI) and between prevalent HIV-1 and recent RAI among the women is a “red flag.” Please note that our point estimate for the odds of HIV-1 infection among women admitting recent RAI is 1.2, and has a 95% confidence interval compatible with odds ranging from 0.5 to 2.5. The association between prevalent syphilis on enrolment and recent RAI was based on only 11 cases in women, and the importance of this finding should not be exaggerated.

Your suggestion that the treponemal disease we have diagnosed in our female sex workers is not syphilis, but rather another treponemal species, is intriguing but very unlikely. Pinta is limited to the Americas, and endemic syphilis (bejel) is not found in Kenya [3]. Yaws is very uncommon in Kenya [3], and we have seen none of the chronic skin or bone lesions typical of this infection in our clinic population. Other spirochetal illnesses that can lead to positive nontreponemal and treponemal tests (e.g. relapsing fever, rat-bite fever) are also uncommon and were unsuspected in the clinical context [4]. A non-specific test such as the RPR, followed by a specific treponemal test (TPHA) is the commonly accepted means of diagnosing syphilis [5]. At the same time, syphilis is the most likely diagnosis in these sexually active young women [6].

Your remark that we did not assess nonsexual (blood) exposures is true, since the focus of this article was on screening for sexually transmitted genital and anorectal infections. While some HIV infections we diagnosed at enrolment into our study population may be due to unsafe injections, including injection drug use, the prevalence of injection drug use in our population is only 1.4% among MSM [2] and was not reported among women. In a prevalence study, history of any medical injection is not useful because lifetime exposure is very common. Having received a medical injection in the 3 months preceding enrolment was reported by an equal proportion of HIV negative and positive women (35 vs. 36%). We have included data collection on both injection drug use and a number of other non-sexual exposures (medical injections, blood transfusion, traditional practices) in our ongoing study of incident HIV-1 infections in this cohort, and hope that your curiosity regarding this factor will be satisfied in an upcoming publication.

The second remark concerns the fact that “a strong association between anal sex and prostitution might mask the association between anal sex and prevalent HIV-1 in female participants in our cohort”. It is correct that the majority of women (89%) reporting recent RAI, identified themselves as sex workers. We have also included this in our paper in section ‘results’, in the paragraph on RAI. Please note that table 4 presenting associations between prevalent HIV-1 and RAI are adjusted for age, transactional sex, partner numbers, and unprotected sex.

Finally, a remark was made on the fact that “unprotected receptive anal intercourse is probably not confined to high-risk persons and that broader community prevention messages might more usefully fit overall HIV prevention objectives.” We agree on the importance of addressing unprotected (receptive) anal intercourse as a potential risk factor for HIV-1 transmission. We did not mean to imply that this was not important on a population level, but meant to highlight the urgency of addressing this risk in a targeted setting such as ours.

We trust these answers have addressed the concerns you have raised.

Sincerely,

Marlous Grijsen, MD,
Susan Graham, MD MPH,
Eduard Sanders, MD PhD.

References

1. Grijsen ML, Graham SM, Mwangome M, et al. Screening for genital and anorectal sexually transmitted infections in HIV prevention trials in Africa. Sex Transm Inf (doi: 10.1136/sti2007.028852)

2. Sanders EJ, Graham SM, Okuku HS, et al. HIV-1 infection in high risk men who have sex with men in Mombasa, Kenya. AIDS 2007;21: 2513-20.

3. Meheus A, Antal GM. The endemic treponematoses: not yet eradicated. World Health Stat Q. 1992;45(2-3): 228-37.

4. Mandell GL, Bennett JE, Dolin R. Principles and practice of infectious diseases. 6th edn. Philadelphia, Pa.: Elsevier Churchill Livingstone; 2005.

5. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2006. MMWR 2006;55(RR-11): 1-94.

6. Holmes KK, Sparling PF, Stamm WE, et al. Sexually transmitted diseases. 4th edn. New York: the McGraw-Hill Companies; 2008.

Dear Editor,

The report by Grijsen and colleagues documenting the high frequency of unprotected receptive anal intercourse (RAI) in young Kenyans at high risk for HIV infection (1) is a welcome contribution to the small but growing number of studies investigating RAI as a specific risk for HIV in sub-Saharan Africa (2-7). Their study, however, presents us with yet another anomaly unlikely to be resolved by the assessment of sexual risk factors alone (8). Although the authors found that “RAI was strongly associated with HIV-1 in men (adjusted odds ratio = 3.8)", they also reported that among women, RAI was not associated with prevalent HIV infection, but that those practicing RAI were much more likely to have syphilis (adjusted odds ratio 12.9). Puzzled, the authors note: “It is not clear why this difference was found…” None of the possible reasons they propose for this anomaly includes nonsexual HIV transmission. That these women were 10 times more likely to have serological markers of HIV (a sexually transmissible infection) than of current or past syphilis (a sexually transmitted) infection should be viewed as a red flag, even considering their nonspecific diagnostic criteria for syphilis (classification based on qualitative rapid plasma reagin test and the Treponema pallidum haemagglutination assay, neither of which rules out nonsexually transmitted treponematoses). The magnitude of the difference between HIV and “syphilis” markers alone suggests that sexual factors may have played a lesser role in observed HIV prevalence than nonsexual ones.

Because the authors apparently did not also assess nonsexual (blood) exposures, this possibility cannot be explored with their data -- a frustratingly common shortcoming in epidemiologic studies conducted in Africa (9). In addition, a strong association between anal sex and prostitution might mask the association between anal sex and HIV in their women participants. Thus, given a strong association between RAI and prostitution, it is important to report the bivariate relationships among all predictors and their relationship with prevalent HIV infection. Lastly, Grijsen and colleagues stress the importance of prevention messages about the dangers of unprotected RAI to those high- risk persons reporting it. Yet because RAI is probably not confined to “high-risk” persons (2), broader community prevention messages might more usefully fit overall HIV prevention objectives. Anal intercourse is common in sub- Saharan Africa populations (2-7) and is often perceived as involving no risk for HIV transmission (4,7).

John J. Potterat, BA, Colorado Springs, CO, USA
Stuart Brody PhD, University of the West of Scotland, UK
Devon D. Brewer, PhD, Interdisciplinary Scientific Research, Seattle, WA, USA
Stephen Q. Muth, BA, Colorado Springs, CO, USA

References

1. Grijsen MI, Graham SM, Mwangome M, et al. Screening for genital and anorectal sexually transmitted infections in HIV prevention trials in Africa. Sex Transm Infect (doi: 10.1136/sti2007.028852)

2. Brody S, Potterat JJ. Assessing the role of anal intercourse in the epidemiology of AIDS in Africa. Int J STD AIDS 2003; 14: 431-436.

3. Lane T, Pettifor A, Pascoe S, Fiamma A, Rees H. Heterosexual anal intercourse increases risk of HIV infection among young South African men. AIDS 2006; 20: 123-125.

4. Ramjee G, Gouws E, Andrews A, Myer L, Weber AE. The acceptability of a vaginal microbicide among South African men. Int Fam Plan Persp 2001, 27: 164-170.

5. Simbayi LC, Kalichman SC, Jooste S, Cherry C, Mfecane S, Cain D. Risk factors for HIV-AIDS among youth in Cape Town, South Africa. AIDS Behav 2005; 9: 53-61.

6. Mwakagile D, Mmari E, Makwaya C, et al. Sexual behaviour among youths at high risk for HIV-1 infection in Dar es Salaam, Tanzania. Sex Transm Infect 2001; 77: 255-259.

7. Stadler JJ, Delany S, Matambo M. Sexual coercion and sexual desire: ambivalent meanings of heterosexual anal sex in Soweto, South Africa. AIDS Care 2007; 19: 1189-1193.

8. Brewer DD, Brody S, Drucker E, et al. Mounting anomalies in the epidemiology of AIDS in Africa: cry the beloved paradigm. Int J STD AIDS 2003; 14: 144-147.

9. Gisselquist D, Potterat JJ, Brody S, Vachon F. Let it be sexual: how health care transmission of AIDS in Africa was ignored. Int J STD AIDS 2003; 14: 148-161.