Timing-Specific Factors Associated With Early, Intermediate, and Late Readmission After Heart Failure Hospitalisation

Why clinicians should know about it

• Picked for Internal Medicine (paper of the day, 22 August 2026): Factors linked to early, intermediate, late HF readmission

Abstract

AIMS: To identify factors associated with early, intermediate, and late readmission after heart failure hospitalisation. METHODS AND RESULTS: This secondary analysis included 1,992 patients discharged alive from a single-centre Chinese heart failure cohort. Median age was 75 years (interquartile range 65-85), and 58.2% were female. Within 6 months, 776 patients (39.0%) were readmitted: 140 (7.0%) early (≤28 days), 358 (18.0%) intermediate (29-90 days), and 278 (14.0%) late (91-180 days). Early readmission was associated with biventricular heart failure (odds ratio 1.72; P = 0.022), New York Heart Association class IV (1.47; P = 0.043), estimated glomerular filtration rate <60 mL/min/1.73 m2 (1.64; P = 0.021), and hyponatraemia (1.71; P = 0.008). Intermediate readmission was associated with biventricular heart failure (1.68; P < 0.001), systolic blood pressure <110 mmHg (1.68; P < 0.001), and B-type natriuretic peptide ≥1,000 pg/mL (1.32; P = 0.029). Late readmission was associated with dementia (2.30; P < 0.001), biventricular heart failure (1.69; P = 0.002), chronic kidney disease (1.56; P = 0.030), and diabetes (1.44; P = 0.043). Apparent areas under the curve were 0.677, 0.641, and 0.629 for early, intermediate, and late readmission, respectively. CONCLUSIONS: Associated factors differed across post-discharge intervals but overlapped. Early readmission was associated with advanced heart failure severity and cardiorenal-electrolyte instability. Intermediate readmission was associated with haemodynamic compromise and cardiac stress. Late readmission demonstrated stronger associations with chronic comorbidity burden, alongside cognitive and systemic disease. Biventricular heart failure was associated with all three intervals. These findings require external validation.

Abstract as published, via PubMed.