A randomised controlled trial of metronidazole for the prevention of preterm birth in women positive for cervicovaginal fetal fibronectin: the PREMET Study

Author’s Reply


We would like to thank Lamont and Dravonic for their interest in our article and their valuable contribution.1 We agree that the PREMET article should not be cited as evidence that antibiotics have no benefit in preventing preterm birth. However, we do believe that there is little evidence to support antibiotic use for this purpose in asymptomatic women.

Lamont and Dravonic report three trials that show benefit with clindamycin.1 However, one must be sceptical regarding the antimicrobial benefit from one of these studies2 as there was a statistically significant 40% reduction in preterm birth not related to antibiotic use, and this benefit was still apparent in the 80% of the ‘revealed swab result’ arm not receiving antibiotics, who had normal flora. Benefit is therefore more likely to be related to other factors, as suggested by the authors. Moreover, the majority of those with abnormal flora only had candida, not previously recognised as a risk factor for preterm birth.

To our knowledge, there are at least six additional randomised controlled trials using clindamycin, all showing no benefit, including women treated early for abnormal vaginal flora. These, when combined in meta-analysis with the previous ‘beneficial’ trials which treated abnormal flora, result in there being no overall statistically significant benefit.3

There is a sound rationale for considering different antimicrobials at different gestations in different risk groups, given the wide heterogeneity of outcome in many trials in this area. We therefore have recently completed a meta-analysis of randomised controlled trials in asymptomatic pregnant women, investigating the use of antibiotics for the prevention of preterm birth (submitted to this journal). We have specifically addressed whether the timing of treatment, antimicrobial used, or risk status (e.g. abnormal vaginal flora) of the women treated affects the results. Unfortunately, our results do not demonstrate any evidence that antibiotics are of benefit, even when these confounding factors are taken into account.

Clinicians have a choice; either to give antibiotics to subgroups of women where there is contradictory evidence of benefit or to await further evidence. We are concerned that clinicians feel compelled to offer interventions in an attempt to help under the premise that ‘something is better than nothing’. ‘First do no harm’ would be our favoured approach, rather than selective interpretation of the literature to support the use of antimicrobials. PREMET shows us that harm is possible.

There may be 1b grade A evidence suggesting benefit (i.e. at least one randomised controlled trial), but there is 1a grade A evidence suggesting no benefit (i.e. meta-analysis of randomised controlled trials).3 We believe that it is currently reasonable not to screen and treat asymptomatic pregnant women with antimicrobials until there is further evidence to support their use and confirm their safety.


  1. Top of page
  2. References
  • 1
    Lamont RF, Dragovic B. A randomised controlled trial in metronidazole for the prevention of preterm birth in women positive for cervicovaginal fetal fibronectin: the PREMET Study by Shennan et al. BJOG 2006. doi: 10.1111/j.1471-0528.2006.00958.x.
  • 2
    Kiss H, Petricevic L, Husslein P. Prospective randomised controlled trial of an infection screening programme to reduce the rate of preterm delivery. BMJ 2004;329:371.
  • 3
    Okun N, Gronau KA, Hannah M. Antibiotics for bacterial vaginosis or Trichomonas vaginalis in pregnancy: a systematic review. Obstet Gynecol 2005;105:857868.