Prophylactic antibiotics for inhibiting preterm labour with intact membranes

The aetiology of preterm birth is complex and there is evidence that subclinical genital tract infection influences preterm labour in some women but the role of prophylactic antibiotic treatment in the management of preterm labour is controversial. Since rupture of the membranes is an important factor in the progression of preterm labour, it is important to see if the routine administration of antibiotics confers any benefit or causes harm, prior to membrane rupture.

To assess the effects of prophylactic antibiotics administered to women in preterm labour with intact membranes, on maternal and neonatal outcomes.

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 August 2013).

Randomised trials that compared antibiotic treatment with placebo or no treatment for women in preterm labour (between 20 and 36 weeks' gestation) with intact membranes.

Two review authors independently assessed trial eligibility, and undertook quality assessment and data extraction. We contacted study authors for additional information. Results are presented using risk ratio (RR) for categorical data and mean difference (MD) for data measured on a continuous scale with their respective 95% confidence intervals (CI). The number needed to treat to benefit (NNTB) and the number needed to treat to harm (NNTH) was calculated where appropriate.

In this update (2013), with the addition of three trials (305 women), the large ORACLE II 2001 trial continues to dominate the results of this review. This review now includes a total of 14 studies randomising 7837 women. No significant difference was shown in perinatal or infant mortality for infants of women allocated to any prophylactic antibiotics compared with no antibiotics. However, an increase in neonatal deaths was shown for infants of women receiving any prophylactic antibiotics when compared with placebo (RR 1.57, 95% CI 1.03 to 2.40; NNTH 149, 95% CI 2500 to 61). No reduction in preterm birth or other clinically important short-term outcomes for the infant were shown.Long-term child outcomes to seven years of age were available for infants in the UK enrolled in the ORACLE II trial. Comparing any antibiotics with placebo, a marginally non-statistically significant increase was shown in any functional impairment (RR 1.10, 95% CI 0.99 to 1.23) and cerebral palsy (CP) (RR 1.82, 95% CI 0.99 to 3.34). In subgroup analysis, CP was statistically significantly increased for infants of women allocated to macrolide and beta-lactam antibiotics combined compared with placebo (RR 2.83, 95% CI 1.02 to 7.88; NNTH 35, 95% CI 333 to 9).Further, exposure to any macrolide antibiotics (including erythromycin alone or erythromycin plus co-amoxiclav) versus no macrolide antibiotics (including placebo and co-amoxiclav alone) was shown to increase neonatal death (RR 1.52, 95% CI 1.05 to 2.19; NNTH 139, 95% CI 1429 to 61), any functional impairment (RR 1.11, 95% CI 1.01 to 1.20; NNTH 24, 95% CI 263 to 13) and CP (RR 1.90, 95% CI 1.20 to 3.01; NNTH 64, 95% CI 286 to 29). Exposure to any beta-lactam (beta-lactam alone or in combination with macrolide antibiotics) versus no beta-lactam antibiotics resulted in more neonatal deaths (RR 1.51, 95% CI 1.06 to 2.15; NNTH 143, 95% CI 1250 to 63) and CP (RR 1.67, 95% CI 1.06 to 2.61; NNTH 79, 95% CI 909 to 33), however no difference was shown in functional impairment.Maternal infection was reduced with the use of any prophylactic antibiotics compared with placebo (RR 0.74, 95% CI 0.63 to 0.86; NNTB 34, 95% CI 24 to 63) and any beta-lactam compared with no beta-lactam antibiotics (RR 0.80, 95% CI 0.69 to 0.92; NNTB 47, 95% CI 31 to 119). However, caution should be exercised with this finding due to the possibility of bias shown by funnel plot asymmetry. Any beta-lactam compared with no beta-lactam antibiotics was associated with an increase in maternal adverse drug reaction (RR 1.61, 95% CI 1.02 to 2.54; NNTH 17, 95% CI 526 to 7).

This review did not demonstrate any benefit in important neonatal outcomes with the use of prophylactic antibiotics for women in preterm labour with intact membranes, although maternal infection may be reduced. Of concern, is the finding of short- and longer-term harm for children of mothers exposed to antibiotics. The evidence supports not giving antibiotics routinely to women in preterm labour with intact membranes in the absence of overt signs of infection.Further research is required to develop sensitive markers of subclinical infection for women in preterm labour with intact membranes, as this is a group that might benefit from future novel interventions, including new modalities of antibiotic therapy. The results of this review demonstrate the need for future trials in the area of preterm birth to include assessment of long-term neurodevelopmental outcome.