BACKGROUND The question of which colloid (albumin or synthetic colloids) used for plasma expansion following paracentesis or other complications requiring fluid loading in patients with cirrhosis remains controversial. AIMS To compare outcome and hospital-related cost in patients with cirrhosis treated with 20% human albumin with those treated with a synthetic colloid (3. 5% polygeline). METHODS The primary end point was
occurrence of a first liver-related complication. RESULTS When the trial was prematurely discontinued because of safety concerns about bovine-derived products, 30 patients were assigned to receive albumin and 38 were assigned to receive a synthetic colloid. Sixty-three patients were included for ascites removal by paracentesis and five patients for ascites removal by paracentesis and renal impairment. The median time to first liver-related complication was not significantly longer in the albumin group (20 vs. 7 days). However, the total number of liver-related complications adjusted to a 100-day period was significantly lower in the albumin group. The median hospital cost for a 30-day period was significantly lower in the albumin group (1915 euros vs. 4612 euros). CONCLUSIONS In patients with cirrhosis and ascites, human albumin appears to be more effective in preventing liver-related complications than synthetic colloid. This may be associated with decreased hospital costs.