A Self-Controlled Case Series to Evaluate Risk of Ischemic Stroke Following Bivalent COVID-19 Vaccine Administration in a US Veteran Population

Background: A recent FDA study reported an association between the bivalent COVID-19 vaccines and an increased risk of ischemic stroke in patients aged >= 65 years. To further evaluate this risk, the US Department of Veterans Affairs (VA) conducted a self-controlled case series study evaluating the risk of ischemic stroke events following bivalent COVID-19 vaccines in a US Veteran population.

Objectives: To evaluate the risk of ischemic stroke in patients who received bivalent COVID-19 vaccines.

Methods: Data from the VA Corporate Data Warehouse were used to identify patients who were vaccinated with bivalent COVID-19 and/or influenza vaccines from September 1, 2022 to March 9, 2023 using CPT and CVX codes. Patients were restricted to VA healthcare users and aged 18 years or older. Post-vaccination ischemic stroke was identified using ICD-10 codes (G45.8, G45.9, and I63.x) in inpatient and emergency department settings. The modified self-controlled case series (SCCS) method was employed. Relative Incidence (RI) was calculated in the risk period (1-21 days after vaccination) versus the non-risk periods (22-42 days and 22-90 days). Analysis was performed by vaccine type and stratified by age ( < 65 years and >= 65 years). Ischemic stroke was also evaluated restricting to those who received bivalent and influenza vaccines on the same day.

Results: Approximately, 768,000 patients, 390,000 mRNA-1273 (Moderna) and 378,000 BNT162b2 (Pfizer-BioNTech), were vaccinated with bivalent COVID-19 vaccines. Overall, 72% were aged 65 years or older; 91% were male, 21% were Black, and 68% were White. In total, 555 patients had ischemic stroke within 90 days of bivalent vaccination. Comparing 1-21 days vs 22-42 days, RIs were 0.60 (0.22, 1.65) for mRNA-1273 and 1.10 (0.47, 2.59) for BNT162b2 for age < 65 years and 0.88 (0.58, 1.34) for mRNA-1273 and 1.18 (0.83, 1.68) for BNT162b2 for age >= 65 years. Comparing 1-21 days vs 22-90 days, RIs were 0.70 (0.28, 1.71) for mRNA-1273 and 0.81 (0.42, 1.59) for BNT162b2 for age < 65 years and 0.74 (0.52, 1.04) for mRNA-1273 and 1.03 (0.78, 1.36) for BNT162b2 for age >= 65 years. In 269,000 patients who received co-administration of bivalent and influenza vaccines there were 238 ischemic stroke events, and the analyses restricted to this group showed similar results.

Conclusions: No increased risk of ischemic stroke was found by vaccine type and by age subgroups. The analysis on patients with co-administration yielded the same results. Ischemic stroke events are rare. Further investigation is needed with limitations of the SCCS method.