Applicability and prognostic value of frailty assessment tools among hospitalized patients with advanced chronic liver disease

Abstract

AimTo assess and compare the feasibility and prognostic value of various frailty assessment tools among decompensated cirrhosis inpatients.MethodsOur prospective observational registry included consecutive patients admitted for cirrhosis between June 2017 and July 2018. Exclusion criteria were intensive-care unit admission, hepatocellular carcinoma outside of the Milan criteria, and other malignancies. Frailty at baseline was assessed with the Liver Frailty Index (LFI), Clinical Frailty Scale (CFS), Fried Frailty Score (FFS), and Short Physical Performance Battery test (SPPB). The follow-up lasted for at least 180 days.ResultsThe study enrolled 168 patients (35.1% women, median age 57.9 years). The most frequent primary etiology was alcohol-related liver disease (78.6%). The Median Model for End-Stage Liver Disease (MELD) was 16. The 80th percentile of frailty scores was LFI>5.4, CFS>4, FFS>3, and SPPB<5, and it identified patients with higher mortality. LFI and CFS had the highest numerical prognostic value for in-hospital, and 90- and 180-day mortality. In a bivariate analysis of the risk of death or liver transplantation, the combination of MELD and LFI had the highest concordance (0.771 ± 0.04). In a multivariate model, MELD score (HR 1.17, 95% CI 1.12-1.22), overt encephalopathy (2.39, 1.27-4.48), infection at baseline (2.32, 1.23-4.34), and numerical LFI (1.41, 1.02-1.95) were independent predictors of overall mortality.ConclusionFrailty assessment using the evaluated tools is feasible among hospitalized cirrhotic patients, identifying those with worse prognosis. CFS had the highest applicability and accuracy for the initial assessment and LFI for the initial and follow-up assessments.