COUNTERPOINT: Should Intravenous Albumin Be Used for Volume Resuscitation in Severe Sepsis/Septic Shock? No Angel O. Coz Yataco, MD, FCCP Alexander H. Flannery, PharmD, BCPS Lexington, Kentucky Steven Q. Simpson, MD, FCCP Kansas City, KS Hypoalbuminemia is common in patients with severe sepsis and septic shock and is associated with poor outcomes. 1 The use of albumin for resuscitation in this population remains controversial. Despite theoretical benefits to the use of albumin, multiple clinical trials and systematic reviews have failed to demonstrate an outcome benefit. We believe that the use of albumin in the resuscitation of patients with severe sepsis and septic shock is not warranted, given the lack of evidence of outcome benefits and its high cost. 2 We present a few important questions. Does the Use of Albumin as a Resuscitation Agent Improve Clinical Outcomes? On the basis of a meta-analysis indicating that critically ill patients receiving albumin had a higher mortality than patients who received plasma protein fraction or crystalloids, the US Food and Drug Administration released a letter urging clinicians to exercise discretion in the use of albumin. 3,4 The Saline vs Albumin Fluid Evaluation (SAFE) trial compared albumin 4% and normal saline as volume expanders and found that there was no mortality difference in the overall population and the subgroup with severe sepsis. 5 Although multivariate analysis after adjustment for baseline characteristics showed a decreased odds ratio for death at 28 days favoring albumin, these results represent the post hoc analysis of a trial not designed to assess the benefits of albumin in patients with sepsis. 6 The Therapy in the Colloids vs Crystalloids for the Resuscitation of the Critically Ill (CRISTAL) trial showed no benefit in 28- or 90-day mortality in septic patients treated with albumin (4% and 20% solutions) vs those treated with normal saline. 7 The Albumin Italian Outcome Sepsis (ALBIOS) trial assessed whether albumin replacement with a 20% solution in septic patients for 28 days to a target of 3 g/dL was beneficial. 8,9 Despite achieving higher serum albumin levels in the treatment group, there was no difference in 28-day or 90-day mortality. Post hoc analysis of the septic shock subgroup showed no difference in 28-day mortality but suggested a 90-day mortality benefit favoring albumin, an effect that lost significance when adjusted for clinically relevant variables. ALBIOS did not show that albumin provides a benefit in the occurrence of new organ failure or duration of mechanical ventilation. Although lower heart rates and higher mean arterial pressures were reported in the albumin-treated group, such differences do not represent patient-centered outcomes. Multiple meta-analyses have shown that albumin does not provide a benefit in mortality or the need for renal replacement therapy in critically ill patients, including those with hypoalbuminemia and sepsis. 10-13 A meta- analysis, using data dominated by the ALBIOS trial, found no mortality benefit in patients with severe sepsis. 12 In the septic shock group, a 90-day mortality benefit was reported, but 28-day mortality was not analyzed. In summary, albumin used either as a resuscitation agent or to normalize albumin levels in septic patients has not been shown to provide a benefit in mortality, new organ failure, or duration of mechanical ventilation. Furthermore, the hemodynamic benefits should be interpreted with caution because they are not associated with beneficial outcomes. Is There a Potential for Harm With the Use of Albumin as a Resuscitation Agent? Albumin is typically considered a drug product, but it is in reality a blood product. Albumin is obtained from pooled human plasma, using a fractionation process and pasteurization to reduce the transmission of microorganisms. It seems prudent to limit albumin administration with the same vigilance used with other blood products. AFFILIATIONS: From the Division of Pulmonary, Critical Care and Sleep Medicine (Dr Coz Yataco), University of Kentucky; College of Pharmacy (Dr Flannery), University of Kentucky HealthCare; and the Division of Pulmonary and Critical Care (Dr Simpson), University of Kansas. FINANCIAL/NONFINANCIAL DISCLOSURES: None declared. CORRESPONDENCE TO: Angel O. Coz Yataco, MD, FCCP, University of Kentucky, KY Clinic, 740 S Limestone L-543, Lexington, KY 40536; e-mail: angel.coz@uky.edu Copyright Ó 2016 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved. DOI: http://dx.doi.org/10.1016/j.chest.2016.03.050 1368 Point and Counterpoint [ 149#6 CHEST JUNE 2016 ]