Article Text

Download PDFPDF

Beware of simple explanations
  1. Tabassome Simon1,2,
  2. Nicolas Danchin3
  1. 1 Department of Clinical Pharmacology and Clinical Research Platform (URCEST-CRC-CRB), APHP.SU; Hôpital Saint Antoine, Assistance Publique- Hôpitaux de Paris, FACT French Alliance for Cardiovascular clinical Trials, Paris, France
  2. 2 Clinical Pharmacology, Sorbonne Universite, Paris, Île-de-France, France
  3. 3 Department of Cardiology, Hôpital Européen Georges Pompidou, Assistance Publique-Hôpitaux de Paris, and Université de Paris, Paris, France
  1. Correspondence to Dr Tabassome Simon, Department of Clinical Pharmacology and Clinical Research Platform (URCEST-CRC-CRB), APHP.SU; Hôpital Saint Antoine, Assistance Publique-Hôpitaux de Paris, FACT French Alliance for Cardiovascular clinical Trials, Paris, France; tabassome.simon{at}

Statistics from

Faced with the emergence of a new disease, and particularly of a disease of pandemic proportions, understanding the pathophysiological basis of the disease is fundamental and is the very essence of research. But the danger, particularly in an urgent situation, is to take therapeutic measures or to take public health measures, based solely on basic research data. The debate on the interaction between renin-angiotensin-system (RAS) blockers and COVID-19 is an excellent illustration of this risk.

ACE inhibitors and angiotensin receptor blockers (ARB) play a central role in the management of cardiovascular disease, including hypertension, diabetes, coronary artery disease and heart failure and are used in a considerable number of patients. Myocardial injury is common in patients with COVID-19 disease, particularly those with diabetes or hypertension, and may contribute to increased risk in these patients.1

As RAS inhibitors are commonly prescribed in patients with cardiovascular conditions, who are precisely at increased risk of complications in the case of SARS-COV-2 infection, it has been suggested that those medications could have a deleterious role in this context. Among the mechanisms most frequently put forward is the link between the ACE2 receptors and host cell infection.2

ACE2 is present in the organism as membrane-bond and in a soluble form. It is a membrane glycoprotein found in the heart, vascular endothelium, kidney and also in lung parenchyma, bronchus and small intestine. Experimental studies suggest that RAS blockers can upregulate ACE2 expression.2 3 It has been shown that membrane-bound ACE2 mediates SARS-COV-2 infection via spike (S) protein binding, facilitating viral entry into host cells, which, however, requires the preliminary action of another enzyme, transmembrane protease serine S2 (TMPRSS2). On the other hand, the main substrate of ACE-2 is angiotensin II. ACE2 converts angiotensin II to angiotensin 1–7 and angiotensin 1 to angiotensin 1–9. Angiotensin 1–7 has anti-inflammatory, vasodilatory and antioxidant properties, and increased expression of ACE2 might also have beneficial effects, notably against inflammation and thrombosis.

The interference between SARS-COV 2 and ACE2 might therefore play a role in the pathogenesis of the disease by facilitating virus propagation and also in the protection against cardiovascular damage resulting from infection or both prescribing RAS inhibitors.

Based on these data, antagonist conclusions have been drawn by clinicians, that is, some asking for systematic prescription while others for stopping RAS inhibitors, and released in public media. Several studies are currently investigating whether RAS inhibition can improve prognosis in patients with COVID-19 disease (Hypertension in Patients Hospitalized With COVID-19 (HT-COVID19), NCT04318301; ACE Inhibitors, Angiotensin II Type-I Receptor Blockers and Severity of COVID-19 (COVID-ACE), NCT04318418); Losartan for Patients With COVID-19 Requiring Hospitalization, NCT04312009) and those who are ambulatory(Losartan for Patients With COVID-19 Not Requiring Hospitalization, NCT04311177). In opposite, others have proposed that upregulation of ACE2 by RAS blockers might facilitate disease progression and severity, irrespective of the fact that RAS blockers have no action on TMPRSS2,4 and investigate the effect of RAS medication replacement or withdrawal (Coronavirus [COVID-19] ACEi/ARB Investigation (CORONACION), NCT04330300; ACE inhibitors or ARBs discontinuation in Context of SARS-CoV-2 Pandemic [ACORES-2], NCT04329195; The Randomized Elimination or Prolongation of Angiotensin Converting Enzyme Inhibitors and Angiotensin Receptor Blockers in Coronavirus Disease 2019 trial (REPLACE COVID-19), NCT04338009), hypothesising that RAS antagonist drugs withdrawal or substitution might improve patient outcome.

In this context, and pending the results of randomised trials, Flacco and colleagues provide an important meta-analysis on the association between ACE inhibitor or ARB treatment and the risk of severe or lethal COVID-19.5 Ten studies, all of them observational, were included. The studies were a mix of patient cohorts and case-control analyses and included mostly patients who were hospitalised, although some were population-based and included office patients. The data retrieved by Flacco et al were adjusted for potential confounders. The conclusion seems straightforward: ACE inhibitors or ARBs do not appear to have any deleterious nor beneficial effects in patients with COVID-19 disease in terms of disease severity, whatever the stage of the disease assessed. One of the studies included in the meta-analysis,6 however, analysed separately testing positive for SARS-COV-2 (versus testing negative), hospital admission (versus no admission) and among hospitalised patients, admission to intensive care unit (ICU) (versus no ICU admission); the use of ACE inhibitors or ARBs was not associated with the results of SARS-COV-2 testing, while both hospital admissions and ICU admissions were more frequent in patients receiving ACE inhibitors or ARBs. Recently, a yet unpublished analysis of the Kaiser Permanente Washington database7 confirmed that use of ACE inhibitors or ARBs was not associated with an increased risk of being tested positive (adjusted OR 0.94, 95% CI 0.75 to 1.16), whatever the dose of ACE inhibitors or ARBs used. In contrast with the Cleveland Clinic cohort,6 however, use of ACE inhibitors or ARBs was not associated with hospitalisation for COVID-19 infection (adjusted OR 0.92, 95% CI 0.57 to 1.49); for patients on the highest doses of RAS inhibitors, the adjusted OR for hospitalisation was 0.86, 95% CI 0.47 to 1.57). Another study using the Danish database showed results similar to those of the Kaiser Permanente:8 ACE inhibitor or ARB use was not associated with an increased risk of severe disease in patients with COVID-19 disease (HR 0.83, 95% CI 0.67 to 1.03) and was not associated with an increased risk of developing COVID-19 in patients treated for hypertension (adjusted HR 1.05, 95% CI 0.60 to 1.30).

Although the data used for the meta-analysis were adjusted for potential confounders, it should be kept in mind that the quality of adjustments may have varied form one study to another, mainly depending on the granularity of the clinical information available in each database. For instance, left ventricular function, a major prognostic factor for cardiovascular outcomes, highly correlated with the prescription of ACE inhibitors, was not available. Maybe more importantly still, in most instances, the precise time course of the infection could not be recorded; in particular, in cohorts of hospitalised patients, the time from disease onset to hospitalisation (as well as the potential risk of out-of-hospital death before hospitalisation or in-hospital death before diagnosis could be formally established) were not known, with the possibility of an immortal time bias (ie, patients dying before inclusion were de facto excluded from the study). In other words, even the most careful analysis of observational data, using up-to-date adjustment techniques, cannot achieve the level of certainty of a randomised controlled trial within a finite population perimeter. This obviously does not mean that observational data are worthless, since when analysed carefully, they provide information far superior to subjective clinical judgement.

In conclusion, jumping to therapeutic conclusions on the sole basis of pathophysiological or experimental considerations is hazardous. Nature can be tricky, and pathways that are blocked by a therapeutic intervention are often short-circuited by other, secondary, pathways that will result in the lack of clinical effect of the said therapeutic intervention. Here, in spite of the recognised interference between ACE2 and SARS-COV-2, and of the possible link between RAS blockers and ACE2 in humans, these medications appeared clinically neutral. Until we get the results of the ongoing randomised trials in different types of COVID-19-positive populations, it therefore seems prudent, as suggested by Flacco and colleagues, not to withdraw treatments that have demonstrated cardiovascular benefits, such as ACE inhibitors, ARBs or mineralocorticoid receptor antagonists (MRAs) in COVID-19 patients. This is precisely what learnt societies propose.3 9



  • Editor's note The linked article to which this Editorial refers can be found in Volume 106, Issue 19 of Heart.

  • Contributors The authors contributed equally in drafting this editorial.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests TS reports grants from AstraZeneca, Bayer, Daichi-Sankyo, GSK, MSD, Novartis and personnel fees from AstraZeneca, BMS, Novartis and Sanofi. ND has received grants, personal fees and non-financial support from Amgen, AstraZeneca, Bayer, BMS, Sanofi, personal fees from Boehringer Ingelheim, Intercept, MSD, Novo Nordisk, Pfizer, Servier and UCB, all outside the submitted work.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles