Abstract 10837: Prognostic Importance of Atrial Fibrillation in Heart Failure According to Time Elapsed Since Diagnosis

Abstract

Introduction: Mortality rates in heart failure (HF) are greatest early after onset and decline rapidly with time (with the survival of the less sick patients). However, it is not known if the presence of atrial fibrillation (AF), a well-known risk factor for mortality in this population, alters such course. Methods: From the Danish nationwide administrative database, all first time HF patients were identified between 1998-2018 using ICD coding and followed for all cause mortality until the end of 2018. AF was categorized as occurring antecedent (before), concomitant (with), or after HF diagnosis. Time-dependent Poisson regression model was used to determine mortality rate ratios, using ‘no AF’ as referent, with multivariable adjustment. Results: A total of 252,988 HF patients (45% women, mean age 74 ±13 years) were included. Of these, 54,064 (21%) had AF before HF onset and 27,651 (11%) patients had AF diagnosed concomitantly with HF. During follow-up, 30,565 additional patients developed AF. The cumulative mortality and mortality rates were highest early after onset for all groups, but greater for those with AF versus without AF, regardless of the time of onset, Figure A and B. Of note, the rates declined substantially more and faster for patients without AF, rendering AF greater prognostic importance the longer out from initial diagnosis, p for interaction between AF groups and HF duration <0.0001, Figure C. Conclusion: While mortality declined in all patients with HF over time, patients with AF experienced substantially less decline in mortality than the no AF group, rendering AF of greater prognostic importance later in the clinical course of HF.