Researchers publishing in the journal PLOS ONE have investigated a potential link between use of antibiotics during pregnancy, and cerebral palsy and epilepsy in offspring. A link has been suggested previously, and the authors suggest that it is macrolide antibiotics which are associated with these illnesses. A Before the Headlines analysis accompanied this roundup.
Prof. Dame Valerie Beral, Nuffield Department of Population Health, University of Oxford, said:
“The findings from this study demonstrate clearly that use of antibiotics during pregnancy does not increase the risk of cerebral palsy or epilepsy in children.
“The author’s claim that one group of antibiotics (macrolides) may increase the risk is not supported by their own results. They subdivided their data into many different types of antibiotics, but failed to use standard statistical methods to correct for apparently spurious results that can often be produced with such subdivisions. Once appropriate statistical tests are done, the findings for macrolides are unremarkable.
“In other words results from this study indicate that use of antibiotics in pregnancy does not increase the risk of cerebral palsy or epilepsy, and that there is little to suggest that use of macrolides is associated with greater risks than use of other types of antibiotics.”
Prof. Marian Knight, NIHR Professor of Maternal and Child Population Health, University of Oxford, said:
“Women can be reassured that taking antibiotics in pregnancy does not appear to be associated with cerebral palsy or epilepsy in their children.
“This study included a secondary comparison of antibiotic types, which appears to suggest a possible increased number of children with cerebral palsy or epilepsy amongst women prescribed macrolide antibiotics (such as erythromycin) in pregnancy compared with women prescribed penicillin. However, the study cannot determine whether it is the underlying infection or the antibiotic that may be responsible for the association. In addition, the group of women prescribed erythromycin is likely to be different from those prescribed penicillin in ways not able to be identified in a study like this. In particular, as the authors acknowledge, we know that erythromycin has side effects, which might make women stop taking it, and hence their foetus might be exposed to infection for longer than the foetuses of women who are prescribed penicillin. There may thus be other explanations for their findings.
“The study does not add substantially to the information already known from the ORACLE trial (Kenyon S, Pike K, Jones DR, Brocklehurst P, Marlow N, Salt A, et al. ‘Childhood outcomes after prescription of antibiotics to pregnant women with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial’. Lancet. 2008;372: 1319-1327). In that trial, women with premature labour whose waters had not broken, who were treated with erythromycin, had an associated increase in cerebral palsy in their children compared with women who had no antibiotics. Women treated with a penicillin had a similar increase. In both groups the number of children affected was small.
“The messages to women, their doctors and midwives should remain unchanged from that of the ORACLE trial, i.e. that antibiotics are not advisable in pregnancy in the absence of overt signs of infection. However, it is clear that infections in pregnancy are an important cause of maternal and infant death and severe illness. It is vital that pregnant women with signs of severe infection are seen without delay by senior staff and receive antibiotics, if needed, quickly. The antibiotics chosen should be the most appropriate to treat each individual woman’s infection.”
Prof. Sir David Spiegelhalter, Winton Professor of the Public Understanding of Risk, University of Cambridge, said:
“This study demonstrates the value of a good clinical data-base, which allowed an apparent association between antibiotics given to pregnant women and cerebral palsy or epilepsy in their babies to be reassuringly explained away – there was no difference in the overall incidence of cerebral palsy or epilepsy in babies of women who didn’t take any antibiotics, and of women who did take an antibiotic. However the authors point out the possible increased risks from macrolides, and this suggests further investigation, although this class accounted for less than 7% of antibiotics prescribed.”
Dr Christoph Lees, Clinical Reader in Obstetrics, Imperial College London, said:
“This is an interesting paper that considers antibiotic usage in pregnant women in primary care-in other words GP prescriptions. Nearly 200,000 women’s records linked to prescriptions were accessed and 65,000 were prescribed antibiotics
“Women need antibiotics for many reasons in pregnancy, and it is important to say that they are only prescribed if clinically needed and they may prevent serious illness. Reassuringly, overall, antibiotics were not linked to cerebral palsy and epilepsy in their infants.
“However on further analysis in women prescribed one particular type of antibiotic (macrolides), the most well-known of which is erythromycin, there was a higher risk of their infants developing cerebral palsy or epilepsy. Also, women who had three or more courses of antibiotic ran this risk. However, these outcomes were extremely rare-well below 1%.
“The authors are careful to say that this study does not show a causal relationship: in other words, we do not know if macrolides do actually cause a higher incidence of epilepsy or cerebral palsy. We just know that there is an association between their use and the conditions.
“There could be several explanations:
“So, this study is important and raises several very pertinent questions-but as the authors suggest, we don’t yet know what the answers are. One thing is for sure: pregnant women should definitely take antibiotics if prescribed them as the risk of an untreated infection to the pregnancy can be catastrophic, and to put things in context the extra risk to their baby of being prescribed macrolides-if indeed there is a risk-is very tiny.”
Prof. Alan Cameron, Royal College of Obstetricians and Gynaecologists (RCOG) Vice President for Clinical Quality, says:
“We welcome any new research which sheds light on the safety of antibiotics in pregnancy. The secondary analysis in this paper has shown a small increase in cerebral palsy and/or epilepsy with prenatal use of macrolides but it does not consider the severity of the maternal infection which, in turn, has an effect on the outcome.
“This level of evidence does not warrant a change in current clinical guidelines. Doctors should continue to prescribe erythromycin during pregnancy where indicated.”
Dr June Raine, MHRA Director of Vigilance and Risk Management of Medicines said:
“Antibiotics continue to be vital for women in treating infection and it is important that healthcare professionals continue to prescribe these medicines. This study found no overall evidence of harm when antibiotics are prescribed in pregnancy. Any pregnant woman that is prescribed antibiotics should continue to take them as instructed and speak to their healthcare professional if they have any questions.
“The suggestion of a risk with use of macrolide antibiotics during pregnancy should be treated with great caution. The conclusion is based on small numbers and incomplete data such as information on the type and seriousness of the infection. MHRA has sought independent expert advice which confirms that the study is insufficient to suggest that use of macrolides in pregnancy is associated with a particular risk.
“Infection in pregnancy can cause serious harm, both to the mother and baby, and it is essential that pregnant women continue to receive treatment with an appropriate antibiotic when necessary.”
Professor Albert Ferro, King’s College London and Fellow of the British Pharmacological Society, said:
“The major strength of this study is its size: the largest so far looking at the issue of the possible relationship between antibiotic prescribing in pregnancy and the occurrence of cerebral palsy or epilepsy in offspring. It followed a cohort of nearly 200,000 women, a third of whom (just under 65,000) had received one or more courses of antibiotics during pregnancy.
“By the very nature of the study, it can only infer association not causation: it is an observational study, rather than the ‘gold standard’ prospective randomised controlled study. However, a prospective study would be ethically difficult if not impossible to do; and it would certainly not be possible to perform a prospective study on the kind of numbers of patients studied in this paper.
“Antibiotics of the macrolide class are often used for the treatment of respiratory infections. They are also sometimes used for the treatment of oral infections and for diarrhoea secondary to campylobacter enteritis. They are also used as an alternative to penicillin for a variety of infections if the patient is penicillin-allergic.
“In this study, although overall there was no association between prescribing of any antibiotic and cerebral palsy / epilepsy, there was an apparent increase in risk of cerebral palsy or epilepsy (apparently driven by an apparent doubling in risk of epilepsy) in offspring of women receiving macrolide as compared with penicillin antibiotics. Also, there was an apparent 40% increase in risk of cerebral palsy or epilepsy in offspring of women receiving 3 or more courses of antibiotics during pregnancy as compared with those receiving none. In both cases, it is entirely possible that the nature of the underlying infection(s) – and specifically the type of infections for which macrolides were being prescribed in preference to penicillins – could be the root cause of the neurological issues seen in the offspring; and this could not be sufficiently controlled for. It should also be noted that the number of women receiving macrolides was relatively small – only 6.7% of the totality of women receiving antibiotics, i.e. less than 3,000 out of just over 40,000 women receiving single antibiotics – and this small number may have generated a significant association being found where in reality there is no association.
“A previous trial, the Oracle Childhood Study (OCS) II, had found that offspring of women who had received erythromycin (a macrolide) or co-amoxiclav (a penicillin) had an increase in risk of cerebral palsy and epilepsy, compared to offspring of women who had received no antibiotics during pregnancy. This was a small trial however, much smaller than the present one. In preclinical studies, there has been no clear association found between macrolide exposure in pregnancy and neurological problems in offspring in animal models.
“Although macrolide exposure in this study was found to relate to a statistically significant increase in risk of cerebral palsy or epilepsy – especially epilepsy – in children, the major caveats are:
– small numbers may have generated a significant association being found where in reality there is no association it is entirely unclear if the relationship is simply association (so that, for example, the underlying infection being treated is the causative factor) or is causative (i.e. that macrolides cause these neurological problems)
– despite the increase in risk of cerebral palsy or epilepsy, the absolute numbers affected were very small: so the risk increased from around 140 cases per 100,000 child years at risk to around 250 cases per 100,000 child years at risk.
“In my opinion, this should not deter APPROPRIATE prescribing of a macrolide in pregnancy if clinically indicated and if a macrolide would be more appropriate than another antibiotic class.”
Prof. Mark Fielder, Professor of Medical Microbiology Kingston University, Hon Gen Sec of Society for Applied Microbiology, said:
“The use of antibiotics in pregnancy remains a decision facing physicians on a regular basis. There are some circumstances where the use of antibiotics in a pregnant patient are deemed appropriate such as in the case of some urinary tract infection where treatment of the initial infection can prevent the infection spreading further and causing more serious issues such as kidney infections. Some antibiotics are routinely avoided during pregnancy such as streptomycin and tetracycline. However studies have shown that broad spectrum antibiotics and the cephalosporins are generally considered to be safe options when treating infections during pregnancy.
“The current study indicates that generally there was no association between prescribing antibiotics in pregnancy and adverse outcomes for the child. The authors did note that prenatal prescribing of macrolides versus penicillins did suggest an increased relative risk of the conditions examined in the study. However it is important to note that the absolute risk still remains low overall. The use of antibiotics should only be used when there is a definite clinical need as there is a need to protect these drugs from the increasing levels of resistance. This study also highlights the importance of examining the use of certain medicines in specified patients groups to ensure the best treatment outcome for the patient as well.”
‘Association between antibiotic prescribing in pregnancy and cerebral palsy or epilepsy in children born at term: A cohort study using The Health Improvement Network’ by Wilhelmine Hadler Meeraus et al. published in PLOS ONE on Wednesday 25 March.
Prof. Dame Valerie Beral: I am a member of the Board of the Medicines and Healthcare Products Regulatory Agency.
Prof. Marian Knight is a researcher at the National Perinatal Epidemiology Unit, University of Oxford. She is employed by the University of Oxford and holds an honorary contract with Public Health England. She is the principal investigator of a National Institute for Health Research trial investigating whether an antibiotic (co-amoxiclav, a penicillin), given to women after they have had a baby delivered by forceps or ventouse, prevents maternal infection.
Prof. Alan Cameron is also Director of the Ian Donald Fetal Medicine Department at the Queen Mother’s Maternity Hospital, Glasgow, and an honorary Professor of fetal Medicine at the University of Glasgow
No other interests declared