The efficacy and safety profile of albumin administration for patients with cirrhosis at high risk of hepatorenal syndrome is dose dependent

Background: Albumin is a critical component in the standard therapeutic approach to acute renal failure (ARF) and spontaneous bacterial peritonitis (SBP) in the setting of ascites. However, data regarding the safety and minimum effective dose are limited. Methods: We conducted a retrospective review of patients with decompensated cirrhosis who received albumin within the first 48 hours of hospitalization at Beth Israel Deaconess Medical Center between 2010 and 2013. Outcomes included 90-day risk of death or transplantation (primary) and (secondary) complications of albumin infusion (length of stay (LOS) and need for critical care)), all adjusted for comorbidity and severity of illness. Results: We included 169 patients with ARF and 88 patients with SBP. The optimal doses of albumin for a survival benefit were found to be 87.5 g and 100 g in the ARF and SBP cohorts, respectively. The odds ratio (OR) for the 90-day risk of death or liver transplantation associated with the optimal loading dose was 0.36 (95% CI: 0.17-0.76, P = 0.008) and 0.28 (95% CI: 0.07-0.97, P = 0.04) for the ARF and SBP cohorts, respectively. This effect persisted for patients with ARF who had neither hepatorenal syndrome (HRS) nor SBP (OR: 0.13, 95% CI: 0.007-0.79, P = 0.02). LOS (beta coefficient per log albumin dose: 1.69; 95% CI: 0.14-3.24, P = 0.03) and risk of critical care (OR/g albumin: 1.03; 95% CI: 1.01-1.05, P = 0.01) were also dose dependent. Conclusion: Albumin has a dose-dependent effect on both survival and complications in patients with cirrhosis with ARF (HRS and otherwise) and/or SBP.