Electronic health record analysis reveals no association between COVID-19 vaccines and ischemic stroke

Abstract

Background: We aimed to conduct a comprehensive investigation of the association between COVID-19 vaccination and incident ischemic stroke. Methods: Using data from the National COVID Cohort Collaborative (N3C) database, we identified 119,275 patients (n = 39,922 vaccinated, n = 79,353 unvaccinated) who were at least 16 years of age and suffered ischemic stroke between February 2021 and April 2023. We applied the Self-Controlled Case Series (SCCS) to estimate the association between multiple vaccine-type-dose combinations and risk of incident ischemic stroke. We calculated incident rate ratios (IRRs) to compare incidence of ischemic stroke during the 28-day risk period following vaccine or SARS-CoV-2 exposure versus baseline periods of no vaccine exposure. Because patients who experience strokes are unlikely to receive the first or subsequent vaccinations, we applied SCCS methods capable of handling such event-dependent exposure to ensure valid results. Results: Using the SCCS method robust to event-dependent exposures, no associations between any of the vaccine exposures and incident ischemic stroke were found. When we applied the standard SCCS, however, relative to baseline, there was elevated risk of incident ischemic stroke following first doses of BNT162b2 (IRR = 1.42, 95 % CI: 1.24–1.62), mRNA1273 (IRR =1.37, 95 % CI: 1.16–1.62), and Ad26.COV2·S (IRR = 1.40, 95 % CI: 1.08–1.81) vaccine types. No significant results emerged for the Bivalent vaccines. Effect sizes were larger in patients vaccinated once versus multiple times, implying ischemic stroke incidence likely prevented patients receiving initial and/or subsequent vaccinations. Conclusions: Results indicate no association between mRNA or adenovirus COVID-19 vaccines and ischemic stroke.