Abstract

American Journal of TransplantationVolume 16, Issue 6 p. 1779-1787 Original ArticleFree Access Liver Transplantation After Ex Vivo Normothermic Machine Preservation: A Phase 1 (First-in-Man) Clinical Trial R. Ravikumar, R. Ravikumar Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK Institute of Biomedical Engineering, University of Oxford, Oxford, UKJoint first authors.Search for more papers by this authorW. Jassem, W. Jassem Liver Unit, Kings College Hospital, London, UKJoint first authors.Search for more papers by this authorH. Mergental, H. Mergental orcid.org/0000-0001-5480-9380 Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorN. Heaton, N. Heaton Liver Unit, Kings College Hospital, London, UKSearch for more papers by this authorD. Mirza, D. Mirza Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorM. T. P. R. Perera, M. T. P. R. Perera orcid.org/0000-0002-5417-3850 Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorA. Quaglia, A. Quaglia Liver Unit, Kings College Hospital, London, UKSearch for more papers by this authorD. Holroyd, D. Holroyd Institute of Biomedical Engineering, University of Oxford, Oxford, UKSearch for more papers by this authorT. Vogel, T. Vogel Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UKSearch for more papers by this authorC. C. Coussios, C. C. Coussios Institute of Biomedical Engineering, University of Oxford, Oxford, UKSearch for more papers by this authorP. J. Friend, Corresponding Author P. J. Friend Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UKCorresponding author: Peter J. Friend, peter.friend@nds.ox.ac.ukSearch for more papers by this author R. Ravikumar, R. Ravikumar Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK Institute of Biomedical Engineering, University of Oxford, Oxford, UKJoint first authors.Search for more papers by this authorW. Jassem, W. Jassem Liver Unit, Kings College Hospital, London, UKJoint first authors.Search for more papers by this authorH. Mergental, H. Mergental orcid.org/0000-0001-5480-9380 Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorN. Heaton, N. Heaton Liver Unit, Kings College Hospital, London, UKSearch for more papers by this authorD. Mirza, D. Mirza Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorM. T. P. R. Perera, M. T. P. R. Perera orcid.org/0000-0002-5417-3850 Liver Unit, University Hospital Birmingham, Birmingham, UKSearch for more papers by this authorA. Quaglia, A. Quaglia Liver Unit, Kings College Hospital, London, UKSearch for more papers by this authorD. Holroyd, D. Holroyd Institute of Biomedical Engineering, University of Oxford, Oxford, UKSearch for more papers by this authorT. Vogel, T. Vogel Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UKSearch for more papers by this authorC. C. Coussios, C. C. Coussios Institute of Biomedical Engineering, University of Oxford, Oxford, UKSearch for more papers by this authorP. J. Friend, Corresponding Author P. J. Friend Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UKCorresponding author: Peter J. Friend, peter.friend@nds.ox.ac.ukSearch for more papers by this author First published: 11 January 2016 https://doi.org/10.1111/ajt.13708Citations: 302 ISRCTN 14355416. AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract The number of donor organs suitable for liver transplantation is restricted by cold preservation and ischemia–reperfusion injury. We present the first patients transplanted using a normothermic machine perfusion (NMP) device that transports and stores an organ in a fully functioning state at 37°C. In this Phase 1 trial, organs were retrieved using standard techniques, attached to the perfusion device at the donor hospital, and transported to the implanting center in a functioning state. NMP livers were matched 1:2 to cold-stored livers. Twenty patients underwent liver transplantation after NMP. Median NMP time was 9.3 (3.5–18.5) h versus median cold ischaemia time of 8.9 (4.2–11.4) h. Thirty-day graft survival was similar (100% NMP vs. 97.5% control, p = 1.00). Median peak aspartate aminotransferase in the first 7 days was significantly lower in the NMP group (417 IU [84–4681]) versus (902 IU [218–8786], p = 0.03). This first report of liver transplantation using NMP-preserved livers demonstrates the safety and feasibility of using this technology from retrieval to transplantation, including transportation. NMP may be valuable in increasing the number of donor livers and improving the function of transplantable organs. Abbreviations AST aspartate transaminase DBD donation after brain death DCD donation after circulatory death IRI ischemia–reperfusion injury IVC inferior vena cava MELD model for end-stage liver disease NHSBT National Health Service Blood and Transplant NMP normothermic machine perfusion Introduction The availability of donor organs for transplantation is restricted by the limitations of current cold preservation techniques. Despite rising numbers of organ donors in many countries, the gap between demand and availability of donor livers is increasing, with waiting lists and pretransplant mortality growing in many countries 1, 2. On April 1, 2014, 552 patients were on the UK liver transplant waiting list. During the previous 12 months, 20% of patients died while waiting or were removed from the list (typically having become too sick to transplant) 3. Despite this, only 65% of solid organ donors culminated in a liver transplant 3. Much of the increase in deceased donor numbers is in donors that would once have been declined as unsuitable—including older donors and those with medical comorbidities (cardiovascular disease, diabetes, obesity) 3. Also, organ donation is increasingly offered following cardiovascular/circulatory, rather than neurological, determination of death: this inevitably implies a period of warm ischemic injury prior to preservation 3. The use of these marginal organs is associated with a much higher risk of immediate graft failure and later complications 4, 5. Notably, while 87% of all donation after brain death (DBD) livers in the United Kingdom were utilized, only 28% of donation after circulatory death (DCD) livers were transplanted 3. The standard technique for the preservation of donor organs between recovery and implantation is static cold storage in a specialist preservation solution. Although effective for ideal donor organs, this method is less suitable for marginal (high-risk) organs, not only because such organs experience greater ischemia–reperfusion effects, but also because the lack of an effective means of viability assessment is so much more problematic in this group. Investigators around the world are exploring novel preservation methods in an attempt to enable greater use of such high-risk organs without compromising outcomes. Recent advances include the demonstration of the benefit of cold continuous machine perfusion for the kidney 6 and the first use of a similar strategy in the liver 7, 8. Early clinical investigations by the Zurich group suggest that oxygenated hypothermic liver perfusion at the end of cold preservation is beneficial 9, 10. However, there has been no widely adopted change in clinical practice since the introduction of University of Wisconsin solution in the late 1980s. There is mounting evidence that only perfusion under more physiological conditions of temperature and oxygen delivery will enable a step change in the utilization of marginal donor organs. Recent clinical studies have tested varying periods of normothermic perfusion (“reconditioning”) of donor organs transported in a cold state for the kidney 11 and lung 12, 13. However, these studies have not achieved normothermic preservation throughout the period from explantation to implantation. The feasibility of maintaining physiological temperature throughout the period of lung preservation was recently demonstrated 14, but clinical success has not yet been reported in the context of other organs. We have previously demonstrated in a pig model that even a short (4 h) period of cold preservation is markedly deleterious to the liver following an ischemic injury and concluded that, if normothermic machine perfusion (NMP) is to be beneficial in the transplantation of marginal donor organs, then the technology must be transportable to the donor hospital 15. We present here, in a Phase 1 trial designed to test safety and feasibility rather than efficacy, the results of the first clinical series of transplants carried out using a novel normothermic liver perfusion device that enables transport, storage, and assessment of a liver in a fully functioning state. Methods In this Phase 1, nonrandomized, prospective trial, the outcomes of the recipients of 20 consecutive NMP donor livers were compared to those of matched control patients who received conventionally cold-stored donor livers. Approvals were obtained from National Health Service Blood and Transplant (NHSBT), National Research Ethics Committee, and the Medicines and Healthcare Products Regulatory Authority. The trial was registered with the ISRCTN (14355416). Patient and donor selection Adult patients with end-stage liver disease on the King's College Hospital and University Hospital Birmingham liver transplant waiting list were approached for consent to take part in the study. The families of suitable organ donors were approached for consent. All adult donor organs, aged over 18 years, including DBD and DCD were potentially eligible, except those undergoing splitting for two recipients. All adult recipients, aged over 18 years, were potentially eligible, except those undergoing transplantation for fulminant liver failure (because of the marginal of non-graft-related mortality) or transplantation of more than one organ. Matching Patients undergoing NMP liver transplantation were matched retrospectively 1:2 to patients undergoing transplantation of conventional cold-stored livers at the same centers between January 2011 and December 2013 (a broader time span than the trial recruitment in order to achieve the desired level of matching). Anonymized, matched control patients were identified by applying the following criteria hierarchically, in the following order: (i) graft type (DBD, DCD); (ii) donor age (within 5 years); (iii) recipient MELD (model of end-stage liver disease) score (within 2 points); (iv) recipient age (within 10 years). Matching criteria limits were extended when no suitable matches were identified. Ten donor livers were within standard criteria and the other 10 donor livers were specifically selected as high-risk, using criteria based on the Eurotransplant Donor Risk Index 16. End-points The primary end-point was 30-day graft survival. Secondary end-points included biochemical measures of liver function/injury (bilirubin, aspartate aminotransferase [AST], alkaline phosphatase [ALP], international normalized ratio [INR]) during the first 7 days, patient and graft survival, and graft function at 6 months. Early allograft dysfunction was defined by the occurrence of one or more of the following: bilirubin >170 μmol/L on day 7 posttransplant; INR >1.6 on day 7 posttransplant; peak AST >2000 IU/L within the first 7 days posttransplant 17. Analysis Statistical analysis was conducted using SPSS® 22 (IBM®, New York), with data expressed as medians and ranges. Continuous numerical data were compared using a Mann–Whitney U test for nonparametric data or Kruskal–Wallis test of multiple variance; Fisher's exact test was used to compare categorical data. For categorical outcomes, absolute differences between the NMP and control groups, expressed as percentage points with 95% confidence intervals (CI), are provided. Differences were considered to be of statistical significance when a p-value of <0.05 was achieved. Machine perfusion When a suitable donor liver with research consent was allocated to a consenting recipient, the perfusion device was transported to the donor hospital. The perfusion team prepared the device and set up the surgical back-table during the retrieval process. Standard multiorgan retrieval was carried out with the addition of in-situ cholecystectomy (to reduce the risk of bleeding during perfusion). The liver was cooled in situ with University of Wisconsin solution and transferred to the back-table. The suprahepatic inferior vena cava (IVC) was prepared by excising attached diaphragmatic tissue and oversewing the orifices of the phrenic veins, and then closed using a linear vascular stapler (Covidien, Hampshire, UK). The infrahepatic IVC was cannulated (28F Sorin, Gloucester, UK). The hepatic hilum was dissected, taking care to ligate all tributaries. Cannulae were secured in the portal vein (24F Sorin), celiac artery (10F Sorin), and common bile duct (12–18Fr Summit Medical, Cheltenham, UK). In three cases, an accessory right hepatic artery was anastomosed to the gastroduodenal artery. The liver was flushed with 500 mL colloid solution (Gelofusine®, B Braun, South Yorkshire, UK) to remove preservation solution, and then transferred to the perfusion device. The OrganOx metra liver perfusion device (Figure 1) provides automated pumping, oxygen/air delivery, and heat exchange, in order to maintain the perfusate at normal temperature, within physiological ranges for pO2, pCO2, pH, and at physiological pressures in the vascular inflows and outflow of the liver (hepatic artery pressure 60 to 75 mmHg; IVC pressure (−1 to 2 mmHg.). The portal pressure was not monitored (it is effectively fixed by the height of the portal venous reservoir), but portal flow is continuously measured. Hemodynamic parameters and blood gas data are continuously recorded during preservation. Cannulation of the bile duct enables collection and automated monitoring of hourly bile production. The machine also continuously infuses (i) bile salt (sodium taurocholate, New Zealand Pharmaceuticals, Palmerston North, New Zealand); (ii) insulin (Actrapid®, Novo Nordisk, West Sussex, UK); (iii) heparin (CP Pharmaceuticals, Wrexham, UK); (iv) prostacyclin (Flolan®, Glaxo, Middlesex, UK). A variable rate infusion of glucose and amino acids (Nutriflex, B Braun, Sheffield, UK) is regulated by 4-hourly manually inputted glucose levels. Figure 1Open in figure viewerPowerPoint Schematic representation of the OrganOx metra circuit. The device was primed with three units of packed red blood cells, sourced from the blood bank and cross-matched to the donor, and one unit of colloid solution (Gelofusine®, B Braun), with addition of calcium gluconate (B Braun), heparin (CP Pharmaceuticals), cefuroxime (GSK), and 30 mL of sodium bicarbonate (B Braun). During priming, the perfusate was allowed to reach operating conditions: temperature (37°C); pO2 (12 kPa); pCO2 (5 kPa); pH (7.35). The cannulated organ was then connected and blood flow started. Once perfusion was established, minor bleeding points were controlled surgically and the liver container was then closed. Transport and implantation The organ was transported by road to the transplant hospital and remained on the perfusion device until the transplanting team was ready to implant the organ. Perfusion was then stopped and the organ was cooled by rapid perfusion of 2 L of cold HTK solution (Custodiol®-HTK, Essential Pharmaceuticals, Ewing, NJ). The cannulae were removed and the organ was transferred to the recipient for immediate revascularization, using the unit's standard surgical technique. Postoperative management was conducted according to standard local protocols, which included tacrolimus-based immunosuppression. Results Donor and recipient characteristics Between February and December 2013, 20 patients underwent liver transplantation using donor organs preserved from recovery to implantation by normothermic perfusion (Table 1). There were no device failures leading to organs not being transplanted. The cases reported here represent 20 consecutive perfusions, without omission. Sixteen livers (80%) were from DBD and four (20%) were from DCD (Maastricht category III: circulatory arrest following withdrawal of support) donors. The indication for transplantation was chronic liver failure except in one recipient who underwent retransplantation for hepatic artery thrombosis. The underlying etiology of liver disease was the following: hepatitis C virus infection (n = 6); alcoholic liver disease (n = 5); primary sclerosing cholangitis (n = 3); primary biliary cirrhosis (n = 2); α1-antitrypsin deficiency (n = 1); nonalcoholic steatohepatitis (n = 1); chronic autoimmune hepatitis (n = 1); other cholangiopathic disease (choledocholithiasis and biliary cirrhosis) (n = 1). Table 1. Characteristics of normothermic liver perfusion (NMP) and control livers Patient Graft type Donor age (years) MELD Recipient age (years) Preservation time (h) NMP Control NMP Control NMP Control NMP Control NMP Control 1 DBD DBD 62 61 18 18 62 61 4.5 7.83 DBD 59 18 59 9 2 DBD DBD 44 42 25 25 54 48 9.87 5 DBD 46 23 46 9.5 3 DBD DBD 64 68 Re-Tx HAT Re-Tx CR 43 48 10.75 13.35 DBD 65 ReTx RD 40 8.78 4 DBD DBD 41 46 27 23 38 46 3.5 9.5 DBD 42 25 48 5 5 DBD DBD 53 56 18 20 54 53 6.15 10.1 DBD 55 16 58 9.38 6 DBD DBD 50 49 12 13 61 62 14.1 9.83 DBD 49 13 62 10.5 7 DBD DBD 68 66 9 8 46 47 15.8 9 DBD 66 10 45 8.16 8aa Livers where the matching criteria had to be extended. DBD DBD 77 73 7 6 47 59aa Livers where the matching criteria had to be extended. 9.75 4.23 DBD 72 8 35aa Livers where the matching criteria had to be extended. 8.33 9 DBD DBD 59 59 11 7 60 61 12.5 5.83 DBD 56 14 58 8 10 DBD DBD 78 79 15 15 62 60 11.58 5.5 DBD 75 17 57 6.67 11 DCD, 14 min DCD, 22 min 64 65 11 9 52 51 5.92 5.6 DCD, 23 min 65 12 48 9.23 12 DBD DBD 46 46 14 13 58 60 8.87 6.4 DBD 45 16 65 7 13 DBD DBD 61 64 18 19 55 50 4.75 7.6 DBD 63 19 52 7.45 14aa Livers where the matching criteria had to be extended. DBD, mod steatosis 3 kg DBD 47 48 11 14 57 62 8.83 7.2 DBD 48 16aa Livers where the matching criteria had to be extended. 56 10.2 15 DBD, fibrotic, retrieval ALT 1300 DBD 21 21 16 18 48 49 7.17 11.42 DBD 23 18 44 10.08 16 DCD, 27 min DCD, 20 min 53 58 11 12 66 62 5.5 5.78 DCD, 23 min 56 7 63 8.87 17aa Livers where the matching criteria had to be extended. DBD DBD 85 82 12 8 54 54 18.5 6.17 DBD 79aa Livers where the matching criteria had to be extended. 14 60 5.5 18 DCD, 31 min DCD, 18 min 67 65 12 13 64 62 17.82 9.83 DCD, 23 min 64 11 47 7.9 19aa Livers where the matching criteria had to be extended. DBD, fibrotic, retrieval ALT 2300 DBD 27 33aa Livers where the matching criteria had to be extended. 12 17aa Livers where the matching criteria had to be extended. 33 27 10.5 7.62 DBD 26 16 37 11.08 20 DCD, 15 min DCD, 10 min 57 57 9 10 57 54 8 5.78 DCD, 9 min 59 7 61 5.97 For graft type, numbers expressed adjacent to DCD indicate warm ischemia time in minutes. MELD, model for end-stage liver disease; DBD, donation after brain death, DCD, donation after circulatory death; Re-Tx, retransplantation; HAT, hepatic artery thrombosis; CR, chronic rejection; RD, recurrent disease; ALT, alanine aminitransferase. a Livers where the matching criteria had to be extended. Matched control patients were identified as described above; to find two matched controls, preset criteria were extended in some cases, as follows: donor age in liver 17 (85 years, matched liver 79 years), and 19 (27 years, matched 33 years); recipient age in liver 8; MELD score in livers 14 and 19. Median donor age was 58.0 (21–85) years in NMP versus 58.5 (21–82) years in matched controls (p = 0.93). Median recipient age was 54.4 (33–66) years in NMP versus 55.0 (27–65) years in matched controls (p = 0.99). In DCD transplants, donor median warm ischemic time was 21 (range 14–31) min in NMP versus 15 (9–23) min in the matched controls (p = 0.53). Median recipient MELD was 12 (7–27) in NMP versus 14 (6–25) in matched controls (p = 0.55). Assessment during NMP Median NMP time was 9.3 (range from 3.5 to 18.5) h (Figure 2A). Median cold ischemia time in the matched controls was 8.9 (range 4.2–11.4) h (Table 2). The period of NMP was governed by logistic considerations (mainly other transplants). Figure 2Open in figure viewerPowerPoint Assessment during normothermic machine perfusion (NMP). (A) NMP duration; (B) bile production; (C) perfusate pH during NMP; (D) hepatic arterial and portal venous flow during NMP. Table 2. Clinical outcomes of normothermic machine perfusion (NMP) and control livers Outcomes Total NMP (n = 20) Control (n = 40) Risk ratio/effect size (95% CI) p-value 30-day graft survival, n (%) 20 (100) 39 (97.5) 1.03 (0.98–1.08) RR 1.00 PNF, n (%) 0 0 1.000 EAD, n (%) 3 (15) 9 (22.5) 0.67 (0.20–2.19) RR 0.734 Peak AST within 7 days (IU/L), median (range) 417 (84–4681) 902aa N = 39 as 1 death on day 0. (218–8786) −0.44 (−0.98 to 0.11) ES 0.034 Bilirubin on day 7 (μmol/L), median (range) 25 (8–211) 30aa N = 39 as 1 death on day 0. (9–221) −0.23 (−0.77 to 0.32) ES 0.203 INR on day 7, median (range) 1.05 (0.88–1.40) 1.03 (0.90–2.22)aa N = 39 as 1 death on day 0. −0.16 (−0.70 to 0.38) ES 0.922 ALP on day 7 (U/L) 245 (81–568) 243 (76–743)aa N = 39 as 1 death on day 0. −0.11 (−0.65 to 0.43) ES 0.798 ITU stay (days), median (range) 3 (1–8) 3 (1–41)aa N = 39 as 1 death on day 0. −0.42 (−0.96 to 0.13) ES 0.459 Hospital stay (days), median (range) 12 (6–34) 14 (8–88)aa N = 39 as 1 death on day 0. −0.44 (−0.98 to 0.11) ES 0.100 30-day mortality (%) 0 (0) 1 (2.5) 1.000 6-month survival, n (%) 20 (100) 39 (97.5) 1.03 (0.98–1.08) RR 1.000 ALP, alkaline phosphatase; AST, aspartate aminotransferase; INR, international normalized ratio; ITU, intensive therapy unit; DBD, donation after brain death; DCD, donation after circulatory death; PNF, primary nonfunction; EAD, early graft dysfunction; RR, relative risk; ES, effect size; CI, confidence interval. a N = 39 as 1 death on day 0. Table 3. Complications of normothermic machine perfusion (NMP) livers Patient Graft type NMP complications 1 DBD 2 DBD Anastomotic biliary stricture, stented 3 DBD Anastomotic biliary stricture, biliary sepsis, stented 4 DBD 5 DBD 6 DBD 7 DBD Escherichia coli sepsis and acute kidney injury 8 DBD 9 DBD 10 DBD Anastomotic biliary stricture, stented 11 DCD Urinary sepsis and diabetes 12 DBD 13 DBD 14 DBD 15 DBD 16 DCD 17 DBD Death from recividism at 9 months 18 DCD 19 DBD Sepsis 20 DCD CMV +ve donor into CMV −ve recipient; recipient converted DBD, donation after brain death, DCD, donation after circulatory death; CMV, cytomegalovirus. There was evidence of stable hemodynamic, synthetic, and metabolic function throughout all perfusions (Figure 2) with maintenance of pH between 7.2 and 7.4 (Figure 2C), without pharmacological correction. Bile production commenced after the first hour and was maintained throughout NMP (Figure 2B). Hepatic arterial and portal venous flows were consistent throughout (Figure 2D). Outcomes (Table 2) All grafts and patients in the NMP group survived the first 30 days but one recipient of a DBD liver in the matched control group died on day 0 from a cardiovascular event (100% NMP vs. 97.5% Control, Absolute Difference −2.5, 95% CI −7.5–2.5; p = 1.00). There was a statistically significant difference in peak AST levels (417 vs. 902 IU/L, p = 0.034), numerically more pronounced in the DCD cohort (422 vs. 1894 IU/L, p = 0.283). There was no primary nonfunction in either group. Three patients (15%) demonstrated early graft dysfunction (EAD) in the NMP group compared to nine (23%) in the control group. This difference was more pronounced in the DCD subset (one [25%] vs. four [50%] patients). EAD in the NMP group was due to the following: day 7 bilirubin of 211 (liver 8, donor age 77); peak AST of 2158 IU/L (liver 15 prerecovery AST of 1300 IU/L); peak AST of 4681 IU/L (liver 16, DCD, age 53, warm ischemia time 27 min). Median intensive therapy unit and hospital stays were similar between the two groups overall and when analyzed as DBD and DCD subsets. All patients and grafts in the NMP group survived 6 months. One-year patient survival in the NMP group was 95% (one death at 9 months as a result of alcohol recividism). Figure 3 demonstrates actual biochemical parameters for each NMP liver compared to matched controls. Differences are seen in peak AST and bilirubin, particularly in the last 10 liver grafts (a higher-risk group). Figure 3Open in figure viewerPowerPoint Postoperative biochemical results of the normothermic machine perfusion (NMP) and control livers. AST, aspartate transaminase; INR, international normalized ratio; ALP, alkaline phosphatase. Four anastomotic biliary strictures in the trial patients underwent stenting, all in recipients of DBD grafts (Table 3), with the following primary diagnoses: one cholangiopathic disease, one retransplant for hepatic artery thrombosis, one alcoholic liver disease, and one primary biliary cirrhosis. There were no vascular complications within this study. One trial patient developed postoperative hepatic parenchymal infarcts, which resolved on subsequent computed tomography imaging. Notably, this patient received a liver with poor NMP arterial flow. One patient developed noncirrhotic portal hypertension with ascites but well maintained synthetic function. A 6-month posttransplant biopsy demonstrated perivenular fibrosis, mild portal-lobular hepatitis, and injury to small interlobular bile ducts attributed to a veno-occlusive disease in the absence of alternative overt pathology. Perfusate cultures (not carried out routinely) were all negative for bacterial growth. No significant difference was seen in the pre- and postperfusion biopsies, with all showing scattered neutrophils in keeping with mild preservation injury. Safety, feasibility, and logistics All livers selected for this trial underwent NMP throughout the period of preservation. The results reported were of 20 consecutive NMP livers with no exclusions (e.g. due to aborted perfusions). No livers were excluded due to perfusions. The 40–60 min of back-table preparation at the donor hospital occurred in parallel with the 30 min required to prime the normothermic perfusion device. Connection of the organ to the device and confirmation of stable flows typically took less than 15 min. With the exception of one liver (retrieved within the transplanting center), all grafts were then transported by road, with journey times of up to 3 h. A VW Transporter minibus with mains power outlet was used for transport. One member of the team oversaw the perfusion parameters and the device function throughout the perfusion. The only technical complication during transport was an airlock in the fluid sensing system, which necessitated a brief stop during transit to rectify the problem. Subsequent minor modification of the circuit design prevented any recurrence; there were no other transport-related complications and no liver became unsuitable for transplantation due to perfusion problems. The importance of meticulous back-table preparation of the liver was noted at an early stage, in order to avoid the need for hemostatic procedures once on the device. Although the need did not arise, cold preservation solution was always carried, with a tubing set for quick connection to the cannulated liver to enable rapi