Abstract
BACKGROUND AND AIM: Long-term human albumin (HA) infusions improve survival in cirrhotic patients with diuretic resistant ascites. We aimed to determine whether there is a significant benefit in a more unwell real-world cohort. METHODS: This is a single-center retrospective cohort study. Patients received outpatient HA between April 2017 and June 2021. Inclusion criteria were age ≥18 years, cirrhosis with ascites, and received at least 1 month of HA. Patients with significant comorbidities and ongoing alcohol use were not excluded. Outcomes assessed were transjugular intrahepatic portosystemic shunt (TIPS)/transplant-free survival (TTFS), and biochemical and prognostic outcomes. RESULTS: Twenty-four patients were included. Median age was 59.5 years. Seven were female (29.2%). Etiology included were alcohol (50%), non-alcoholic steatohepatitis (16.7%), and viral/alcohol (12.5%). Median model for end-stage liver disease-sodium (MELD-Na) was 18.5, with Child-Pugh scores (CPS) A (4.2%), B (50%), and C (45.8%). Improvements in serum sodium (P = 0.014), albumin (P = 0.003), and CPS (P = 0.017) were observed. Reduction in hospitalizations (P = 0.001), particularly portal hypertensive related admissions was observed (relative risk 0.39; 95% confidence interval [CI] 0.21-0.69, P = 0.003), needed to treat 2.09 (95% CI 1.25-3.67). There was a reduction in total paracentesis requirements (P = 0.005). On multivariate analysis, type 2 diabetes mellitus significantly increased risk of TIPS/transplant/death (hazard ratio 6.16; 95% CI 1.23-30.84, P = 0.027). Median TTFS improved in patients with a change in MELD-Na ≤1 at 1 month: 29.4 months versus 7.7 months (P = 0.011). CONCLUSION: Outpatient HA infusions decrease portal hypertensive related hospital admissions, improve serum sodium, albumin levels, and CPS. Type 2 diabetes mellitus and change in MELD-Na score help discriminate those likely to benefit most.