BACKGROUND: Human albumin solutions are used in a range of medical and surgical problems. Licensed indications are the emergency treatment of shock and other conditions where restoration of blood volume is urgent, burns, and hypoproteinaemia. Human albumin solutions are more expensive than other colloids and crystalloids. OBJECTIVES: To quantify the effect on mortality of human albumin and plasma protein fraction (PPF) administration in the management of critically ill patients. SEARCH STRATEGY: We searched the Cochrane Injuries Group trials register, Cochrane Controlled Trials Register, Medline, Embase and BIDS Index to Scientific and Technical Proceedings. Reference lists of trials and review articles were checked, and authors of identified trials were contacted. The search was last updated in November 1999. SELECTION CRITERIA: Randomised controlled trials comparing albumin/PPF with no albumin/PPF, or with a crystalloid solution, in critically ill patients with hypovolaemia, burns or hypoalbuminaemia. DATA COLLECTION AND ANALYSIS: We collected data on the participants, albumin solution used, mortality at the end of follow up, and quality of allocation concealment. Analysis was stratified according to patient type. We assessed publication bias using the regression test for funnel plot asymmetry. MAIN RESULTS: We found 30 trials meeting the inclusion criteria and reporting death as an outcome. There were 156 deaths among 1419 trial participants. For each patient category the risk of death in the albumin treated group was higher than in the comparison group. For hypovolaemia the relative risk of death following albumin administration was 1.46 (95% confidence interval 0.97 to 2.22), for burns the relative risk was 2.40 (1.11 to 5.19), and for hypoalbuminaemia the relative risk was 1.69 (1.07 to 2.67). The pooled relative risk of death with albumin administration was 1.68 (1.26 to 2.23). Overall, the risk of death in patients receiving albumin was 14% compared to 8% in the control groups, an increase in the risk of death of 6% (3% to 9%). These data suggest that for every 17 critically ill patients treated with albumin there is one additional death. REVIEWER'S CONCLUSIONS: There is no evidence that albumin administration reduces the risk of death in critically ill patients with hypovolaemia, burns or hypoalbuminaemia, and a strong suggestion that it may increase the risk of death. These data suggest that the use of human albumin in critically ill patients should be urgently reviewed and that it should not be used outside the context of a rigorously conducted randomised controlled trial.