COVID-19: international results add evidence to safety of RAS inhibitors

  • Mehra MR & al.
  • N Engl J Med
  • 1 May 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.

This article has been retracted.

Takeaway

  • Renin-angiotensin system (RAS) inhibitor use in patients with cardiovascular (CV) disease and hospitalized with COVID-19 was not associated with mortality risk in this large international study.
  • Use of angiotensin-converting enzyme inhibitors (ACEis) was tied to a greater likelihood of discharge, but use of angiotensin receptor blockers (ARBs) was not.

Why this matters

  • Editorial: these and other findings provide further support for continuing RAS inhibitor therapy in patients with COVID-19.

Key results

  • 8910 patients, 169 hospitals. 
  • Mean age: 49 years; 40.0% women.
  • Mean length of stay: 10.7±2.7 days.
  • Overall hospital mortality: 5.8% (n=515).
  • On multivariable analysis, in-hospital death was associated with (ORs; 95% CIs):
    • Age >65 years: 1.93 (1.60-2.41).
    • Coronary artery disease: 2.70 (2.08-3.51).
    • Congestive heart failure: 2.48 (1.62-3.79).
    • Cardiac arrhythmia: 1.95 (1.33-2.86).
    • COPD: 2.96 (2.00-4.40). 
    • Current smoking: 1.79 (1.29-2.47).
  • Immunosuppression, race/ethnicity, hyperlipidemia, and diabetes were not independently predictive of hospital death.
  • Discharge odds (ORs; 95% CIs) were increased with:
    • Female sex: 0.79 (0.65-0.95);
    • Statin use: 0.35 (0.24-0.52); and
    • ACEis: 0.33 (0.20-0.54).
    • However, use of ARBs was not significantly associated: 1.23 (0.87-1.74).

Study design

  • Multinational, multicenter, retrospective analysis of associations of CV disease and related drug therapies with COVID-19 mortality among hospitalized patients.
  • Funding: None disclosed.

Limitations

  • Retrospective.
  • Unidentified confounders.
  • Primary hypothesis lacking.