Association and Prognostic Value of Multi-Domain Frailty Defined by Cumulative Deficit and Phenotype Models in Patients with Heart Failure

Background: Frailty is associated with a poor prognosis in older patients with heart failure (HF). However, multidomain frailty assessment tools have not been established in patients with HF, and the association between the frailty phenotype and the deficit-accumulation frailty index in these patients is unclear. We aimed to understand this relationship and evaluate the prognostic value of the deficit-accumulation frailty index in older patients with HF. Methods: We retrospectively analyzed FRAGILE-HF cohort, which consisted of prospectively registered hospitalized patients with HF aged >= 65 years. The frailty index was calculated using 34 health-related items. The physical, social, and cognitive domains of frailty were evaluated using a phenotypic approach. The primary endpoint was all-cause mortality. Results: Among 1027 patients with HF (median age, 81 years; male, 58.1%; median frailty index, 0.44), a higher frailty index was associated with a higher prevalence in all domains of cognitive, physical, and social frailty defined by the phenotype model. During the 2-year follow-up period, a higher frailty index was independently associated with all-cause death even after adjustment for Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) score plus log B-type natriuretic peptide (per 0.1 increase: hazard ratio, 1.21; 95% confidence interval, 1.07-1.37; P = 0.002). The addition of the frailty index to the baseline model yielded statistically significant incremental prognostic value (net reclassification improvement, 0.165; 95% confidence interval, 0.012-0.318; P = 0.034). Conclusions: A higher frailty index was associated with a higher prevalence of all domains of frailty defined by the phenotype model and provided incremental prognostic information with pre-existing risk factors in older patients with HF.