Umeå universitets logga

Background and Aim: COVID-19 is a multiorgan disease and there has been increasing reports of cardiovascular complications. However, previous studies have shown conflicting results and have mainly included hospitalized individuals with severe disease. The aim of this thesis was to estimate the risk of incident cardiovascular disease following COVID-19. 

Material and Methods: This project was based on Swedish national register data from all individuals who tested positive for SARS-CoV-2 between February 1^st, 2020, and May 25^th, 2021. Outcomes were events of incident cardiovascular disease, recorded as ICD-10 codes in the National Patient Register. Self-controlled case series (SCCS) studies and matched cohort studies were performed to determine the relative risks for a new onset cardiovascular event following COVID-19. Moreover, a data-simulation study was performed to investigate features that could introduce bias in the SCCS studies: the "day zero-effect", i.e., a high incidence of events at the COVID-19 date; and the increase in mortality due to cardiovascular events.

Results: In the SCCS studies, the risk of cardiovascular disease was significantly increased compared to the control period as follows: up to 14 days after COVID-19 for acute myocardial infarction; up to 1 month for ischemic stroke; up to 3 months for deep vein thrombosis; up to 6 months for pulmonary embolism; up to 2 months for bleeding and for atrial tachycardias; up to 6 months for paroxysmal supraventricular tachycardias; and up to 14 days for bradyarrhythmias. In the matched cohort studies, COVID-19 was associated with an approximately 3- and 4-fold increase in the risk of acute myocardial infarction and ischemic stroke, respectively, during day 1-14 after the infection. During day 1-30 following the infection, the increase in risk was 5-fold for deep vein thrombosis; 33-fold for pulmonary embolism; 2-fold for bleeding; 12-fold for atrial tachycardias; 5-fold for paroxysmal supraventricular tachycardias; and 3-fold for bradyarrhythmias. The relative risks were higher in older individuals with comorbidities, with more severe COVID-19, and during the first months of the pandemic. Unvaccinated individuals had a higher risk of arrhythmias. In the data-simulation study, bias was introduced by including "day-zero events" in the analyses. Moreover, the extended rather the traditional SCCS model was more appropriate to minimize possible bias introduced by the increase in mortality due to cardiovascular events.

Conclusion: There is an increased risk of cardiovascular complications in individuals with COVID-19, especially in individuals with severe disease. These findings highlight the value of diagnostic and prophylactic strategies in individuals with COVID-19, such as risk factor control or thromboprophylaxis, and the value of vaccination.