A new study published in the journal BMJ in February 2020 reports that macrolide or penicillin antibiotic consumption in pregnancy could be linked to several adverse health outcomes in children. These include major malformations, cerebral palsy, attention deficit hyperactivity disorder, and autism spectrum disorder.
Macrolides have long been used as substitutes for penicillin antibiotics in women with sensitivity to the latter. However, in 2005, a large study showed that heart defects were linked to erythromycin use in the first three months of pregnancy, and warnings against azithromycin and clarithromycin use have also been issued in the UK.
The retrospective study was based on over 100,000 children born in the period 1990 to 2016 to mothers who had been prescribed macrolide antibiotics at any single time during pregnancy at any time from the fourth week of pregnancy to delivery. Only one macrolide (erythromycin, clarithromycin, or azithromycin) or one penicillin was prescribed. One or more successive prescriptions for the same antibiotic with less than 30 days between prescriptions were counted as a single episode.
There were two control groups, one containing over 53,000 children, siblings of the children in the study, and another containing over 82,000 children whose mothers were prescribed macrolides or penicillins before pregnancy began.
The researchers were looking for evidence of any major malformations, both overall and specific to each system, following the prescription of either of these drug categories in each trimester. The first trimester was measured from 4 to 13 weeks of gestation, is the period when most major organs are formed, and the second and third trimester period from 14 weeks to birth. The risks of a birth defect were also measured for the prescription of these antibiotics at any point in pregnancy. The investigators also measured the risks of four neurodevelopmental disorders: cerebral palsy, epilepsy, attention deficit hyperactivity disorder, and autism spectrum disorder.
Altogether, almost a third of the children were born to mothers who had at least one antibiotic in pregnancy. Of these prescriptions, penicillin made up about 70%, and macrolides about 10%. Malformations occurred in almost 190 of about 8,600 children, which comes to about 21.55 per 1,000 children whose mothers were put on prescribed macrolides, and nearly 1,700 of 100,000, that is, 13.7 per 1,000, when penicillin was prescribed in pregnancy.
Macrolide prescription in the first 13 weeks of pregnancy was linked to a 55% higher risk of congenital malformation compared to penicillin prescription. The rates were about 28/1,000 vs 18/1,000 children.
Specifically, erythromycin in the first trimester was linked to a 50% increase in major malformations (28/1,000 vs. 18/1,000, respectively). Malformations of the heart and blood vessels were increased by 62% in the macrolide group vs the penicillin group when prescribed in the first trimester (10.6/1,000 vs 6.6/1,000).
If macrolides were prescribed in any trimester, genital malformations went up by 58%, mainly in the form of hypospadias, at 4.7/1,000 vs. 3/1,000. This supports the theory that defects of the genital tract are later to occur than with the rest of the body.
There was no observable linkage between macrolide prescribing and neurodevelopmental disorders, which contrasts with the findings in earlier research. This could be because there are many different causes of neurodevelopmental disorders. Genetic causes are responsible for up to 70% of the risk for epilepsy, attention deficit hyperactivity disorder, and autism spectrum disorder.
Macrolide antibiotic prescription during pregnancy in the first trimester of pregnancy was linked to a higher risk of major birth defects, and specifically, defects of the cardiovascular system when compared to penicillin antibiotics. At any time during pregnancy, macrolide prescriptions were associated with a higher risk of genital defects.
The mechanism of macrolide-mediated cardiovascular defects could be the arrhythmogenic effect of macrolides, which is known to cause a higher risk of cardiovascular events and death in adults at high risk of these conditions. Experiments have shown that fetuses exposed to drugs that cause arrhythmia may suffer hypoxia, which could explain the occurrence of birth defects.
The strengths of this study include the large sample size, the use of children whose mothers were put on macrolides vs. penicillin to compensate for the effects of infection on the child’s development since both these drugs are treated as identical in their indications and effects on the infectious agent. Only a single episode of treatment with one drug is considered since severe or recurrent infection can itself be a risk factor for poor developmental outcomes.
It was not possible to compare the risk of exposure with the risk of defective development during periods known to be linked to the critical window of development of various organs and organ systems. Again, antibiotic prescription rather than use or dispensing was used, which could dilute the results.
The study concludes: “These findings show that macrolides should be used with caution during pregnancy and if feasible alternative antibiotics should be prescribed until further research is available.”
Fan, H., Gilbert, R., O’Callaghan, F., and Li, L. Associations between macrolide antibiotics prescribing during pregnancy and adverse child outcomes in the UK: population-based cohort study. BMJ 2020;368:m331 | doi: 10.1136/bmj.m331