This article is available to subscribers. Subscribe now. Already have an account? Sign in

Original ArticleFree Preview

A Randomized Trial of Albumin Infusions in Hospitalized Patients with Cirrhosis

List of authors.
  • Louise China, Ph.D.,
  • Nick Freemantle, Ph.D.,
  • Ewan Forrest, M.D.,
  • Yiannis Kallis, Ph.D.,
  • Stephen D. Ryder, D.M.,
  • Gavin Wright, Ph.D.,
  • Andrew J. Portal, M.D.,
  • Natalia Becares Salles, Ph.D.,
  • Derek W. Gilroy, Ph.D.,
  • and Alastair O¡¯Brien, Ph.D.
  • for the ATTIRE Trial Investigators*



Infection and increased systemic inflammation cause organ dysfunction and death in patients with decompensated cirrhosis. Preclinical studies provide support for an antiinflammatory role of albumin, but confirmatory large-scale clinical trials are lacking. Whether targeting a serum albumin level of 30 g per liter or greater in these patients with repeated daily infusions of 20% human albumin solution, as compared with standard care, would reduce the incidences of infection, kidney dysfunction, and death is unknown.


We conducted a randomized, multicenter, open-label, parallel-group trial involving hospitalized patients with decompensated cirrhosis who had a serum albumin level of less than 30 g per liter at enrollment. Patients were randomly assigned to receive either targeted 20% human albumin solution for up to 14 days or until discharge, whichever came first, or standard care. Treatment commenced within 3 days after admission. The composite primary end point was new infection, kidney dysfunction, or death between days 3 and 15 after the initiation of treatment.


A total of 777 patients underwent randomization, and alcohol was reported to be a cause of cirrhosis in most of these patients. A median total infusion of albumin of 200 g (interquartile range, 140 to 280) per patient was administered to the targeted albumin group (increasing the albumin level to ¡Ư30 g per liter), as compared with a median of 20 g (interquartile range, 0 to 120) per patient administered to the standard-care group (adjusted mean difference, 143 g; 95% confidence interval [CI], 127 to 158.2). The percentage of patients with a primary end-point event did not differ significantly between the targeted albumin group (113 of 380 patients [29.7%]) and the standard-care group (120 of 397 patients [30.2%]) (adjusted odds ratio, 0.98; 95% CI, 0.71 to 1.33; P=0.87). A time-to-event analysis in which data were censored at the time of discharge or at day 15 also showed no significant between-group difference (hazard ratio, 1.04; 95% CI, 0.81 to 1.35). More severe or life-threatening serious adverse events occurred in the albumin group than in the standard-care group.


In patients hospitalized with decompensated cirrhosis, albumin infusions to increase the albumin level to a target of 30 g per liter or more was not more beneficial than the current standard care in the United Kingdom. (Funded by the Health Innovation Challenge Fund; ATTIRE EudraCT number, 2014-002300-24; ISRCT number, N14174793.)

Digital Object ThumbnailQUICK TAKE VIDEO SUMMARY
Albumin Infusions in Hospitalized Patients with Cirrhosis

Funding and Disclosures

Supported by a grant (HICF reference HICF-R8-439, WT grant number WT102568, to Dr. O¡¯Brien) from the Health Innovation Challenge Fund (a partnership between the Wellcome Trust and the Department of Health and Social Care).

Disclosure forms provided by the authors are available with the full text of this article at

Dr. Freemantle reports receiving consulting fees from Allergan, AstraZeneca, Grifols Biologicals, Ipsen Biopharmaceuticals, and Novo Nordisk, and consulting fees and fees for serving as an expert witness from Sanofi US Services. No other potential conflict of interest relevant to this article was reported.

A data sharing statement provided by the authors is available with the full text of this article at

Author Affiliations

From the Institute for Liver and Digestive Health (L.C., N.B.S., A.O.), the Comprehensive Clinical Trials Unit (N.F.), and the Division of Medicine, University College London (D.W.G.), Barts and the London School of Medicine and Dentistry, Queen Mary University of London (Y.K.), London, the Glasgow Royal Infirmary and the University of Glagow, Glasgow (E.F.), the National Institute for Health Research Nottingham Biomedical Research Centre at Nottingham University Hospitals NHS Trust and the University of Nottingham, Queens Medical Centre, Nottingham (S.D.R.), the Mid and South Essex NHS Foundation Trust and the Basildon and Thurrock University Hospitals NHS Foundation Trust, Basildon (G.W.), and the Bristol Royal Infirmary, Bristol (A.J.P.) ¡ª all in the United Kingdom.

Address reprint requests to Dr. O¡¯Brien at the University College London Institute for Liver and Digestive Health, Upper 3rd Fl., Division of Medicine, Royal Free Campus, Rowland Hill St., London NW3 2PF, United Kingdom, or at .

A complete list of the ATTIRE Trial Investigators is provided in the Supplementary Appendix, available at