The incidence of preterm birth (<37 weeks’ gestation) is rising. Preterm birth currently occurs in 7.6% of live births in England and Wales and 12.5% of births in the United States, where the annual cost exceeds $26.2bn (£17.5bn; €20bn).1 Up to three quarters of these births have a spontaneous onset. Evidence that labour is associated with an inflammatory process is increasing.2

Very preterm birth (<32 weeks) is commonly associated with infection3; micro-organisms usually gain access to the sterile uterine cavity by ascending from the vagina. Other routes include haematogenous spread, iatrogenic introduction, and retrograde spread through the fallopian tubes. The risk of preterm birth is higher with pyelonephritis and bacterial vaginosis. Intrauterine activation of prostaglandins and phospholipase A2 by micro-organisms may cause contractions or preterm rupture of the membranes.4 Infection can also cause neurological damage and cerebral palsy.3

It would seem logical to use antibiotics to tackle this insult and thereby reduce preterm birth and improve outcome. Their benefit is not always clear, however, and evidence of adverse effects is increasing. Follow-up of participants in the ORACLE II …