The impact of atrial fibrillation on prognosis in aortic stenosis

Abstract

Abstract Background Atrial fibrillation (AF) and aortic stenosis (AS) are both highly prevalent and increasing with age. Various studies have focused on the complex relationship between these entities that frequently coexist. AS might induce adverse cardiac remodelling, which is associated with poor prognosis in severe AS. Left atrial remodelling, especially left atrial enlargement, is also an important risk factor for AF. Purpose To evaluate the additive prognostic value of AF besides markers of left atrial and left ventricular remodelling in patients with AS, irrespective of severity of AS and left ventricular ejection fraction (LVEF). Methods Patients with moderate and severe AS were selected and history of AF was assessed. Subgroups were defined according to LVEF (reduced (<50%) vs. preserved (≥50%)) and severity of AS (moderate vs. severe). The endpoint was all-cause mortality. Unadjusted Kaplan-Meier survival curves were plotted. Four multivariable Cox regression models were constructed. Results In total, 2849 patients with moderate and severe AS (mean age 72±12 years, 54.7% men) were evaluated of whom 686 (24.1%) had a history of AF. Regarding the defined subgroups, 1091 (38.3%) patients had severe AS and 2207 (77.5%) patients had preserved LVEF. During a median follow-up time of 60 months (interquartile range 30 to 97), 1182 (41.5%) patients died. Ten-year mortality rate in patients with AF was 46.8% compared with 36.8% in patients with sinus rhythm (SR) (Figure 1) (p<0.001). In subgroup analysis, patients with AF and severe AS, moderate AS or preserved LVEF had worse survival than those who maintained SR (p=0.015, p<0.001 and p<0.001 respectively). On univariable (HR: 1.42; 95% CI: 1.25 to 1.62; p<0.001) and multivariable Cox regression analysis (HR: 1.19; 95% CI: 1.02 to 1.38; p=0.026) adjusting for age, body mass index, hypertension, diabetes mellitus, coronary artery disease, chronic obstructive pulmonary disease, kidney function, New York Heart Association class, aortic valve replacement as a time-dependent covariate, left ventricular mass index, left ventricular end-diastolic volume index, LVEF, mean aortic valve gradient, tricuspid annular plane systolic excursion, AF is independently associated with mortality (Table 1; model 1). However, when correcting for LAVI, E/e' or both, AF is no longer independently associated with all-cause mortality (Table 1; model 2–4). Conclusion Patients with moderate or severe AS and AF have a significantly higher 10-year mortality rate than patients with SR. This finding is irrespective of AS severity and also apparent in the subgroup with preserved LVEF. Nonetheless, when correcting for markers of diastolic dysfunction, AF is not independently associated with outcome in patients with moderate or severe AS. Funding Acknowledgement Type of funding sources: None.