Portal de Periódicos da CAPES

Preventing HIV-1: lessons from Mwanza and Rakai

1999; Elsevier BV; Volume: 353; Issue: 9163 Linguagem: Inglês

10.1016/s0140-6736(05)75132-1

1474-547X

Maria J. Wawer, Nelson Sewankambo, David Serwadda, Ronald H. Gray,

HIV/AIDS Research and Interventions

Authors' reply Sir—Angus Nicoll and colleagues suggest that the divergent results of the Mwanza and Rakai STD control trials for AIDS prevention were due to the interventions used: syndromic treatment in the former study and mass treatment in the latter. It is unlikely that a syndromic approach would have had substantial effects on HIV-1 in Rakai, where most HIV-1 transmission occurred independently of STD symptoms or laboratory diagnoses. In addition, the sensitivity and specificity of symptoms of STD screening were poor1Paxton LA Sewankambo N Gray R et al.Asymptomatic non-ulcerative genital tract infections in a rural Ugandan population.Sex Transm Infect. 1998; 74: 421-425Crossref PubMed Scopus (39) Google Scholar (as has also been reported in Mwanza).2Grosskurth H Mayaud P Mosha F et al.Asymptomatic gonorrhea and chlamydia infection in rural Tanzanian men.BMJ. 1996; 312: 277-280Crossref PubMed Scopus (66) Google Scholar Finally, many symptoms in the Rakai population were not due to treatable STDs: 42% of genital ulcers were herpes simplex virus-2, and only 7% were identified as syphilis or chancroid; over 50% of women had bacterial vaginosis, a disease which is not amenable to cure and is associated with risk of HIV-1.3Taha T Hoover DR Dallabetta GA et al.Bacterial vaginosis and disturbances of vaginal flora: association with increased acquisition.AIDS. 1998; 12: 1699-1706Crossref PubMed Scopus (523) Google Scholar The hypothesis that the Mwanza trial achieved success by reducing symptom duration is attractive, but data on duration were not reported in that study. Finally, reintroduction of STDs in Rakai may have diluted an effect; however, in a substudy of pregnant women in whom STDs were significantly reduced in the intervention compared with the control group, we observed no reduction in HIV-1 incidence. Gunnar Kvåle suggests that the Rakai results may have been due to lack of comparability between groups, or to the services offered to the control population. Absolute differences between groups in the distribution of key variables were small, and were adjusted for in analyses. Condom use was low in both groups, and only 16% of patients with symptoms in the control group reported seeking effective treatment. Thus, ethically mandated services cannot explain the negative results in the overall population or in all subgroups. Francine Matthys and Marleen Boelaert raise issues of drug resistance and costs. Medications were provided as single, directly observed treatment, to keep inadequate compliance, a main cause of selective resistance, to a minimum. Gonorrhoea sensitivity testing identified no resistant strains. The drug costs estimated by Matthys and Boelaert are excessively high: metronidazole, ciprofloxacin, and penicillin together cost under US$1 in Kampala pharmacies. These drugs are also included in the Uganda Ministry of Health standard drug regimen, where their combined cost is even lower. Azithromycin is being used in mass treatment trachoma campaigns in developing countries, and prices are falling. Matthys and Boelaert do not distinguish between efficacy trials that show proof of concept and operations research that test programmatic strategies. Efficacy trials identify which therapeutic or preventive measures have desired health effects and warrant additional feasibility research; such trials are crucial for sound health policy and resource allocation. Matthys and Boelaert also raise ethical questions. The Rakai trial was conducted by the Uganda Ministry of Health/Uganda Virus Research Institute, and researchers at Makerere, Columbia, and Johns Hopkins Universities. The trial was approved by human subjects boards in all collaborating instutions and was fully explained to individuals and communities. Participation was completely voluntary. As discussed by Mbidde,4Mbidde E Bioethics and local circumstances.Science. 1998; 279: 155Crossref PubMed Scopus (20) Google Scholar African researchers and policy makers are fully aware of ethical issues and of the need to conduct studies that will provide data appropriate to local circumstances. Our results from Rakai, including the beneficial effects of STD mass treatment in pregnancy on maternal and infant health, are under consideration by the Uganda Ministry of Health to determine policy implications. In accordance with previous studies, Rakai showed associations between STDs and HIV-1 infection at the individual level. We agree that STD control is important to reduce the risk of HIV-1 for individuals, provide reproductive health benefits, and in some settings potentially to reduce population incidence of HIV-1 infection. In this context, it should be noted that neither Rakai nor Mwanza achieved optimum STD results. We are currently collaborating with Mwanza researchers to model better approaches to STD control; among the models being looked at are combined syndromic and mass treatment approaches. Preventing HIV-1: lessons from Mwanza and RakaiObservational studies early in the HIV-1 epidemic showed the strong relation between sexually transmitted disease (STDs) and HIV-1 infection.1 By causing inflammation, ulceration, or both, in the genital tract, STDs increase the shedding of HIV-1 in men and women and may also increase susceptibility to HIV-1 infection. Full-Text PDF Preventing HIV-1: lessons from Mwanza and RakaiIn their Feb 13 commentary, Penny Hitchcock and Lieve Fransen1 discuss the seeming discrepancy between the results of the Mwanza study2 which showed an effect on HIV-incidence of STD treatment, and the findings by Maria Wawer and colleagues3 from Rakai where no effect on HIV-1 infection of such treatment was observed. They suggest a possible explanation relating to the difference in the maturity of the infection, the Mwanza results being an example of a short-term impact of STD prevention and control in an immature epidemic (HIV prevalence 4%). Full-Text PDF Preventing HIV-1: lessons from Mwanza and RakaiPenny Hitchcock and Lieve Fransen1 provide a plausible epidemiological explanation for the strikingly negative results of Maria Wawer and colleagues' STD prevention trial2 on HIV-1 incidence rates in Rakai Uganda. Nevertheless, Hitchcock and Fransen go on to recommend that STD prevention programmes should be equally implemented in similar contexts with mature HIV-1 epidemics. They are of course right, for Wawer's negative result in a community intervention trial will not withhold us from preventing and treating STDs. Full-Text PDF