Human Albumin Infusion for the Management of Liver Cirrhosis and Its Complications: An Overview of Major Findings from Meta-analyses.

Abstract

The role of human albumin (HA) infusion in cirrhotic patients has been increasingly recognized. This paper aims to summarize the evidence from meta-analyses regarding HA infusion for the management of cirrhosis and its complications. A systematic search in the PubMed, EMBASE, and Cochrane library databases, and in reference lists was conducted. All relevant meta-analyses were identified and their findings were reviewed. The Assessment of Multiple Systematic Reviews 2 (AMSTAR-2) checklist was used to evaluate the methodological quality and the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) system to assess the quality of evidence for significant outcomes. Among 300 papers initially identified, 18 meta-analyses have been included. Short- and long-term HA infusion at high doses decreased the mortality of patients with decompensated cirrhosis. In cirrhotic patients with ascites, long-term HA infusion reduced the recurrence of ascites, but not mortality. In cirrhotic patients undergoing large-volume paracentesis (LVP), HA infusion reduced the incidence of post-paracentesis circulatory dysfunction and hyponatremia, but not mortality or renal impairment. In cirrhotic patients with overt hepatic encephalopathy (HE), HA infusion improved the severity of overt HE, but not overall mortality. In cirrhotic patients with spontaneous bacterial peritonitis (SBP), but not those with non-SBP infections, HA infusion reduced the mortality and renal impairment. In cirrhotic patients with type-1 hepatorenal syndrome (HRS), an increment of 100g in cumulative HA dose increased 1.15-fold survival, but not HRS reversal. In these meta-analyses, the quality of methodology was low or critically low, and that of the evidence was from very low to moderate. Based on the limited evidence from these meta-analyses, HA infusion appears to be beneficial in cirrhotic patients with ascites, overt HE, and SBP and in those undergoing LVP, but not in those with non-SBP infections.