key: cord-0870179-npzzycis authors: Pirola, Carlos J.; Sookoian, Silvia title: Estimation of RAAS-Inhibitor effect on the COVID-19 outcome: A Meta-analysis date: 2020-05-28 journal: J Infect DOI: 10.1016/j.jinf.2020.05.052 sha: 7811d81b40da4aba995ff31049b9f404c3272c29 doc_id: 870179 cord_uid: npzzycis BACKGROUND AND RATIONALE: Some studies of hospitalized patients suggested that the risk of death and/or severe illness due to COVID-19 is not associated with the use of angiotensin-converting enzyme inhibitors (ACEIs) and/or angiotensin II receptor type 1 blockers (ARBs). Nevertheless, some controversy still exists and there is limited information of the ACEIs/ARBs effect size on COVID-19 prognosis. AIM AND METHODS: We aimed to measure the effect of ACEIs and/or ARBs on COVID-19 severe clinical illness by a meta-analysis. Literature search included all studies published since the COVID-19 outbreak began (December 2019) until May 9, 2020. We analyzed information from studies that included tested COVID-19 patients with arterial hypertension as comorbidity prior to hospital admission and history of taking ACEIs, ARBs, or ACEIs/ARBs. RESULTS: We included 16 studies that involved 24,676 COVID-19 patients, and we compared patients with critical (n = 4134) vs. non-critical (n = 20,542) outcomes. The overall assessment by estimating random effects shows that the use of ACEIs/ARBs is not associated with higher risk of in-hospital-death and/or severe illness among hypertensive patients with COVID-19 infection. On the contrary, effect estimate shows an overall protective effect of RAAS inhibitors/blockers (ACEIs, ARBs, and/or ACEIs/ARBs) with ∼ 23 % reduced risk f death and/or critical disease (OR: 0.768, 95%CI: 0.651-0.907, p=0.0018). The use of ACEIs (OR:0.652, 95%CI:0.478-0.891, p=0.0072) but not ACEIs/ARBs (OR:0.867, 95%CI:0.638-1.179, p =NS) or ARBs alone (OR:0.810, 95%CI:0.629-1.044, p=NS) may explain the overall protection displayed by RAAS intervention combined. CONCLUSION: RAAS inhibitors might be associated with better COVID-19 prognosis. Although COVID-19 pandemic is only a few months old, the magnitude of clinical information regarding the disease spectrum is overwhelming. ACE2 (angiotensin-converting enzyme 2) is presumably the host receptor of the novel SARS-CoV-2 coronavirus (1) . Although the effect is significantly reduced by adjusting by age (2) , arterial hypertension seems to be one of the most common risk factors associated with COVID-19 mortality, (3;4) . In fact, 56.6% of a case series of 5700 patients with COVID-19 admitted to 12 hospitals in New York City (3) and 30% of patients with COVID-19 in Wuhan, China (4) presented arterial hypertension as comorbidity. Therefore, the effect/s of angiotensin-converting enzyme inhibitors (ACEIs) and/or angiotensin II receptor type 1 blockers (ARBs) on the clinical course of the disease have been on the top of clinical debates owing to the putative regulation of ACE2 exerted by these drugs (5) . Four large studies, including hospitalized patients from Europe and USA (6) (7) (8) (9) convincingly demonstrated that the risk of severe COVID-19 and/or in-hospital death among those infected is not associated with the use of ACEIs and/or ARBs. Likewise, results from large studies from Asia suggested that it is unlikely that in-hospital use of ACEIs/ARBs is associated with increased COVID-19 mortality risk (10;11) . While the evidence shows consistent results, there is limited information on the ACEIs/ARBs effect size on the COVID-19 prognosis. Hence, the primary objective of the current study is to provide a quantitative estimation of the effect of ACEIs and/or ARBs, alone or in combination on COVID-19 severe clinical illness in patients with arterial hypertension by a meta-analysis. We followed the appropriate method for conducting a meta-analysis of observational studies (MOOSE) ( Supplementary Table S1 ). The literature search included all studies published since the COVID- 19 "COVID-19 AND hypertension AND RAAS". Figure S1 . The authors (CJP and SS) reviewed all abstracts independently to determine the alignment with the eligibility criteria, or to establish the appropriateness of the research topic. If these criteria were met, the article was retrieved and reviewed in its entirety. There were no discrepancies in this process. The following meta-analysis inclusion criteria were considered when assessing the eligibility of the identified studies: Observational studies of hospitalized patients with confirmed COVID-19 infection that: 1) included COVID-19 patients with arterial hypertension as comorbidity prior to hospital admission and history of taking ACEIs, ARBs or ACEIs/ARBs (the Authors did not disclose individual drug information) at the time of COVID-19 testing, and 2) disclosed information on clinical outcomes defined as critical or fatal versus non-critical disease. A random effect model was adopted when summarising statistical synthesis; this model assumes that the treatment effect is not the same across all studies included in the analysis. For each analysis, a forest plot was generated to display results. Heterogeneity was evaluated via the Q statistic and I 2 statistic, which is a transformation of Q that estimates the percentage of the variation in effect sizes that is due to heterogeneity. As an I 2 value of 0% indicated no observed heterogeneity, greater values denoted increasing heterogeneity. Subgroup analyses were performed to determine the presence of potential heterogeneity sources. We identified characteristics that allowed the studies to be stratified into subsets with homogeneous effects. As we hypothesized that the RAAS inhibitor class, ethnicity, and peer-reviewed process may provide an important source of variability, the estimate of the average effect of the studies was additionally stratified by these moderator variables. To identify studies yielding findings that had a disproportionately significant influence on the effect estimate, we repeated the analysis after removing one study at a time. We performed a visual inspection of funnel plots, but publication bias was formally tested by using the Begg and Mazumdar's rank correlation test and Egger's method. Statistical significance was assumed for p  0.05. All calculations were performed using the Comprehensive Meta-Analysis computer program (Biostat, Englewood, NJ, USA). The quality of the studies included in the meta-analysis was assessed using The Table S2 ). Following the previously described search strategy, 29 articles were initially identified as potentially relevant for the present investigation, based on the assessment of the titles and abstracts. We excluded thirteen studies because they did not meet all the inclusion criteria (Supplementary Figure S1) . Thus, the remaining 16 studies were included in the metaanalysis (3;6-20) , which scored well in terms of the selection criteria, comparability of critical and non-critical COVID-19 on the basis of the design or analysis, and ascertainment of exposure (Supplementary Table S2 ). We included 16 studies that involved 24,676 COVID-19 patients, and we compared patients with critical (n = 4134) vs. non-critical (n = 20,542) outcomes. The study characteristics, including the clinical criterion used for the differentiation between critical and non-critical patients, are shown in Table 1 and one study included data extracted from an international registry (7). Complete details of the study design and sample sizes are fully disclosed in Table 1 . The overall assessment by estimating random effects shows that the use of ACEIs/ARBs is not associated with a higher risk of in-hospital death and/or severe illness among Based upon the results yielded by a comprehensive analysis of the results reported by 16 published studies, we presented robust evidence on the lack of association between the use of RAAS inhibitors/blockers and COVID-19 severe clinical illness. In addition, our findings demonstrated that the use of ACEIs/ARBs is associated with potential protective effects on the COVID-19 prognosis. The analysis focused on the estimation of the individual effect size of each group of drugs, including ACEIs, ARBs, or indistinct drug (ACEIs/ARBs) suggested that the protective effect of RAAS inhibitors against severe COVID-19 illness may be explained by the use of ACEIs. It is worth noting, however, that none of the studies included in this meta-analysis were randomized trials. Thus, many unmeasured confounding factors could not be assessed. Finally, one could speculate on any key differential effect of ACEIs on the pathophysiology of severe COVID-19. Nevertheless, the lack of complete knowledge on the mechanism/s behind critical COVID-19 illness jeopardizes the plausibility of any biological hypothesis, including the question of whether the ACEIs or ARBs-mediated reduction of the angiotensin II production or the AT1R activation might explain the clinical observations. Both drug classes seem to up-regulate ACE2 expression in relevant organs (22) , and its implications in COVID-19 outcomes have been largely discussed (5;23). There is one remarkable aspect that could not be specifically weighted in our meta-analysis, which is the analysis of comorbidities and effect sizes for the individual treatment or the coadministration of ACEIs and ARBs in elderly patients. Patients with suboptimal control of blood pressure with any of the drug classes, included those in the reference groups (non-RAAS inhibitors), might also influence the explored outcomes. Some limitations of our study, which are implicit in the studies included, have been mentioned but should be emphasized. Indeed, there are limitations and potential sources of heterogeneity imposed by the quality of the observational data. For instance, although many reports used age and sex-matched patients, potential confounders and selection bias, not only regarding the patients but also treatment comparisons could not be assessed because of insufficient information. By meta-regression, the average age of the studied populations did not explain the results, but a nondisclosure difference between the age of treated and untreated with RAAS inhibitor groups cannot be ruled out. Notably, substantial heterogeneity was present within most studies from North America and Europe but not among studies from China. We could not identify the sources of heterogeneity among studies involving non-Asian COVID-19 patients. However, there are many potential explanations, from differences in doses of antiviral drugs and/or interventions for the treatment of severe COVID-19 to differences in recruitment and timing of outcomes measurements. Furthermore, characteristics of the studies (for example, methodological differences in the study design), or even differences at the population level (such as unknown environmental factors and/or underlying disease comorbidities), are certainly highly important variables that may explain the heterogeneity of the dataset as a whole. Unfortunately, as the authors of a large majority of studies included in the meta-analysis did not report the findings for male and female patients separately, we were unable to perform stratification of the results by sex. Consequently, the potential presence of sexual dimorphism could not be explored. Likewise, the effect of potential confounder risk factors, such as obesity and/or type 2 diabetes, which might probably co-exist with arterial hypertension, could not be assessed as the potential source of heterogeneity because of lack of information in the original studies. Finally, the quality of the studies retrieved from online repositories might be compromised because preprints are preliminary reports of work that have not been certified by peer review. Surprisingly, sensitivity analysis of studies published in peerreviewed journals vs. preprint reports showed no heterogeneity between the latter. More studies are needed to ensure that our results can be generalizable to all populations. It seems relevant to replicate and confirm these findings in well-controlled studies with clear disclosure of co-variables to provide not only accurate clinical recommendations for patients with COVID-19 but also precise estimates of the treatment effects. patients receiving ACEIs, ARBs, or ACEIs/ARBs without discrimination. For the dichotomous variable (critical / non-critical), the effect denotes odds ratio (OR) and corresponding 95% confidence interval (CI). Because of the presence of heterogeneity, a random effect model was adopted to estimate the pooled ORs. This model assumes that the treatment effect is not the same across all studies included in the analysis. The first author of the study is shown under the sub-heading "study name." Popul: indicates the use of ACEIs, ACEIs/ARBs, or ARBs. In the graph, the filled squares denote the effect of individual studies, and filled diamonds express combined fixed and random effects. The first author of the removed study is shown under the sub-heading "study name." Popul: indicates the use of ACEIs, ACEIs/ARBs, or ARBs. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor Age but not sex may explain the negative effect of arterial hyptertension and diabetes on COVID-19 prognosis Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Renin-Angiotensin System Blockers and the COVID-19 Pandemic: At Present There Is No Evidence to Abandon Renin-Angiotensin System Blockers Renin-Angiotensin-Aldosterone System Blockers and the Risk of Covid-19 Cardiovascular Disease, Drug Therapy, and Mortality in Covid-19 Association of Use of Angiotensin-Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers With Testing Positive for Coronavirus Disease 2019 (COVID-19) Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Covid-19 Association of Renin-Angiotensin System Inhibitors With Severity or Risk of Death in Patients With Hypertension Hospitalized for Coronavirus Disease 2019 (COVID-19) Infection in Wuhan, China Association of Inpatient Use of Angiotensin Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers with Mortality Among Patients With Hypertension Hospitalized With Hypertension and Renin-Angiotensin-Aldosterone System Inhibitors in Patients with Covid-19. medRxiv preprint ACEinhibitors and Angiotensin-2 Receptor Blockers are not associated with severe SARSCOVID19 infection in a multi-site UK acute Hospital Trust. medRxiv RAAs inhibitors and outcome in patients with SARS-CoV-2 pneumonia. A case series study COVID-19 with Different Severity: A Multi-center Study of Clinical Features Cardiovascular Implications of Fatal Outcomes of Patients With Coronavirus Disease 2019 (COVID-19) Anti-hypertensive Angiotensin II receptor blockers associated to mitigation of disease severity in elderly COVID-19 patients. medRxiv preprint Renin-angiotensin system inhibitors improve the clinical outcomes of COVID-19 patients with hypertension Effects Of ARBs And ACEIs On Virus Infection, Inflammatory Status And Clinical Outcomes Patients With Hypertension: A Single Center Retrospective Study. YANG2020. Hypertension Hypertension in patients hospitalized with COVID-19 in Wuhan, China: A single-center retrospective observational study Angiotensin-converting enzyme 2 (ACE2) mediates influenza H7N9 virus-induced acute lung injury Effect of angiotensin-converting enzyme inhibition and angiotensin II receptor blockers on cardiac angiotensin-converting enzyme 2 Angiotensin receptor blockers as tentative SARS-CoV-2 therapeutics