ENGAGE AF–TIMI 48 HF Subscore: Background, meaning and references The ENGAGE AF–TIMI 48 randomized trial (edoxaban vs warfarin) enrolled over 21,000 patients with atrial fibrillation. A prespecified analysis examined clinical predictors of a composite endpoint of heart-failure hospitalization, death due to heart failure, or sudden cardiac death among patients without a prior history of heart failure. Multivariable Cox analysis identified higher age, impaired renal function (creatinine clearance ≤50 mL/min), reduced LVEF <50%, current diuretic use, higher heart rate, diabetes, heart valve disease, prior stroke/TIA, coronary artery disease, and hypertension as independent predictors. The hazard ratios for those predictors were used to derive the log-coefficients implemented in this calculator. Biomarkers (NT-proBNP, troponin) provided additional prognostic value when available, improving discrimination beyond the clinical model. This calculator returns a relative risk multiplier (derived from the published coefficients) and an approximate absolute risk by scaling the cohort event rate (6.6% over median 2.8 years) — this absolute estimate is approximate because the original paper did not publish an individual baseline hazard function.
Reference:
Inciardi RM, Giugliano RP, Nordio F, et al. Heart failure hospitalization and mortality in non-valvular atrial fibrillation: results from ENGAGE AF–TIMI 48 (multivariable clinical predictors and added value of biomarkers). (ENGAGE AF–TIMI 48 analysis). Presented/published summary and slides; cohort event-rate and hazard ratios as reported.
Trial & design reference:
Ruff CT, Giugliano RP, Antman EM, et al. Design and rationale for ENGAGE AF–TIMI 48 (Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation—TIMI 48).
Additional reading / echo substudy (mechanistic insights):
Gupta DK, et al. The ENGAGE AF–TIMI 48 Echocardiographic Substudy — echocardiographic measures and outcome associations.