Angiotensin II receptor blocker or angiotensin-converting enzyme inhibitor use and COVID-19-related outcomes among US Veterans.

BACKGROUND: Angiotensin II receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACEIs) may positively or negatively impact outcomes in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. We investigated the association of ARB or ACEI use with coronavirus disease 2019 (COVID-19)-related outcomes in US Veterans with treated hypertension using an active comparator design, appropriate covariate adjustment, and negative control analyses. METHODS AND
FINDINGS: In this retrospective cohort study of Veterans with treated hypertension in the Veterans Health Administration (01/19/2020-08/28/2020), we compared users of (A) ARB/ACEI vs. non-ARB/ACEI (excluding Veterans with compelling indications to reduce confounding by indication) and (B) ARB vs. ACEI among (1) SARS-CoV-2+ outpatients and (2) COVID-19 hospitalized inpatients. The primary outcome was all-cause hospitalization or mortality (outpatients) and all-cause mortality (inpatients). We estimated hazard ratios (HR) using propensity score-weighted Cox regression. Baseline characteristics were well-balanced between exposure groups after weighting. Among outpatients, there were 5.0 and 6.0 primary outcomes per 100 person-months for ARB/ACEI (n = 2,482) vs. non-ARB/ACEI (n = 2,487) users (HR 0.85, 95% confidence interval [CI] 0.73-0.99, median follow-up 87 days). Among outpatients who were ARB (n = 4,877) vs. ACEI (n = 8,704) users, there were 13.2 and 14.8 primary outcomes per 100 person-months (HR 0.91, 95%CI 0.86-0.97, median follow-up 85 days). Among inpatients who were ARB/ACEI (n = 210) vs. non-ARB/ACEI (n = 275) users, there were 3.4 and 2.0 all-cause deaths per 100 person months (HR 1.25, 95%CI 0.30-5.13, median follow-up 30 days). Among inpatients, ARB (n = 1,164) and ACEI (n = 2,014) users had 21.0 vs. 17.7 all-cause deaths, per 100 person-months (HR 1.13, 95%CI 0.93-1.38, median follow-up 30 days).
CONCLUSIONS: This observational analysis supports continued ARB or ACEI use for patients already using these medications before SARS-CoV-2 infection. The novel beneficial association observed among outpatients between users of ARBs vs. ACEIs on hospitalization or mortality should be confirmed with randomized trials.

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) binds to angiotensin-converting enzyme 2 (ACE2) to enter and infect host cells [1]. Under homeostasis, ACE2 converts angiotensin II to angiotensin-(1–7), reducing blood pressure and inflammation [2-6]. Angiotensin II receptor blockers (ARBs) and ACE inhibitors (ACEIs) inhibit the renin-angiotensin system and may differentially influence ACE2 expression in select organs [7, 8]. During the ongoing pandemic, there has been unprecedented interest in whether ARBs and ACEIs are beneficial or harmful in patients with SARS-CoV-2 infection or coronavirus disease 2019 (COVID-19) [9].

At least twelve randomized clinical trials are testing the effects of ARB or ACEI continuation, initiation, or discontinuation on COVID-19 outcomes, two of which are now completed and reported no increased risk of harm with continuation of ARB or ACEI therapy among hemodynamically stable adults hospitalized for COVID-19 [10, 11]. Numerous retrospective studies examined the association of ARBs and ACEIs with COVID-19 outcomes using case-control and historical cohort study designs and generally reported no excess risk of SARS-CoV-2 infection or COVID-19 severity among ARB or ACEI users compared to non-users, with some suggesting potential benefit from ARBs or ACEIs [12]. However, several of the prior observational studies contain methodological limitations that may reduce their internal validity, and thus, their clinical applicability [13, 14]. Despite the volume of published observational analyses to date (more than 80 as of Feb 15, 2020) [12], focused, high-quality observational studies are needed to complement randomized trials and guide ARB or ACEI use during the growing COVID-19 pandemic. Unfortunately, due to the short duration of the pandemic limiting sample size, it is not feasible to perform a new-user design, which is considered best practice for estimating medication effects outside of randomized trials [15]. However, unlike many prior observational analyses, we utilized several methods to increase robustness of results including an active comparator design, rigorous adjustment for a rich set of covariates, and negative control analyses to assess for residual bias.

ARBs and ACEIs are among the most commonly prescribed medications in the United States (US). The potential risks of abrupt discontinuation are enormous, and their effects on the ongoing COVID-19 pandemic are not clear. Our goal was to estimate the direction and magnitude of association between ARB and ACEI use with COVID-19-related outcomes among US Veterans with treated hypertension in the outpatient and inpatient settings using Veterans Health Administration (VHA) data and robust observational methods.

Methods

Data source and extraction

We used outpatient and inpatient demographic, pharmacy, clinical, and healthcare utilization data from the national VHA to conduct this retrospective cohort study. The study was pre-registered with ClinicalTrials.gov (NCT04467931), where the protocol is publicly available. The University of Utah institutional review board and Salt Lake City Veterans Affairs Health Care System Research and Development Office, approved the study with a waiver of informed consent.

Cohort derivation

We derived separate outpatient and inpatient cohorts for analyses. Veterans were eligible for inclusion as outpatients if they had a positive SARS-CoV-2 test performed between January 19, 2020 (the day before the first confirmed COVID-19 case in the US) and October 15, 2020 but were not hospitalized in the seven days before the positive test (S1 Fig, Panel A); the date of the positive SARS-CoV-2 test was the index date. Veterans were eligible for inclusion as inpatients if they had a positive SARS-CoV-2 test performed between January 19, 2020 and October 15, 2020 and were hospitalized for COVID-19; the index date was the first hospital admission date on or after the date where the Veteran started receiving SARS-CoV-2 testing (S1 Fig, Panel B). We identified COVID-19 cases using the definitions developed by the VHA for public reporting [16, 17]. Nursing home residents could be included in the outpatient and inpatient cohorts if meeting all other eligibility criteria.

We excluded Veterans who did not meet one-year continuous enrollment criteria prior to the index date or who had data inconsistencies (e.g., not alive on index date or multiple death dates). Finally, we restricted the analyses to Veterans with treated hypertension, defined as ≥1 inpatient or ≥2 outpatient encounters with an International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) code (401.x, 403.0x, 403.1x, 403.9x, I10, I12.0, or I12.9) [18] using all available data pre-index, and ≥1 outpatient pharmacy fill for an antihypertensive medication in the 104-day pre-index period (S1 Table). We selected 104 days to account for patients who receive 90-day supply prescriptions, plus a 14-day grace period to account for modest non-adherence.

Exposures

For both outpatient and inpatient cohorts, we assigned Veterans to exposure groups based on ≥1 outpatient pharmacy antihypertensive medication fill in the 104-day pre-index period. We compared two medication exposures among outpatients and inpatients, separately. First, we compared current users of any ARB and/or ACEI, irrespective of concurrent non-ARB/ACEI antihypertensive therapy, to Veterans using neither an ARB nor ACEI (current user of an ARB/ACEI-based regimen vs. non-ARB/ACEI-based regimen, S2 Table). For this comparison, because diagnoses other than hypertension (i.e., compelling indications) for ARB or ACEI use would introduce confounding by indication, we excluded Veterans with compelling indications defined in the 2017 American College of Cardiology/American Heart Association blood pressure guidelines [19] (i.e., diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke). We choose, a priori, to restrict the study population for this comparison to only those without compelling indications because we believe that the confounding by indication for this specific comparison (ACEI/ARB users compared to non-users) would be insurmountable.

Our second exposure compared current users of an ARB vs. ACEI (current user of an ARB vs. an ACEI). For this comparison, we included Veterans with and without compelling indications for ARBs and ACEIs, and we excluded Veterans who were current users of both an ARB and ACEI. For this comparison, restriction to those without compelling indications is not needed because these compelling indications are shared between both medications; ACEIs and ARBs are used interchangeably as they are thought to be equivalent in benefit and safety [19, 20]. As such, for this comparison, we included Veterans with and without compelling indications for ARBs and ACEIs, and we excluded Veterans who were current users of both an ARB and ACEI. Due to the small sample size of inpatient Veterans without compelling indications who were ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen users, we considered comparisons of these two groups to be exploratory.

Outcomes

For outpatients, the primary outcome was first occurrence of all-cause hospitalization or all-cause mortality, and secondary outcomes were all-cause hospitalization, all-cause mortality, and ICU admission, separately (definitions in S1 Table). For outpatients, Veterans were censored at the first occurrence of an outcome or October 15, 2020. For inpatients, the primary outcome was first occurrence of all-cause mortality, and secondary outcomes were ICU admission, dialysis, or mechanical ventilation, separately. For inpatients, Veterans were censored at the first occurrence of an outcome, 30 days’ follow-up, or October 15, 2020.

Covariates

We determined baseline demographics, vital signs, laboratory measurements, and concomitant medications within the one year prior to the index date (i.e., the “pre-index period”; S1 Table). We defined vital signs and laboratory results using the value closest to the index date in the pre-index period, and we defined comorbid conditions using all available claims and medical record data prior to the index date (S1 Table). We also calculated a Charlson Comorbidity Index [21].

Propensity score estimation

To adjust for measured confounding among outpatients and inpatients separately, we generated propensity scores (PS) to estimate the probability of being an ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen user and, separately, of being an ARB user vs. being an ACEI user, using logistic regression with all baseline covariates included in the PS model (S3 Table) [22]. The regions of common support were assessed using histograms of the PS between medication exposure groups (S2-S3 Figs). Next, using the PS, we calculated each Veteran’s matching weight [23] and verified covariate balance in the matching weighted populations using absolute standardized mean differences [24].

Statistical analyses

All analyses were performed in the outpatient and inpatient cohorts, separately. We compared patient characteristics between current users of ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimens and ARB users vs. ACEI users before and after weighting. To account for missing data (S4-S7 Tables), we used multiple imputation with chained equations to provide 10 imputed datasets using Rubin’s formulae [25, 26]. We estimated hazard ratios (HR) for the associations between the medication exposure groups and outcomes using matching-weighted Cox regression. The use of matching-weights provides a weighted analogue to 1:1 paired PS matching. We applied a two-sided alpha of 0.05 for all hypothesis tests, without correction for multiple comparisons.

Subgroup and sensitivity analyses

All analyses, including fitting logistic regression models to estimate PS, were repeated in subgroups according to age, sex, race-ethnicity, body mass index, and the number of antihypertensive medication classes being taken.

In sensitivity analyses, we repeated Cox regression analyses comparing medication regimens using several different covariate adjustment strategies (Supplemental Methods in ). To investigate if the observed associations were sensitive to the definition of the medication exposure window, we repeated all analyses with several alternative definitions (see S1 File). Finally, we repeated analyses using a composite of gastrointestinal bleeding or urinary tract infection as negative control outcomes.

All analyses were completed using R v.4.0.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Patient characteristics

After applying all inclusion and exclusion criteria, a total of 4,969 outpatient Veterans without compelling indications were included for analysis, of which there were 2,482 and 2,487 users of an ARB/ACEI- and non-ARB/ACEI-based antihypertensive regimen, respectively (S4 Fig). Of 13,581 outpatient Veterans with and without compelling indications, 4,877 were current users of an ARB, and 8,704 were current users of an ACEI.

Of 485 inpatient Veterans without compelling indications, there were 210 and 275 users (i.e., filled a prescription in the 104 days prior to the admission) of an ARB/ACEI- and non-ARB/ACEI-based antihypertensive regimen, respectively (S4 Fig). Of 3,178 inpatient Veterans with and without compelling indications, 1,164 were current users of an ARB, and 2,014 were current users of an ACEI.

Among both outpatients and inpatients, the majority of ARB/ACEI, non-ARB/ACEI, ARB, and ACEI users were men, obese, and had comorbid depression (S4-S7 Tables). The majority of outpatients and inpatients who were ARB users and ACEI users also had diabetes and/or chronic kidney disease. There were no differences in baseline characteristics after weighting (S5 Fig, Panels A-D, all absolute standardized mean differences <0.1).

Over a median 87 days of follow-up among outpatients, there were 5.0 vs. 6.0 all-cause hospitalizations or deaths per 100 person-months among ARB/ACEI-based users vs. non-ARB/ACEI-based regimen users (HR 0.85, 95% CI 0.73–0.99; and ). When comparing ARB users vs. ACEI users in outpatients, there were 13.2 vs. 14.8 all-cause hospitalizations or deaths per 100 person-months over a median 85 days of follow-up (HR 0.91, 95% CI 0.86–0.97; and ). There was a 9.0% lower relative risk of all-cause hospitalization (HR 0.91, 95% CI 0.85–0.97) and 16.0% lower relative risk of ICU admission (HR 0.84, 95% CI 0.71–0.98) among outpatient ARB users vs. ACEI users (S6 Fig).

Matching weight-adjusted cumulative hazard curves for all-cause hospitalization or mortality among outpatient Veterans who were current users of an ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen (Panel A) and ARB user vs. ACEI user (Panel B). Cumulative incidence curves demonstrating the risk for all-cause hospitalization or mortality in outpatients between exposure groups were generated from matching-weighted Cox regression models and bootstrap resampling with 2,000 iterations. Veterans are censored at the first occurrence of all-cause hospitalization, all-cause mortality, or October 15, 2020. Numbers at risk in table are unweighted. Abbreviations: ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; CI: confidence interval; HR: hazard ratio.

Among inpatients, over a median 30 days of follow-up, there was no statistically significant difference in all-cause mortality events in the exploratory analysis of ARB/ACEI-based users (i.e., filled a prescription in the 104 days prior to the admission) vs. non-ARB/ACEI-based regimen users (HR 1.25, 95% CI 0.30–5.13; and ). Over a median 30 days of follow-up, there were 21.0 vs. 17.7 all-cause deaths among inpatient ARB users vs. ACEI users (HR 1.13, 95% CI 0.93–1.38; and ). There were no statistically significant differences between exposure groups for secondary outcomes of ICU admission, dialysis, and mechanical ventilation ( and S7 Fig).

Matching weight-adjusted cumulative hazard curves for all-cause mortality among inpatient Veterans who were current users of an ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen (Panel A) and ARB user vs. ACEI user (Panel B). Cumulative incidence curves demonstrating the risk for all-cause mortality in inpatients between exposure groups were generated from matching-weighted Cox regression models and bootstrap resampling with 2,000 iterations. Veterans are censored at the first occurrence of all-cause mortality, 30 days’ follow-up, or October 15, 2020. Numbers at risk in table are unweighted. Due to small sample size, results displayed in Panel A are exploratory. Abbreviations: ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; CI: confidence interval; HR: hazard ratio.

Among outpatients taking less than three antihypertensive medication classes, users of an ARB/ACEI- vs. non-ARB/ACEI-based regimen appeared have a stronger association with all-cause hospitalization or mortality (HR 0.74, 95% CI 0.63–0.88) compared to those taking three or more antihypertensive medication classes (HR 1.52, 95% CI 0.91–2.71, p-value for interaction 0.003, S8 Table). Outpatient non-Hispanic Black Veterans taking an ARB vs. an ACEI also appeared to have a stronger association with all-cause hospitalization or mortality (HR 0.85, 95% CI 0.77–0.93) compared to Veterans of other race-ethnicities (HR 0.94, 95% CI 0.87–1.02, p-value for interaction 0.036, S9 Table). Conversely, inpatient non-Hispanic Black Veterans appeared to have a 48% higher relative risk of all-cause mortality if taking an ARB vs. an ACEI (HR 1.48, 95% CI 1.06–2.08), which was not observed among other race-ethnicities (HR 0.95, 95% CI 0.72–1.24, p-value for interaction 0.025, S10 Table). Sample size was not adequate to allow for subgroup analyses for the inpatient ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen comparison.

Sensitivity analysis using different covariate adjustment strategies and varying definitions of the medication exposure window did not materially change the results in either outpatients or inpatients (S11-S15 Tables). There was no association between any of the main exposures in either outpatients or inpatients for the negative control outcomes (S13-S14 Tables).

Discussion

This historical cohort study of ARB and ACEI use among outpatient and inpatient Veterans with treated hypertension positive for SARS-CoV-2 offers at least two main findings. First, among outpatients, current use of an ARB/ACEI was associated with a 15% lower relative risk of COVID-19-related outcomes compared to a non-ARB/ACEI-based antihypertensive medication regimen. Second, this is one of the first reports demonstrating that among outpatients, ARB users had a 9% lower relative risk of all-cause hospitalization or mortality compared to ACEI users. These findings support continued use of renin-angiotensin system blockade for (1) outpatients already using an ARB or ACEI, and (2) continued use of an ARB for outpatients using an ARB prior to testing positive for SARS-CoV-2. For current outpatient users of ACEIs prior to testing positive for SARS-CoV-2 in clinical practice, the HR of 0.91 for all-cause hospitalization or mortality with ARB vs. ACEI is insufficiently far from 1.0 to recommend converting from an ACEI to an ARB given the prevalent user design limitations. Randomized trials are needed to confirm the observed potential beneficial effects of ARBs compared to ACEIs on COVID-19-related outcomes.

The mechanistic basis for the potential effects of ARB or ACEI use on COVID-19-related outcomes is centered around the physiology of ACE2, the protein which SARS-CoV-2 uses for host cell entry. ACE2 upregulation in the lungs by ARBs or ACEIs could enhance SARS-CoV-2 interaction with its receptor, increase infection of type II alveolar epithelial cells, and increase COVID-19 severity. The contrasting hypothesis maintains that ARBs or ACEIs could mitigate viral-induced inflammatory responses by reducing available angiotensin II or by blocking angiotensin II’s actions via the angiotensin II type 1 receptor (AT1R), thus shifting the balance of the renin-angiotensin system back towards ACE2 and angiotensin-(1–7). This would, thereby, promote anti-inflammatory and anti-fibrotic effects in the lungs and possibly mitigate SARS-CoV-2-induced acute lung injury [2, 27]. Animal experiments and human studies have not conclusively demonstrated the presence, direction, or magnitude of the association between ARBs or ACEIs and ACE2 levels in the circulation or tissue expression, especially in the lungs [7, 8, 28–36].

Two small, randomized trials reported no increased risk of harm with continuation of ARB or ACEI therapy among hemodynamically stable adults hospitalized for COVID-19 [10, 11]. The current report adds to these smaller trials by reinforcing the randomized trial results in a larger scale observational study in a diverse population. Also, our analysis evaluated SARS-CoV-2+ outpatients’ risk of developing COVID-19, a population not tested in these trials. Furthermore, our inpatient cohort was older, and our analysis directly compared ARB vs ACEI users, which these smaller trials were not adequately powered to evaluate. High-quality observational studies have potential to augment trial results and provide helpful insights into the association of ARB and ACEI use and COVID-19 outcomes.

The current analysis may be the first observational study to show a protective association among outpatients between current users of an ARB compared to ACEI on COVID-19-related outcomes. Current users of an ARB in the outpatient cohort had a 9% lower relative risk of all-cause hospitalization or mortality and 16% lower relative risk of ICU admission compared to ACEI users. There is a mechanistic basis for the differential effects between ARBs and ACEIs. It has been hypothesized that ARBs and ACEIs have differential effects on ACE2 [7, 8]. In healthy adult humans, ACEIs, but not ARBs, may increase duodenal ACE2 expression [37]. In animals, ARB-mediated AT1R antagonism reduces peripheral and pulmonary inflammation, fibrosis, and edema [38, 39]. In addition, angiotensin II can induce beneficial effects through the AT2R, and ARBs and ACEIs differentially activate secondary pathways within the renin-angiotensin system, leading to alternative angiotensin II generation and metabolism [2]. Recent data suggest that bradykinin-mediated inflammation may contribute to adverse outcomes in patients with COVID-19 [40]. Because ACEIs inhibit bradykinin metabolism, this is another plausible mechanism for differences in COVID-19-related outcomes between ACEI users and ARB users.

Understanding the extent to which use of ARBs and ACEIs is beneficial or harmful in COVID-19 is a complicated yet critical knowledge gap because hypertension is highly prevalent among Americans with more severe COVID-19-related outcomes [9]. Ideally, we would have larger randomized trials to address these gap. The recently completed trials and prior observational investigations have generally found no association between users of an ARB/ACEI vs. non-ARB/ACEI and COVID-19-related outcomes [12]. However, in contrast to the prior analyses largely conducted among inpatients or those with and without COVID-19, the current analysis identified a significant 15% reduction in all-cause hospitalization or mortality among SARS-CoV-2-positive outpatients with hypertension taking ARB/ACEI- vs. non-ARB/ACEI-based regimens. The difference in results may be due to a larger sample size in the current analysis (i.e., increased power), more specific population (i.e., only those with treated hypertension), use of a PS matching-weighted analysis, or differences in exposure definitions compared to prior analyses. The heterogeneity in study design, study populations, exposure definitions, and statistical analyses may explain contrasting results between observational research investigating ARB and ACEI use and COVID-19-related outcomes.

In analyses evaluating ARB vs. ACEI use among inpatients, we did not find evidence of a significant association between ARB vs. ACEI users and all-cause mortality. This is in contrast to the significantly lower risk of all-cause hospitalization or mortality among ARB users compared to ACEI users in outpatients. One explanation for this inconsistency is the concept of “depletion of the susceptibles,” in which ARB users who are susceptible to the beneficial effects of ARBs on all-cause hospitalization or mortality were less likely to be hospitalized, and therefore, were less likely to be included in the inpatient analysis [14]. The restriction of the inpatients to those who did not benefit from ARBs as outpatients could bias results towards observing a benefit in the ACEI group. Accordingly, we caution against over-interpretation of the directionality of the relationship between ARBs vs. ACEIs in inpatients with COVID-19. Additionally, because we ascertained ARB and ACEI use in the three months prior to SARS-CoV-2 infection, our results should not be applied to acute changes in medication therapy at the time of COVID-19 infection or during hospitalization.

The current analysis has several strengths. This was a highly focused report with two primary exposures and two cohorts utilizing detailed and validated electronic health record and pharmacy data from the largest integrated health care delivery system in the US. We included several sensitivity analyses to assess for residual confounding and robustness of results. The current study also has several limitations. The external validity of our findings may be limited by a predominantly male and white population. Because we restricted the analysis comparing ARB/ACEI- vs. non-ARB/ACEI-based regimens users to only those without a compelling indication for an ARB or an ACEI to increase internal validity, our results may not be generalizable to adults with diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke. The study question necessitated a prevalent-user design, which could introduce bias. Our active-comparator design and methods of covariate adjustment mitigate these sources of bias, as is evident through the balance in exchangeability achieved between exposure groups before and after weighting. In addition, the results of the negative control analyses suggested a low risk of substantial residual bias. We were not able to estimate the impact of initiating ARB or ACEI in patients not previously receiving treatment, stopping treatment in inpatients already receiving treatment, or, finally, switching patients between ARB and ACEI after developing COVID-19; our findings can be interpreted within the clinical question of continuing an ARB vs. continuing an ACEI during infection with SARS-CoV-2. We did not capture healthcare utilization external to the VHA, however, we do not believe there would be differential ascertainment by exposure groups. We did not adjust for multiple comparisons. Instead, consistent with the recommendations from Rothman [41], Perneger [42], Feise [43], and Althouse [44], we provided extensive information and detail on the study design, methods, and statistical analyses and report effect sizes, confidence intervals, and p-values so that the readers can make their own interpretation of our findings in the context of the other recent reports on this topic. Data on symptoms at the time of SARS-CoV-2 testing were unavailable. Due to regional differences in the acceleration of COVID-19 cases and resource availability in the US, decisions on which patients with COVID-19 were admitted likely varied by facility and over time according to clinical judgment, local bed availability and patient acuity. Therefore, the severity of hospitalized patients with COVID-19 included in the inpatient analysis may have varied by region and over time.

Conclusion

COVID-19 is a persistent public health crisis that has infected more than 27 million and killed more than 475,000 US adults as of February 15, 2020. More than 45 million US adults use ARBs and ACEIs annually. Unlike prior randomized trials and observational analyses, we provide novel evidence that for SARS-CoV-2-positive outpatients, use of an ARB compared to ACEI in the three months prior to infection was associated with significantly lower risk of all-cause hospitalization or mortality. Randomized trials are needed to confirm the signal of potential beneficial effects of ARBs compared to ACEIs on COVID-19-specific outcomes. We also confirmed results of prior randomized trials and observational reports which found no evidence of a harmful association of ARBs or ACEIs with adverse COVID-19-related outcomes compared to use of non-ARB/ACEI antihypertensive medications. In light of the serious risks of discontinuation of ARBs and ACEIs in patients with hypertension, our findings support professional society recommendations to continue ARB or ACEI treatment among patients with hypertension in the absence of an established indication to discontinue them.

(DOCX)

Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

10 Feb 2021

PONE-D-21-01248

Angiotensin II receptor blocker or angiotensin-converting enzyme inhibitor use and COVID-19-related outcomes among US Veterans

PLOS ONE

Dear Dr. Bress,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Mar 26 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. Thank you for stating the following in your Competing Interests section:

"None"

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

6. Please amend the manuscript submission data (via Edit Submission) to include authors Catherine G. Derington, PharmD, MS; Jordana B. Cohen, MD, MSCE; April F. Mohanty, MPH, PhD; Tom H. Greene, PhD; James Cook, MS; Jian Ying, PhD; Guo Wei, MS; Jennifer S. Herrick, MS; Vanessa W. Stevens, PhD; Barbara E. Jones, MD, MS; Libo Wang, MD; Alexander R. Zheutlin, MD, MS; Andrew M. South, MD, MS; Thomas C. Hanff, MD, MSCE; Steven M. Smith, PharmD, MPH; Rhonda M. Cooper-DeHoff, PharmD, MS; Jordan B. King, PharmD, MS; G. Caleb Alexander, MD;l Dan R. Berlowitz, MD, MPH; Faraz S. Ahmad, MD, MS; M. Jason Penrod, MD; Rachel Hess, MD, MS; Molly B. Conroy, MD, MPH; James C. Fang, MD; Michael A. Rubin, MD, PhD, MS; Srinivasan Beddhu, MD; Alfred K. Cheung, MD; Weiming Xia, PhD; Lewis Kazis, ScD; William S. Weintraub, MD

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I commend you on this extensive work and nicely performed analysis. I personally agree with your conclusion that the decision on ACEi/ARB should be made very carefully. The risks vs benefits of RAAS should be weighed very carefully prior to considering in discontinuation. Yet, the controversy continues on this topic. I think it will be useful if the authors also cite the below paper:

Harky A, Chor CYT, Nixon H, Jeilani M. The controversy of using angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in COVID-19 patients. J Renin Angiotensin Aldosterone Syst. 2021 Jan-Dec;22(1):1470320320987118. doi: 10.1177/1470320320987118. PMID: 33412991; PMCID: PMC7797594.

Reviewer #2: This retrospective study from the large Veterans Health Administration database compared outcomes in users of Angiotensin II type 1 receptor blockers (ARB) or Angiotensin converting enzyme inhibitors (ACEI) users (ARB/ACEI users) vs. non-ARB/ACEI users, and ARB users vs ACEI users, among SARS-CoV-2+ outpatients and COVID-19 hospitalized inpatients. Only hypertensive treated Veterans were included, and patients with compelling indications for ARBs or ACEIs were excluded. Adjusted comparisons between groups were done using propensity score-weighted Cox regression.

Among SARS-CoV-2+ outpatients, the rate of primary outcome (all-cause hospitalization or death) was lower in the ARB/ACEI users as compared to the non-users (adjusted HR 0.85, 95% CI 0.73-0.99), and it was also lower in ARB users vs ACEI users (HR 0.91, 95%CI 0.86-0.97).

Among COVID-19 hospitalized inpatients, the primary outcome (all-cause mortality) was not significantly different between groups.

The article is well written. The rationale for the objectives are accurate, and the hypothesis tested are important. The methods used are mostly appropriate and accurate, data were appropriately measured and analysed.

The results are only partly original. Many concordant studies have already showed that prognosis of patients hospitalized for COVID-19 is not affected by a prior exposure to ARB/ACEI, and the present study appears underpowered to be conclusive, due to the low number of inpatients ARB/ACEI users (n=210), and non-ARB/ACEI users (n=275).

The potential protective effect of a prior treatment with ARB/ACEI in SARS-CoV-2+ outpatients is a more original finding. However, several issues need to be addressed:

MAJOR

1) In the introduction, authors wrote: “However, most prior observational studies contained significant methodological limitations that reduce their overall validity, and thus, their clinical applicability and generalizability, including lack of a well-defined question, study population, or specific focus/exposure of interest which can lead to confounding and residual bias.”

This statement is severe for previous observational studies. I fear that the present study partly shares criticisms made of previous studies, particularly on the definition of populations, and the generalizability of the results and conclusions.

2) It is unclear why authors decided to exclude the "compelling indications" of ARB/ACEI for comparisons between ARB/ACEI users vs non users, and to include the compelling indications for comparisons between ARB vs ACEI.

This results in surprising numbers of patients included into analyses: the total number of patients treated with either an ARB or an ACEI (n=2,482 in the comparisons between ARB/ACEI users vs non users) appears 5 times lower than the sum of patients with an ARB and patients with an ACEI (n=13,581 in the comparisons between ARB vs ACEI). This may be confusing.

3) In the ARB/ACEI vs non ARB/ACEI comparison, the choice to exclude Veterans who had “compelling indications” for an ARB or an ACEI resulted in the exclusion of 78% of screened outpatients Veterans (18,067/23,036). What are the clinical applicability and generalizability of results observed in a highly selected population of hypertensive Veterans having no diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke (pages 7-8)?

4) Was multiple testing considered in choosing the significance levels? Is a P value of 0.035 (Table 1) significant considering multiple testing?

MINOR

1. How many of outpatients Veterans have been tested because of symptoms, and how many have been tested systematically without symptoms (screening, contact subject, travel)?

2. What were the criteria for hospitalization in the inpatients group? Did the criteria for hospitalization change from January to October 2020?

3. How many Veterans were common to the out- and in-patients cohorts?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Feb 2021

REVIEWER #1

1. I commend you on this extensive work and nicely performed analysis. I personally agree with your conclusion that the decision on ACEi/ARB should be made very carefully. The risks vs benefits of RAAS should be weighed very carefully prior to considering in discontinuation. Yet, the controversy continues on this topic. I think it will be useful if the authors also cite the below paper: Harky A, Chor CYT, Nixon H, Jeilani M. The controversy of using angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in COVID-19 patients. J Renin Angiotensin Aldosterone Syst. 2021 Jan-Dec;22(1):1470320320987118. doi: 10.1177/1470320320987118. PMID: 33412991; PMCID: PMC7797594.

Response: Thank you for your review and comment. We now cite the paper you mention both in the Introduction section and the Discussion section on pages 14 and 15, respectively. We also added citations for the recent published reports of REPLACE-COVID and BRACE CORONA randomized trials in the context of how our study adds value to the trial results that are now available.

Introduction section page 5:

“During the ongoing pandemic, there has been unprecedented interest in whether ARBs and ACEIs are beneficial or harmful in patients with SARS-CoV-2 infection or coronavirus disease 2019 (COVID-19) [1].”

Discussion section page 15:

“Understanding the extent to which use of ARBs and ACEIs is beneficial or harmful in COVID-19 is a complicated yet critical knowledge gap because hypertension is highly prevalent among Americans with more severe COVID-19-related outcomes [1].”

Discussion section page 14:

“Two small, randomized trials reported no increased risk of harm with continuation of ARB or ACEI therapy among hemodynamically stable adults hospitalized for COVID-19 [2, 3]. The current report adds to these smaller trials by reinforcing the randomized trial results in a larger scale observational study in a diverse population. Also, our analysis evaluated SARS-CoV-2+ outpatients’ risk of developing COVID-19, a population not tested in these trials. Furthermore, our inpatient cohort was older, and our analysis directly compared ARB vs ACEI users, which these smaller trials were not adequately powered to evaluate. High-quality observational studies have potential to augment trial results and provide helpful insights into the association of ARB and ACEI use and COVID-19 outcomes.”

REVIEWER #2

1. This retrospective study from the large Veterans Health Administration database compared outcomes in users of Angiotensin II type 1 receptor blockers (ARB) or Angiotensin converting enzyme inhibitors (ACEI) users (ARB/ACEI users) vs. non-ARB/ACEI users, and ARB users vs ACEI users, among SARS-CoV-2+ outpatients and COVID-19 hospitalized inpatients. Only hypertensive treated Veterans were included, and patients with compelling indications for ARBs or ACEIs were excluded. Adjusted comparisons between groups were done using propensity score-weighted Cox regression. Among SARS-CoV-2+ outpatients, the rate of primary outcome (all-cause hospitalization or death) was lower in the ARB/ACEI users as compared to the non-users (adjusted HR 0.85, 95% CI 0.73-0.99), and it was also lower in ARB users vs ACEI users (HR 0.91, 95%CI 0.86-0.97). Among COVID-19 hospitalized inpatients, the primary outcome (all-cause mortality) was not significantly different between groups. The article is well written. The rationale for the objectives are accurate, and the hypothesis tested are important. The methods used are mostly appropriate and accurate, data were appropriately measured and analysed. The results are only partly original. Many concordant studies have already showed that prognosis of patients hospitalized for COVID-19 is not affected by a prior exposure to ARB/ACEI, and the present study appears underpowered to be conclusive, due to the low number of inpatients ARB/ACEI users (n=210), and non-ARB/ACEI users (n=275). The potential protective effect of a prior treatment with ARB/ACEI in SARS-CoV-2+ outpatients is a more original finding. However, several issues need to be addressed:

In the introduction, authors wrote: “However, most prior observational studies contained significant methodological limitations that reduce their overall validity, and thus, their clinical applicability and generalizability, including lack of a well-defined question, study population, or specific focus/exposure of interest which can lead to confounding and residual bias.” This statement is severe for previous observational studies. I fear that the present study partly shares criticisms made of previous studies, particularly on the definition of populations, and the generalizability of the results and conclusions.

Response: We understand the point the reviewer is making and agree that sentence should be modified. We have tempered the statement by modifying the first sentence and deleting the second sentence entirely.

The sentence now reads as follows on page 5:

“However, several of the prior observational studies contain methodological limitations that may reduce their internal validity, and thus, their clinical applicability.”

We provide specific examples of the methodological limitations we refer to later in the same paragraph.

2. It is unclear why authors decided to exclude the "compelling indications" of ARB/ACEI for comparisons between ARB/ACEI users vs non users, and to include the compelling indications for comparisons between ARB vs ACEI. This results in surprising numbers of patients included into analyses: the total number of patients treated with either an ARB or an ACEI (n=2,482 in the comparisons between ARB/ACEI users vs non users) appears 5 times lower than the sum of patients with an ARB and patients with an ACEI (n=13,581 in the comparisons between ARB vs ACEI). This may be confusing.

Response: We understand and agree that excluding those with compelling indications for the comparison of ARB/ACEI- vs. non-ARB/ACEI-based regimen users but not for the direct comparisons of ARB vs. ACEIs users led to significantly different samples sizes which may be confusing. However, we felt excluding those with compelling indications was critical for the comparison of ARB/ACEI- vs. non-ARB/ACEI-based regimen users and unnecessary for the direct ARB vs. ACEI comparison for the following reasons. We chose, a priori, to restrict the study population for the comparison of current users of ARB/ACEI- vs. non-ARB/ACEI-based regimen to only those without compelling indications because we believe that the confounding by indication for this specific comparison (ARB/ACEI users compared to non-users) would be insurmountable. We felt that patients with hypertension treated with ACEIs or ARBs are systematically different than those who are not on ACEIs or ARBs and that no level of statistical adjustment could overcome this systematic bias by indication. Instead, we chose to account for this bias through study design by restricting the analysis to only those without a history of compelling indications for ACEIs or ARBs. We felt that the restriction approach was a superior strategy to adjusting for the individual compelling indications as covariates in the full population because the latter approach cannot account for unmeasured confounders that likely coexist with the compelling indications that would go unaccounted for with statistical adjustment alone. We understand that although this approach, we believe, increases internal validity, it comes at the cost of external validity and generalizability to those with a history of compelling indications, which represents a large group of patients.

We added the following sentence to the Limitations section of the manuscript on page 16 to address this limitation:

“Because we restricted the analysis comparing ARB/ACEI- vs. non-ARB/ACEI-based regimen users to only those without a compelling indication for an ARB or an ACEI to increase internal validity, our results may not be generalizable to adults with diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke.”

We felt that this restriction was not needed for the comparison of ARB vs. ACEI users because these compelling indications are shared between both medications; ACEIs and ARBs are used interchangeably as they are thought to be equivalent in benefit and safety [4, 5]. Therefore, we believe there is no need to restrict the population as the distribution of these conditions and the associated unmeasured factors are expected to be similar among ACEI and ARB users.

We added the following sentences to the Methods section to state our rationale more clearly on page 7:

“For this comparison, because diagnoses other than hypertension (i.e., compelling indications) for ARB or ACEI use would introduce confounding by indication, we excluded Veterans with compelling indications defined in the 2017 American College of Cardiology/American Heart Association blood pressure guidelines [5] (i.e., diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke). We chose, a priori, to restrict the study population for the comparison of current users of ARB/ACEI- vs. non-ARB/ACEI-based regimen to only those without compelling indications because we believe that the confounding by indication for this specific comparison (ARB/ACEI users compared to non-users) would be insurmountable.

“Our second exposure compared current users of an ARB vs. ACEI (current user of an ARB vs. an ACEI). For this comparison, we included Veterans with and without compelling indications for ARBs and ACEIs, and we excluded Veterans who were current users of both an ARB and ACEI. For this comparison, restriction to those without compelling indications is not needed because these compelling indications are shared between both medications; ARBs and ACEIs are used interchangeably as they are thought to be equivalent in benefit and safety [4, 5]. As such, for this comparison, we included Veterans with and without compelling indications for ARBs and ACEIs, and we excluded Veterans who were current users of both an ARB and ACEI.”

3. In the ARB/ACEI vs non ARB/ACEI comparison, the choice to exclude Veterans who had “compelling indications” for an ARB or an ACEI resulted in the exclusion of 78% of screened outpatients Veterans (18,067/23,036). What are the clinical applicability and generalizability of results observed in a highly selected population of hypertensive Veterans having no diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke (pages 7-8)?

Response: Please see our response to Reviewer #2, Comment #2. We choose, a priori, to restrict the analysis comparing ARB/ACEI- vs. non-ARB/ACEI-based regimens users to only those without a compelling indication for an ARB or an ACEI to increase internal validity. By doing so, we understand this comes at the cost of external validity or generalizability of our results to groups with a history of compelling indications.

We added the following sentence to the Limitations section to address this limitation on page 16:

“Because we restricted the analysis comparing ARB/ACEI- vs. non-ARB/ACEI-based regimen users to only those without a compelling indication for an ARB or an ACEI to increase internal validity, our result may not be generalizable to adults with diabetes, chronic kidney disease, coronary heart disease, heart failure with reduced ejection fraction, or prior stroke.”

4. Was multiple testing considered in choosing the significance levels? Is a P value of 0.035 (Table 1) significant considering multiple testing?

Response: We agree with the reviewer that multiple testing may increase the likelihood of false-positive findings. We did consider multiple testing in choosing the significance levels. In the current analysis, we have two main exposures assessed in two cohorts (i.e., the inpatient and outpatient cohorts). In each cohort for each exposure, we designate a single primary outcome and three secondary outcomes and multiple sensitivity analyses. In this case, we feel adjustment for multiple comparisons is not useful when presenting secondary outcomes or sensitivity analyses surrounding a single "main analysis." The secondary outcomes and sensitivity analyses are playing a supportive role to the main analysis, and generally provide the opportunity for greater caution in interpreting if the results are inconsistent between analyses, rather than an opportunity for declaring a result "significant" if any of the sensitivity analyses has a nominally significant p-value. As such, consistent with the recommendations from Rothman [6], Perneger [7], Feise [8], and Althouse [9], instead of performing an adjustment for multiple comparisons, we chose to provide extensive information and detail on the identification of the main analyses vs. sensitivity analyses in study design, methods, and statistical analyses and report effect sizes, confidence intervals, and p-values so that the readers can make their own assessments about the quality of our study and the interpretation of our findings in the context of the other recent reports on this topic. We posted our statistical analysis plan on clinicatrials.gov and we make note that additional dedicated studies are needed to confirm the results.

We added the following statement to the Limitations of the manuscript on beginning on page 16:

“We did not adjust for multiple comparisons. Instead, consistent with the recommendations from Rothman [6], Perneger [7], Feise [8], and Althouse [9], we provided extensive information and detail on the study design, methods, and statistical analyses and report effect sizes, confidence intervals, and p-values so that the readers can make their own interpretation of our findings in the context of the other recent reports on this topic.”

5. How many of outpatients Veterans have been tested because of symptoms, and how many have been tested systematically without symptoms (screening, contact subject, travel)?

Response: Data on symptoms at the time of SARS-CoV-2 testing were unavailable. We added this to the limitations section on page 17:

“Data on symptoms at the time of SARS-CoV-2 testing were unavailable.”

6. What were the criteria for hospitalization in the inpatients group? Did the criteria for hospitalization change from January to October 2020?

Response: There were no standardized VA-system wide criteria for COVID-19 hospitalization during the pandemic. COVID-19 patient admission decisions likely varied by facility and over time according to clinical judgment, local bed availability, and patient acuity.

We added the following sentence to the Limitation section to address this on page 17:

“Due to regional differences in the acceleration of COVID-19 cases and resource availability in the US, decisions on which patients with COVID-19 were admitted likely varied by facility and over time according to clinical judgment, local bed availability and patient acuity. Therefore, the severity of hospitalized patients with COVID-19 included in the inpatient analysis may have varied by region and over time.”

7. How many Veterans were common to the out- and in-patients cohorts?

Response: There were 452 Veterans who were included for analyses in both the outpatient and inpatient cohorts for the comparison of ARB/ACEI- vs. non-ARB/ACEI-based regimens users. For the comparison of ARB vs. ACEI users, there were 3,110 Veterans who were included for analyses in both the outpatient and inpatient cohorts.

Editorial Requests

1. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

Response: We corrected the grant numbers.

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

Response: There are legal restrictions on sharing data publicly. The dataset used for this study contains sensitive patient information that is potentially identifying. The Department of Veterans Affairs (VA) restricts access to such data unless researchers meet specific criteria per VHA Directive 1605.01, Privacy and Release of Information, section 13e which includes a written request for the data, IRB approval with waiver for HIPAA authorization prior to the request for individually identifiable information, and approval by the VA.

3. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Response: We added Dr. Bress’s ORCID iD.

4. Please amend the manuscript submission data (via Edit Submission) to include authors Catherine G. Derington, PharmD, MS; Jordana B. Cohen, MD, MSCE; April F. Mohanty, MPH, PhD; Tom H. Greene, PhD; James Cook, MS; Jian Ying, PhD; Guo Wei, MS; Jennifer S. Herrick, MS; Vanessa W. Stevens, PhD; Barbara E. Jones, MD, MS; Libo Wang, MD; Alexander R. Zheutlin, MD, MS; Andrew M. South, MD, MS; Thomas C. Hanff, MD, MSCE; Steven M. Smith, PharmD, MPH; Rhonda M. Cooper-DeHoff, PharmD, MS; Jordan B. King, PharmD, MS; G. Caleb Alexander, MD;l Dan R. Berlowitz, MD, MPH; Faraz S. Ahmad, MD, MS; M. Jason Penrod, MD; Rachel Hess, MD, MS; Molly B. Conroy, MD, MPH; James C. Fang, MD; Michael A. Rubin, MD, PhD, MS; Srinivasan Beddhu, MD; Alfred K. Cheung, MD; Weiming Xia, PhD; Lewis Kazis, ScD; William S. Weintraub, MD

Response: We amended the manuscript submission via “Edit Submission” to include all co-authors.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Response: We included captions for our supporting information files at the end of our manuscript and updated in-text citations to match accordingly.

REFERENCES

1. Harky A, Chor CYT, Nixon H, Jeilani M. The controversy of using angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in COVID-19 patients. J Renin Angiotensin Aldosterone Syst. 2021;22(1):1470320320987118. Epub 2021/01/09. doi: 10.1177/1470320320987118. PubMed PMID: 33412991; PubMed Central PMCID: PMCPMC7797594.

2. Lopes RD, Macedo AVS, de Barros ESPGM, Moll-Bernardes RJ, Dos Santos TM, Mazza L, et al. Effect of Discontinuing vs Continuing Angiotensin-Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers on Days Alive and Out of the Hospital in Patients Admitted With COVID-19: A Randomized Clinical Trial. JAMA. 2021;325(3):254-64. Epub 2021/01/20. doi: 10.1001/jama.2020.25864. PubMed PMID: 33464336; PubMed Central PMCID: PMCPMC7816106.

3. Cohen JB, Hanff TC, William P, Sweitzer N, Rosado-Santander NR, Medina C, et al. Continuation versus discontinuation of renin-angiotensin system inhibitors in patients admitted to hospital with COVID-19: a prospective, randomised, open-label trial. Lancet Respir Med. 2021. Epub 2021/01/11. doi: 10.1016/S2213-2600(20)30558-0. PubMed PMID: 33422263; PubMed Central PMCID: PMCPMC7832152.

4. Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021-104. Epub 2018/08/31. doi: 10.1093/eurheartj/ehy339. PubMed PMID: 30165516.

5. Whelton PK, Carey RM, Aronow WS, Casey DE, Jr., Collins KJ, Dennison Himmelfarb C, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):1269-324. Epub 2017/11/15. doi: 10.1161/HYP.0000000000000066. PubMed PMID: 29133354.

6. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990:43-6.

7. Perneger TV. What's wrong with Bonferroni adjustments. Bmj. 1998;316(7139):1236-8.

8. Feise RJ. Do multiple outcome measures require p-value adjustment? BMC medical research methodology. 2002;2(1):1-4.

9. Althouse AD. Adjust for Multiple Comparisons? It's Not That Simple. Ann Thorac Surg. 2016;101(5):1644-5. Epub 2016/04/24. doi: 10.1016/j.athoracsur.2015.11.024. PubMed PMID: 27106412.

19 Feb 2021

Angiotensin II receptor blocker or angiotensin-converting enzyme inhibitor use and COVID-19-related outcomes among US Veterans

PONE-D-21-01248R1

Dear Dr. Bress,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

13 Apr 2021

PONE-D-21-01248R1

Angiotensin II receptor blocker or angiotensin-converting enzyme inhibitor use and COVID-19-related outcomes among US Veterans

Dear Dr. Derington:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Corstiaan den Uil

Academic Editor

PLOS ONE

Table 1

Matching weight-adjusted incidence rates and hazard ratios for the primary and secondary outcomes among outpatient Veterans with treated hypertension who were current users of an ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen and ARB user vs. ACEI user, separately.

	ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen comparison (n = 4,969)				ARB user vs. ACEI user comparison(n = 13,581)
Outcome	ARB/ACEI	Non-ARB/ACEI	Hazard Ratio		ARB	ACEI	Hazard Ratio
	(n = 2,482)	(n = 2,487)	(95% CI)	p-value	(n = 4,877)	(n = 8,704)	(95% CI)	p-value
Primary Outcome
All-cause hospitalization or all-cause mortality	314 (5.0)	395 (6.0)	0.85 (0.73–0.99)	0.035	1,435 (13.2)	2,467 (14.8)	0.91 (0.86–0.97)	0.002
Secondary Outcomes
All-cause hospitalization	273 (4.3)	344 (5.1)	0.86 (0.73–1.01)	0.07	1,254 (10.9)	2,146 (12.3)	0.91 (0.85–0.97)	0.005
All-cause mortality	61 (0.8)	65 (0.9)	0.95 (0.66–1.36)	0.77	386 (2.7)	628 (2.8)	0.95 (0.84–1.08)	0.44
ICU admission	61 (0.9)	49 (0.7)	1.29 (0.87–1.92)	0.21	257 (1.8)	491 (2.1)	0.84 (0.71–0.98)	0.024

Numbers in table are expressed as unweighted frequency of event (weighted rate per 100 person-months).

ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; CI: confidence interval; ICU: intensive care unit

Table 2

Matching weight-adjusted incidence rates and hazard ratios for the primary and secondary outcomes among inpatient Veterans with treated hypertension who were current users of an ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen and ARB user vs. ACEI user, separately.

	ARB/ACEI- vs. non-ARB/ACEI-based antihypertensive regimen comparison* (n = 485)				ARB user vs. ACEI user comparison (n = 3,178)
Outcome	ARB/ACEI	Non-ARB/ACEI	Hazard Ratio		ARB	ACEI	Hazard Ratio
	(n = 210)	(n = 275)	(95% CI)	p-value	(n = 1,164)	(n = 2,014)	(95% CI)	p-value
Primary Outcome
All-cause mortality	8 (3.4)	6 (2.0)	1.25 (0.30–5.13)	0.76	168 (21.0)	242 (17.7)	1.13 (0.93–1.38)	0.23
Secondary Outcomes
ICU admission	48 (33.0)	40 (19.5)	1.05 (0.54–1.94)	0.94	264 (26.4)	497 (24.5)	0.92 (0.79–1.08)	0.29
Dialysis	0 (0.0)	2 (0.2)	n/a	n/a	33 (4.6)	448 (4.5)	1.08 (0.84–1.40)	0.54
Mechanical ventilation	11 (4.8)	16 (5.9)	0.67 (0.22–2.03)	0.47	185 (24.5)	313 (23.9)	0.96 (0.80–1.15)	0.65

Numbers in table are expressed as unweighted frequency of event (weighted rate per 100 person-months).

*Exploratory due to small sample size.

ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; CI: confidence interval; ICU: intensive care unit