Macrolide Antibiotics in Pregnancy Linked to Birth Defects

Liam Davenport

February 21, 2020

Women who take macrolide antibiotics during the first trimester have an increased risk of their child having a major malformation, say UK researchers, who urge caution in the use of the drugs throughout pregnancy.

Macrolide antibiotics include erythromycin, clarithromycin, and azithromycin, and are among the most frequently prescribed antibiotics during pregnancy.

The study included primary care data on more than 100,000 children born to women prescribed antibiotics during pregnancy, as well as more than 50,000 of their siblings, and a further 80,000 children born to mothers who took antibiotics before conception.

Call for Cautious Use

Heng Fan, a PhD candidate at University College London (UCL) Great Ormond Street Institute of Child Health, and colleagues, found that compared with taking penicillin, taking a macrolide during the first trimester increased the risk of major malformations by 55%, rising to 62% for cardiovascular malformations.

The research, published by the BMJ on 19th February, showed that erythromycin use in the first trimester was linked to a 50% increased risk of any major malformation.

The team said that the findings: "call for cautious use of macrolides during pregnancy.

"Drug safety leaflets should report that there are concerns about the safety of macrolides, including erythromycin, and recommend the use of alternative antibiotics when feasible until further research is available."

Study author Professor Ruth Gilbert, also at UCL Great Ormond Street Institute of Child Health, said in a news release that "it would be better to avoid macrolides during pregnancy if alternative antibiotics can be used".

However, she underlined that "women should not stop taking antibiotics when needed, as untreated infections are a greater risk to the unborn baby".

Drugs Regulator Agrees With Assessment

Dr Sarah Branch, director of Vigilance and Risk Management of Medicines at the UK Medicines and Healthcare products Regulatory Agency (MHRA), who was not involved in the study, agreed with that assessment.

"Untreated infection in pregnancy can cause serious harm, both to the mother and baby, and it is essential that pregnant women receive treatment with an appropriate antibiotic when necessary," she said in a statement.

"Pregnant women who have been prescribed a macrolide antibiotic should continue to take it, and discuss any concerns they have with their healthcare professional."

Dr Branch added: "We are reviewing the findings of this study in the context of similar studies which have not found this association. Macrolides should continue to be used in pregnancy where there is a clinical need."

Stephen Evans, professor of Pharmacoepidemiology, London School of Hygiene & Tropical Medicine, UK, was among experts commenting via the Science Media Centre. He said that, while the study was "well-conducted and analysed", he believes the "authors may have had a tendency to over-interpret their findings".

He said that this is "often the case with observational studies", and pointed to the lack of consistency in the pattern of cardiovascular anomalies as being "compatible" with over-interpretation.

Nevertheless, he said that "even if the evidence for harm of macrolides in general is not as strong as these authors suggest, there is no real evidence of absence of harm".

Prof Evans therefore suggests that "caution is certainly reasonable", particularly when prescribing for respiratory infections, "where the benefits may be largely absent if the primary infection is viral".

He added: "Antibiotics are vital in protecting individuals against serious consequences of infection, but their over-use has led to resistance. It is not simple to reduce their use but being careful in the choice in pregnancy is entirely sensible."


The researchers note that, despite the widespread use of macrolide antibiotics in pregnancy, there is a lack of consistency in the advice on how they should be used in pregnancy.

Uncertainties over the effects of the drugs in this period had already been noted, with one recent systematic review indicating consistent evidence for an increased risk of miscarriage, but less consistent evidence for congenital malformations and other potential adverse effects.

The current study used data from the UK Clinical Practice Research Datalink (CPRD), a primary care database that covers almost 7% of the UK population, to identify all babies born alive between 1990 and 2016.

Children with known chromosomal abnormalities were excluded, as well as those whose mothers were prescribed known teratogenic drugs during pregnancy, such as warfarin and angiotensin converting enzyme inhibitors.

The team focused on 104,605 children whose mothers were prescribed macrolide (n=8632, 8.3%) or penicillin (n=95,973, 91.7%) monotherapy during any trimester of the pregnancy. The median follow-up was 5.8 years.

They were compared with a negative control cohort of 82,314 children born to mothers prescribed macrolides (n=11,874) or penicillins (70,440) before pregnancy.

A second negative control cohort consisted of 53,735 siblings of the children prenatally exposed to macrolides (n=4512) or penicillins (49,223).

The results showed that the prevalence of major nervous, cardiovascular, gastrointestinal, genital, and urinary malformations was 27.7 per 1000 livebirths in mothers prescribed macrolides in the first trimester, and 19.5 per 1000 livebirths among those prescribed the drugs in the second to third trimester.

The prevalence of major malformations was 17.7 per 1000 live births among mothers prescribed penicillins in the first trimester, and remained stable at 17.3 per 1000 livebirths in the second to third trimester.

It was calculated that macrolide use in the first trimester was associated with a significantly increased risk of any major malformation, with an adjusted risk ratio of 1.55 (p=0.001), and of cardiovascular malformations in particular, with a risk ratio of 1.62 (p=0.030).

While there was no significantly increased risk of major malformations overall with macrolide prescribing in the second and third trimesters (p=0.182), there was a link of borderline significance with gastrointestinal malformations specifically, at a risk ratio of 1.89 (p=0.050).

The team also identified an increased risk of genital malformations with macrolide prescribing in any trimester, with a risk ratio of 1.58 (p=0.006), mainly due to an increased incidence of hypospadias.

Further analysis suggested that erythromycin use in the first trimester was associated with an increased risk of any major malformation, with a risk ratio versus penicillin of 1.50.

There was, however, no link between macrolide prescribing and the risk of cerebral palsy, epilepsy, attention deficit hyperactivity disorder, or autism spectrum disorder.

There was also no association between the prescription of macrolide antibiotics before pregnancy and adverse child outcomes, and there was no link between sibling prenatal exposure and major malformations.

'High Quality' Investigation

Dr Sarah Stock, senior clinical lecturer, Maternal and Fetal Medicine, University of Edinburgh Usher Institute, UK, who was not involved in the study, said that it was a "high quality" investigation, with "real care" taken to ensure the findings are robust.

Consequently, she found the conclusion of "a small increased risk of some birth defects pretty convincing".

Dr Stock added: "Using an alternative antibiotic wherever possible in early pregnancy seems sensible.  However, if macrolides are the only treatment option, women can be reassured that the absolute risk of a problem is low.

"The highest risk was seen with early pregnancy prescriptions, where the additional risk of a heart defect associated with macrolide use was less than half a percent."

Dr Pat O’Brien, consultant obstetrician and vice president of the Royal College of Obstetricians and Gynaecologists, London, UK, noted: "In some cases, for example in women allergic to penicillin and where the infecting bacteria are resistant to other antibiotics, erythromycin may be the only suitable antibiotic.

"In this situation, it seems very likely that the risks of an untreated infection will be greater than the possible risks of using erythromycin.

"As the authors say, further research is urgently needed to clarify this issue. In particular, an analysis pooling the information from all the good studies of this question (including the current one) should be carried out without delay."

Heng Fan received funding from Child Health Research CIO Trust and China Scholarship Council. Ruth Gilbert receives funding from Health Data Research UK. The study was supported by the National Institute for Health Research.

Stock receives funding from the Wellcome Trust to study the effects of medicines given in pregnancy. No other conflicts of interest declared.

BMJ 2020;368:m331 doi: 10.1136/bmj.m331


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: