From the *Department of Epidemiology, School of Public Health and Tropical Medicine, Tulane University, New Orleans, LA; †Health Sciences Center–Department of Medicine, Section of Infectious Diseases, Louisiana State University, Baton Rouge, LA; ‡Department of Medicine, University of Mississippi Medical Center, Jackson, MS; and §Baylor College of Medicine, Houston, TX
Conflict of interest: None declared.
Correspondence: Patricia Kissinger, PhD, Department of Epidemiology, School of Public Health and Tropic Medicine, Tulane University, SL-18, 1440 Canal St, New Orleans, LA 70012. E-mail: firstname.lastname@example.org.
Received for publication February 28, 2013, and accepted March 20, 2013.
High repeat infection rates (8%–20%1–6) among women receiving 2-g dose of metronidazole (MTZ) for the treatment of Trichomonas vaginalis (TV) have brought into question the effectiveness of single-dose treatment. Repeat infection rates seem to be particularly high among HIV+ women ranging from 18.3% to 36.9%,2,7,8 suggesting that some factors specific to HIV+ women may be interfering with single-dose MTZ therapy.
In our multicentered randomized trial among HIV+ women attending 2 southern US HIV outpatient clinics, we found that the 7-day 500-mg twice daily MTZ regimen was superior to the 2-g single-dose regimen for TV infection.9 In a subsequent stratified analysis, we discovered that superiority of multidose over single dose was restricted to women with bacterial vaginosis (BV), suggesting that this condition interfered in some way with single-dose MTZ therapy.10
In this issue of the journal, Balkus et al.11 report their observations of TV treatment outcomes in a cohort of HIV− and HIV+ female sex workers in Kenya. They found that the 2-g single-dose MTZ treatment success rate among HIV+ women was significantly lower for those with BV, corroborating our findings. However, in addition to lower treatment success rates for women with BV, they also found that women who were receiving antiretroviral therapy (ART) had lower treatment success.
To compare our finding with that of Balkus et al.,11 we performed a secondary analysis of TV treatment outcomes for women randomized to the 2-g MTZ arm in our published study. Both ART and BV status were determined at baseline. Antiretroviral therapy was elicited by abstracting prescribed medication from the medical records and then confirming with the patient’s self-report of whether or not they were taking the medicine. Patients’ were asked which HIV medications they used by a trained study staff person using a chart with pictures of medications available during the study period (May 2006–July 2009). When ART status was discrepant between medical records and patient report, the patient report was used. Bacterial vaginosis was diagnosed in women who had a Nugent score of 7 or higher. Women were tested for TV using InPouch culture at both baseline and at the 7-day posttreatment test of cure (TOC) visit. Treatment was administered and directly observed at the clinic, and all women were offered 2-g MTZ treatment to deliver to their sex partner(s).
Of the 130 women in this arm, 113 (86.9%) returned for their TOC visit and had complete data for Nugent score, CD4 cell count, and ART and were included in the analyses. At baseline, 54.9% of the women had BV per Nugent score, CD4 counts were as follows: less than 200/mm3 (35.7%), 200 to 499/mm3 (53.1%), 500/mm3 or greater (21.2%). Viral loads were as follows: less than 50 copies/mL (43.8%), 50 to 9999 copies/mL (21.4%), greater than 10,000 copies/mL (34.8%). Of the 76 (67.3%) women receiving ART, the classes of drugs as documented on the medical record and confirmed by the participant were the following: nucleoside/nucleotide reverse transcriptase inhibitors (94.8%), non-nucleoside reverse transcriptase inhibitors (31.1%), PI (58.1%), and other (2.0%). As a measure of ART adherence, we asked “did you take this medication as prescribed the day before,” and 94.7% said yes to this question for each ART they were taking. Unlike the Balkus cohort, where all who were taking ART were receiving nevirapine-containing therapies, only 1 of our participants received nevirapine.
At TOC, 18.6% of those included in the analysis were TV positive via culture. Those receiving ART were as likely as those not receiving ART to have reinitiated sex (24.3% vs. 14.5%, P = 0.20). Women who were receiving ART were more likely to have CD4 cell count less than 200/mm3 (32.9% vs. 10.8%, P = 0.008) and viral loads less than 50 copies/mL (54.7% vs. 21.6%, P = 0.002) compared with women not receiving ART. These 2 factors were similar by BV status. In multivariable logistic regression, after adjusting for the presence of BV and CD4 cell count, women who were receiving ART were 4.58 times more likely to be TV+ at TOC than women not receiving ART (1.09–19.36, P = 0.038) (Table 1). After removing the 4 women who did not adhere to their ART regimen the day before from the analysis, the effect measures on ART did not change but had a slightly narrower confidence interval (odds ratio, 4.72; 95% confidence interval, 1.12–19.82; P = 0.034).
To examine if the association between ART and TV treatment failure could be explained by BV, a stratified analysis was performed. The association between ART and TV+ at TOC by BV status had a Cochrane χ2 of 3.69 (P = 0.06), demonstrating that after adjusting for BV, ART was associated with treatment failure. The Breslow-Day χ2 was 0.77 (P = 0.38), indicating no interaction between BV and ART, thus supporting an independent association of ART on the repeat infection rate.
There were some differences between these 2 studies, and our analyses address several of the limitations identified by Balkus et al. in their study. We used culture rather than the less sensitive wet prep for TV detectible and therefore likely included more asymptomatic women. Future studies should use more sensitive nucleic acid amplification techniques. Follow-up assessment in our study was 7 days postcompletion of treatment compared with 1 month posttreatment in the study of Balkus et al. Nearly 50% of the Balkus cohort had sexual exposure compared with 20% in our cohort. This difference in sexual exposure is likely attributed to differences in the cohorts (Balkus cohort includes sex workers) and a longer follow-up time between baseline and TOC visit (mean of 33 days vs. 10 days). Our study, therefore, was able to reduce but not eliminate the confounding effect of sexual exposure. Because sexual exposure was similar by ART status in our study, we think that confounding was minimal. Finally, a number of ARTs were used in our cohort compared with the Balkus cohort, which was exclusively nevirapine-containing therapies, demonstrating that the ART effect is not a phenomenon specific to nevirapine.
Although our study addresses some of the limitations of the study of Balkus et al., neither study can establish causality because they both lacked experimentation. It is possible that ART may be a marker for some other factor that interferes with single-dose MTZ treatment of TV. To better understand the mechanism whereby ART influences MTZ treatment, pharmacokinetic studies are necessary.
In summary, our secondary analyses of a previously published study support the findings of Balkus et al. in a US cohort with exposure to a wider variety of HIV retroviral drugs. Because 83.2% of our cohort either was receiving ART or had BV or both, most of our cohort would have been affected. These findings lend strength to the recommendation that the single 2-g MTZ regimen should not be used for treating HIV+ women.
1. Schwebke JR, Desmond RA. A randomized controlled trial of partner notification methods for prevention of trichomoniasis in women. Sex Transm Dis 2010; 37: 392–396.
2. Kissinger P, Secor WE, Leichliter JS, et al. Early repeated infections with Trichomonas vaginalis
among HIV-positive and HIV-negative women. Clin Infect Dis 2008; 46: 994–999.
3. Krashin JW, Koumans EH, Bradshaw-Sydnor AC, et al. Trichomonas vaginalis
prevalence, incidence, risk factors and antibiotic-resistance in an adolescent population. Sex Transm Dis 2010; 37: 440–444.
4. Tidwell BH, Lushbaugh WB, Laughlin MD, et al. A double-blind placebo-controlled trial of single-dose intravaginal versus single-dose oral metronidazole in the treatment of trichomonal vaginitis. J Infect Dis 1994; 170: 242–246.
5. Spence MR, Harwell TS, Davies MC, et al. The minimum single oral metronidazole dose for treating trichomoniasis: A randomized, blinded study. Obstet Gynecol 1997; 89: 699–703.
6. duBouchet L, Spence MR, Rein MF, et al. Multicenter comparison of clotrimazole vaginal tablets, oral metronidazole, and vaginal suppositories containing sulfanilamide, aminacrine hydrochloride, and allantoin in the treatment of symptomatic trichomoniasis. Sex Transm Dis 1997; 24: 156–160.
7. Magnus M, Clark R, Myers L, et al. Trichomonas vaginalis
among HIV-infected women: Are immune status or protease inhibitor use associated with subsequent T. vaginalis
positivity? Sex Transm Dis 2003; 30: 839–843.
8. Niccolai LM, Kopicko JJ, Kassie A, et al. Incidence and predictors of reinfection with Trichomonas vaginalis
in HIV-infected women. Sex Transm Dis 2000; 27: 284–288.
9. Kissinger P, Mena L, Levison J, et al. A randomized treatment trial: Single versus 7-day dose of metronidazole for the treatment of Trichomonas vaginalis
among HIV-infected women. J Acquir Immune Defic Syndr 2010; 55: 565–571.
10. Gatski M, Martin DH, Levison J, et al. The influence of bacterial vaginosis on the response to Trichomonas vaginalis
treatment among HIV-infected women. Sex Transm Infect 2011; 87: 205–208.
11. Balkus JE, Richardson BA, Mochache V, et al. A prospective cohort study comparing the effect of single-dose 2 g metronidazole on trichomonas vaginalis infection in HIV-seropositive versus HIV-seronegative women. Sex Transm Dis 2013; 40: 499–505.