COCHRANE FOR CLINICIANS: PUTTING EVIDENCE INTO PRACTICE

What Is the Best Way to Treat Trichomoniasis in Women?

Am Fam Physician. 2001 Oct 1;64(7):1241-1245.

Clinical Scenario

A 26-year-old woman presents with a yellow vaginal discharge, vaginal itching and soreness, and mild dyspareunia. No cervicitis is found but, on wet mount, numerous motile protozoa consistent with Trichomonas vaginalis are seen.

Clinical Question

What is the most effective treatment strategy for women diagnosed with vaginal trichomoniasis?

Evidence-Based Answer

Nitroimidazoles (such as metronidazole [Flagyl]) are the most effective treatment for parasitological cure of trichomoniasis. Single-dose treatment is as effective as longer term treatment but increases side effects.

Cochrane Abstract

REVIEW 1. INTERVENTIONS IN WOMEN

Background. Approximately 120 million women worldwide are infected with Trichomonas vaginalis every year.1 The infection is sexually transmitted and believed to facilitate transmission of human immunodeficiency virus (HIV) infection.

Objectives. The objective of this review was to assess the effects of various treatment strategies for trichomoniasis in women.

Search Strategy. The authors searched the Cochrane Controlled Trials Register, MED-LINE and EMBASE. Trials were also identified from reference lists of reviews, through pharmaceutical companies and by informal discovery. Only published data were used in this review.

Selection Criteria. Randomized or quasirandomized trials of different treatment strategies in women with trichomoniasis. Various antitrichomonal drugs or dosages were eligible, as were comparisons of treatment with no treatment or placebo.

Data Collection and Analysis. Trial quality was assessed and data extracted by two reviewers independently, using standard criteria.

Primary Results. Fifty-three trials were included. Nitroimidazole drugs seem to be effective in achieving parasitological cure in short-term follow-up. Partner treatment can be effective in decreasing longer-term reinfection rates.

Reviewers' Conclusions. Parasitological cure can be achieved with a single oral dose of a nitroimidazole. Further research should focus on developing effective partner treatment strategies to prevent reinfection and reduce trichomoniasis prevalence.

REVIEW 2. INTERVENTIONS IN PREGNANCY

Background. Vaginitis caused by T. vaginalis is one of the most common sexually transmitted diseases.2 Trichomoniasis affects women during pregnancy, but it is not clearly established whether it causes preterm birth or other pregnancy complications.

Objectives. The objective of this review was to assess the effects of various treatments for trichomoniasis during pregnancy.

Search Strategy. The Pregnancy and Childbirth Group trials register and the Cochrane Controlled Trials Register were searched.

Selection Criteria. Randomized trials comparing antitrichomonal agents during pregnancy.

Data Collection and Analysis. Eligibility and trial quality were assessed by one reviewer.

Primary Results. One study was included in which benzoylmetronidazole was compared to no treatment in low-risk symptomatic and asymptomatic women. Of the treated women, 95 percent (97 of 102) and 93 percent (77 of 83) were free of infection on day 7 and week 4, respectively. This equates to a relative risk of 0.11 (95 percent confidence interval [CI], 0.05 to 0.24). About 75 percent of partners were reported to have taken the treatment.

Reviewers' Conclusions. Metronidazole, given in a single dose, is likely to provide parasitological cure for trichomoniasis, but it is not known whether this treatment will have any effect on pregnancy outcomes. The cure rate could probably be higher if more partners also used the treatment.


These summaries have been derived from Cochrane reviews published in the Cochrane Database of Systematic Reviews in the Cochrane Library. Their content has, as far as possible, been checked with the authors of the original reviews, but the summaries should not be regarded as an official product of the Cochrane Collaboration; minor editing changes have been made to the text (http://www.cochrane.org)

Reading the Numbers

Confidence Interval

In reading the medical literature, it is common to focus on the one number given for each result of a study. For instance, the relative risk of persistent trichomonas infection in pregnant women after a course of nitroimidazole was 0.11 (11 percent). It is important to remember, however, that this number is derived from a sample of the population and, statistically, we cannot be certain that this is the true relative risk for the population. To express the level of confidence in our estimate of the “true” relative risk, we can calculate a 95 percent CI (in this case, 0.05 to 0.24). Conceptually, this means that if we repeated this trial ad infinitum in various similar populations, the result would fall within the range of numbers given in the interval 95 percent of the time. The smaller the confidence interval, the better we feel about the results' precision, or closeness to the truth.

In addition, we must ensure that the confidence interval does not include zero for absolute risk reduction or 1 for odds ratio or relative risk. Inclusion of these values would mean the results were not significant. The confidence interval is increasingly recognized as a powerful description of the statistical strength of a study because of its ability to address both the significance and the precision of the result.

Reading the Numbers

View Table

Reading the Numbers

Confidence Interval

In reading the medical literature, it is common to focus on the one number given for each result of a study. For instance, the relative risk of persistent trichomonas infection in pregnant women after a course of nitroimidazole was 0.11 (11 percent). It is important to remember, however, that this number is derived from a sample of the population and, statistically, we cannot be certain that this is the true relative risk for the population. To express the level of confidence in our estimate of the “true” relative risk, we can calculate a 95 percent CI (in this case, 0.05 to 0.24). Conceptually, this means that if we repeated this trial ad infinitum in various similar populations, the result would fall within the range of numbers given in the interval 95 percent of the time. The smaller the confidence interval, the better we feel about the results' precision, or closeness to the truth.

In addition, we must ensure that the confidence interval does not include zero for absolute risk reduction or 1 for odds ratio or relative risk. Inclusion of these values would mean the results were not significant. The confidence interval is increasingly recognized as a powerful description of the statistical strength of a study because of its ability to address both the significance and the precision of the result.

Cochrane Critique

Did the authors address focused clinical questions? Yes.

How precise were the results? Where a difference was shown, the confidence intervals were narrow enough to suggest reasonable precision of results.

Can the results be applied to patient care? Yes.

Do the conclusions make biologic and clinical sense? Yes.

Are the benefits worth the harms and costs? Yes.

Were the criteria used to select articles for inclusion appropriate? Yes. In review 1, the authors searched for “randomized or “quasi-randomized” trials. They categorized the trials based on whether the randomization scheme was hidden from the investigators (“allocation concealment”).

Is it likely that important relevant articles were missed? Possibly. In review 2, the search does not appear to be as exhaustive as the search in review 1.

Was the validity of the individual articles appraised and were the appraisals reproducible? In review 2, the single author summarized a methodologic assessment similar to that used in the first study, but the assessment was performed by only one reviewer and without the quality assurance step. The lack of explicit detail regarding the methodologic assessment used in review 2 is concerning because it hinders the reader's ability to quickly determine the adequacy of the review.

Were the results similar from study to study? In review 1, the individual trials were generally very small, and the interventions and outcomes in each were sufficiently different that there was significant heterogeneity in the methods and results of the trials. Because of this, the authors could not meaningfully combine the results from these trials in a meta-analysis. In review 2, only one study met the inclusion criteria.

What are the overall results of the reviews? In review 1, the following significant results were found: treatment with nitroimidazoles was better than no treatment in achieving parasitological cure. Short-term treatment (usually as a single dose) was comparable to longer treatment in terms of parasitological cure. The most favorable mode of delivery of medication was oral plus intravaginal, followed by oral alone. Intravaginal nitroimidazoles alone were the least effective mode.

In review 1, a small trial of partner treatment showed that women were more likely to reacquire trichomonas infection if their partners were not also treated. The most powerful conclusion related by the reviewers in review 1 was that in most trials, a single dose of any nitroimidazole resulted in parasitological cure rates above 90 percent, but side effects were common and dose-related.

In review 2, only one study was found that satisfied the inclusion criteria. Low-risk symptomatic and asymptomatic pregnant women were randomized to receive ben-zoylmetronidazole (50 mL of benzoylmetronidazole equals 2 g of metronidazole) or placebo. The treatment group had a relative risk of 0.11 (95 percent CI, 0.05 to 0.24) of continued parasitological infection, but there was no effect on birth weight or gestational age at delivery.

Practice Pointers

In women with trichomoniasis, physicians can treat both the patient and her partner with a single dose of metronidazole (e.g., one 2-g oral dose) in return for a 90 percent chance of parasitological cure in one month. The use of single-dose therapy should improve compliance, but patients should be warned about the nausea and vomiting that can accompany this high dosage.

Physicians who treat patients with a longer course of lower dose metronidazole (e.g., 250 mg orally three times daily for seven days) should not feel pressured to change their regimen. The lack of more significant data on patient-oriented outcomes, such as symptom relief, infection recurrence and pregnancy outcomes, means that these reviews support single-dose therapy but not as a clearly better choice.

There has historically been concern about the use of metronidazole in the first trimester of pregnancy because of teratogenicity. Review 2 cites two narrative review articles that recommend withholding treatment during the first trimester. The Physicians' Desk Reference3 lists a contraindication to metronidazole in the first trimester (because of carcinogenicity in rats and potentiation of teratogenicity related to alcohol intake) but notes that the risk and benefits of use at any point in pregnancy should be weighed carefully.

Two meta-analyses4,5 of cohort, case-control and case series studies analyzing 150,000 and 200,000 patients, respectively, failed to find any evidence of teratogenicity with early pregnancy exposure to metronidazole. As with most pharmacologic therapy in pregnancy, the risks of the infection (including symptoms and risk of HIV transmission) and the risks of the treatment should be explained to the patient, and a decision should be reached with informed consent.

John Epling, M.D., is an assistant professor of family medicine in the Center for Evidence-Based Practice in the Department of Family Medicine at State University of New York–Upstate Medical University, Syracuse.

Address correspondence to John Epling, M.D., Department of Family Medicine, SUNY–Upstate Medical University, 475 Irving Ave., Suite 200, Syracuse, NY 13210 (e-mail: eplingj@upstate.edu). Reprints are not available from the author.

REFERENCES

1. Forna F, Gulmezoglu AM. Interventions for treating trichomoniasis in women. Cochrane Database Syst Rev. 2000;(3):CD000218.

2. Gulmezoglu AM. Interventions for trichomoniasis in pregnancy. Cochrane Database Syst Rev. 2000;(2):CD000220.

3. Physicians' desk reference, 2000. 54th ed. Montvale, N.J.: Medical Economics, 2000:2917–9.

4. Caro-Paton T, Carvajal A, Martin de Diego I, Martin-Arias LH, Alvarez Requejo A, Rodriguez Pinilla E. Is metronidazole teratogenic? A meta-analysis. Br J Clin Pharmacol. 1997;44:179–82.

5. Burtin P, Taddio A, Ariburnu O, Einarson TR, Koren G. Safety of metronidazole in pregnancy: a meta-analysis. Am J Obstet Gynecol. 1995;172(2 pt 1):525–9.

The Cochrane Abstract is a summary of a review from the Cochrane Library. It is accompanied by an interpretation to help clinicians put the evidence into practice. John Epling, M.D., presents a clinical scenario and a clinical question based on the Cochrane Abstract, along with the evidence-based answer and a full critique of the abstract.


Copyright © 2001 by the American Academy of Family Physicians.
This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP. Contact afpserv@aafp.org for copyright questions and/or permission requests.

Want to use this article elsewhere? Get Permissions


Article Tools

  • Print page
  • Share this page
  • AFP CME Quiz

Information From Industry

More in Pubmed

Navigate this Article