Risk of Ischemic Stroke after COVID-19 Bivalent Booster Vaccination in an Integrated Health System (preprint)

Abstract: Purpose: What is the absolute occurrence of ischemic stroke and transient ischemic attack after a COVID-19 bivalent vaccination? Methods: We conducted a retrospective cohort study of Kaiser Permanente Northwest (KPNW) patients 18 years and older who were vaccinated with either the Pfizer or Moderna formulation of the COVID19 bivalent vaccine between September 1, 2022 and March 1st 2023. Patients were included in the study if they had KP membership at the time of vaccination and through the 21-day follow up period. We replicated the Vaccine Safety Datalink (VSD) rapid cycle analysis methodology and searched for possible cases of ischemic stroke or TIA in the 21 days following vaccination using ICD10CM diagnosis codes in both the primary position and any position. We waited 90 days from the end of the follow up (March 21, 2023) for complete non KP data accrual before analyzing the data to account for the lag in processing outside hospital insurance claims. Two physicians adjudicated possible cases by reviewing the clinical notes in the electronic health record. The analyses were stratified by age 65 years and older to allow for comparisons with VSDs reporting at the Advisory Committee on Immunization Practices (ACIP) meeting of incidence of ischemic stroke or TIA (VSD reported incidence; 24.6 cases of ischemic stroke or TIA per 100,000 patients vaccinated). Results: The incidence of ischemic stroke or TIA was 34.3 per 100,000 (95% CI, 17.7 to 59.9) in patients 65 years or older who received the bivalent Pfizer vaccine, based on a diagnosis code in the primary position of the emergency department or hospital discharge. The incidence increased to 45.7 per 100,000 (95% CI 26.1 to 74.2) when we expanded the search to a diagnosis in any position and did not adjudicate to confirm. However, most of those additional apparent stroke or TIA diagnoses were false-positive diagnoses based on physicians adjudications. Estimating the incidence based on the primary position agreed closely with estimating the incidence based on any position and physician adjudication: 37.1 per 100,000 (95% CI 19.8 to 63.5). Seventy-nine percent of the ischemic stroke cases were admitted to hospitals that are not owned by the integrated delivery system. Conclusion: We identified a 50% increase in the incidence of ischemic stroke per 100,000 patients ages 65 and older vaccinated with the Pfizer bivalent vaccine, compared to the data presented by the VSD. Seventy-nine percent of the ischemic stroke cases were admitted to hospitals that are not owned by the integrated delivery system and a delay in processing outside hospital insurance claims was likely responsible for the discrepancy in case ascertainment of ischemic stroke. Physician adjudication of all cases in this study allowed accurate absolute incidence estimates of stroke per 100,000 vaccine recipients and is helpful in calculation of net benefit for policy recommendations and shared decision-making.

Risk of Myopericarditis following COVID-19 mRNA vaccination in a Large Integrated Health System: A Comparison of Completeness and Timeliness of Two Methods (preprint)

Purpose: How completely do hospital discharge diagnoses identify cases of myopericarditis after an mRNA vaccine? Methods: We assembled a cohort 12 to 39 years old patients, insured by Kaiser Permanente Northwest, who received at least one dose of an mRNA vaccine (Pfizer BioNTech or Moderna) between December 2020 and October 2021. We followed them for up to 30 days after their second dose of an mRNA vaccine to identify encounters for myocarditis, pericarditis or myopericarditis. We compared two identification methods: A method that searched all encounter diagnoses using a brief text description (e.g., ICD-10-CM code I40.9 is defined as acute myocarditis, unspecified). We searched the text description of all inpatient or outpatient encounter diagnoses (in any position) for myocarditis or pericarditis. The other method was developed by the Centers for Disease Control and Preventions Vaccine Safety Datalink (VSD), which searched for emergency department visits or hospitalizations with a select set of discharge ICD-10-CM diagnosis codes. For both methods, two physicians independently reviewed the identified patient records and classified them as confirmed, probable or not cases using the CDCs case definition. Results: The encounter methodology identified 14 distinct patients who met the confirmed or probable CDC case definition for acute myocarditis or pericarditis with an onset within 21 days of receipt of COVID-19 vaccination. Three of these 14 patients had an ICD-10 code of I51.4 Myocarditis, Unspecified which was overlooked by the VSD algorithm. The VSD methodology identified 11 patients who met the CDC case definition for acute myocarditis or pericarditis. Seven (64%) of the eleven patients had initial care for myopericarditis outside of a KPNW facility and their diagnosis could not be ascertained by the VSD methodology until claims were submitted (median delay of 33 days; range of 12-195 days). Among those who received a second dose of vaccine (n=146,785), we estimated a risk as 95.4 cases of myopericarditis per million second doses administered (95% CI, 52.1 to 160.0). Conclusion: We identified additional valid cases of myopericarditis following an mRNA vaccination that would be missed by the VSDs search algorithm, which depends on select hospital discharge diagnosis codes. The true incidence of myopericarditis is markedly higher than the incidence reported to US advisory committees. The VSD should validate its search algorithm to improve its sensitivity for myopericarditis.