Clinical and Therapeutic Phenotypic Clustering and Prognostic Stratification in Heart Failure Patients With Atrial Fibrillation
Masamichi Yano, Yasuyuki Egami, Kazushi Nagai, Haya Matsubara, Ryo Muroya, Taichi Mukai, Noriyuki Kobayashi, Ayako Sugino, Masaru Abe, Hiroaki Nohara, Shodai Kawanami, Koji Yasumoto, Naotaka Okamoto, Yasuharu Matsunaga‐Lee, Masami NishinoABSTRACT
Background
Atrial fibrillation (AF) commonly coexists with heart failure (HF), but its prognostic impact remains uncertain. This study aimed to identify phenotypic subgroups of HF patients with AF and assess their clinical outcomes.
Methods
A total of 407 hospitalized HF patients with AF on admission were analyzed. Key prognostic factors (age, sex, NT‐proBNP, creatinine, hemoglobin, albumin, and use of β‐blockers, renin‐angiotensin‐aldosterone system inhibitors) were standardized and subjected to hierarchical cluster analysis. The primary endpoint was a composite of all‐cause death and HF readmission.
Results
Four phenotypes with distinct clinical profiles were identified. Phenotype 1 comprised younger, predominantly male patients with relatively reduced left ventricular ejection fraction but preserved systemic condition and the highest use of guideline‐directed medical therapy (GDMT), resulting in the most favorable outcomes. Phenotype 2 showed intermediate frailty and moderate GDMT use. Phenotype 3 was characterized by advanced age, systemic impairment, renal dysfunction, and anemia. Phenotype 4 represented the oldest and most frail patients with severe malnutrition and minimal GDMT use. During a median follow‐up of 612 days, Cox analysis demonstrated progressively increasing risk for the composite endpoint compared with Phenotype 1 (hazard ratios: 1.87, 3.80, and 4.60; p for trend < 0.001). Kaplan–Meier curves confirmed significant divergence in event‐free survival among groups.
Conclusions
HF patients with AF exhibit four distinct phenotypes associated with progressively worsening outcomes. Phenotypic stratification may improve individualized risk assessment and therapeutic decision‐making in this high‐risk population.
Trial Registration
Data were obtained from the Acute Heart Failure Registry in the Osaka Rosai Hospital (AURORA; UMIN000045096)