Clarithromycin plus DOACs in older patients: increased major hemorrhage risk

  • Hill K & al.
  • JAMA Intern Med
  • 08/06/2020

  • Liz Scherer
  • Clinical Essentials
L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano

Takeaway

  • Clarithromycin use in older patients taking direct oral anticoagulants (DOACs) increases risk for major hemorrhage requiring hospitalization vs using azithromycin.

Why this matters

  • Clinicians should assess individual hemorrhage risk, anticoagulation needs, and antibiotic substitutes when co-prescribing DOACs and macrolides.
  • In situations in which concomitant clarithromycin/DOACs are inevitable, monitor to prevent DOACs supratherapeutic levels.

Key results

  • 24,943 patients; 36.2% (9025) ages 66-75 years.
  • 26.4% received clarithromycin, 73.6% azithromycin.
  • DOACs: rivaroxaban 40.0%, apixaban 31.9%, dabigatran 28.1%.
  • Mean (standard deviation) duration of DOAC use before antibiotic administration: 390 days (0.11) before azithromycin vs 353 days (0.11) before clarithromycin.
  • Incident hemorrhage rate (95% CI) was higher with clarithromycin vs azithromycin:
    • 204.8 (191.3-219.7) vs 133.7 (127.0-140.8).
  • In a self-controlled case series analysis, for periods of clarithromycin use (145 events) vs nonuse (1615 events), the rate ratio was 1.64 (95% CI, 1.35-1.98).
  • 30-day major hemorrhage rate requiring hospital admission in patients taking clarithromycin vs azithromycin:
    • Adjusted HR, 1.71 (95% CI, 1.20-2.45).

Study design

  • Retrospective, population-based cohort evaluation of elevated bleeding risk among older patients age ≥66 years taking DOACs with concomitant clarithromycin vs azithromycin, June 23, 2009 to December 31, 2016.
  • Funding: Heart and Stroke Foundation of Canada. 

Limitations

  • Limited generalizability.
  • Small number of bleeding events.
  • Dosage adjustments adherence unaccounted for. 
  • Unmeasured confounding.