There has been some debate about whether particular antibiotics present a risk when used in pregnancy. In early 2020, a study published in the British Medical Journal (BMJ, 2020) found that children of mothers who were prescribed macrolide antibiotics during early pregnancy were at an increased risk of major birth defects, namely heart defects, when compared with children of mothers who received penicillin.
Macrolide antibiotics include erythromycin, clarithromycin, and azithromycin, and are often used to treat common bacterial infections, particularly as an alternative for patients with allergies to penicillin.
The BMJ (2020) noted that their study findings meant that macrolides should be used with caution throughout pregnancy and that it would be best where possible to prescribe appropriate alternative options until further research provides further evidence regarding the apparent risk presented by macrolides. The authors stated at the time that past research suggested evidence of rare yet serious adverse outcomes of macrolide use, particularly in unborn babies. These outcomes were also found mainly to consist of heart rhythm problems, but the BMJ commented that policy advice at the time about macrolide use in pregnancy was variable (BMJ, 2020).
Since the potential risk was identified, however, a new study by Worm et al (2021), also published in the BMJ, seems to provide further clarity on the matter. Here we will examine both studies and their findings regarding the potential risk of macrolides during pregnancy.
2020 study
A team of researchers at University College of London analysed the association between macrolide antibiotics prescribed during pregnancy and major malformations, as well as four neurodevelopmental disorders (cerebral palsy, epilepsy, ADHD, and autism spectrum disorder) in children.
Data were analysed from 104 605 children born in the UK between 1990 and 2016 with a median follow-up of 5.8 years after birth. A further 82 314 children whose mothers were prescribed macrolides or penicillins before pregnancy, and 53 735 children who were siblings of children in the study group acted as negative control cohorts. The study found that major malformations were recorded in 186 of 8632 children whose mothers were prescribed macrolides at any point during pregnancy and 1666 of 95 973 children whose mothers were prescribed penicillins during pregnancy. Potentially influential factors were considered yet despite this, the researchers concluded that macrolide prescribing during the first trimester of pregnancy in particular was associated with an elevated risk of major malformation in comparison with the use of penicillin. The difference was large, finding the defects in 28 of 1000 babies of the erythromycin group and 18 per 1000 in the penicillin group. Cardiovascular malformations specifically were found in 11 per 1000 of the erythromycin group and seven per 1000 of those whose mothers received penicillin.
The research did not find that the risk was in children of mothers whose macrolides were prescribed in the second to third trimester, however. The research at the time also found that macrolide prescribing in any trimester was associated in addition to this with a slightly elevated risk of genital malformations, finding five in 1000 for those born to mothers who had erythromycin and three per 1000 whose mother received penicillin. The BMJ (2020) reported that no statistically significant associations were discovered for other system-specific malformations or for any of the four neurodevelopmental disorders.
The study was observational, and therefore could not establish cause, and of course treatment exposure could not be examined due to the historical data gathering design, which would have benefited the analysis of use of such antibiotics in known critical periods for specific malformations and neurodevelopmental disorders.
The BMJ (2020) stated however that the results were largely unchanged after further analyses, stating the results would withstand scrutiny. They also stated that if the associations are shown to be causal, on estimation a further four children with cardiovascular malformations would occur for every 1000 children exposed to macrolides instead of penicillins in the first trimester. They therefore concluded use with caution of such antibiotics during pregnancy until further research becomes available.
2021 study
Worm et al (2021) also examined the association between use of macrolides in pregnancy and risk of major birth defects. It was a nationwide register-based cohort study thus carrying some of the same limitations, set in Denmark between 1997 and 2016.
Of 1 192 539 live birth pregnancies, Worm et al (2021) compared pregnancies during which macrolides had been used (13 019) with those during which penicillin had been used, matching the groups in a 1:1 ratio on propensity scores. Other comparative groups were pregnancies where macrolides had been used recently but prior to the pregnancy (matched 1:1) and pregnancies where no antibiotics had been used at all (matched 1:4). The team identified the main outcomes measures to be the association with an outcome of any major birth defect and specific subgroups of birth defects. These were assessed by relative risk ratios and absolute risk differences.
Worm et al (2021) found that in matched comparisons, 457 infants were born with major birth defects to women who had used macrolides during pregnancy (35.1 per 1000 pregnancies) compared with 481 infants (37.0 per 1000 pregnancies) to women who had used penicillin, corresponding to an absolute risk difference of −1.8 per 1000 pregnancies. The risk of major birth defects was found to not be significantly increased for women who had received macrolides during pregnancy in comparison to women who had used macrolides recently but before becoming pregnant, showing an absolute risk difference of −0.1, and similarly there was no significant risk increase found when comparing he macrolides in pregnancy group with women who did not use any antibiotics at all. For all three comparative group analyses and in the analyses of use of individual macrolides, the team found there to be no significant increased risk of specific subgroups of birth defects associated with the use of macrolides.
Worm et al (2021) therefore concluded that in their nationwide cohort study, the use of macrolide antibiotics in pregnancy was not associated with an increased risk of major birth defects, therefore completely disagreeing in effect with the outcome of the other study a year earlier. Analyses of the associated risk of 12 specific subgroups of birth defects with the use of macrolides in pregnancy were not significant.
Conclusion
These conflicting findings serve as a reminder of how ongoing research into the same topic can provide further clarity, or perhaps a lack of clarity, but also influence potential changes in practice (yet to be confirmed) and a change in the knowledge base surrounding use of certain medications in certain types of patients.
Of course, practitioners will always follow current national guidelines regardless, as some research carries with it some limitations, and when considering a huge mass of research on the same topic, a more reliable conclusion can be formed.
The BMJ is considered a reputable journal that publishes reliable evidence from higher quality studies carried out using best practice. Alongside professional journals, journals such as the BMJ are important to read but as shown here, we must never be too heavily influenced by one study, even from a highly reputable journal, as the evidence base is ever-changing.
Currently, erythromycin is the only macrolide that can be used during pregnancy. This would then be taken at the usual doses throughout the pregnancy or period breastfeeding. According to the NHS (2021), a woman can take most penicillins during pregnancy and breastfeeding at the usual doses, but cephalosporins despite their wide use may present some risk in a pregnant or breastfeeding woman if she has a rare inherited blood disorder known as acute porphyria. Tetracyclines are not usually recommended unless it is deemed absolutely necessary in a pregnant or breastfeeding woman and fluoroquinolones are not normally suitable for women who are pregnant or breastfeeding.
It is recommended however as always to keep up to date with NHS and BNF recommendations in case of any change to guidelines and best practice in future.