key: cord-0710369-w1rlhlnv
authors: Jung, Sun-Young; Choi, Jae Chol; You, Seung-Hun; Kim, Won-Young
title: Association of renin-angiotensin-aldosterone system inhibitors with COVID-19-related outcomes in Korea: a nationwide population-based cohort study
date: 2020-05-22
journal: Clin Infect Dis
DOI: 10.1093/cid/ciaa624
sha: 83cec622d480f0475cdf40565dcc7954f0b8c9b7
doc_id: 710369
cord_uid: w1rlhlnv

BACKGROUND: Renin-angiotensin-aldosterone system (RAAS) inhibitors may facilitate host cell entry of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or attenuate organ injury via RAAS blockade. We aimed to assess the associations between prior use of RAAS inhibitors and clinical outcomes among Korean patients with coronavirus 2019 (COVID-19). METHODS: We performed a nationwide population-based cohort study using the Korean Health Insurance Review and Assessment database. Claim records were screened for 66793 individuals who were tested for COVID-19 until April 8, 2020. Adjusted odds ratios (ORs) were used to compare the clinical outcomes between RAAS inhibitor users and nonusers. RESULTS: Among 5179 confirmed COVID-19 cases, 762 patients were RAAS inhibitor users and 4417 patients were nonusers. Relative to nonusers, RAAS inhibitor users were more likely to be older, male, and have comorbidities. Among 1954 hospitalized patients with COVID-19, 377 patients were RAAS inhibitor users and 1577 patients were nonusers. In-hospital mortality was observed for 33 RAAS inhibitor users (9%) and 51 nonusers (3%) (p<0.001). However, after adjustment for age, sex, Charlson Comorbidity Index, immunosuppression, and hospital type, the use of RAAS inhibitors was not associated with a higher risk of mortality (adjusted OR, 0.88; 95% confidence interval, 0.53–1.44; p=0.60). No significant differences were observed between RAAS inhibitor users and nonusers in terms of vasopressor use, modes of ventilation, extracorporeal membrane oxygenation, renal replacement therapy, and acute cardiac events. CONCLUSIONS: Our findings suggest that prior use of RAAS inhibitors was not independently associated with mortality among COVID-19 patients in Korea.

A c c e p t e d M a n u s c r i p t 5 

In late 2019, several pneumonia cases with an undetermined cause were identified in Wuhan (Hubei province, China), and similar cases were rapidly observed in other Chinese provinces and other countries [1, 2] . On January 3, 2020, a novel coronavirus (severe acute respiratory syndrome coronavirus 2, SARS-CoV-2) was isolated from these patients [1] , and the corresponding disease was subsequently named coronavirus 2019 (COVID-19) by the World Health Organization. Severe cases, including cases that resulted in death, have been observed in older patients and patients with comorbidities, including hypertension, coronary artery disease, diabetes, and chronic kidney disease [3] [4] [5] .

Both SARS-CoV-1 and SARS-CoV-2 require human angiotensin-converting enzyme 2 (ACE2) to enter the host cell [6, 7] . Interestingly, ACE2 expression is increased in patients with diabetes [8] and after treatment using renin-angiotensin-aldosterone system (RAAS) inhibitors, such as ACE inhibitors and angiotensin receptor blockers (ARBs) [9] . Given that mechanism, and the fact that several Chinese studies have identified more severe outcomes in COVID-19 patients with comorbidities that meet the indications for ACE inhibitors and ARBs [3] [4] [5] , there are concerns regarding the use of these drugs in COVID-19 cases [10] . In addition, SARS-CoV-2 appears to down-regulate ACE2 expression, which is a key enzyme that counteracts RAAS by degrading angiotensin II to angiotensin-(1-7) and thereby attenuates organ injury [11] . Blockage of RAAS and increased ACE2 expression attenuated lung injury induced by SARS-CoV in preclinical model [11, 12] . Nevertheless, discontinuation of RAAS inhibitor treatment could lead to clinical deterioration in patients with cardiovascular diseases, and clinical data are lacking regarding the theoretical relationships between RAAS inhibitor use and COVID-19 severity. Recent Chinese studies demonstrated that inpatient use of RAAS inhibitors was not associated with mortality in A c c e p t e d M a n u s c r i p t 6 COVID-19 patients with hypertension [13, 14] . However, prehospital drug use was not evaluated in these studies, thereby questioning adequate exposure to RAAS inhibitors. Therefore, this nationwide population-based cohort study evaluated records from the Korean Health Insurance Review and Assessment (HIRA) database to determine whether prior use of RAAS inhibitors was associated with clinical outcomes among patients with COVID-19.

Limited real-world clinical data are available to physicians and policymakers during the COVID-19 pandemic. Therefore, the Republic of Korea's government decided to share the world's first nationwide database of de-identified COVID-19 patient data with domestic and international researchers [15] . The database contains claim records (including for a 5-year period before hospitalization) for 69793 individuals who were tested for COVID-19 and for whom hospitals issued claims to the HIRA until April 8, 2020. The database includes detailed information regarding demographic characteristics, diagnoses, prescriptions, procedures, and patient outcomes. The diagnostic codes were assigned on the basis of the Korean Classification of Diseases seventh revision (KCD-7), which is a modified version of the tenth revision of the International Classification of Diseases (ICD-10). Use of drugs was identified using Anatomical Therapeutic Chemical codes and HIRA general name codes. The study protocol for analysis of de-identified patient data was exempted from review by the Institutional Review Board of Chung-Ang University (1041078-202004-HR-082-01).

We included patients with COVID-19 who were ≥18 years old. Confirmed COVID-19 cases were identified using the following KCD-7/ICD-10 codes: B34.2 (coronavirus A c c e p t e d M a n u s c r i p t 7 infection, unspecified site), B97.2 (coronavirus as the cause of diseases classified to other chapters), U18 (provisional assignment of new diseases of uncertain etiology or emergency use), U18.1 (novel coronavirus infection), and U07.1 (coronavirus disease 2019 ). These diagnostic codes were given only if a definitive diagnosis of COVID-19 was made based on positive nasopharyngeal swab specimens tested with real-time reverse transcription-polymerase chain reaction assays [16] . The index date was defined as the date of the COVID-19 diagnosis. All patients were followed until the first instance of death or April 8, 2020.

This study identified all RAAS inhibitors, including ACE inhibitors and ARBs, that were prescribed within 1 year before the index date. For the present study, RAAS inhibitor users were defined as patients with RAAS inhibitor use at 1-30 days before the index date, whereas nonusers were defined as patients who had never received RAAS inhibitors or had received them at 31-365 days before the index date. A prescription duration of ≥7 days was required to define drug use. The detailed types and codes for RAAS inhibitors are described in Supplementary Table 1 .

Comorbidities were defined on the basis of claim codes within 1 year before the index date and evaluated on the basis of the Charlson Comorbidity Index [17] . The presence of hypertension, which is not included in the Charlson Comorbidity Index, was separately identified using the ICD-10 codes (see Supplementary Table 2 ). Data were also collected M a n u s c r i p t 8 regarding previous use of steroids (oral or intravenous for ≥30 days during the previous year) and immunosuppressants (see Supplementary Table 3 ). Immunosuppression was identified on the basis of an underlying disease or condition that affected the immune system (e.g., human immunodeficiency virus infection or malignancy) or if the patient was receiving immunosuppressive therapy. The type of hospital in which a diagnosis of COVID-19 has been established was defined according to the center's number of beds and medical specialties (see Supplementary Material).

Data were extracted regarding antibiotics, antivirals, hydroxychloroquine, steroids, and vasopressors that were used during the hospitalization (see Supplementary Tables 3 and   4 ). In-hospital steroid treatment was defined as only intravenous corticosteroid treatment (methylprednisolone or hydrocortisone). Vasopressor use was defined as the administration of norepinephrine, epinephrine, vasopressin, dopamine, or dobutamine during the hospitalization. Procedure codes were also used to identify cases that involved conventional oxygen therapy, high flow nasal cannula, mechanical ventilation, extracorporeal membrane oxygenation (ECMO), and renal replacement therapy (see Supplementary Table 5 ).

In-hospital mortality was defined as the primary study outcome. The secondary outcomes included vasopressor use, modes of ventilation, ECMO, renal replacement therapy, and acute cardiac events (cardiac arrest, myocardial infarction, and acute heart failure).

A c c e p t e d M a n u s c r i p t 9

Logistic regression models were used to calculate odds ratios (ORs) and 95% confidence interval (CIs) for in-hospital mortality according to RAAS inhibitor use. The association of RAAS inhibitor use with in-hospital mortality was also assessed in the subgroup of patients with hypertension, given the fact that previous studies have identified more severe outcomes in these patients. Multivariate logistic regression analysis was performed to adjust for age, sex, Charlson Comorbidity Index, immunosuppression, and hospital type. A separate analysis was also performed to ascertain the effects of in-hospital medication exposures, vasopressor use, and mechanical ventilation. The rate at which excess ACE2 would disappear from host cell membranes after withdrawal of RAAS inhibitors is unclear. Thus, as sensitivity analysis, the patients who received RAAS inhibitors at any time during the 365 days before the index date were analyzed to compare the clinical outcomes between RAAS inhibitor users and nonusers. The associations between RAAS inhibitor use and mortality were assessed on the basis of stratification according to age, sex, Charlson Comorbidity Index, diabetes, hypertension, myocardial infarction, cerebrovascular disease, chronic pulmonary disease, chronic kidney disease, and immunosuppression. Further, we evaluated whether prior RAAS inhibitor use was significantly associated with any of the secondary outcomes. All tests were two-tailed and differences were considered statistically significant at p-values of <0.05. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary,

A c c e p t e d M a n u s c r i p t

Among the 69793 individuals who were tested, 5179 patients were confirmed to have COVID-19. These patients with COVID-19 were classified as RAAS inhibitor users (762 patients) or nonusers (4417 patients). 732 users (96%) received ARBs, and 110 patients received RAAS inhibitors at 31-365 days before the index date and were classified as nonusers (see Supplementary Table 6 ). The patients' baseline characteristics are described in Table 1 . The RAAS inhibitor users were more likely to be older, male, and have the comorbidities listed in Table 1 . Moreover, RAAS inhibitor users were more likely to be immunosuppressed.

Among the 5179 patients with COVID-19, hospitalization was observed for 1954 patients (38%). Based on the absence of claims data, we assume that the remaining patients stayed at home or received out-of-hospital treatment, such as at community treatment centers.

The hospitalized patients included 377 patients who used RAAS inhibitors and 1577 patients who were nonusers. 365 users (97%) received ARBs, and 40 patients received RAAS inhibitors at 31-365 days before the index date and were classified as nonusers (see Supplementary Table 6 ). The in-hospital treatments included antibiotics (599 patients, 31%), antivirals (782 patients, 40%), hydroxychloroquine (412 patients, 21%), and intravenous steroids (75 patients, 4%) (see Supplementary Table 7) . Each treatment type was more common among RAAS inhibitor users.

The overall mortality rate for the entire cohort was 1.6% (84/5179 patients), and inhospital mortality was observed for 33/377 RAAS inhibitor users (9%) and for 51/1577 nonusers (3%) (p<0.001) ( Table 2 ). The survival curves for the entire cohort and the hospitalized patients are shown in Supplementary Figures 1 and 2 Table 2) .

Relative to nonusers, RAAS inhibitor users were more likely to develop cardiac arrest and acute heart failure, and these observations were not influenced when we evaluated subjects who received RAAS inhibitors at any time during the 365 days before the index date (see Supplementary Table 8 ).

However, multivariate analysis that adjusted for baseline imbalances revealed that RAAS inhibitor use was not independently associated with a higher risk of mortality among COVID-19 patients (adjusted OR, 0.88; 95% CI, 0.53-1.44; p=0.60) ( Table 3 ). This finding was not influenced when we evaluated subjects who received RAAS inhibitors at any time during the 365 days before the index date (see Supplementary Table 9 ). The analysis that additionally adjusted for in-hospital medications, vasopressor use, and mechanical ventilation was consistent with the overall results (see Supplementary Table 10 ). The multivariate results were generally consistent across all subgroups (Figure 1) , and there were generally no significant differences in the secondary outcomes between the RAAS inhibitor users and nonusers (Table 4) .

Among the 1157 patients with hypertension and COVID-19, there were 719 patients who used RAAS inhibitors and 438 patients who were nonusers. The RAAS inhibitor users were more likely to have diabetes, and nonusers were more likely to have malignancy and be immunosuppressed (see Supplementary Table 11 ). In-hospital mortality was observed for 31/348 RAAS inhibitor users (9%) and for 25/194 nonusers (13%) (p=0.14) (see Supplementary Table 12 ). Multivariate analysis that adjusted for baseline imbalances revealed that RAAS inhibitor use was not independently associated with a higher risk of Supplementary Table 13 ).

The present study revealed a significantly higher mortality rate among patients with COVID-19 who were using RAAS inhibitors, relative to patients who were not receiving RAAS inhibitors. However, RAAS inhibitor users were older, had more comorbidities, and were more likely to receive in-hospital treatments. The elevated risk of mortality among RAAS inhibitor users disappeared after adjusting for these confounding factors.

The use of RAAS inhibitors is thought to up-regulate ACE2 expression and facilitate host cell entry of SARS-CoV [6, 7] . Furthermore, ACE2 is predominantly expressed in lung alveolar epithelial cells and in many extrapulmonary tissues, including the heart and kidneys [18] . However, preclinical models have revealed inconsistent findings regarding the effects of RAAS inhibitors on ACE2 expression [9, 19] . Moreover, ACE2 may protect against organ injury by reducing angiotensin II activity, and experimental models have indicated that decreased ACE2 activity may contribute to respiratory virus-induced lung injury [20] . A recent study also suggested that patients with COVID-19 appeared to exhibit RAAS activation, increased viral load levels, and subsequent lung injury [21] . The present study revealed that, relative to nonusers, RAAS inhibitor users had a higher mortality rate and more advanced ventilatory support. Nevertheless, the potential for confounding precludes causal inference regarding the relationship between RAAS inhibitor use and lung injury. In addition, the effects of ARBs on ACE2 expression are inconsistent in animal models [9, 19] , while ACE inhibitors do not affect directly the activity of ACE2 [22] . Thus, it would be interesting to perform subgroup analysis according to specific RAAS inhibitors. However, our study had A c c e p t e d M a n u s c r i p t 13 a relatively small sample of ACE inhibitor users, relative to ARB users (see Supplementary Three Chinese studies of patients with severe COVID-19 revealed that hypertension, diabetes, and coronary artery disease were the most common comorbidities, and that older age was a risk factor for mortality [3] [4] [5] . An Italian cohort of critically ill patients were predominantly older men and hypertension was the most common comorbidity, which was followed by cardiovascular disorders, hypercholesterolemia, and diabetes [23] . In Korea, the case fatality rate was higher among older men with comorbidities that included cardiovascular disease, diabetes, and neurologic disease [24] . These poor outcomes are likely related to age-dependent defects in immune function and an excessive host response to viral infection [25] . The protective effects of both sex hormones and a second X chromosome, which carries several genes encoding for innate immune molecules, may explain the low susceptibility of female patients to viral infections [26] . Moreover, ACE2 is located on the X chromosome, and there is a possibility that the dosage effect of sex chromosome may impact ACE2 activity due to escape from X-inactivation [27] . The present study revealed that 22% of patients were ≥60 years old and 44% of patients were men, which are similar to the official statistics from the Korea Centers for Disease Control and Prevention as of April 5, 2020 (≥60 years old: 24%, men: 40%). We did not separately estimate the individual risk factors for mortality, although it would be reasonable to assume that baseline factors (e.g., age, sex, and comorbidities) did affect the study outcomes, given that the associations with RAAS inhibitor use disappeared after adjustment for these factors.

A c c e p t e d M a n u s c r i p t 14 In clinical practice, RAAS inhibitors are typically used to treat cardiovascular diseases, diabetes, and chronic kidney disease [28, 29] . Unfortunately, the withdrawal of RAAS inhibitors may result in severe clinical decompensation in patients with heart failure [30] . In addition, myocardial dysfunction has been observed during the course of COVID-19 [31] . Dysregulation of ACE2 may play a role in myocardial injury, as ACE2 knockout in animal models led to angiotensin II-induced injury and adverse left ventricular remodeling [32] . Furthermore, ACE2 expression was reduced in viral RNA-positive heart samples from patients who died from SARS [33] . The present study revealed RAAS inhibitor users to have higher proportions of vasopressor use, cardiac arrest, and acute heart failure, although these relationships were not significant after adjustment for confounding factors. Therefore, our findings suggest that RAAS inhibitors should be continued in COVID-19 patients, especially patients with high risks of heart failure and myocardial infarction. Future studies are needed to evaluate the association between RAAS inhibitor discontinuation and outcomes among COVID-19 patients.

To the best of our knowledge, ours is the first comprehensive study to evaluate whether prior RAAS inhibitor use was associated with outcomes in a nationwide cohort of Asian COVID-19 patients. However, the findings suggest that RAAS inhibitor use was not independently associated with COVID-19 outcomes after adjusting for baseline demographic characteristics, comorbidities, and in-hospital treatments. These data support the recommendations of major medical societies, which suggest continuation of RAAS inhibitors in patients with COVID-19 [34, 35] . Thus, our "real-world" evidence regarding this relationship is crucial, given the concerns regarding ill-advised discontinuation among patients who rely on these drugs [36] . City cohorts [37] [38] [39] . Although these studies provide information similar to ours, the evaluation of the associations of RAAS inhibitors with COVID-19-related outcomes in an Asian population is clinically relevant, given that the East Asian populations have higher ACE2 expression in tissues than other populations under the similar conditions [40] .

The present study has several limitations. First, the accuracy of diagnostic codes may be limited and there is always a possibility of over-or under-coding. However, the Korean HIRA Service vigorously audits insurance claims and numerous peer-reviewed publications have been based on HIRA data. It is also unlikely that there were discrepancies between the database disease codes and the actual diseases, as epidemic investigations were performed for all COVID-19 cases. Moreover, RAAS inhibitors are not available as over-the-counter drugs or via self-pay prescriptions in Korea. Second, the retrospective observational design cannot exclude the possibility of residual confounding and precludes causal inference regarding the relationship between RAAS inhibitors and COVID-19-related outcomes. Thus, randomized controlled studies would be needed to establish causality, although it would not likely be feasible to enroll a sufficient number of patients in a short period. Thus, observational data are currently the best available evidence regarding RAAS inhibitor use among patients with COVID-19. Third, database limitations made it impossible to obtain information regarding a potential control group (e.g., patients who were not sick enough to be tested for COVID -19) .

In addition, the database did not include information regarding smoking status and laboratory results. Nevertheless, our primary objective was to evaluate whether RAAS inhibitor use was associated with outcomes among patients with COVID-19. Finally, additional information regarding the cause of death was not available in the database. In this study, the probability of survival dropped abruptly in the first few days for the entire cohort and the hospitalized patients. It is possible that early mortality may be related to acute cardiac event or acute pulmonary embolism rather than COVID-19-related respiratory disease. M a n u s c r i p t 25 

A Novel Coronavirus from Patients with Pneumonia in China

First Case of 2019 Novel Coronavirus in the United States

Clinical Characteristics of Coronavirus Disease 2019 in China

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Clinical Features of 85 Fatal Cases of COVID-19 from Wuhan: A Retrospective Observational Study

Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Receptor Recognition by the Novel Coronavirus from Wuhan: an Analysis Based on Decade-Long Structural Studies of SARS Coronavirus

Effect of angiotensin-converting enzyme inhibition and angiotensin II receptor blockers on cardiac angiotensin-converting enzyme 2

Are patients with hypertension and diabetes mellitus at increased risk for COVID-19 infection?

A crucial role of angiotensin converting enzyme 2 (ACE2) in SARS coronavirus-induced lung injury

Angiotensin-converting enzyme 2 protects from severe acute lung failure

Association of Inpatient Use of Angiotensin Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers with Mortality Among Patients With Hypertension Hospitalized With 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

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

A new method of classifying prognostic comorbidity in longitudinal studies: development and validation

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis

Combination reninangiotensin system blockade and angiotensin-converting enzyme 2 in experimental myocardial infarction: implications for future therapeutic directions

Angiotensin-converting enzyme 2 protects from lethal avian influenza A H5N1 infections

Clinical and biochemical indexes from 2019-nCoV infected patients linked to viral loads and lung injury

Evaluation of angiotensinconverting enzyme (ACE), its homologue ACE2 and neprilysin in angiotensin peptide metabolism

Baseline Characteristics and Outcomes of 1591

Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region

Korean Society of Infectious Diseases and Korea Centers for Disease Control and Prevention

The immunopathogenesis of sepsis in elderly patients

Sexual Dimorphism in Innate Immunity

Sex differences in renal angiotensin converting enzyme 2 (ACE2) activity are 17β-oestradiol-dependent and sex chromosome-independent

Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A

The RAAS in the pathogenesis and treatment of diabetic nephropathy

Withdrawal of pharmacological treatment for heart failure in patients with recovered dilated cardiomyopathy (TRED-HF): an open-label, pilot, randomised trial

Association of Cardiac Injury With Mortality in Hospitalized Patients With COVID-19 in Wuhan

Loss of angiotensin-converting enzyme 2 accelerates maladaptive left ventricular remodeling in response to myocardial infarction

SARS-coronavirus modulation of myocardial ACE2 expression and inflammation in patients with SARS

HFSA/ACC/AHA statement addresses concerns re: using RAAS antagonists in COVID-19

Position statement of the ESC Council on Hypertension on ACE-inhibitors and angiotensin receptor blockers

Antihypertensive drugs and risk of COVID-19?

Cardiovascular Disease, Drug Therapy, and Mortality in Covid-19

Renin-Angiotensin-Aldosterone System Blockers and the Risk of Covid-19

Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Covid-19

Comparative genetic analysis of the novel coronavirus (2019-nCoV/SARS-CoV-2) receptor ACE2 in different populations