Macrolide prescriptions in first trimester linked to major malformations in children

  • Fan H, et al
  • The BMJ
  • 19 Feb 2020

  • curated by Liz Scherer
  • Clinical Essentials
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • First-trimester macrolide prescriptions are associated with increased risk for any major malformations (especially cardiovascular) in children. 
  • Risk for genital malformations (hypospadias) persists after first trimester.

Why this matters

  • Prescribe macrolide alternatives whenever possible in pregnant women and use macrolides with caution.

Key results

  • Of 104,605 children, 8.3% (8632) were born to mothers prescribed 1 macrolide prescription and 91.7% (95,973) to mothers prescribed 1 penicillin prescription during pregnancy. 
  • Negative controls: 
    • 82,314 children: mothers prescribed preconception macrolides (11,874) or penicillins (70,440).
    • 53,735 study cohort siblings with prenatal macrolide (4512) or penicillin (49,223) exposure. 
  • Major malformations prevalence per 1000 live births:
    • First-trimester macrolides: 27.65. 
    • Second/third-trimester macrolides: 19.5.
    • First-trimester penicillin: 17.65.
    • Second/third-trimester penicillin: 17.27.
  • With first-trimester macrolides, adjusted risk ratios (aRRs; 95% CIs):
    • Any malformation: 1.55 (1.19-2.03);
    • Cardiovascular malformation: 1.62 (1.05-2.51).
  • With macrolides in any trimester:
    • Genital malformations: aRR, 1.58 (95% CI, 1.14-2.19).
  • With erythromycin in first trimester: 
    • Any malformation: aRR, 1.50 (95% CI, 1.13-1.99).

Study design

  • Retrospective, population-based cohort analysis evaluating association between macrolide antibiotics during pregnancy and risk for major malformations, neurodevelopmental disorders in their children. 
  • Funding: National Institute for Health Research.

Limitations

  • Bias toward the null, effect dilution.
  • Inability to capture antibiotic dispensing/use.
  • Outcomes misclassifications, birth bias.