Obesity Decreases Liver-Related Death in Patients With Cirrhosis: A Retrospective Propensity Score-Matched Study

To determine the impact of obesity on the prognosis of patients with liver cirrhosis.

This present retrospective observational study enrolled 482 patients without hepatocellular carcinoma (HCC). The patients were evaluated for obesity during the subsequent 8-year follow-up period. Obesity was defined as a body fat percentage of ≥ 25.8% for men and ≥ 36.5% for women as measured using a bioelectrical impedance analysis. After propensity score matching was adjusted for age, sex, etiology, and severity of liver disease, 185 patients were assigned to either the obese or nonobese group. The prognostic impact of obesity in patients with cirrhosis was evaluated by differentiating between liver-related and nonliver-related deaths.

The median observation period was 4.2 years. The cumulative incidence of liver-related deaths was significantly higher in the nonobese group than in the obese group (P = 0.02), whereas there was no significant difference in the cumulative incidence of nonliver-related deaths between the obese and nonobese groups (P = 0.70). Multivariate competing risk analyses revealed that a nonviral etiology (hazard ratio [HR], 5.43), decompensated cirrhosis (HR, 5.22), sarcopenia (HR, 2.28), and obesity (HR, 0.49) were independently associated with liver-related death. Among patients with a nonviral etiology, female patients sarcopenia, and patients with decompensated cirrhosis, the cumulative incidence of liver-related death was significantly higher in nonobese patients than in obese patients (P < 0.01, 0.04, 0.03, and 0.01, respectively).

Obesity decreased liver-related death in patients with cirrhosis, particularly in female patients, patients with a nonviral etiology, and patients with sarcopenia.