A randomised controlled trial of metronidazole for the prevention of preterm birth in women positive for cervicovaginal fetal fibronectin: the PREMET Study
We would like to thank McParland et al.1 for their interest in our article. We agree that fetal fibronectin is not a reliable test for the presence of bacterial vaginosis (BV), nor did we intend it to be. It is, however, efficacious as a predictor of preterm birth, much more so than the rather weaker association between BV and preterm birth. It is not known whether fetal fibronectin positive status reflects the consequences of intrauterine infection, but it is likely that infection will result in a positive test as both are related to delivery. Our conclusion of metronidazole therapy not reducing the incidence of preterm birth refers to those women identified as being high risk from their obstetric history and positive fibronectin status and not from their vaginal flora status.
Metronidazole had no influence on the prevalence of BV in our study population (although the number of women with BV was small); however, we would still advise caution when prescribing metronidazole to asymptomatic women as it may increase the risk of preterm birth in a high-risk population.
We do not believe that there are sufficient grounds to treat women with BV until more evidence is available and agree with McParland et al.1 about the need for further investigation. If clinicians do feel compelled to treat BV (e.g. if women are symptomatic), there are safer alternatives.
As the authors allude to in their letter, the trials in the literature pertaining to antibiotic therapy for abnormal vaginal flora are very heterogeneous, and we too would recommend caution in the use of antibiotics in pregnancy until more research is performed.