INTRODUCTION: Decompensation with ascites portends a poor prognosis in cirrhosis. The aim of this study was to compare the outcomes of patients with nonalcoholic steatohepatitis (NASH) with hepatitis B virus (HBV) cirrhosis after decompensation with ascites. METHODS: We conducted a retrospective study to evaluate the outcomes of patients with NASH and HBV cirrhosis who were admitted to hospital for first-onset ascites from January 1, 2004, to June 30, 2015. They were followed up until death, liver transplantation, or loss to follow up. RESULTS: Patients with NASH had lower median (interquartile range) Model for End-Stage Liver Disease score (11 [9-14] vs 14 [11-17], P < 0.001). Over 60 months, patients with NASH cirrhosis had higher cumulative incidence of dilutional hyponatremia (P < 0.001) and refractory ascites (P = 0.028). They also had higher cumulative incidence of cirrhosis-related deaths and liver transplantation compared with HBV cirrhosis (65.7%; [95% confidence interval (CI) 53.6-75.4] vs 42.5% [95% CI 32.4-55.2], P = 0.008). Multivariable competing risk analysis showed that NASH (subdistribution hazard ratio [sHR] 1.88 [95% CI 1.14-3.11], P = 0.014), non-Chinese ethnicity (sHR 1.63 [95% CI 1.06-2.50], P = 0.027), history of hepatocellular carcinoma (sHR 1.76 [95% CI 1.05-2.95], P = 0.033), estimated glomerular filtration rate <60 mL/min/1.73 m2 (sHR 1.70 [95% CI 1.09-2.65], P = 0.020), and Model for End-Stage Liver Disease score ≥15 (sHR 3.26 [95% CI 2.11-5.05], P < 0.001) were independent predictors of poor transplant-free survival. DISCUSSION: Patients with decompensated cirrhosis due to NASH had much poorer prognosis compared with HBV with more complications and greater healthcare resource utilization. Greater awareness is necessary for early diagnosis of NASH before decompensation.
INTRODUCTION: Decompensation with ascites portends a poor prognosis in cirrhosis. The aim of this study was to compare the outcomes of patients with nonalcoholic steatohepatitis (NASH) with hepatitis B virus (HBV) cirrhosis after decompensation with ascites. METHODS: We conducted a retrospective study to evaluate the outcomes of patients with NASH and HBV cirrhosis who were admitted to hospital for first-onset ascites from January 1, 2004, to June 30, 2015. They were followed up until death, liver transplantation, or loss to follow up. RESULTS:Patients with NASH had lower median (interquartile range) Model for End-Stage Liver Disease score (11 [9-14] vs 14 [11-17], P < 0.001). Over 60 months, patients with NASH cirrhosis had higher cumulative incidence of dilutional hyponatremia (P < 0.001) and refractory ascites (P = 0.028). They also had higher cumulative incidence of cirrhosis-related deaths and liver transplantation compared with HBV cirrhosis (65.7%; [95% confidence interval (CI) 53.6-75.4] vs 42.5% [95% CI 32.4-55.2], P = 0.008). Multivariable competing risk analysis showed that NASH (subdistribution hazard ratio [sHR] 1.88 [95% CI 1.14-3.11], P = 0.014), non-Chinese ethnicity (sHR 1.63 [95% CI 1.06-2.50], P = 0.027), history of hepatocellular carcinoma (sHR 1.76 [95% CI 1.05-2.95], P = 0.033), estimated glomerular filtration rate <60 mL/min/1.73 m2 (sHR 1.70 [95% CI 1.09-2.65], P = 0.020), and Model for End-Stage Liver Disease score ≥15 (sHR 3.26 [95% CI 2.11-5.05], P < 0.001) were independent predictors of poor transplant-free survival. DISCUSSION: Patients with decompensated cirrhosis due to NASH had much poorer prognosis compared with HBV with more complications and greater healthcare resource utilization. Greater awareness is necessary for early diagnosis of NASH before decompensation.