Background:Sepsis affects millions of people and imposes a substantial economic and social burden worldwide. However, the role of human albumin in the management of septic patients remains unclear.Methods:EMBASE, PubMed, and Cochrane Library databases were searched. Randomized controlled trials regarding the use of human albumin in septic patients were eligible. The overall mortality and the intensive care unit (ICU), in-hospital/28-day, and 90-day mortality were pooled, respectively. Subgroup analyses were performed according to target population, type and dose of human albumin, and type of control group. Risk ratios (RRs) was calculated.Results:Twenty-four randomized controlled trials were finally included. Meta-analysis showed that human albumin cannot decrease the overall (RR = 1.02, P = .56), ICU (RR = 1.06, P = .65), in-hospital/28-day (RR = 1.01, P = .68), and 90-day (RR = 1.01, P = .65) mortality of total patients. Subgroup analyses showed that human albumin both cannot significantly decrease the overall, ICU, in-hospital/28-day, and 90-day mortality of sepsis and septic patients. Additionally, 20% human albumin (RR = 0.89, P = .03) and high daily dose of human albumin (RR = 0.90, P = .03) might benefit for the survival of patients with septic shock.Conclusions:Based on the current evidence, the general use of human albumin to improve the survival of septic patients cannot be recommended.