To our knowledge, this study is the first multisite evaluation conducted to assess the prevalence of in vitro T. vaginalis resistance in the United States. Although metronidazole has been used to treat T. vaginalis infections for ≈40 years, we found a low prevalence of in vitro metronidazole resistance. MLCs of tinidazole were lower than MLCs of metronidazole, and we did not detect tinidazole resistance.
The prevalence of in vitro metronidazole and tinidazole resistance is consistent with previously published US estimates. Three studies conducted in the southeastern United States among women attending STD or gynecology clinics from 1997 through 2005 found a metronidazole-resistance prevalence of 2.4%–9.5% (9–11
). Most metronidazole-resistant isolates in these studies exhibited low-level resistance. In 2 of these studies tinidazole resistance also was tested: Krashin et al. did not detect tinidazole resistance (11
), and Schwebke and Barrientes detected 1 isolate (0.6%) that exhibited low-level tinidazole resistance (MLC 50 µg/mL) among the 178 isolates tested (10
). Among 91 isolates collected in Spain during 1995 and 1999, 2.2% exhibited low-level resistance to metronidazole (14
). A small study conducted among women from Papua New Guinea found 21 (91%) of 23 studied isolates had MLCs of metronidazole of >
50 µg/mL, including 4 (17%) with MLCs of 200 µg/mL (15
). However, the sampling method used to enroll women was not described and may not have been systematic, thus substantially limiting the ability to estimate the population-level prevalence of resistance. Investigators in the United Kingdom reported that 1.7% of women treated for trichomoniasis during 1998–2002 appeared to have not responded to treatment and denied re-exposure; in vitro susceptibility data were not available (16
Inconsistency does exist between in vitro susceptibility results and clinical outcomes of treatment, particularly for infections with low-level in vitro resistance. Clinical resistance and treatment failure have occurred with T. vaginalis
isolates for which MLCs of nitroimidazoles were as low as 12.5 µg/mL, and treatment success has occurred in infections with T. vaginalis
isolates for which MLCs of nitroimidazoles were 100–200 µg/mL (13
). In general, however, elevated MLCs are associated with a greater likelihood of treatment failure. A recent evaluation of the utility of susceptibility testing in women for whom clinical treatment has failed found that treatment recommendations based on susceptibility results may have a beneficial role in informing the clinical management of some women with persistent infection (17
). T. vaginalis
susceptibility testing is not available routinely; such testing should be conducted by a qualified laboratory and is available at CDC (1–800-CDC-INFO).
As a cross-sectional evaluation of in vitro antimicrobial drug susceptibility, the study was not designed to detect clinical treatment failures. Multiple isolates were collected from 17 women. However, we did not systematically collect data on sexual re-exposure after treatment or adherence, so we were not able to determine whether any of these cases resulted from treatment failure. The MLCs of metronidazole for the initial isolates were low, suggesting that clinical resistance was unlikely. In 2 cases, the initial isolate exhibited low-level resistance (MLC of metronidazole 50 µg/mL) and a second isolate was later collected. Both of these cases were probably re-infections.
For isolates in our study, MLCs of tinidazole were lower than those of metronidazole, which supports the idea that tinidazole should be prescribed for patients whose infections do not respond clinically to metronidazole. This finding is consistent with results of previous studies which showed that tinidazole had better in vitro activity than metronidazole at similar molar concentrations (18
). Tinidazole has a longer serum half-life than metronidazole and exhibits good tissue penetration (19
), yet is more expensive than metronidazole. Although tinidazole and metronidazole are the only nitroimidazoles available in the United States, ornidazole, tenonitrazole, and nimorazole are available in Europe and could be alternatives to metronidazole. These agents are of the same drug class as metronidazole, however, and the emergence of clinically notable nitroimidazole resistance would be expected to adversely influence the treatment effectiveness of each of these agents.
This study had several limitations. First, the sample was limited to women attending STD clinics participating in SSuN; thus, our findings are not representative of the general population. In addition, symptomatic women were likely to have been overrepresented because participating women were seeking care in STD clinics, and 2 of the sites sampled only symptomatic women. Also, although we believe this is the largest study of its kind, the sample size may not have been large enough to detect significant differences across sites, nor to detect tinidazole resistance or high-level metronidazole resistance. That we did not detect isolates with such resistance suggests that its prevalence in this population is low. CDC occasionally receives isolates that are highly resistant to metronidazole or tinidazole, however (W.E. Secor, pers. comm.).
Although the prevalence of resistance is currently low in the United States, reliance on a single class of antimicrobial drugs heightens vulnerability if clinical T. vaginalis nitroimidazole resistance becomes widespread. Market forces alone are unlikely to spur the development of new anti-trichomonal drugs. Further evaluation of existing compounds and development of novel systemic treatment options are needed, and efforts to promote and support antimicrobial drug development and evaluation are warranted.
In summary, we found a 4% prevalence of low-level metronidazole resistance among T. vaginalis isolates from women attending several STD clinics throughout the United States. Periodic sentinel surveillance evaluations of T. vaginalis antimicrobial drug susceptibility should be carried out to monitor the possible emergence of resistance.