Renin-angiotensin system blockers and susceptibility to COVID-19: a multinational open science cohort study

Daniel R. Morales, Mitchell M. Conover, Seng Chan You, Nicole Pratt, Kristin Kostka, Talita Duarte‐Salles, Sergio Fernández‐Bertolín, María Aragón, Scott L. DuVall, Kristine E. Lynch, Thomas Falconer, Kees van Bochove, Cynthia Sung, Michael E. Matheny, Christophe G Lambert, Fredrik Nyberg, Thamir M. Alshammari, Andrew E. Williams, Rae Woong Park, James Weaver, Anthony G. Sena, Martijn J. Schuemie, Peter R. Rijnbeek, Ross D. Williams, Jennifer C. E. Lane, Albert Prats‐Uribe, Lin Zhang, Carlos Areia, Harlan M. Krumholz, Daniel Prieto‐Alhambra, Patrick Ryan, George Hripcsak, Marc A. SuchardmedRxiv 2020 · DOI

Abstract

There is no clinically significant increased risk of COVID-19 diagnosis or hospitalization with ACE or ARB use. Users should not discontinue or change their treatment to avoid COVID-19.

Introduction

Early in the COVID-19 pandemic, a biological hypothesis suggested that drugs interacting with the Renin-Angiotensin System (RAS)—specifically Angiotensin-Converting Enzyme inhibitors (ACEs) and Angiotensin Receptor Blockers (ARBs)—might increase susceptibility to the SARS-CoV-2 virus. The concern was that these medications could upregulate ACE2 expression, the receptor used by the virus to enter human cells. Given the prevalence of hypertension and the widespread use of these drugs, clarifying this risk was a global health priority.

Methods

This study employed an open-science, federated cohort design to compare ACE/ARB users against patients taking other first-line antihypertensives, such as Calcium Channel Blockers (CCBs) and Thiazide Diuretics (THZs). This active comparator design helps mitigate "confounding by indication" by ensuring compared groups have similar underlying health profiles.

Study design schematic illustrating the timelines for cohort eligibility, exposure assessment, covariate adjustment, and outcome follow-up.

Key methodological components included:

  • Large-Scale Propensity Score (PS) Adjustment: Regularized regression was used to balance cohorts across thousands of potential confounding variables (age, sex, comorbidities, etc.).
  • Empirical Calibration: The researchers used 76 "negative control" outcomes to detect and adjust for systematic bias, providing a higher level of statistical confidence than nominal p-values.
  • Federated Analytics: Using the OHDSI network and the OMOP Common Data Model, identical analysis code was executed across international databases without sharing sensitive patient-level data.

Datasets

The analysis utilized three major electronic health record (EHR) and claims databases, encompassing over 1.1 million antihypertensive users:

  1. SIDIAP (Spain): Primary care records covering ~80% of Catalonia's population.
  2. CUIMC (USA): EHR data from Columbia University Irving Medical Center in New York.
  3. VA-OMOP (USA): Data from the US Department of Veterans Affairs, covering millions of veterans across 170 medical centers.

Results

The study found no evidence that ACE or ARB use increases the risk of COVID-19 infection or severe outcomes.

  • Infection Risk: The meta-analytic Hazard Ratio (HR) for COVID-19 diagnosis comparing ACE/ARB users to CCB/THZ users was 0.98 (95% CI 0.84 - 1.14), indicating no significant difference.
  • Disease Severity: No significant differences were observed regarding COVID-19-related hospitalization or pneumonia.
  • ACE vs. ARB: Direct comparison showed a marginally lower risk of diagnosis for ACE users (HR 0.88), though this was inconsistent across outcomes and likely reflects residual bias rather than a true clinical advantage.

Discussion and Conclusion

While the "ACE2 upregulation" theory was biologically plausible in laboratory settings, this large-scale study demonstrates it does not translate into clinical risk for patients. The consistency of results across diverse healthcare systems in Spain and the US reinforces the generalizability of these findings.

Clinical guidelines advising patients to continue their hypertension treatments during the pandemic are strongly supported by this evidence. Future research may still explore the basic biological mechanisms of RAS therapy on tissue-specific ACE2 expression, but the clinical question of susceptibility appears resolved.

Summary updated

Abstract

There is no clinically significant increased risk of COVID-19 diagnosis or hospitalization with ACE or ARB use. Users should not discontinue or change their treatment to avoid COVID-19.
BibTeX
@misc{morales2020renin,
  author = {Daniel R. Morales and Mitchell M. Conover and Seng Chan You and Nicole Pratt and Kristin Kostka and Talita Duarte‐Salles and Sergio Fernández‐Bertolín and María Aragón and Scott L. DuVall and Kristine E. Lynch and Thomas Falconer and Kees van Bochove and Cynthia Sung and Michael E. Matheny and Christophe G Lambert and Fredrik Nyberg and Thamir M. Alshammari and Andrew E. Williams and Rae Woong Park and James Weaver and Anthony G. Sena and Martijn J. Schuemie and Peter R. Rijnbeek and Ross D. Williams and Jennifer C. E. Lane and Albert Prats‐Uribe and Lin Zhang and Carlos Areia and Harlan M. Krumholz and Daniel Prieto‐Alhambra and Patrick Ryan and George Hripcsak and Marc A. Suchard},
  title = {{Renin-angiotensin system blockers and susceptibility to COVID-19: a multinational open science cohort study}},
  year = {2020},
  doi = {10.1101/2020.06.11.20125849},
  howpublished = {medRxiv},
}
Figure