Thursday September 18 2008
The original ORACLE studies were followed up seven years later
“Antibiotics to delay premature birth may harm babies” is the headline in The Independent. A warning about the "indiscriminate prescribing of antibiotics to pregnant women to delay premature labour was issued to all doctors” following research that has unexpectedly revealed long-term harm, the newspaper says.
Two studies looked at the use of antibiotics in two different groups of women: those who had started labour early (preterm labour) and those women whose waters had broken early (premature rupture of membranes). The study found that at age seven, children born to mothers who were given antibiotics for early labour (but whose waters hadn’t broken), had an increased risk of cerebral palsy. There was no increased risk for those given antibiotics for early labour when their waters had broken. The reason for this difference is unknown.
The Government's chief medical officer, Sir Liam Donaldson, has said that antibiotics should continue to be given to women in early labour where there was evidence of infection or risk of infection because their waters had broken early. Several other commentators including the Royal College of Obstetricians say that, "These findings do not mean that antibiotics are unsafe for use in pregnancy. Pregnant women showing signs of infection should be treated promptly with antibiotics."
Where did the story come from?
Sara Kenyon is the first author for two studies from the Reproductive Sciences section, Cancer Studies and Molecular Medicine and the Health Sciences Department, all at the University Of Leicester. The studies were co-authored by other professors from Nottingham, Oxford and Great Ormond Street Hospital in the UK. The studies were funded by the UK Medical Research Council and sponsored by the University Hospitals Of Leicester and approved by their research and development directorate. Both studies – ORACLE I and ORACLE II – were published with an accompanying editorial in the peer-reviewed medical journal: The Lancet.
What kind of scientific study was this?
The ORACLE I study (published in 2001) was a randomised controlled trial that compared the use of two antibiotics, erythromycin and/or co-amoxiclav, with that of placebo for women with preterm rupture of the membranes (PROM) without obvious signs of infection. The early results of this trial showed that erythromycin was associated with prolongation of pregnancy and a reduction in problems in the newborn baby. The prescribing of erythromycin is now recommended practice in this situation. There is also a Cochrane Review on the topic by the same author as this trial. The aim of the present study – the ORACLE Children Study I – was to determine the long-term effects of these antibiotics on the children born to mothers who took part in the ORACLE I study.
Seven years after the study, the researchers assessed the children born to the 4,148 women who had enrolled using a structured parental questionnaire that asked about child health status. They only included children who were eligible for follow-up and some parents did not complete the questionnaire. Of the 4,378 children who were eligible for follow-up, outcomes were known for 3,298 (75%) and full questionnaire data was available for 3,171 (72%) of the children. By the time the results were analysed at seven years, 37 children (1%) had died.
Based on the responses to the questionnaires, the researchers assessed any impairment of function (severe, moderate, or mild) based on a validated system – the Mark III Multi-Attribute Health Status classification system. They also assessed educational outcomes with support from the UK Qualifications and Curriculum Authority with access to the national curriculum test results at age seven (key stage one) for all children resident in England.
The ORACLE II study (also published in 2001) was similar in design – a randomised controlled trial – but this looked at the use of the same antibiotics compared with placebo for women in spontaneous preterm labour with intact membranes, without obvious signs of infection. The results of this trial found that there was no benefit to using antibiotics in this condition, as there was no difference in the length of the pregnancy or problems in the newborn baby.
Again, the ORACLE Children Study II looked at the long-term effects of antibiotic exposure on the children born during the ORCALE II study. Researchers assessed the children (at age seven years) born to the 4,221 women who had completed the ORACLE II study using a parental questionnaire of the child’s health status. The functional and educational outcomes were assessed in the same way as described above.
What were the results of the study?
For the 3,298 (75%) eligible children in the ORACLE I trial (those with preterm rupture of the membranes – PROM), there was no difference in the proportion of children with any functional impairment after prescription of erythromycin, with or without co-amoxiclav (594 [38.3%] of 1,551 children) compared with those born to mothers who received no erythromycin (655 [40.4%] of 1,620 children). A similar, non-significant difference was shown when the results were analysed the other way round, i.e. co-amoxiclav, with or without erythromycin, compared with those born to mothers who received no co-amoxiclav. Neither antibiotic had a significant effect on the overall level of behavioural difficulties experienced, on specific medical conditions or on the proportions of children achieving each level in reading, writing or mathematics at key stage one.
For the 3,196 (71%) eligible children in the ORACLE II trial (those with premature labour without membrane rupture), there were some statistically significant differences between groups. Overall, a greater proportion of children whose mothers had been prescribed erythromycin, with or without co-amoxiclav, had functional impairment (658 [42.3%] of 1,554 children) than those whose mothers had received no erythromycin (574 [38.3%] of 1,498 children). The odds ratio for this was 1.18 (95% CI 1.02–1.37), suggesting a small but statistically significant effect. However, co-amoxiclav (with or without erythromycin) had no significant effect on the proportion of children with any functional impairment, compared with those who did not get co-amoxiclav (624 [40.7%] of 1,523 vs 608 [40.0%] of 1,520).
No effects were seen with either antibiotic on the number of deaths, other medical conditions, behavioural patterns or educational attainment. However, more children whose mothers had received erythromycin or co-amoxiclav developed cerebral palsy than the children born to mothers who received no erythromycin or no co-amoxiclav (53 [3.3%] of 1,611 whose mothers were given erythromycin versus 27 [1.7%] of 1,562 who did not get erythromycin; 50 [3.2%] of 1,587 whose mothers were given co-amoxiclav versus 30 [1.9%] of 1,586 who did not receive co-amoxiclav).
What interpretations did the researchers draw from these results?
The prescription of antibiotics for women with preterm rupture of the membranes (PROM) seems to have little effect on the health of children at seven years of age.
The prescription of erythromycin for women in spontaneous preterm labour with intact membranes was associated with an increase in functional impairment among their children at seven years of age. The risk of cerebral palsy was increased by either antibiotic, although the overall risk of this condition was low.
What does the NHS Knowledge Service make of this study?
These are both reliable and valid studies in which, in ORACLE II, the researchers report a finding that was unexpected. The researchers say that the excess of children with cerebral palsy born to mothers who received both antibiotics is clear enough to suggest that this should not be dismissed as a chance result of multiple testing. They mention some cautions and some features that support the idea that they were observing a true effect:
- There was no evidence of an interaction between the two antibiotics, which was expected, given that an increased risk was associated with the use of either.
- The power of the study (number of children whose outcomes could be analysed to detect these interactions) was low and this might explain the lack of a significant interaction effect.
- They say that data from another source (four counties in the UK) suggest that 7.5 cases would have been expected in this population, compared with the 12 cases observed. The fact that the overall rate of cerebral palsy is similar in their trial suggests that the result was not simply due to a low rate of cerebral palsy in the placebo group.
The editorial in the same edition of The Lancet comments that the prescribing of erythromycin during labour has been increasing over recent years and that unfortunately there has been no specific monitoring (microbiological surveillance) of the consequences. Nationally collected data shows a large rise in the number of erythromycin-resistant bacteria (Strep B) isolated in laboratories, from 6.4% in 2002 to 11.2% in 2006. The author of the editorial highlights this as a potential danger of prescribing antibiotics, concluding that they are not risk-free. The danger to individual children of an increasing rate of cerebral palsy seems clear, although the risk is small and the mechanism for the effect is currently unclear. In general, pregnant women should not worry, the problems were quite specific to one group of women and do not apply to all antibiotics or all situations in which they might be given.
Sir Muir Gray adds...
The message is clear; antibiotics should not be given, or taken, ‘just in case’ but only when there is a clear clinical need.