Physical frailty predicts adverse outcomes in hospitalized patients with acute decompensation of cirrhosis

ABSTRACT BACKGROUND AND AIM: Patients with end-stage liver disease are at high risk of malnutrition, sarcopenia and frailty. Recent evidence indicates that physical frailty, as assessed by liver frailty index (LFI), is an independent predictor of mortality in patients with cirrhosis. However, majority of available data are from North American cohorts in the outpatient setting. We prospectively investigated the relationship between physical frailty and sarcopenia in hospitalized patients with cirrhosis and their impact on clinical outcomes. PATIENTS AND METHODS: All patients with cirrhosis admitted to our unit between October 1st, 2021 and March 31st, 2023 were prospectively screened for recruitment. Exclusion criteria were: age<18 or>70 years; known musculoskeletal disease; history of liver transplantation. LFI was calculated at admission. Frailty was defined according to previously established thresholds: <3.2 (robust); between 3.2 and 4.5 (pre-frail); >4.5 (frail). Skeletal muscle mass was assessed via CT scan when this performed during hospital stay. Sarcopenia was defined by skeletal muscle index (SMI) <50 cm²/m² in males and <39 cm²/m² in females. All patients were prospectively followed for liver-related death. Cox regression analysis was used to identify predictors of primary outcome. RESULTS: We included 127 patients with cirrhosis (median age 59 years; 68.5% male). The median MELD score was 22 (IQR 17-27); 70% of patients had Child-Pugh C cirrhosis. Alcohol-related liver disease was the most common aetiology of cirrhosis (62%). At inclusion, 48.8% of patients were frail and 60% were sarcopenic. Compared to robust and non- sarcopenic individuals, those with frailty and sarcopenia had higher risk of further decompensation with increased need of ICU care during hospitalization. Frailty – but not sarcopenia – was associated with a longer hospital stay (15 [9-27] vs. 6.5 [4-15] in robust patients; p=0.001). Rate of re-hospitalization was comparable between patients with and without frailty (p=0.589) and sarcopenia (p=0.065). During a median follow-up of 96 days (IQR 37-384), 39.4% patients died. Both frailly and sarcopenia were associated with a higher mortality (60% vs 20% and 58% vs. 38%, respectively; p < 0.001 and p<.009). In multivariate analysis (Table 1), frailty (HR:3,377; 95% CI 1,729-6,597; p<0.001) and ACLF (HR:2,354, 95% CI 1.124-4.929; p=0.023) were the only independent predictors of death. The association between frailty and mortality was confirmed in a post-hoc analysis in both patients with and without ACLF separately. CONCLUSIONS: Frailty and sarcopenia are highly prevalent in hospitalized patients with cirrhosis and are both associated with higher rates of ICU admission and health care costs. However, frailty emerges as the strongest, independent predictor of early mortality after index hospitalization. Thus, the assessment of frailty should be considered in all patients with cirrhosis hospitalized due to acute decompensation.

BACKGROUND AND AIM: Patients with end-stage liver disease are at high risk of malnutrition, sarcopenia and frailty. Recent evidence indicates that physical frailty, as assessed by liver frailty index (LFI), is an independent predictor of mortality in patients with cirrhosis. However, majority of available data are from North American cohorts in the outpatient setting. We prospectively investigated the relationship between physical frailty and sarcopenia in hospitalized patients with cirrhosis and their impact on clinical outcomes. PATIENTS AND METHODS: All patients with cirrhosis admitted to our unit between October 1st, 2021 and March 31st, 2023 were prospectively screened for recruitment. Exclusion criteria were: age<18 or>70 years; known musculoskeletal disease; history of liver transplantation. LFI was calculated at admission. Frailty was defined according to previously established thresholds: <3.2 (robust); between 3.2 and 4.5 (pre-frail); >4.5 (frail). Skeletal muscle mass was assessed via CT scan when this performed during hospital stay. Sarcopenia was defined by skeletal muscle index (SMI) <50 cm²/m² in males and <39 cm²/m² in females. All patients were prospectively followed for liver-related death. Cox regression analysis was used to identify predictors of primary outcome. RESULTS: We included 127 patients with cirrhosis (median age 59 years; 68.5% male). The median MELD score was 22 (IQR 17-27); 70% of patients had Child-Pugh C cirrhosis. Alcohol-related liver disease was the most common aetiology of cirrhosis (62%). At inclusion, 48.8% of patients were frail and 60% were sarcopenic. Compared to robust and non- sarcopenic individuals, those with frailty and sarcopenia had higher risk of further decompensation with increased need of ICU care during hospitalization. Frailty – but not sarcopenia – was associated with a longer hospital stay (15 [9-27] vs. 6.5 [4-15] in robust patients; p=0.001). Rate of re-hospitalization was comparable between patients with and without frailty (p=0.589) and sarcopenia (p=0.065). During a median follow-up of 96 days (IQR 37-384), 39.4% patients died. Both frailly and sarcopenia were associated with a higher mortality (60% vs 20% and 58% vs. 38%, respectively; p < 0.001 and p<.009). In multivariate analysis (Table 1), frailty (HR:3,377; 95% CI 1,729-6,597; p<0.001) and ACLF (HR:2,354, 95% CI 1.124-4.929; p=0.023) were the only independent predictors of death. The association between frailty and mortality was confirmed in a post-hoc analysis in both patients with and without ACLF separately. CONCLUSIONS: Frailty and sarcopenia are highly prevalent in hospitalized patients with cirrhosis and are both associated with higher rates of ICU admission and health care costs. However, frailty emerges as the strongest, independent predictor of early mortality after index hospitalization. Thus, the assessment of frailty should be considered in all patients with cirrhosis hospitalized due to acute decompensation.