Sexually Transmitted Diseases:
Effect of Trichomoniasis Therapy on Genital HIV Viral Burden Among African Women
Anderson, Brenna L. MD, MSc*; Firnhaber, Cynthia MS, MD†,‡; Liu, Tao PhD§; Swarts, Avril MSc†; Siminya, Maureen PHC‡; Ingersoll, Jessica MS, MP(ASCP)¶; Caliendo, Angela M. MD, PhD¶; Cu-Uvin, Susan MD*
From the *Department of Obstetrics and Gynecology, Women & Infants Hospital/Alpert Medical School at Brown University, Providence, RI; †Clinical HIV Research Unit, Faculty of Health Science Center, Department of Medicine, University of Witwatersrand, Johannesburg, South Africa; ‡Right to Care, Johannesburg, South Africa; §Department of Biostatistics, Brown University, Providence, RI; ¶Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA
Conflicts of Interest and Sources of Funding: This research was funded in part by a 2007 developmental grant from the Lifespan/Tufts/Brown Center for AIDS Research. The project described was supported by Grant Number P30AI042853 from the National Institute of Allergy and Infectious Diseases (NIAID). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAID or the National Institute of Health. The work was also partially supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Women's Reproductive Health Research Scholar program, K12 HD050108 and NICHD K23 HD062340–01as well as National Institute of Allergy and Infectious Diseases K24 AI066884. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAID, NICHD, or the National Institute of Health.
Correspondence: Brenna Anderson, MD, Women and Infants Hospital, 101 Dudley Street, Providence, RI 02905. E-mail: firstname.lastname@example.org.
Received for publication, November 28, 2011, and accepted March 8, 2012.
Background: Our objective was to test the hypothesis that treatment for trichomoniasis among HIV-infected women not taking antiretrovirals in South Africa would be associated with decreased HIV genital shedding.
Methods: HIV-infected women presenting for routine HIV care were screened for trichomoniasis using self-collected vaginal swabs with a rapid point-of-care immunochromatographic antigen test. Women testing positive were offered enrollment into a prospective cohort study, if they had documented HIV infection, were aged 18 to 50 years, and were not receiving antiretroviral therapy. Recent use of postexposure prophylaxis or antibiotic therapy, active genital ulcers, or systemic illness were exclusion criteria. Cervical swabs were collected for gonococcal and chlamydial testing, and those testing positive were excluded. Women were treated with directly observed oral therapy with 2 g of oral metronidazole. A follow-up visit was scheduled 1 month after therapy, and partner letters were provided. Paired cervical wicks and plasma were collected for viral load measurement.
Results: In all, 557 women were screened. Sixty tested positive for trichomoniasis, 10 subsequently met exclusion criteria, and 4 were lost to follow-up. Of 46 women evaluated at follow-up, 37 (80.4%) were cured. Plasma viral load was not significantly different after therapy (P = 0.93). Genital tract viral load decreased by 0.5 log10 (P < 0.01). The mean genital tract viral load (log10) decreased from 4.66 (<3.52–6.46) to 4.18 (<3.52–6.48) (P < 0.01) after therapy.
Conclusions: Screening and treatment of vaginal trichomoniasis decrease genital shedding of HIV among South African women not receiving antiretrovirals at 1 month after therapy.
Trichomonas vaginalis is the most common nonviral sexually transmitted infection worldwide.1–3 In general, reports of Trichomonas vaginalis prevalence in sub-Saharan Africa range from 11% to 34%,4–6 and in selected populations, the prevalence are as high as 74%.7,8 At least 16% of Zimbabwean women with trichomoniasis were asymptomatic on direct questioning, and some reports suggest that nearly half of women with trichomoniasis are asymptomatic.9,10 Despite the high prevalence of the infection and the frequency of asymptomatic infection, trichomoniasis is not routinely screened for in most settings. T. vaginalis infection is believed to enhance HIV transmission.9 An estimate of the annual number of new HIV infections in the United States attributable to trichomoniasis assumed a cofactor effect of 2- to 5-fold increased risk.11,12 Both HIV and trichomoniasis are associated with alterations in the vaginal inflammatory cytokine milieu.13–17 In a study conducted in Malawi, men with urethritis had higher HIV viral RNA concentrations in semen if infected with T. vaginalis than did control subjects with urethritis of unidentified cause.18
In a study of Kenyan women performed by Wang et al, there was a significant decrease in HIV-1 RNA genital shedding after a 10-day course of metronidazole.19 Were screening for trichomoniasis and treatment with simple, inexpensive therapy to decrease genital HIV viral load, it could potentially translate into decreased transmission and acquisition of HIV. This is particularly true in resource-limited settings in which antiretroviral therapy (ART) use for HIV is not widespread. Work performed by the Rakai Study Group has clearly shown that there is an increase in HIV transmission risk with increased viral load. The odds of HIV transmission increase by 2.45 for each log10 increase in serum HIV viral load.20 It is also well established that there is good correlation between plasma and genital shedding, with a 2.6 odds increase in the likelihood of detecting genital shedding with each log10 increase in plasma viral load.21,22 And recent work shows that higher genital tract shedding does, in fact, increase sexual transmission risk.23 The objective of this study was to estimate whether single-dose metronidazole therapy for trichomoniasis given to HIV-infected women not taking ART would decrease genital HIV shedding.
MATERIALS AND METHODS
This was a prospective cohort study performed at Thembu Lethu clinic in Johannesburg, South Africa. The site holds a wellness clinic for HIV-infected women recently engaged in care. The study protocol was approved by the University of Witswatersrand Ethics Committee in South Africa and at the Institutional Review Board at Women & Infants Hospital of Rhode Island. Eligible women were aged 18 to 50 years, nonpregnant with CD4 cell counts >200/mm3, not receiving ART, willing to avoid use of intravaginal products during the approximately 1-month study period, and subject and partner willing to use condoms with sexual intercourse during the study period. Women were offered education about sexually transmitted infections and this study. Full informed consent was signed, if the woman was interested in participating in the study. They were given a brochure and instructed on method of self collection of a vaginal swab. The research assistant collected the swabs from the subject and performed a rapid immunochromatographic test for trichomoniasis (Genzyme Diagnostics, Cambridge, MA) while the subject waited. Those testing positive for trichomoniasis were offered enrollment into the cohort.
At enrollment, the subject was interviewed and underwent a pelvic/speculum examination to collect study specimens. A blood sample was collected for determination of plasma viral load. The speculum examination involved a routine method of specimen collection and collection of samples for polymerase chain reaction testing for Neisseria gonorrhoeae and Chlamydia trachomatis. Women with positive tests for gonococcal or chlamydial infection were excluded. Cervical viral load samples were collected with a wick (TearFlo strips, Sigma Pharmaceuticals, Monticello, IA), as was a dry slide for gram staining for the diagnosis of bacterial vaginosis via the method of Nugent.24 Nugent Gram stain score result accuracy was confirmed with >90% concordance with the reference laboratory of Sharon Hillier, PhD (Pittsburgh, PA).
Infected women were treated at enrollment on site with directly observed therapy with 2 g of oral metronidazole. The subjects were also given condoms and asked to use them for intercourse during the study period. Subjects were given partner letters to seek medical care, but uptake of partner therapy was not measured. The follow-up examination was performed 1 month later, with specimen collection performed in a manner identical to the first examination, as well as a follow-up rapid trichomoniasis test for confirmation of cure.
The cervical wick specimens were prepared for shipment by placing the strips in 9.0 mL of NASBA Lysis Buffer, and frozen at −20°C until the assays were run. The samples were shipped to Emory University where the genital viral load measurements were performed using bioMérieux Nuclisens EasyQ HIV-1 v1.1 (Durham, NC). The lower limit of detection of the assay was 3300 copies/mL. The plasma viral loads were drawn at the same visit and were conducted in the local South African reference laboratory using the Roche linear assay (Pretoria, South Africa), with a lower limit of detection of 400 copies/mL.
The cohort of study participants was characterized by summarizing the baseline demographic covariates using mean/median and range for continuous variables and count and percentage for categorical and count variables. Changes after trichomoniasis therapy from the baseline were compared using McNemar tests for categorical outcome variables, such as detectable viral load and contraception use, and paired t tests for continuous outcome variables, such as log10 transformed plasma/genital tract viral load. Two versions of plasma/genital tract viral loads were compared: (1) a dichotomized version as “detectable” and “undetectable,” and (2) a log10 transformed version to achieve normal approximation (undetectable viral loads were imputed by midvalues of the respective lower limits of detection before the log10 transformation). To adjust for the effect of potential confounders, we also conducted a regression analysis. Age, trichomoniasis cure, contraceptive use, and bacterial vaginosis (BV) were included in the regression as independent covariates to account for their potential confounding effects. Detectable viral load changes from baseline to follow-up were examined using conditional logistic regression model; log10 transformed viral loads were examined using linear mixed-effect model.25 In these 2 models, participant ID was used as a stratifying factor to account for repeated measures. In this article, we present the findings from the mixed-effect model, which are consistent with the finding from the conditional logistic regression model (in both the direction of association and significance). Two-sided P values of <0.05 were regarded as statistically significant. All statistical analyses were conducted using R (version 2.12, Vienna, Austria).
The planned study sample size (N = 57) was calculated based on the hypothesis that (1) trichomoniasis treatment by metronidazole would result in 1 log10 decrease in the genital shedding of HIV viral load at follow-up compared with the baseline; (2) genital viral load would have a standard deviation of 2 log10; (3) within-person correlation is 0.15; and (4) 5% loss to follow-up. This sample size would lend us a power of 0.80 for a 2-sided significance (α) of 0.05 based on a paired t test.
A total of 557 women were screened for vaginal trichomoniasis at the clinic. Of these, 60 (10.7%) (95% confidence interval [CI], 7.5–12.6) tested positive. All of these women agreed to enrollment and were provided with directly observed therapy with 2 g of oral metronidazole. Ten women were subsequently excluded because of meeting the prespecified exclusion criteria, whereas an additional 4 were lost to follow-up (Fig. 1). The majority of the study participants reported being single, and only 21/46 (46%) reported condom use (Table 1).
In this cohort of women not taking ART, mean log10 plasma viral load was unchanged among the participants after metronidazole administration (P = 0.93) (Table 2), and there were comparable proportions of women with detectable virus at baseline and after therapy (P = 0.99). Genital tract viral load decreased significantly (P < 0.01) and was less likely to be detectable after therapy (P = 0.04) (Fig. 2). The frequency of BV, defined as a Nugent score ≥7, was not statistically different after therapy.
To assess the independent effect of therapy on genital tract viral load and to assess for the effect of potential confounding variables, we conducted a regression analysis. The outcome variable was log transformed to approximate a normal distribution, and the regression model included random effects to account for the effects of repeated measures. Plasma viral load was the most significant predictor of genital tract viral load. However, the only other significant predictor of genital tract viral load in the model was trichomoniasis therapy. It remained significant even after controlling for the potential confounding variables, such as age, contraceptive use, and BV (Table 3). And, although follow-up after therapy was associated with a decrease in genital shedding of HIV, cure of trichomoniasis was not in this regression model.
We found that single-dose directly observed therapy with oral metronidazole for vaginal trichomoniasis at point-of-care decreases genital shedding of HIV-1 among South African women not taking ART. Plasma viral load was not affected by therapy, as anticipated because metronidazole has no known activity against HIV-1. This finding supports the theory that trichomoniasis causes an inflammatory state, predisposing to the increased genital shedding of HIV-1. The trichomoniasis cure rate at 1 month was >80% even in the absence of partner therapy or known reliable condom use. The reason that cure of trichomoniasis was not associated with decrease in HIV shedding is unknown. This may be related to small cell sizes in the regression model but could also indicate that metronidazole decreases genital shedding by some other mechanism than eradication of trichomonads. The anti-inflammatory properties of metronidazole could play a role. The implications for the results of this study are that there is a potential for a screening and treatment program to decrease HIV transmission. It will be important for trials of screening and treatment to not only document change in shedding of virus but also show how that translates into transmission risk.
There have been recent disappointing findings from large trials aimed at decreasing risk of HIV transmission by treating genital herpes after much promise from early studies. Nagot et al reported a randomized controlled trial aimed at reducing both plasma and genital HIV-1 viral load.26 The study showed a promising decrease in both genital and plasma viral load, and thus trials aimed at decreasing HIV acquisition and transmission were undertaken. Unfortunately, the subsequent high-quality randomized trial of HSV suppression therapy by infected partners did not reduce the risk of transmission to uninfected partners.27 Genital herpes and its therapy may not be comparable with trichomoniasis for a number of reasons. Of most concern for therapy, herpes is not something that can be cured. It requires twice-daily therapy in a patient that is likely to be asymptomatic. Trichomoniasis can be treated with a single dose of a medication that is inexpensive. Further, the genital tract viral load decrease that is required to decrease transmission is not clearly known. Female-to-male transmission increases with a hazard ratio of 2.38 (95% CI, 1.13–4.78), with each log10 increase of endocervical HIV-1 viral load.23 It may be that the 0.29 log10 decrease seen with herpes suppression26 reported in the Nagot et al study was not adequate to decrease transmission risk.
Those trials that have included therapy for trichomoniasis have been predominantly syndromic therapy or aimed at multiple sexually transmitted infections.28 Similar to the herpes trials, these may have resulted in inadequate improvement in genital viral shedding. Alternatively, they may have included sexually transmitted infections that do not respond to therapy by diminished shedding.
Our study has several limitations. It was a single-arm treatment trial in which each subject served as her own control without a separate control group. Because of the relatively small sample size, we used genital tract shedding as a surrogate marker for transmission risk. We cannot draw any causal relationship between therapy for vaginal trichomoniasis and risk of HIV transmission. Because of the local social environment, we were not permitted to provide therapy for partners, but only partner letters or measure partner uptake of therapy. Therefore, we had to rely on the subjects to both give the letters to their partners and abstain from intercourse or use condoms until their partner was treated. This may have lead to a number of reinfections.
Strengths of the study include its prospective nature and the use of directly observed therapy in an HIV treatment government clinic, eradicating any concern about compliance with medication that is problematic in many studies. Because there is no medication to take home and because the therapy was taken at the time of the study visit, compliance was 100%. Thus, we can exclude noncompliance as a reason for not seeing a larger decrease in genital shedding. As has been seen in other studies examining HIV-infected women, the prevalence of BV was high at 43%. Another strength of the study was our ability to control for BV, which has been previously shown to increase shedding of HIV.29–33
Point-of-care testing for vaginal trichomoniasis is currently somewhat expensive, but uptake of therapy was good in this general HIV clinic, with a prevalence that was about what was expected even with the use of self-collected samples. Directly observed therapy with metronidazole is both convenient and inexpensive. If future studies are able to demonstrate efficacy in decreasing HIV transmission risk with therapy for trichomoniasis, an approach using point-of-care testing and directly observed therapy should be considered.
1. Gerbase AC, Rowley JT, Heymann DH, et al.. Global prevalence and incidence estimates of selected curable STDs. Sex Transm Infect 1998; 74(suppl 1):S12–S16.
2. Johnston VJ, Mabey DC. Global epidemiology and control of Trichomonas vaginalis
. Curr Opin Infec Dis 2008; 21:56–64.
3. Bachmann LH, Hobbs MM, Seña AC, et al.. Trichomonas vaginalis
genital infections: Progress and challenges. Clin Infect Dis 2011; 53(suppl 3):S160–S172.
4. Watson-Jones D, Mugeye K, Mayaud P, et al.. High prevalence of trichomoniasis in rural men in Mwanza, Tanzania: Results from a population based study. Sex Transm Infect 2000; 76:355–362.
5. Wilkinson D, Ndovela N, Harrison A, et al.. Family planning services in developing countries: An opportunity to treat asymptomatic and unrecognised genital tract infections? Genitourin Med 1997; 73:558–560.
6. Mhlongo S, Magooa P, Muller EE, et al.. Etiology and STI/HIV coinfections among patients with urethral and vaginal discharge syndromes in South Africa. Sex Transm Dis 2010; 37:566–570.
7. Anorlu RI, Fagbenro Beyioku AF, Fagorala T, et al.. Prevalence of Trichomonas vaginalis
in patients with vaginal discharge in Lagos, Nigeria. Niger Postgrad Med J 2001; 8:183–186.
8. Buve A, Weiss HA, Laga M, et al.. The epidemiology of trichomoniasis in women in four African cities. AIDS 2001; 15 (suppl 4):S89–96.
9. Schwebke JR. Trichomoniasis in adolescents: A marker for the lack of a public health response to the epidemic of sexually transmitted diseases in the United States. J Infect Dis 2005; 192:2036–2038.
10. Swygard H, Sena AC, Hobbs MM, et al.. Trichomoniasis: Clinical manifestations, diagnosis and management. Sex Transm Infect 2004; 80:91–95.
11. Shafir SC, Sorvillo FJ, Smith L. Current issues and considerations regarding trichomoniasis and human immunodeficiency virus in African-Americans. Clin Microbiol Rev 2009; 22:37–45.
12. Chesson HW, Blandford JM, Pinkerton SD. Estimates of the annual number and cost of new HIV infections among women attributable to trichomoniasis in the United States. Sex Transm Dis 2004; 31:547–551.
13. Narimatsu R, Wolday D, Patterson BK. IL-8 increases transmission of HIV type 1 in cervical explant tissue. AIDS Res Hum Retroviruses 2005; 21:228–233.
14. Al-Harthi L, Kovacs A, Coombs RW, et al.. A menstrual cycle pattern for cytokine levels exists in HIV-positive women: Implication for HIV vaginal and plasma shedding. AIDS 2001; 15:1535–1543.
15. Crowley-Nowick PA, Ellenberg JH, Vermund SH, et al.. Cytokine profile in genital tract secretions from female adolescents: Impact of human immunodeficiency virus, human papillomavirus, and other sexually transmitted pathogens. J Infect Dis 2000; 181:939–945.
16. Olaitan A, Johnson MA, Reid WM, et al.. Changes to the cytokine microenvironment in the genital tract mucosa of HIV+ women. Clin Exp Immunol 1998; 112:100–104.
17. Mitchell C, Hitti J, Paul K, et al.. Cervicovaginal shedding of HIV type 1 is related to genital tract inflammation independent of changes in vaginal microbiota. AIDS Res Hum Retroviruses 2011; 27:35–39.
18. Price MA, Zimba Da, Hoffman IF, et al.. Addition of treatment for trichomoniasis to syndromic management of urethritis in Malawi: A randomized clinical trial. Sex Transm Dis 2003; 30:516–522.
19. Wang CC, McClelland RS, Reilly M, et al.. The effect of treatment of vaginal infections on shedding of human immunodeficiency virus type 1. J Infect Dis 2001; 183:1017–1022.
20. Quinn TC, Wawer MJ, Sewankambo N, et al.. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med 2000; 342:921–929.
21. Cu-Uvin S, Snyder B, Harwell JI, et al.. Association between paired plasma and cervicovaginal lavage fluid HIV-1 RNA levels during 36 months. J Acquir Immune Defic Syndr 2006; 42:584–587.
22. Anderson BL, Cu-Uvin S. Determinants of HIV shedding in the lower genital tract of women. Curr Infect Dis Rep 2008; 10:505–511.
23. Baeten JM, Kahle E, Lingappa JR, et al.. Genital HIV-1 RNA predicts risk of heterosexual HIV-1 transmission. Sci Transl Med 2011; 3:77ra29.
24. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation. J Clin Microbiol 1991; 29:297–301.
25. Diggle PJ, Heagerty P, Liang KY, et al.. Analysis of Longitudinal Data. 2nd ed. Oxford, England: Oxford University Press, 2002.
26. Nagot N, Ouedraogo A, Foulongne V, et al.. Reduction of HIV-1 RNA levels with therapy to suppress herpes simplex virus. N Engl J Med 2007; 356:790–799.
27. Celum C, Wald A, Lingappa JR, et al.. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362:427–439.
28. Hayes R, Watson-Jones D, Celum C, et al.. Treatment of sexually transmitted infections for HIV prevention: End of the road or new beginning? AIDS 2010; 24(suppl 4):S15–26.
29. Sha BE, Zariffard MR, Wang QJ, et al.. Female genital-tract HIV load correlates inversely with lactobacillus species but positively with bacterial vaginosis and mycoplasma hominis. J Infect Dis 2005; 191:25–32.
30. Cu-Uvin S, Hogan JW, Caliendo AM, et al.. Association between bacterial vaginosis and expression of human immunodeficiency virus type 1 RNA in the female genital tract. Clin Infect Dis 2001; 33:894–896.
31. Spinillo A, Debiaggi M, Zara F, et al.. Factors associated with nucleic acids related to human immunodeficiency virus type 1 in cervico-vaginal secretions. BJOG 2001; 108:634–641.
32. Cohn JA, Hashemi FB, Camarca M, et al.. HIV-inducing factor in cervicovaginal secretions is associated with bacterial vaginosis in HIV-1-infected women. J Acquir Immune Defic Syndr 2005; 39:340–346.
33. Coleman JS, Hitti J, Bukusi EA, et al.. Infectious correlates of HIV-1 shedding in the female upper and lower genital tracts. AIDS 2007; 21:755–759.
© Copyright 2012 American Sexually Transmitted Diseases Association
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Data is temporarily unavailable. Please try again soon.