The prescription of antibiotics for women in spontaneous preterm labour whose waters have not broken and who have no evidence of infection has been associated with an increase in functional impairment and cerebral palsy in their children, according to a recent study. However, no association was found in women whose waters had broken. The Commission on Human Medicines (CHM) has reviewed the results and advises that any association is unlikely to be directly to the antibiotic but rather due to a number of factors involved in the natural history of pre-term labour.
Pregnant women should not be concerned about taking antibiotics to treat infections. However, women in spontaneous pre-term labour whose waters have not broken and who have no evidence of infection should not routinely be given antibiotics (i.e. no change from current good practice). Women in pre-term labour with spontaneous premature rupture of membranes should continue to receive antibiotics, even if they have no evidence of infection.
Update – 22/9/08
The Medicines and Healthcare products Regulatory Agency MHRA, the Royal College of Obstetrics and Gynaecology RCOG and the Royal College of Paediatrics and Child Health RCPCH have all issued statements emphasising that “antibiotics save lives (of mothers and their unborn and newly born babies), and pregnant women with possible or obvious infections must be considered for treatment with antibiotics.
What is the background to this?
The original Oracle Trials aimed to find out whether premature labour (before 37 weeks) is linked to underlying maternal infection that can be treated using erythromycin and/or co-amoxiclav. Two groups of women in premature labour, with no overt signs of clinical infection, were analysed separately: those whose waters had broken, and those whose waters had not broken. The study found that erythromycin (but not co-amoxiclav) had short-term benefits in women in premature labour whose waters had broken (e.g. onset of labour was delayed and infections were reduced in babies). The Royal College of Obstetricians and Gynaecologists subsequently recommended a ten-day course of the antibiotic for these women. No benefit was seen in women whose waters had not broken and antibiotics were not recommended for them.
What do the studies claim?
The Oracle Children Studies I and II looked at the progress of children born to the original trial mothers at seven years. Oracle Children Study I included data from 3298 (75%) of the 4378 children born to women in premature labour whose waters had broken. It found that there was no difference between the any of the groups in the proportion of children with functional impairment or degree of impairment. It concluded that, in spite of the short-term benefits, the prescription of antibiotics for these women has little effect on the health of children at 7 years of age.
Oracle Children Study II included data from 3196 (71%) of the 4473 children born to women in premature labour whose waters had not broken. A greater proportion of children whose mothers had been prescribed erythromycin (with or without co-amoxiclav) had at least some functional impairment than did those whose mothers had received no erythromycin (42·3% vs. 38·3%, odds ratio [OR] 1·18, 95% CI 1·02–1·37). However, a significant difference was seen only in the proportion of children with mild impairment. These functioning problems included very mild difficulties such as poor coordination. Most of the functioning problems reported by parents were minor. There was no difference in how the children in the two groups did at educational tests (SATs) which primary school children take at the end of key stage 1 of the National Curriculum
In addition, more children whose mothers had received erythromycin or co-amoxiclav developed cerebral palsy than did those born to mothers who received neither of the drugs (erythromycin: 3·3% vs. 1·7%, OR 1·93, 95% CI 1·21 to 3·09; co-amoxiclav: 3·2% vs. 1·9%, OR 1·69, 95% CI 1·07 to 2·67). The number needed to harm was 64 (95% CI 37 to 209) with erythromycin and 79 (95% CI 42 to 591) with co-amoxiclav.
As the Chief Officers’ letter points out, these findings require further study. Although ORACLE I and II were randomised controlled trials, the results discussed here are, effectively, those of observational studies. Observational studies are subject to bias and can only show association, not causation i.e. this study does not prove that erythromycin caused cerebral palsy. Similarly the study may have been subject to recall bias: mothers whose children were not functionally impaired may not have responded to the questionnaire. In Oracle II responses were not obtained for 29% of children.
It’s also worth noting that the women included in the study had no evidence of infection and in Oracle II their waters had not broken, so they would not routinely be given antibiotics. It is unclear why this association occurred, especially as there was no increased risk of functional impairment in children of women whose waters had broken.
The letter from the Chief Officers provides further information on the Oracle studies, functional disability and cerebral palsy, and ‘questions and answers’ with links to further sources of information. Advice has been sent to the mothers included in the study and a helpline has been set up.