Treatment recommendations for trichomoniasis in women

Multidose metronidazole has been found to be superior to single-dose metronidazole for the treatment of trichomoniasis in a randomised controlled trial done in women with HIV infection,¹ in a meta-analysis of published comparisons,² and in our randomised controlled trial done in women without HIV infection, published in The Lancet Infectious Diseases.³ The abundance of data indicates that multidose metronidazole should be the first line of treatment for trichomoniasis in women.

Zhen-Zhou Luo and colleagues⁴ suggest that false positives from nucleic acid amplification tests (NAAT), re-infection, and non-adherence to treatment could have influenced the results of our study,³ and suggest that more data are needed before changing treatment recommendations to multidose metronidazole. We disagree. In addition to our intention-to-treat analysis, we did numerous sensitivity analyses that showed that the findings of multidose metronidazole superiority were robust. In one of the analyses, we used culture alone as the outcome, to remove possible misclassifications that could occur from detection of remnant RNA during NAAT, and these findings corroborated our intention-to-treat analysis. Additionally, no test of cure was done before 3 weeks,⁵ therefore we do not think that the use of NAAT as an outcome biased our results.

Regarding sexual exposure and medication non-adherence, 33·5% of women reported having sex between enrolment and test of cure, and 2·6% reported not taking all of the medication. Of 540 women followed, 137 (25·4%) had sex during follow-up with a partner they had at baseline, 33 (6·1%) with a new partner, and 15 (2·8%) with a partner from baseline and a new partner. There was no difference between groups in sexual exposure during follow-up (86 [31·9%] of 270 women in the single-dose group vs 95 [35·2%] of 270 in the multidose group; p=0·412). Unsurprisingly, women in the single-dose group were more likely to adhere to their medication (264 [99·2%] of 266 women) than those in the multidose group (253 [95·5%] of 265; p=0·007). However, when re-examining our data to remove those who had sex with any partner during follow-up and who did not adhere to their medication, we still found that multidose remained superior (37 [20·8%] of 179 women in the multidose group vs 16 [9·8%] of 164 in the single-dose group; relative risk 1·70, 95% CI 1·11-2·59; p=0·005]. Earlier data suggest that most repeat infections were due to treatment failure and not reinfection.⁶ Sexual activity with an untreated partner could be driving high re-infection rates post-treatment, though it is unlikely that this occurrence explains the consistent finding of multidose superiority.

We agree that the high rate of test-of-cure positives, even among those receiving multidose metronidazole (not explained by organism insusceptibility), is concerning and a better understanding of factors that interfere with metronidazole treatment and development of more efficacious low-cost treatments for trichomoniasis are needed.