| Sumario: | Background In older patients with heart failure, frailty assessment tools potentially have different prognostic value for all-cause mortality. Objective To compare the Frailty Screening Index (FSI), FRAIL scale and Fried phenotype. Design Post hoc analysis of the FRAGILE-HF study. Subjects Hospitalised patients (age ≥ 65 years) with heart failure. Methods We assessed frailty status at discharge, considered preadmission conditions of patients, using the FSI, FRAIL scale and Fried phenotype models and evaluated the primary outcome (2-year all-cause mortality), prognostic performance using Cox regression analysis (adjusted for the Meta-analysis Global Group in Chronic Heart Failure risk score and log-transformed B-type natriuretic peptide) and discriminative and reclassification abilities (area under the receiver-operating characteristic curve and continuous net reclassification improvement, respectively). Results Among 961 patients [median age: 80 (IQR, 74–85) years; 41.5% women], frailty was identified in 48.5%, 44.5% and 55.8% by the FSI, FRAIL scale and Fried models, respectively. During the 2-year follow-up, 187 (19.5%) patients died. On Kaplan–Meier analysis, frailty defined by the FRAIL and Fried was associated with significantly lower survival; FSI showed no significant difference. Adjusted Cox models showed that frailty was independently associated [HR (95% CI)] with mortality in FRAIL [1.31 (1.01–1.71)] and Fried [1.35 (1.04–1.75)] but not FSI [1.11 (0.84–1.37)]. The Fried model, incorporated into the baseline model, yielded the highest incremental prognostic value for 2-year all-cause mortality. Conclusions Frailty assessment using different tools yields varying results, with notable differences in their prognostic value for 2-year all-cause mortality. The frailty assessment tool influences both frailty classification and risk stratification and warrants careful clinical selection.
|
|---|