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



While we commend the efforts of the authors and contributors, the PREMET Study1 confirmed what we already know: that the wrong antibiotics (metronidazole) given to the wrong women (those without abnormal genital tract flora) too late in pregnancy (more than 22 weeks of gestation) are ineffective in reducing the incidence of preterm birth (PTB).2 In contrast, clindamycin administered early in pregnancy even to low-risk unselected women with abnormal genital tract flora reduces the incidence of PTB by 40 to 60%.2 PREMET1 Study cites one reference to support this but two other randomised controlled trials (RCTs) have also confirmed the benefits of early treatment with clindamycin.3,4 Compared with metronidazole, clindamycin has similar activity against anaerobes but is far superior with respect to broad-spectrum activity, group B streptococcus (which is associated with PTB when present as a heavy colonisation) and many other bacterial vaginosis (BV)-related organisms, particularly those more fastidious organisms such as Mycoplasma hominis. Only 16% of women who received metronidazole had BV.1 Like ORACLE II, it is a self-fulfilling prophecy that antibiotics given to women at risk of PTB for reasons other than abnormal genital tract flora (in this case, detection of fetal fibronectin [fFN] but this caveat also applies to twins and low body mass index) are unlikely to be successful. These data also suggest that the positive fFN detected at 24 or 27 weeks of gestation is likely to be due to noninfective causes.

PREMET Study confirms that antibiotics used late in pregnancy in women without abnormal flora may cause more harm than good. This is probably because the antibiotics are decimating normal flora rather than correcting abnormal flora. The earlier in pregnancy at which preterm labour and PTB occurs, the more likely this is to be due to a pathological trigger such as infection.2 The earlier in pregnancy at which abnormal genital tract flora is detected, the greater is the risk of a subsequent adverse outcome.2 Abnormal flora detected early in pregnancy, even if this reverts to normal, is associated with a degree of adverse outcome comparable with untreated BV.2 This suggests that the damage caused by infections occurs early and persists. It follows that if antibiotics are to be successful in reducing the incidence of PTB, these should be used early in pregnancy before the inflammatory response to infection results in fetomaternal tissue damage.

The oral clindamycin study cited in PREMET Study and the two uncited studies using clindamycin vaginal cream3,4 had early, mean and maximum gestational ages at screening of 16 to 17 weeks and 20 to 22 weeks, respectively, and two of the studies included re-screening and re-treating.3,4 All three RCTs showed a statistically significant 40 to 60% reduction in the incidence of PTB in a low-risk unselected population. We fear that the PREMET Study will be cited as indicating that antibiotics are of no benefit in preventing PTB; yet, level 1 grade A evidence exists which suggests that the future lies in the screening of all women for BV in early pregnancy and treating those who screen positive, irrespective of their risk of PTB by other associations.