key: cord-1021099-km4vrhuc authors: Braude, Philip; Carter, Ben; Short, Roxanna; Vilches-Moraga, Arturo; Verduri, Alessia; Lyndsay Pearce, Miss; Angeline Price, Miss; Quinn, Terence J.; Stechman, Michael; Collins, Jemima; Bruce, Eilidh; Einarsson, Alice; Rickard, Frances; Mitchell, Emma; Holloway, Mark; Hesford, James; Barlow-Pay, Fenella; Clini, Enrico; Kyaw Myint, Phyo; Moug, Susan; McCarthy, Kathryn; Hewitt, Jonathan title: The influence of ACE inhibitors and ARBs on hospital length of stay and survival in people with COVID-19 date: 2020-10-15 journal: Int J Cardiol Heart Vasc DOI: 10.1016/j.ijcha.2020.100660 sha: 75f1d6e54c35c676dbf6caa06a519b46b8f57f7d doc_id: 1021099 cord_uid: km4vrhuc Objective During the COVID-19 pandemic the continuation or cessation of angiotensin-converting enzyme inhibitors (ACEi) and angiotensin receptor blockers (ARBs) has been contentious. Mechanisms have been proposed for both beneficial and detrimental effects. Recent studies have focused on mortality with no literature having examined length of hospital stay. The aim of this study was to determine the influence of ACEi and ARBs on COVID-19 mortality and length of hospital stay. Methods COPE (COVID-19 in Older People) is a multicenter observational study including adults of all ages admitted with either laboratory or clinically confirmed COVID-19. Routinely generated hospital data were collected. Primary outcome: mortality; secondary outcomes: Day-7 mortality and length of hospital stay. A mixed-effects multivariable Cox’s proportional baseline hazards model and logistic equivalent were used. Results 1371 patients were included from eleven centres between 27th February to 25th April 2020. Median age was 74 years [IQR 61-83]. 28.6% of patients were taking an ACEi or ARB. There was no effect of ACEi or ARB on inpatient mortality (aHR=0.85, 95%CI 0.65-1.11). For those prescribed an ACEi or ARB, hospital stay was significantly reduced (aHR=1.25, 95%CI 1.02-1.54, p=0.03) and in those with hypertension the effect was stronger (aHR=1.39, 95%CI 1.09-1.77, p=0.007). Conclusions Patients and clinicians can be reassured that prescription of an ACEi or ARB at the time of COVID-19 diagnosis is not harmful. The benefit of prescription of an ACEi or ARB in reducing hospital stay is a new finding. adults of all ages admitted with either laboratory or clinically confirmed COVID-19. Routinely generated hospital data were collected. Primary outcome: mortality; secondary outcomes: Day-7 mortality and length of hospital stay. A mixed-effects multivariable Cox's proportional baseline hazards model and logistic equivalent were used. Patients and clinicians can be reassured that prescription of an ACEi or ARB at the time of COVID-19 diagnosis is not harmful. The benefit of prescription of an ACEi or ARB in reducing hospital stay is a new finding. Title: The influence of ACE inhibitors and ARBs on hospital length of stay and survival in people with COVID-19 Throughout the coronavirus pandemic there has been speculation that recovery from COVID-19 may be influenced by drugs inhibiting the renin-angiotensin-aldosterone system (RAAS) including angiotensin-converting enzyme inhibitors (ACEi) and angiotensin receptor blockers (ARBs) (1, 2) . Risk factors have been identified as predisposing to poor outcome with COVID-19 including: hypertension, cardiovascular disease, chronic kidney disease and diabetes (3) (4) (5) , all of which have clear indications for ACEi or ARB prescription (6) (7) (8) . Due to the numerous indications for these drugs in chronic disease management ramipril has become the fourth most commonly prescribed drug in the UK at 27 million prescriptions (9) . Of inpatients testing positive for SARS-CoV-2 in New York 8.3% and 10.5% were taking an ACEi or ARB respectively (10) . Teasing out the effect on COVID-19 recovery due to these comorbidities, or due to the drugs prescribed for them, has not been straightforward. Opinions on whether to withhold or continue community prescriptions of these medications during the pandemic have both been debated (11, 12) . However, due to a paucity of trial or observational evidence, organisations including the European Society of Cardiology and the American College of Cardiology and the American Heart Association have advocated continuing ACEi and ARB therapy to avoid deterioration of a person's underlying health issues (13) (14) (15) . Patients receiving these medications have been hypothesised to be detrimentally predisposed to infection; SARS-CoV-2 spike protein attaches to target cells by binding to the ACE2, which is upregulated with ACEi and ARBs in animal models (11, 16) . However, this remains controversial in humans, for example in a cohort of patients with heart failure, ACE2 was not found to be altered by ACEi or ARBs (17) . In contrast, beneficial effects of ACEi and ARBs have been proposed mediated via changes in both the innate and adaptive immune responses occurring within the RAAS (18, 19) . A New York database study and an Italian registry study have shown no increased rate of admissions in those taking an ACEi or ARB in propensity matched cohorts of admitted patients, inferring that there is no increased predisposition to COVID-19 (10) (20) . One Chinese multicentre study has shown a mortality benefit in COVID-19 positive patients who were taking ACEi or ARB for hypertension (21) . A recent metaanalysis including 9,890 patients across 10 studies showed a similar risk of dying from COVID-19 in those taking and not taking these drugs (22) . None of these studies have examined the effects of ACEi and ARB on non-mortality process outcomes such as length of stay. The primary aim was to investigate the influence of ACEi and ARB on mortality and hospital length of stay in patients diagnosed with COVID-19. These data were obtained as part of a prospective multicentre observational study: the COPE study (COVID-19 in Older People study). Authority in the United Kingdom Patients ≥ 18 years old admitted to hospital with laboratory or clinically diagnosed COVID-19 were eligible for inclusion. There were no exclusion criteria. Convenience sampling was undertaken from hospital admission lists which were screened by the local clinical teams. Data were collected from 27 th February to 25 th April 2020. The primary outcome was inpatient mortality. The time to outcome (death or discharge) was measured from patient admission, or the date of diagnosis (if diagnosis was five or more days after admission to take into account hospital acquired COVID-19). Patients' outcomes of discharged and mortality were censored on the date of each. Secondary outcomes included: length of hospital stay and Day-7 mortality. Outcomes were assessed up to 25 th April 2020 using paper and electronic health records. Demographics collected included age, sex, presence of clinical characteristics documented in the patient's health record: coronary artery disease (CAD), diabetes, hypertension, smoking status, and reduced kidney function. Creactive protein (CRP) was collected as a marker of disease severity. Both ACEi and ARB medication doses were recorded and categorised into low and high dose (low dose = as per the British National Formulary initial dose or below maintenance dose; high dose = equal or greater than maintenance dose). Doses for hypertension in adults 18-75 years old were used if the British National Formulary (BNF) stated more than one indication or specific age group for a drug (25, 26) . Sacubitril-valsartan was classified as an ARB. Patients were involved in discussion with the study conception and development of the protocol. A standardised case report form was used at all sites. All study personnel underwent data collection training under the supervision of each site's principal investigator. We summarised the baseline variables and outcomes using descriptive statistics. Missing smoking status was imputed in 19 cases as never smokers, and 31 cases of missing CRP were imputed as not elevated CRP. The primary outcome was analysed as the time to mortality. Secondary outcome events included time in days to discharge and Day-7 mortality. Patients who remained inpatient but had not reached their seventh day of admission by the end of the data collection period were excluded from the Day-7 analysis. Each time to event analysis was reported with a Kaplan-Meier survival plot. The primary outcome of time to mortality was analysed with a mixed-effects multivariable Cox's proportional baseline hazards model. The analysis was fitted with a random effect for site to account for variation occurring at each hospital, and adjusted for patient age group (≤64, 65-79, ≥80 years old), sex, disease severity at presentation (elevated CRP >40mg/L); diabetes (yes / no); hypertension (yes / no); CAD (yes / no), and kidney disease (eGFR >60ml/min/1.73m 2 / 60ml/min/1.73m 2 ). The baseline proportionality assumption was tested visually using log-log residuals. Secondary outcomes were Day-7 mortality (dead / alive), and the length of hospital stay (measured using time-to discharge). Day-7 mortality was analysed using a mixed-effects multivariable logistic regression model, fitting each site as a random effect to account for variation across hospitals, and adjusted with covariates consistent with the primary outcome. The length of stay was analysed using a multivariable Cox model consistent with the primary outcome. Adjusted odds ratio (aOR), and aHR were estimated alongside associated 95% confidence interval. To explore moderating effects in subgroups, the adjusted multivariable analyses were partitioned by: hypertension; diabetes; CAD; kidney disease; patient age; sex and smoking status. Analysis was carried out using Stata version 15. Kaplan Meier survival plots were visualised in R, with packages survival and survminer. We screened 1447 participants from all medical and surgical admissions. 76 participants were excluded due to: no positive laboratory polymerase chain reaction result or clinical diagnosis of COVID-19 found (n=60); and access not granted to records (n=16). The study included a total of 1,371 participants, of which 1124 (89.0%) patients were from the UK and 150 (11%) from Italy (Table 1 ). There were 63 patients still in hospital with less than seven days follow-up that were excluded from the Day-7 mortality analysis. The main study findings have been reported with an analysis looking at frailty (27) . The population median age was 74 years old (IQR, 61-83) with a similar number of participants between the age groups. 560 participants were female (40.9%) and 100 (7.3%) were current smokers ( ACEi or ARBs were not associated with inpatient mortality, Day-7 mortality rate (Table 2 ). However, they were associated with a reduced length of stay. Older age, kidney disease, and elevated CRP were associated with worse outcomes: inpatient mortality, increased Day-7 mortality, and increased length of stay. Presence of CAD was associated with increased Day-7 mortality and increased length of stay. Diabetes mellitus was associated with increased length of stay only. Hypertension and smoking status had no association with any outcome. In the crude analysis mortality there was no crude association between ACEi or ARB prescription and mortality (HR=1.01, 95%CI 0.80-1. 28 Figures 2-4) . A protective effect was demonstrated for ACEi and ARB prescriptions in hypertensive patients with a shorter length of stay (aHR=1.39, 95%CI 1.09-1.77, p=0.007, Supplementary Figure 4 ). There was a suggested finding of an ACEi or ARB prescription being moderated by the influence of smoking status: length of stay was reduced in ex-smokers prescribed an ACEi or ARB (aHR=1.46, 95%CI 1.08-1.98, p=0.015); with a stronger effect seen in current smokers (aHR=3.26, 95%CI 1.16-9.18, p=0.025). However, caution is needed when interpreting all subgroup analyses These data show that ACEi and ARBs were not associated with increased mortality in a hospital population admitted with a diagnosis of COVID-19. Furthermore, patients taking an ACEi or ARB had a reduced length of stay, and this was seen with greater effect in patients with hypertension, independent of age, other comorbidities or disease severity. Our demonstration of no difference in mortality between the ACEi/ARB and non-ACEi/ARB groups admitted with COVID-19 is in keeping with other studies (28) (22) . We have demonstrated a protective effect with a reduction in Day-7 mortality for patients with hypertension taking an ACEi or ARB. This fits with another multi-centre study in China showing similar mortality reductions at 28-day follow-up. However, compared to our study, their reported overall mortality rate was far lower (28.3% vs 8.8% respectively). This may have been due to a due to a younger cohort (median 74 [IQR 61-83] vs 64 [IQR 55-68] respectively) with fewer comorbidities (21) , at a different stage of the pandemic. We are the first to show that ACEi or ARB prescription has been linked to a reduction in the length of stay. Rapid discharge may represent either a marker of better disease recovery, or improvement in unmeasured factors that facilitate discharge from hospital service such as more rapid normalisation of oxygen saturations. The virus likely causes inactivation of ACE2, as has been see for SARS-CoV, and leads to an increase in angiotensin II (Ang II), which in turn acts via the angiotensin II type 1a receptor (AT1aR) to result in pulmonary vasoconstriction and increased lung endothelial permeability. This precipitates acute lung injury and potentially acute respiratory distress syndrome (29) . The reduced length of stay in all patients may be due to the fact that ACEi decreases Ang II production, by blocking the conversion of Ang I to Ang II, and ARBs block AT1aR preventing Ang II's actions, both theoretically resulting in a lower degree of lung injury, and faster recovery. However, despite this faster hospital recovery the overall physiological recovery may not be significant enough to counteract mortality from COVID-19. It is difficult to propose a mechanism that explains the consistent effect of ACEi and ARB improving outcomes for hypertensive patients that is not seen in other comorbidity groups. This may demonstrate better medical optimisation preventing significant inpatient events and so allowing more rapid recovery from COVID-19. In addition, non-prescription of an ACEi or ARB may represent a patient group that has not presented to medical services recently, or has ceased the drug, and therefore has undiagnosed or poorly optimised comorbidities (e.g. CAD); this study would be unable to detect this difference. These data were collected through a collaborative of ten representative hospitals across the UK and included one Italian hospital. Bias of data collection was minimised by the collaborative's established record in collecting multisite observational data (24) , as well as delivery of training to new contributing researchers. Patients were only included in this study if they were admitted to hospital. This would have precluded community cases who never presented to hospital due to COVID-19 being either less severe or fatal. We may have test, and may have been co-prescribed with ACEi and ARB (10), and being prescribed any antihypertensive drug has been shown to be protective of mortality from COVID-19, without ACEi and ARB being significant (31) . These results provide reassurance that patients on an ACEi or ARB at the point of COVID-19 diagnosis is not harmful. As with other studies we found factors that predisposed to higher mortality independent of ACEi or ARB prescription, including being over 65 years old, and having either kidney disease or CAD. In addition, clear differences in mortality outcomes were demonstrated between the age groups of <65, 65-79 and >80 years old. Disease severity at presentation as measured by CRP was associated with higher mortality. Our results have good generalisability across the UK covering three out of the four comprising countries -England, Wales and Scotland. They also represent a large sample of patients. Our inpatient mortality rate is higher than other studies and requires further examination to determine whether this is in relation to a higher threshold for admission to UK hospitals, patients having higher severity of illness, differences in therapy received, or more predisposing factors e.g. older and more comorbid. The prevalence of prescription of ACEi and ARBs in our COVID-19 cohort was higher than other reported populations, potentially representing more comorbidity with a greater frequency for ACEi and ARB indication, and prescribing practice associated with differing healthcare systems (COPE: ACEi 19.8%, ARB 8.8%; Italy: ACEi 23.9%, ARB 22.2% (20) ; New York: 8.3%, ARB 10.5% (10); China: ACEi and ARB grouped 5% (21)). These results may not be generalisable to non-hospitalised COVID-19 patients where a community vs hospital based COVID-19 study reported a lower community comorbidity burden and rate of ACEi / ARB prescription (20) (32) . This study provides reassurance to clinicians to continue ACEi and ARBs despite the risk of exposure to COVID-19. However, those patients that are more likely to receive either an ACEi or ARB -older patients with comorbidity -remain at higher risk of poor outcome from COVID-19. Whilst we have reported a shorter length of stay associated with ACEi or ARB, these results do not endorse the universal prescription of an ACEi or ARB as protective drugs in COVID-19. &&& 63 cases were excluded as the patient was followed up for less than 7 days and was alive and in hospital, and a further 17 cases were not included due to missing covariate data. &&& 29 cases were not included due to patient death on admission, and 16 cases were not included in the analysis due to missing covariate data Drugs and the renin-angiotensin system in covid-19 The BMJ Antihypertensive drugs and risk of COVID-19? -Authors' reply Clinical Characteristics of Coronavirus Disease 2019 in China Clinical features of patients infected with 2019 novel coronavirus in Wuhan Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Clinical characteristics of 113 deceased patients with coronavirus disease 2019: Retrospective study Effects of ramipril on cardiovascular and microvascular outcomes in people with diabetes mellitus: Results of the HOPE study and MICRO-HOPE substudy Effects of an Angiotensin-Converting-Enzyme Inhibitor Cardiovascular Events in High-Risk Patients Prescriptions Dispensed in the Community -Statistics for England Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Covid-19 Are patients with hypertension and diabetes mellitus at increased risk for COVID-19 infection? The Lancet. Respiratory medicine. NLM (Medline); 2020 Position Statement of the ESC Council on Hypertension on ACE-Inhibitors and Angiotensin Receptor Blockers SARS-CoV2: should inhibitors of the renin-angiotensin system be withdrawn in patients with COVID-19? Addresses Concerns Re: Using RAAS Antagonists in COVID-19 -American College of Cardiology Angiotensinconverting enzyme 2 is a functional receptor for the SARS coronavirus Circulating plasma concentrations of angiotensin-converting enzyme 2 inmen and women with heart failure and effects of renin-angiotensinaldosterone inhibitors Renin-Angiotensin-Aldosterone System Inhibitors in Patients with Covid-19 Angiotensinconverting enzyme in innate and adaptive immunity Renin-Angiotensin-Aldosterone System Blockers and the Risk of Covid-19 Association of Inpatient Use of Angiotensin Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers with Mortality Among Patients With Hypertension Hospitalized With COVID-19 Treatment with ACE inhibitors or ARBs and risk of severe/lethal COVID-19: A meta-analysis Study protocol for the COPE study: COVID-19 in Older PEople: the influence of frailty and multimorbidity on survival. A multicentre, European observational study Association between preadmission frailty and care level at discharge in older adults undergoing emergency laparotomy Lower-dose prescriptions in the postmarketing situation and the influencing factors thereon Adverse drug effects, compliance, and initial doses of antihypertensive drugs recommended by the Joint National Committee vs the Physicians' Desk Reference The effect of frailty on survival in patients with COVID-19 (COPE): a multicentre, European, observational cohort study Treatment with ACE-inhibitors is associated with less severe disease with SARS-Covid-19 infection in a multi-site UK acute Hospital Trust. medRxiv A Systematic Review of Cases of Acute Respiratory Distress Syndrome in the Coronavirus Disease Randomised Evaluation of COVID-19 Therapy (RECOVERY) -Health Research Authority Association of hypertension and antihypertensive treatment with COVID-19 mortality: a retrospective observational study Association of Cardiac Injury with Mortality in Hospitalized Patients with COVID-19 in Wuhan, China