Abstract 19059: Risk Factors for the First Cardiopulmonary Hospitalization Differ From Those of Recurrent Cardiopulmonary Hospitalizations in Patients With Cardiovascular Disease Vaccinated Against Influenza — A Secondary Analysis of t
Bahar Behrouzi, Brian Claggett, Thomas Havighurst, Lu Mao, Jacob Joseph, Peter Austin, Douglas Lee, Orly Vardeny, Scott Solomon, Jacob A Udell- Physiology (medical)
- Cardiology and Cardiovascular Medicine
Background: Influenza vaccines are recommended for patients with CVD due to their protective effects and favorable safety profile. Patients with recent MI/HF hospitalization are at high risk of multiple events, but it is not known which flu vaccine type can mitigate recurrent events.
Research Questions: What are predictors of recurrent cardiopulmonary hospitalization and death in patients with CVD and does the high-dose trivalent flu vaccine reduce the rate of recurrent events compared with standard-dose quadrivalent vaccine?
Methods: In this pragmatic, active comparator trial, patients with recent MI/HF hospitalization across 157 sites were randomized to either vaccine type and revaccinated each enrolled season (2016-2019). Recurrent event analysis with a Markov multistate model, negative binomial, and Lin-Wei-Yang-Ying regression models was applied, adjusting for clinical and sociodemographic factors, flu infection and vaccine history, with vaccine type as the main exposure. Cardiopulmonary hospitalization and all-cause death comprised the multistate model’s main states and the composite primary outcome for remaining analyses.
Results: Among 5260 patients (mean age 65.5 years; 28% women; 63% with recent HF), 12% were readmitted ≥1 for cardiopulmonary causes, accounting for half of total cardiopulmonary hospitalization. In the multistate model, a history of comorbidities (e.g., low LVEF and COPD) was associated with recurrent hospitalization (Figure), whereas age and frailty primarily determined a higher rate of death, particularly after a first hospitalization. There were no significant differences between the vaccines on the rate of recurrent events and death in the multistate and negative binomial regression models.
Conclusions: As hospitalizations accrue, the association of certain predictors changes in this population. Alternative strategies beyond vaccine type are needed to mitigate the total burden of cardiopulmonary disease.