Association between renin–angiotensin–aldosterone system blockers and outcome in coronavirus disease 2019: analysing in-hospital exposure generates a biased seemingly protective effect of treatment

OBJECTIVE: The role of renin-angiotensin-aldosterone system (RAAS) blockers during the coronavirus disease 2019 (COVID-19) pandemic is a matter of controversies. Studies based on in-hospital exposure have suggested a beneficial effect of these drugs, unlike those based on chronic exposure. We aimed to analyse RAAS blocker prescription before and during hospital stay in patients with COVID-19, and the corresponding outcomes, to explain these discrepant results. METHODS: In a retrospective cohort study conducted in 347 patients hospitalized for COVID-19 (Bichat Hospital, Paris, France, 23 January-29 April 2020), RAAS blocker exposure, as well as timing and reason for treatment modifications, were collected. The association between exposure and mortality within 30 days of hospital admission was analysed using logistic regression analysis adjusted for age, sex, and comorbidities. RESULTS: Median age was 61 [interquartile range, 51-72] years, 209 (60%) were male, 169 (49%) had a history of treated hypertension, and 117 (34%) received a RAAS blocker prior to hospitalization. RAAS blockers were discontinued within the first 7 days of hospital admission in 33% of previously treated patients (mostly driven by severity of the disease), with a corresponding mortality rate of 33%. Mortality was 8% when treatment was maintained or introduced, and 12% in patients never exposed. Adjusted odds ratios for association between exposure and mortality were 0.62 (95% confidence interval 0.25-1.48) based on chronic exposure and 0.25 (0.09-0.65) based on in-hospital exposure. CONCLUSION: A 'healthy user-sick stopper' bias influences RAAS blocker prescription after hospital admission for COVID-19, and explains the seemingly favourable outcome associated with in-hospital treatment.