Normothermic Machine Perfusion of Donor Livers for Transplantation in the United States: A Randomized Controlled Trial

William C. Chapman, Andrew S. Barbas, Anthony M. D'Alessandro, Rodrigo Vianna, Chandrashekhar A. Kubal, Peter Abt, Christopher Sonnenday, Rolf Barth, Josue Alvarez-Casas, Hasan Yersiz, Devin Eckhoff, Robert Cannon, Yuri Genyk, Linda Sher, Andrew Singer, Sandy Feng, Garrett Roll, Ari Cohen, Maria B. Doyle, Debra L. SudanDavid Al-Adra, Adeel Khan, Vijay Subramanian, Nader Abraham, Kim Olthoff, Akin Tekin, Lynn Berg, Constantin Coussios, Chris Morris, Lucy Randle, Peter Friend, Stuart J. Knechtle

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


Objective: To compare conventional low-temperature storage of transplant donor livers [static cold storage (SCS)] with storage of the organs at physiological body temperature [normothermic machine perfusion (NMP)]. Background: The high success rate of liver transplantation is constrained by the shortage of transplantable organs (eg, waiting list mortality >20% in many centers). NMP maintains the liver in a functioning state to improve preservation quality and enable testing of the organ before transplantation. This is of greatest potential value with organs from brain-dead donor organs (DBD) with risk factors (age and comorbidities), and those from donors declared dead by cardiovascular criteria (donation after circulatory death). Methods: Three hundred eighty-three donor organs were randomized by 15 US liver transplant centers to undergo NMP (n = 192) or SCS (n = 191). Two hundred sixty-six donor livers proceeded to transplantation (NMP: n = 136; SCS: n = 130). The primary endpoint of the study was "early allograft dysfunction" (EAD), a marker of early posttransplant liver injury and function. Results: The difference in the incidence of EAD did not achieve significance, with 20.6% (NMP) versus 23.7% (SCS). Using exploratory, "as-treated" rather than "intent-to-treat," subgroup analyses, there was a greater effect size in donation after circulatory death donor livers (22.8% NMP vs 44.6% SCS) and in organs in the highest risk quartile by donor risk (19.2% NMP vs 33.3% SCS). The incidence of acute cardiovascular decompensation at organ reperfusion, "postreperfusion syndrome," as a secondary outcome was reduced in the NMP arm (5.9% vs 14.6%). Conclusions: NMP did not lower EAD, perhaps related to the inclusion of lower-risk liver donors, as higher-risk donor livers seemed to benefit more. The technology is safe in standard organ recovery and seems to have the greatest benefit for marginal donors.

Original languageEnglish
Pages (from-to)E912-E921
JournalAnnals of surgery
Issue number5
StatePublished - Nov 1 2023


  • liver
  • normothermic
  • perfusion
  • preservation
  • transplantation


Dive into the research topics of 'Normothermic Machine Perfusion of Donor Livers for Transplantation in the United States: A Randomized Controlled Trial'. Together they form a unique fingerprint.

Cite this