R.S. McClelland

A prospective study of risk factors for herpes simplex virus type 2 acquisition among high-risk HIV-1 seronegative women in Kenya

Objectives: Several studies have demonstrated an association between herpes simplex virus type 2 (HSV-2) and HIV-1, but available data on risk factors for HSV-2 acquisition are limited. The objective of this analysis was to determine the incidence and risk factors for HSV-2 acquisition among HIV-1-seronegative female sex workers in Kenya. Methods: Between February 1993 and December 2006, HIV-1-seronegative women attending a municipal sexually transmitted infection (STI) clinic were invited to enroll in a prospective cohort study. Screening for HIV-1 and STIs were done at monthly follow-up visits. Archived blood samples were tested for HSV-2. Results: Of 1527 HIV-1-seronegative women enrolled, 302 (20%) were HSV-2 seronegative at baseline of whom 297 had at least one follow-up visit. HSV-2 incidence was high (23 cases/100 person-years; 115 cases). In multivariate analysis, HSV-2 was significantly associated with more recent entry into sex work, workplace and higher number of sex partners per week. Condom use was protective, although this was statistically significant only for the intermediate strata (25–75% condom use; HR 0.43; p = 0.05). There were statistical trends for bacterial vaginosis to increase HSV-2 risk (HR 1.56; p = 0.07) and for oral contraceptive use to decrease risk (HR 0.50; p = 0.08). The 23% annual HSV-2 incidence in this study is among the highest reported anywhere in the world. Conclusions: Women were at increased risk if they had recently entered sex work, had a higher number of sex partners or worked in bars. HSV-2 risk reduction interventions are urgently needed among high-risk African women.

Screening for genital and anorectal sexually transmitted infections in HIV prevention trials in Africa.

Objectives: To demonstrate the value of routine, basic sexually transmitted infection (STI) screening at enrolment into an HIV-1 vaccine feasibility cohort study and to highlight the importance of soliciting a history of receptive anal intercourse (RAI) in adults identified as “high risk”. Methods: Routine STI screening was offered to adults at high risk of HIV-1 upon enrolment into a cohort study in preparation for HIV-1 vaccine trials. Risk behaviours and STI prevalence were summarised and the value of microscopy assessed. Associations between prevalent HIV-1 infection and RAI or prevalent STI were evaluated with multiple logistic regression. Results: Participants had a high burden of untreated STI. Symptom-directed management would have missed 67% of urethritis cases in men and 59% of cervicitis cases in women. RAI was reported by 36% of male and 18% of female participants. RAI was strongly associated with HIV-1 in men (adjusted odds ratio (aOR) 3.8; 95% CI 2.0 to 6.9) and independently associated with syphilis in women (aOR 12.9; 95% CI 3.4 to 48.7). Conclusions: High-risk adults recruited for HIV-1 prevention trials carry a high STI burden. Symptom-directed treatment may miss many cases and simple laboratory-based screening can be done with little cost. Risk assessment should include questions about anal intercourse and whether condoms were used. STI screening, including specific assessment for anorectal disease, should be offered in African research settings recruiting participants at high risk of HIV-1 acquisition.

Factors associated with herpes simplex virus type 2 incidence in a cohort of human immunodeficiency virus type 1-seronegative Kenyan men and women reporting high-risk sexual behavior.

Background: Herpes simplex virus type 2 (HSV-2) is an important cause of genital ulcers and can increase the risk for human immunodeficiency virus type 1 (HIV-1) transmission. Our objective was to determine the incidence and correlates of HSV-2 infection in HIV-1-seronegative Kenyan men reporting high-risk sexual behavior, compared with high-risk HIV-1-seronegative women in the same community. Methods: Cohort participants were screened for prevalent HIV-1 infection. HIV-1-uninfected participants had regularly scheduled follow-up visits, with HIV counseling and testing and collection of demographic and behavioral data. Archived blood samples were tested for HSV-2. Results: HSV-2 prevalence was 22.0% in men and 50.8% in women (P < 0.001). HSV-2 incidence in men was 9.0 per 100 person-years, and was associated with incident HIV-1 infection (adjusted incidence rate ratio [aIRR], 3.9; 95% confidence interval [CI], 1.3–12.4). Use of soap for genital washing was protective (aIRR, 0.3; 95% CI, 0.1–0.8). Receptive anal intercourse had a borderline association with HSV-2 acquisition in men (aIRR, 2.0; 95% CI, 1.0–4.1; P = 0.057), and weakened the association with incident HIV-1. Among women, HSV-2 incidence was 22.1 per 100 person-years (P < 0.001 compared with incidence in men), and was associated with incident HIV-1 infection (aIRR, 8.9; 95% CI, 3.6–21.8) and vaginal washing with soap (aIRR, 1.9; 95% CI, 1.0–3.4). Conclusions: HSV-2 incidence in these men and women is among the highest reported, and is associated with HIV-1 acquisition. Although vaginal washing with soap may increase HSV-2 risk in women, genital hygiene may be protective in men.

Endothelial activation biomarkers increase after HIV-1 acquisition: plasma vascular cell adhesion molecule-1 predicts disease progression.

Objective:We aimed to determine whether endothelial activation biomarkers increase after HIV-1 acquisition, and whether biomarker levels measured in chronic infection would predict disease progression and death in HIV-1 seroconverters. Design:HIV-1-seronegative Kenyan women were monitored monthly for seroconversion, and followed prospectively after HIV-1 acquisition. Methods:Plasma levels of angiopoietin-1 and angiopoietin-2 (ANG-1, ANG-2) and soluble vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and E-selectin were tested in stored samples from pre-infection, acute infection, and two chronic infection time points. We used nonparametric tests to compare biomarkers before and after HIV-1 acquisition, and Cox proportional-hazards regression to analyze associations with disease progression (CD4 <200 cells/&mgr;l, stage IV disease, or antiretroviral therapy initiation) or death. Results:Soluble ICAM-1 and VCAM-1 were elevated relative to baseline in all postinfection periods assessed (P < 0.0001). Soluble E-selectin and the ANG-2:ANG-1 ratio increased in acute infection (P = 0.0001), and ANG-1 decreased in chronic infection (P = 0.0004). Among 228 participants followed over 1028 person-years, 115 experienced disease progression or death. Plasma VCAM-1 levels measured during chronic infection were independently associated with time to HIV progression or death (adjusted hazard ratio 5.36, 95% confidence interval 1.99–14.44 per log10 increase), after adjustment for set point plasma viral load, age at infection, and soluble ICAM-1 levels. Conclusion:HIV-1 acquisition was associated with endothelial activation, with sustained elevations of soluble ICAM-1 and VCAM-1 postinfection. Soluble VCAM-1 may be an informative biomarker for predicting the risk of HIV-1 disease progression, morbidity, and mortality.

A prospective cohort study of the effect of depot medroxyprogesterone acetate on detection of plasma and cervical HIV-1 in women initiating and continuing antiretroviral therapy

Abstract:Depot medroxyprogesterone acetate (DMPA) use among HIV-1–infected women may increase transmission by increasing plasma and genital HIV-1 RNA shedding. We investigated associations between DMPA use and HIV-1 RNA in plasma and cervical secretions. One hundred two women initiated antiretroviral therapy, contributing 925 follow-up visits over a median of 34 months. Compared with visits with no hormonal contraception exposure, DMPA exposure did not increase detection of plasma (adjusted odds ratio: 0.81, 95% confidence interval: 0.47 to 1.39) or cervical HIV-1 RNA (adjusted odds ratio: 1.41, 95% confidence interval: 0.54 to 3.67). Our results suggest that DMPA is unlikely to increase infectivity in HIV-positive women who are adherent to effective antiretroviral therapy.

Mycoplasma Genitalium infection in sub-Saharan Africa: how big is the problem?

A lthough Mycoplasma genitalium has been clearly associated with nongonococcal urethritis in men and a handful of studies have reported associations with pelvic inflammatory disease and infertility among women, we know relatively little about its natural history. Unfortunately, the carefully designed prospective observational studies in untreated populations that are ideal to study natural history can no longer be done for M. genitalium; there is sufficient evidence suggesting that it causes harm and it can be treated, making such studies unethical. Vandepitte and colleagues have creatively circumvented this barrier by retrospectively testing stored specimens from a completed cohort study and report persistence, clearance, and recurrence rates for M. genitalium in Ugandan female sex workers in this issue of the journal. This doubles the number of studies on this topic among high-risk women in sub-Saharan African and significantly expands our knowledge of the natural history of this emerging pathogen. In some ways, the results of Vandepitte et al. are encouraging. Clearance of M. genitalium infection at the end of 1 year of follow-up was relatively good, and 12-month persistence rates in the Ugandan sex workers studied were substantially lower than those observed in Kenyan sex workers (7% vs. 16%). However, short-term clearance of this organism was modest at best: only 55% of women had cleared their infection after 3 months. Furthermore, a large proportion (40%) had recurrent M. genitalium infections. The low short-term clearance of M. genitalium and high recurrence rate would be cause for concern in any population of women but are doubly so in this high HIV prevalence setting. Evidence that M. genitalium, like many other sexually transmitted organisms, is associated with increased risk of HIV acquisition and transmission continues to build. A recent prospective study found that the risk of HIV acquisition was 2-fold higher among women infected with M. genitalium at the visit before their first positive HIV test result, linking M. genitalium to incident HIV infection for the first time. More prospective studies are needed to confirm this finding, as well as studies among HIV-positive individuals exploring the role of M. genitalium infection in increasing risk for HIV transmission. If M. genitalium, like other sexually transmitted infections (STIs), is a contributor to the spread of HIV, then efforts to reduce its duration of infection and recurrence may prove to be important. The relationship between M. genitalium infection and HIV viral shedding, a marker for infectivity, has been infrequently studied. Among US women infected with HIV-1, there was no relationship between detection of M. genitalium and vaginal shedding of HIV-1 RNA. However, M. genitalium organism burden may be important, and this was not evaluated. When organism burden was measured, cervical shedding of HIV-1 DNAwas more common among Kenyan women with high M. genitalium organism burden than among M. genitaliumYnegative women. In contrast, there was no association between low M. genitalium organism burden and HIV shedding. Whether persistent M. genitalium infections are characterized by high or low organism burden and the degree to which this impacts HIV shedding is unknown and merits further exploration. Not surprisingly, M. genitalium clearance was slower among HIV+ women with CD4 counts less than 350 cells/mm, suggesting that persistent detection of M. genitalium may be due to a failure of the host immune system to clear the organisms. Alternatively, detection of M. genitalium at consecutive visits may be due to rapid reinfection, either from new partners or from reexposure to untreated partners. As noted by the authors, genotypic analysis of the M. genitalium strains EDITORIAL

The post-trial effect of oral periodic presumptive treatment for vaginal infections on the incidence of bacterial vaginosis and Lactobacillus colonization

Background Bacterial vaginosis (BV) is a highly prevalent infection that frequently recurs following standard treatment. In a randomised controlled trial (RCT) of oral periodic presumptive treatment (PPT) to reduce vaginal infections among Kenyan women, we observed a decrease in BV and an increase in Lactobacillus colonisation among women randomised to receive 2 g metronidazole +150 mg fluconazole monthly for 12 months. After the trial, women were invited to continue follow-up in an open cohort study. These post-trial data were analysed to test the hypothesis that the treatment effect would persist in the absence of PPT. Methods Data were obtained from women who completed all 12 RCT visits and attended ≥1 cohort study visit within 120 days of their final RCT visit. We used Andersen-Gill proportional hazards models to estimate the post-trial effect of the intervention vs placebo on the incidence of BV by Gram stain (Nugent score ≥7) and Lactobacillus species by culture on Rogosa agar. Results The RCT enrolled 310 subjects (155 per arm), of whom 165 (83 active and 82 placebo) were included in this analysis. Included subjects were slightly older (median (IQR): 33 years (29–39) vs 30 years (26–35); p<0.001) and reported a longer duration of sex work (median (IQR): 6 years (2–11) vs 3 years (1–6); p<0.001) compared to those excluded. At the final RCT visit, which represented the baseline visit for this analysis, demographic and behavioural characteristics were similar by arm. The prevalence of BV at the final RCT visit was 16% in the active arm and 43% in the placebo arm (p<0.001). The post-trial incidence of BV was 260/100 person-years (p-yrs) in the active arm vs 358/100 p-yrs in the placebo arm (HR=0.76; 95% CI: 0.51% to 1.12%). The prevalence of Lactobacillus colonisation at the final RCT visit was 17% in the active arm and 18% in the placebo arm (p=0.81). The post-trial incidence of Lactobacillus colonisation was 180/100 p-yrs in the active arm vs 127/100 p-yrs in the placebo arm (HR=1.42; 95% CI: 0.85% to 2.71%). Conclusions Despite a decrease in BV and an increase in Lactobacillus colonisation during the RCT, the effect of PPT was not sustained during the 120 days following cessation of the intervention. New interventions that reduce BV recurrence and promote long-term Lactobacillus colonisation without the need for ongoing PPT or suppressive therapy are needed.

Risky Business: condom failures as experienced by female sex workers in Mombasa, Kenya.

Abstract Limited research exists about condom failure as experienced by female sex workers. We conducted a qualitative study to examine how female sex workers in Mombasa, Kenya contextualise and explain the occurrence of condom failure. In-depth, semi-structured interviews were conducted with thirty female sex workers to ascertain their condom failure experiences. We qualitatively analysed interview transcripts to determine how the women mitigate risk and cope with condom failure. Condom failure was not uncommon, but women mitigated the risk by learning about correct use, and by supplying and applying condoms themselves. Many female sex workers felt that men intentionally rupture condoms. Few women were aware of or felt empowered to prevent HIV, STIs, and pregnancy after condom failure. Interventions to equip female sex workers with strategies for minimising the risk of HIV, STIs, and pregnancy in the aftermath of a condom failure should be investigated.

Impact of vaginal washing on the vaginal microbiota

impact of contraceptives on the vaginal microbiome. We hypothesized that women initiating copper IUD would have increased BV and BV-associated microbes with use compared to women initiating and using hormonal contraceptive methods. METHODS: We collected vaginal swabs for Nugent score and quantitative PCR (qPCR) from asymptomatic, healthy women aged 18-35 in Harare, Zimbabwe (N1⁄4234) who we confirmed were in the follicular phase of menses and free of exogenous hormones by mass spectrometry. Contraception was initiated with an injectable (DMPA, norethisterone enanthate (NET), or medroxyprogesterone acetate and ethinyl estradiol (MPA/EE)), implant (LNGor ENGimplant), or copper IUD and repeat vaginal swabs were collected after 30, 90 and 180-days of continuous use. We extracted DNA from vaginal swabs for qPCR detection of Lactobacillus crispatus, L. jensenii, L. gasseri, L. vaginalis, L. iners, Gardnerella vaginalis, Atopobium vaginae and Megasphaera-like bacterium phylotype I. We used Cochran’s Q/Friedman’s test for within group differences and modified Poisson regression/random effects linear models for between group differences to compare marginal prevalence/mean difference in quantity (expressed as gene copies/swab) prior to and during contraceptive use. RESULTS: BV was detected at baseline by Nugent score in 73 (31%) of all women enrolled. BV prevalence increased in women initiating copper IUD from 27% at baseline, 34% at 30 days, 39% at 90 days and 44% at 180 days (p1⁄4.05 compared to baseline and p1⁄4.03 for comparison between groups). Women initiating hormonal methods had no change in BV prevalence over 180 days. The mean increase in Nugent score was 1.0 (95% CI 0.3-1.8, p<.01) in women using copper IUD. Although the frequency and density of beneficial lactobacilli did not change among IUD users over the four visits, there was an increase in the log concentration of G. vaginalis (4.7, 5.3, 5.9, 5.7; p<.01) and A. vaginae (3.1, 3.7, 4.8, 4.8; p1⁄4.01) at baseline and 30, 90 and 180 days after initiation. Among other contraceptive groups, women using Net-En had a decreased concentration of L. jensenii (p1⁄4 .01), and women using MPA/EE had decreased L. iners (p1⁄4.01) over 180 days. CONCLUSIONS: Copper IUD use may increase Nugent score and increase prevalence of BV. Use of most hormonal contraception does not shift vaginal microbiota.

Impact of vaginal bacteria on the presence and clinical presentation of vaginal yeast colonization

9 Validity of self-reported hormonal contraceptive use among women with and at risk for HIV M. Pyra, J. R. Lingappa, D. W. Erikson, S. W. Blue, R. Heffron, N. R. Mugo, K. Nanda, N. Davis, A. P. Kourtis, J. M. Baeten, for the Partners in Prevention HSV/HIV Transmission Study and Partners PrEP Study Teams Department of Epidemiology, Department of Global Health, Department of Medicine, Department of Pediatrics, University of Washington, Seattle, Washington; Endocrine Technologies Support Core, Oregon National Primate Research Center, Beaverton, OR, Department of Obstetrics & Gynaecology, University of Nairobi, Sexual, Reproductive, Adolescent and Child Health Research Program, Kenya Medical Research Institute, Nairobi, Kenya, FHI 360, Integrated Health Sciences, Research Triangle Park, North Carolina, Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia OBJECTIVES: Studies using self-reported hormonal contraception may have biased results due to misclassification. In this analysis, we determined the frequency of misreported hormonal contraceptive use. METHODS: 1,102 archival blood samples from 664 HIV-infected and uninfected women were evaluated using a novel high-performance liquid chromatography-heated electrospray ionizationtandem triple quadrupole mass spectrometry (LC-MS/MS) assay. The women were enrolled as members of serodiscordant couples in three multi-site studies in Africa (Couples Observational Study, Partners in Prevention HSV/HIV Transmission Study, and Partners PrEP Study) and later selected for as controls for analyses of HIV transmission. Women self-reported contraceptive methods at quarterly visits. Hormonal methods assayed were: injectables (medroxyprogesterone acetate, MPA or norethisterone, NET), oral contraceptives (levonorgestrel, LNG or ethinyl estradiol, EE2), and implants (LNG or etonogestrel, ENG), as these were the available agents for each method in these settings. The quantification of any expected exogenous hormone (versus no expected exogenous hormone) as a binary outcome was compared by HIV status using generalized estimating equation models to account for repeated observations, with a Poisson distribution and exchangeable correlation matrix. RESULTS: Injectable use was reported for 23% of samples, oral contraceptives for 6%, implants for 4%, and no method for 68%. Overall, 900/1,102 (82%) samples had hormone testing fully consistent with self-report. Only expected hormones were detected among 81% of injectable users, 27% of oral contraceptive users, and 90% of implant users. No hormones were detected in 17% of samples from women reporting injectable use, 62% with reported oral contraceptive use, and 8% with reported implant use. Among women who reported no hormonal contraceptive use, 14% had exogenous hormones detected; of these, 67% had MPA, 16% had NET, 7% had EE2, 19% had LNG, and 3% had ENG (samples may have had multiple hormones detected). Within each method, there was no statistical difference in concordance by HIV status. CONCLUSIONS: Eighty-two percent of all samples had accurate reporting, though both overand under-reporting was observed for each type of hormonal contraceptive. Among self-reported nonusers, 14% had detected hormones. The low accuracy for oral contraceptive users may reflect the placebo week. For studies of the effects of hormonal contraceptives, evaluations that verify self-reported use may reduce bias. 10 Impact of vaginal bacteria on the presence and clinical presentation of vaginal yeast colonization M. E. Eastment, J. E. Balkus, B. A. Richardson, S. Srinivasan, J. Kimani, O. Anzala, J. Schwebke, T. L. Fiedler, D. N. Fredricks, R. S. McClelland University of Washington, Department of Medicine, Global Health, Biostatistics, Epidemiology, Seattle, WA, Fred Hutchinson Cancer Research Center, Vaccine and Infectious Disease Division, Seattle, WA, University of Nairobi, Institute of Tropical and Infectious Diseases, Nairobi, Kenya, Kenya AIDS Vaccine Initiative, Nairobi, Kenya, University of Alabama, Division of Infectious Diseases, Birmingham, AL OBJECTIVES: This analysis of data from the Preventing Vaginal Infections trial tested whether key vaginal bacteria influence the presence, phenotypic expression, and clinical signs of vaginal yeast. METHODS: A cross-sectional analysis of enrollment data was performed. Taxon-specific quantitative polymerase chain reaction assays measured concentrations of bacterial taxa on vaginal swabs, modeled as undetectable (defined as <95 copies/swab) and detectable. The presence of yeast on wet mount microscopy and/or culture was compared in women with and without each type of bacteria. In women with yeast on wet mount, the relationship between detection of bacteria and the presence of pseudohyphae was examined. Finally, the association between pseudohyphae and clinical signs (vulvar edema, erythema, abnormal discharge) of vulvovaginal candidiasis (VVC) was explored. Logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs). Age and use of hormonal contraception were included as potential confounders. Effect modification by country (United States [US] or Kenya) was examined (stratified analyses were performed where p 0.1). RESULTS: In 221 women at enrollment, detection of several bacterial taxa was strongly associated with lower odds of vaginal yeast detection: Leptotrichia/Sneathia (n1⁄4157, adjusted OR (aOR) 0.32, 95%CI 0.17-0.59), Megasphaera (n1⁄4105, aOR 0.32, 95%CI 0.180.58), Atopobium vaginae (n1⁄4167, aOR 0.38, 95%CI 0.20-0.72), and BVAB2 (n1⁄4112, aOR 0.44, 95%CI 0.25-0.79). There was evidence of effect modification by country (interaction p1⁄40.08) only for Mageeibacillus indolicus and vaginal yeast detection (US, n1⁄452: aOR 1.27, 95%CI 0.37-4.37; Kenya, n1⁄4169: aOR 0.33, 95%CI 0.15-0.78). In 30 women with yeast on wet mount, there were higher odds of pseudohyphae detection when Lactobacillus jensenii was present, but the 95%CI was wide and not statistically significant (n1⁄446, aOR 5.05, 95%CI 0.61-41.58). Clinical signs of VVC were present in 7 of 11 (64%) women with yeast and pseudohyphae present compared to 4 of 11 (36%) women with yeast but absent pseudohyphae (OR 4.90, 95%CI 0.99-24.21). CONCLUSIONS: Vaginal bacteria may influence the likelihood of detection, phenotypic expression, and clinical signs of vaginal yeast. Future work should examine the mechanisms underlying these observations to inform the development of novel interventions to reduce the incidence of symptomatic VVC.