Prognosis and outcome determinants after heart failure diagnosis in patients who underwent aortic valvular intervention

Abstract

Aims: To study clinical phenotype, prognosis for all-cause and cardiovascular (CV) mortality and predictive factors in patients with incident heart failure (HF) after aortic valvular intervention (AVI) for aortic stenosis (AS). Methods and results: In this retrospective, observational study we included patients from the Swedish Heart Failure Registry (SwedeHF) recorded 2003–2016, with AS diagnosis and AVI before HF diagnosis. The AS diagnosis was established according to International Classification of Diseases 10th revision (ICD-10) codes, thus without information concerning clinical or echocardiographical data on the aortic valve disease. The patients were divided into two subgroups: left ventricular ejection fraction (LVEF) ≥ 50% (AS-HFpEF) and <50% (AS-HFrEF). We individually matched three controls with HF from the SwedeHF without AS (control group) for each patient. Baseline characteristics, co-morbidities, survival status and outcomes were obtained by linking the SwedeHF with two other Swedish registries. We used Kaplan–Meier curves to present time to all-cause mortality, cumulative incidence function for time to CV mortality and Cox proportional hazards model to evaluate the relative difference between AS-HFrEF and AS-HFpEF and AS-HF and controls. The crude all-cause mortality was 49.0%, CV mortality 27.9% in AS-HF patients, respectively 44.7% and 26.6% in matched controls. The adjusted risk for all-cause mortality and CV mortality was similar in HF, regardless of LVEF vs. controls. No significant difference in factors predicting higher all-cause mortality was observed in AS-HFrEF vs. AS-HFpEF, except for diabetes (only in AS-HFrEF), with statistically significant interaction predicting death between the two groups. Conclusions: In this nationwide SwedeHF study, we characterized incident HF population after AVI. We found no significant differences in all-cause and CV mortality compared with general HF population. They had virtually the same predictors for mortality, regardless of LVEF.