end-HMPo2, hypothermic machine perfusion preservation; ITT, intention-to-treat; SCS, static cold storage.
end-HMPo2, hypothermic machine perfusion preservation; HR, hazard ratio; SCS, static cold storage.
Trial protocol and statistical analysis plan
eAppendix. Participating Centers
eTable 1. Summary of withdrawals
eTable 2. Number and reasons for cross-overs from End-HMPO2 to SCS
eTable 3. Reasons for graft failure prior to the end of follow-up period
eTable 4. Graft survival at 1 year post-transplant: Per protocol analysis
eTable 5. eGFR measured by the CKD-EPI Equation
eTable 6. Number and causes of death
eTable 7. Incidence of adverse events
eTable 8. Incidence of Serious adverse events
eTable 9. Association between preservation parameters and outcome measures
eTable 10. Association between duration of machine preservation or cold storage and kidney function at 12 months
eTable 11. Number of kidneys per trial arm and length of preservation
eFigure 1. Kaplan-Meier curve for patient survival
eFigure 2. Kaplan-Meier curve for graft survival, stratified by occurrence of DGF
eFigure 3. Correlation analysis between length of End-HMPO2 and graft function at 12 months
eFigure 4. Correlation analysis between cold storage time and graft function at 12 months
Data sharing statement
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Husen P, Boffa C, Jochmans I, et al. Oxygenated End-Hypothermic Machine Perfusion in Expanded Criteria Donor Kidney Transplant: A Randomized Clinical Trial. JAMA Surg. 2021;156(6):517–525. doi:10.1001/jamasurg.2021.0949
Does preimplantation short-term reconditioning of kidney grafts using oxygenated hypothermic machine perfusion for at least 2 hours after an initial period of static cold storage lead to an improvement of 1-year graft survival in kidneys retrieved from expanded criteria donors?
In this randomized clinical trial of 305 kidneys, 1-year graft survival was equal between kidneys that were machine perfused following static cold storage and kidneys that remained on static cold storage prior to implantation without oxygenated hypothermic machine perfusion.
These findings suggest that the use of oxygenated hypothermic machine perfusion prior to implantation and following a period of static cold storage does not improve graft survival or kidney function in kidneys retrieved from donors who are brain dead meeting the expanded donor criteria.
Continuous hypothermic machine perfusion during organ preservation has a beneficial effect on graft function and survival in kidney transplant when compared with static cold storage (SCS).
To compare the effect of short-term oxygenated hypothermic machine perfusion preservation (end-HMPo2) after SCS vs SCS alone on 1-year graft survival in expanded criteria donor kidneys from donors who are brain dead.
Design, Setting, and Participants
In a prospective, randomized, multicenter trial, kidneys from expanded criteria donors were randomized to either SCS alone or SCS followed by end-HMPo2 prior to implantation with a minimum machine perfusion time of 120 minutes. Kidneys were randomized between January 2015 and May 2018, and analysis began May 2019. Analysis was intention to treat.
On randomization and before implantation, deceased donor kidneys were either kept on SCS or placed on HMPo2.
Main Outcome and Measures
Primary end point was 1-year graft survival, with delayed graft function, primary nonfunction, acute rejection, estimated glomerular filtration rate, and patient survival as secondary end points.
Centers in 5 European countries randomized 305 kidneys (median [range] donor age, 64 [50-84] years), of which 262 kidneys (127 [48.5%] in the end-HMPo2 group vs 135 [51.5%] in the SCS group) were successfully transplanted. Median (range) cold ischemia time was 13.2 (5.1-28.7) hours in the end-HMPo2 group and 12.9 (4-29.2) hours in the SCS group; median (range) duration in the end-HMPo2 group was 4.7 (0.8-17.1) hours. One-year graft survival was 92.1% (n = 117) in the end-HMPo2 group vs 93.3% (n = 126) in the SCS group (95% CI, −7.5 to 5.1; P = .71). The secondary end point analysis showed no significant between-group differences for delayed graft function, primary nonfunction, estimated glomerular filtration rate, and acute rejection.
Conclusions and Relevance
Reconditioning of expanded criteria donor kidneys from donors who are brain dead using end-HMPo2 after SCS does not improve graft survival or function compared with SCS alone. This study is underpowered owing to the high overall graft survival rate, limiting interpretation.
isrctn.org Identifier: ISRCTN63852508
Kidney transplant has become the preferred standard form of treatment for most patients with end-stage kidney disease.1,2 To cope with the growing demand for grafts in the setting of kidney transplant, nowadays most centers will accept kidneys that are retrieved from older and higher-risk donors with multiple comorbidities, the so-called expanded criteria donors (ECDs), in an attempt to shorten waiting times in transplant for patients with end-stage kidney disease.3-5 While using grafts from ECDs may address the organ shortage, this policy has the secondary consequence of a higher rate of complications, such as primary nonfunction or delayed graft function (DGF) of the transplanted kidney and ultimately inferior long-term graft survival, when compared with organs retrieved from standard criteria donors.5-8 On the other hand, and despite the overall risk for graft failure is approximately 1.7-times higher in ECD kidneys, the use and transplant of these higher-risk kidneys still offers a significant survival benefit for recipients compared with those patients continuing to receive dialysis.9
In the past decade, it has become clear that reduction of ischemia-reperfusion injury and optimized organ preservation are key in successful transplants, allowing better immediate function and prolonged graft survival, especially of more compromised donor organs.10-12 To date, 2 methods of donor kidney preservation are widely used in the clinical setting of kidney transplant. First, the method of static cold storage (SCS), in which the kidney is flushed at time of procurement using a preservation solution, then submerged in cold preservation solution and kept on melting ice during transport to the recipient center until transplant. Second is the technique of hypothermic machine perfusion (HMP), which is a dynamic preservation method, in which the donor kidney is perfused with a cold machine perfusion solution using a device. SCS has remained the preferred technique worldwide owing to its perceived simplicity and low cost, although HMP has attracted considerable attention in the past decade and has led to changes in policy in some countries since clinical trials and multiple registry analyses found that function and outcomes for recipients are superior for HMP compared with SCS.13,14 In 2009, our group was the first to report that continuous HMP of the donor kidney starting immediately after organ procurement until implantation in the recipient is associated with a reduced risk of DGF and improved kidney graft survival in the first year after transplant.12 In addition, a subsequent subgroup analysis concerned kidneys from ECD donors after brain death and demonstrated a significantly reduced risk of DGF and higher 1-year graft survival in machine-perfused kidneys compared with cold-stored kidneys.15
More recently, a single-center study using SCS followed by a short-term preimplantation period of HMP (ie, end-HMP) suggested that end-HMP was able to reduce the risk for DGF in ECD kidneys. However, no clinical trial data have been available to provide evidence of a potential benefit.16 This evidence is highly relevant because if the simplification by omitting the logistic complexity of sending out devices to donor hospitals could be confirmed, this would improve patient outcomes and cost-effectiveness.
To gain better insight and clarify the effectiveness of the combination strategy of SCS followed by short-term HMP immediately prior to transplant as well as to provide evidence of whether the addition of oxygen during HMP as an optimized form of perfusion enhances outcomes in kidney transplant, we have now compared the current standard SCS preservation of donor kidneys with the regimen propagated by many transplant centers to combine SCS and HMP including oxygenation in a randomized clinical trial using ECD kidneys after brain death.
This investigator-led, prospective, randomized, parallel group, participant-blinded, controlled, multicenter, superiority trial included 10 kidney transplant centers in 5 European countries (Belgium, Germany, Hungary, the Netherlands, and United Kingdom). Approval of the trial protocol, amendments, as well as consent forms was obtained from national research ethics committees for each trial country. The trial protocol and statistical analysis plan are available in Supplement 1. On arrival of the kidney at the trial center (eAppendix in Supplement 2), patients gave written informed consent for participation in the trial and use of follow-up data. The trial was funded by the European Union 7th Framework Programme (Theme Health.2012.1.4-1, grant agreement 305934) and conducted by the Consortium for Organ Preservation in Europe.
Quiz Ref IDKidneys fulfilling the ECD criteria as defined by the United Network for Organ Sharing were eligible for enrollment: a kidney donated for transplant from a donor who was brain dead and older than 60 years or from a donor older than 50 years with 2 of the following: a history of hypertension, the most recent serum creatinine 1.5 mg/dL or more (to convert to micromoles per liter, multiply by 88.4), or death resulting from an cerebrovascular injury.3 Recipients were at least 18 years old and wait-listed for kidney-only transplant with either Eurotransplant or the National Health Service Blood and Transplant in one of the trial centers for a first or retransplant of a kidney. Participants received written and verbal information about the trial in advance while on the waiting list.
Kidneys were randomized between January 2015 and May 2018. Once an eligible kidney had been allocated to a recipient in a trial center, written consent was obtained from the recipient of the organ. Using an online randomization tool, the kidney was randomly assigned to either standard SCS or the combination of SCS with subsequent oxygenated HMP after arrival at the recipient center (end-HMPo2) with a 1:1 allocation per a computer-generated randomization scheme with random permuted block lengths, stratified by trial center.
All kidneys were placed on SCS following retrieval at the donor site and transported to the recipient center. Once a kidney was randomized to SCS on arrival, the kidney remained on SCS and transplanted according to standard local practice.
If the static cold-stored kidney was randomized for machine perfusion (end-HMPo2) after arrival at the recipient transplant center, the kidney was first prepared for implantation and then placed on the Kidney Assist transport device (Organ Assist BV) to be perfused with actively oxygenated University of Wisconsin Machine Perfusion Solution (Bridge to Life) at 1 °C to 4 °C for at least 120 minutes with a set perfusion pressure of 25 mm Hg until implantation into the recipient. Oxygen (100%) was supplemented at 100 mL/min, resulting in partial oxygen tensions of about 600 mm Hg in the perfusate.
Quiz Ref IDThe primary end point of the study was defined as the difference in 1-year graft survival between the 2 treatment arms. Secondary end points included (1) DGF (the need for dialysis within the first 7 days after transplant and preceding the return of kidney function); (2) kidney function at day 7 and months 3, 6, and 12 after transplant, determined by the estimated glomerular filtration rate (measured using the 4-variable Modification of Diet in Renal Disease formula17 and Chronic Kidney Disease Epidemiology Collaboration equation18); (3) functional DGF (the absence of a decrease in serum creatinine levels of at least 10% each day for 3 consecutive days within the first week after transplant, not including patients with biopsy-proven acute rejection or established calcineurin inhibitor toxicity); (4) primary nonfunction (defined as the continued need for dialysis until 3 months after transplant); (5) patient and (death-censored) graft survival; and (6) biopsy-proven acute rejection episodes up to 12 months after transplant. Graft loss was defined as the return to permanent dialysis or the need for retransplant.
In our published previous analysis, we reported that use of HMP vs SCS increases 1-year graft survival for ECD kidneys from 80% to 92%.15 For this trial to have a power of 80% with type I error (α) of 5% in a 2-sided statistical model, the required sample size to detect an improvement in 1-year graft survival from 80% to 92% was 262 kidneys in total, with 131 kidneys in each treatment arm.
Results are reported as an intention-to-treat analysis comparing intervention (end-HMPo2) against control (SCS) for the primary outcome and all secondary outcomes. Kidneys randomized but not transplanted to the consented recipients were excluded from the analysis. Differences between treatment groups are presented as mean and standard deviation, median and range, or percentages. Outcomes are reported with 95% CIs and 2-sided P values. P values less than .05 were regarded as statistically significant.
The treatment effect is described as absolute difference in proportions as well as odds ratio, which is presented together with confidence intervals. Kaplan-Meier curves and log-rank tests were performed for the assessment of time to graft failure and patient death. Any missing creatinine values within the first week of transplant have been imputed using linear interpolation. Here, a total of 10 values were imputed. For statistical analysis, SAS statistical software version 9.4 (SAS Institute) and Stata version 15 (StataCorp) were used. Kidney donor risk index as well as kidney donor profile index were calculated using the respective formula and mapping table provided by Organ Procurement and Transplantation Network.19,20
No interim analyses of end points were carried out. At regular intervals, an independent data monitoring committee reviewed recruitment, accruing data and confidential safety reports. Transplant participants were blinded, and care clinicians were not blinded to the treatment arm. Analysis began May 2019.
A total of 305 kidneys were randomized, with 53 subsequently being excluded (Figure 1). A similar discard and withdrawal rate between the 2 trial arms resulted in 127 end-HMPo2 and 135 SCS kidneys available for primary and secondary outcome analysis (Figure 1; eTable 1 in Supplement 2). Fourteen kidneys of the end-HMPo2 group were cold stored because machine perfusion was found to be impossible (eTable 2 in Supplement 2), and 6 kidneys received machine perfusion for less than 2 hours (for logistical reasons). All these organs are included in the end-HMPo2 arm on an intention-to-treat basis.
Donor and recipient characteristics were well balanced in both treatment arms (Table 1). To summarize donor factors that influence transplant outcome and the relative risk for graft failure after transplant, we calculated both the kidney donor risk index and kidney donor profile index for the transplanted kidneys. Quiz Ref IDIn our trial, the kidney donor risk index and the kidney donor profile index were comparable among treatment arms. Cold ischemia time refers to the total preservation time in both groups. In the end-HMPo2 treatment arm, the median SCS time prior to placement on the perfusion device was 7.97 hours, followed by a median of 4.67 hours of hypothermic oxygenated machine perfusion.
Quiz Ref IDThe primary end point of this trial was graft survival at 1 year after transplant. In the end-HMPo2 group, 92.1% (117 of 127) of kidney grafts were functioning at 1 year, which is similar to 93.3% (126 of 135) of functioning grafts in the control group (SCS) (95% CI, −7.5 to 5.1; P = .71) (Table 2). Death-censored graft survival was similar in both groups at 1 year (Figure 2). Graft losses were due to immunological reasons (n = 3), viral or bacterial infection (n = 3), arterial or venous thrombosis and complications (n = 5), or other reasons (n = 8) (eTable 3 in Supplement 2). To account for the number of crossovers, we performed a per-protocol analysis, which also did not yield any significant difference in terms of graft survival (eTable 4 in Supplement 2).
Estimated GFR was comparable between both treatment groups at all assessed time points and showed a steady increase for both groups over time until 1 year after transplant (Table 2; eTable 5 in Supplement 2). Quiz Ref IDThe rates of DGF were numerically lower within the end-HMPo2 group compared with the SCS group (30 [23.6%] vs 38 [28.1%]); however, this did not reach statistical significance (95% CI, −15.1 to 6.1; P = .40). Similar results were also found for occurrence of functional DGF, with lower rates in the end-HMPo2 group vs the SCS group (76 [59.8%] vs 93 [68.9%]; 95% CI, −22.5 to 2.7; P = .13). Rates of primary nonfunction were the same in both groups (Table 2).
Rates of patient death were higher in the end-HMPo2 group compared with the control group (9 [7.1%] vs 2 [1.5%]; 95% CI, 0.07-10.5; P = .03) (Table 2; eFigure 1 in Supplement 2). Reasons for death over the course of 12 months were myocardial infarction (n = 5), wound infection and subsequent sepsis (n = 3), multiorgan failure (n = 1), unintentional cerebrovascular injury (n = 1), and malignant neoplasms (n = 1) (eTable 6 in Supplement 2). Patients in the end-HMPo2 group who did not survive died with a functioning graft, except for 1 patient. Rates of biopsy-proven acute rejection did not show a significant difference between patients of the end-HMPo2 and SCS groups (Table 2). The rates of patients for whom at least 1 adverse event was reported were similar in both arms (62 [54.9%] in the end-HMPo2 group vs 99 [66.4%] in the SCS group; 95% CI, −23.4 to 0.03) (eTable 7 in Supplement 2). The incidence of at least 1 serious adverse event was also similar in both treatment groups (76 [67.3%] in the end-HMPo2 group vs 93 [62.4%] in the SCS group; 95% CI, −6.8 to 16.5), with none of the serious adverse events being attributable to the storage method (eTable 8 in Supplement 2).
Further exploratory analysis showed that when stratifying graft failure according to study group and the incidence of DGF, once DGF occurred, graft survival was almost identical between kidneys that were either cold stored or machine perfused (eFigure 2 in Supplement 2).
For kidneys that were randomized to the end-HMPo2 group, the duration of machine perfusion did not correlate with kidney function at 12 months (eFigure 3 in the Supplement 2). This was also shown for the length of cold storage time prior to machine perfusion, in which the duration of storage time in general did not show a significant effect on kidney function at 12 months after transplant (eFigure 4 in Supplement 2). The incidence of DGF as well as biopsy-proven acute rejection episodes were also not affected by the length of end-HMPo2 or previous SCS time (eTable 9 in Supplement 2). Extreme hours of machine preservation (ie, >10 hours of machine preservation) did not show differences in outcome (eTable 10 in Supplement 2). For a detailed description of number of kidneys per length of preservation per trial arm, see eTable 11 in Supplement 2.
In the past decades, many transplant centers have adopted the policy of placing donor kidneys on the pump using HMP for a few hours immediately prior to implantation and after a period of SCS during transport to the recipient center.16 To our knowledge, this is the first multicenter randomized clinical trial to evaluate this often-applied strategy that in general is perceived as beneficial and enhancing donor kidney function.
The results of this multicenter trial do not show any improvement in 1-year graft survival or function when higher-risk ECD kidney grafts are first statically cold stored and then exposed to oxygenated HMP prior to implantation. This finding is in contrast to the widespread clinical assumption on the superiority of continuous HMP over SCS in donors with brain death and donation after cardiac death donor kidneys.16,21,22 There are a number of potential explanations for this different outcome we would like to summarize.
First, in our previous report, an improvement of graft survival by 12.1% (from 80.2% to 92.3%) was observed when ECD kidneys were exposed to (nonoxygenated, continuous) HMP instead of SCS.15 The data presented in this recent trial show similar results of graft survival for SCS and end-HMPo2, which exceed 92% in both arms of the trial. This implies that in the past years in the same clinical environment and centers, a substantial improvement of graft survival has occurred compared with our initial publication dating back to 2011.15 Exploring the reasons for improved graft survival at this stage requires much larger sample sizes than the numbers included and on which this analysis is based. A possible trial-specific influencing factor could be the donor age limit that was set in this trial to be 85 years, which was not defined in our underlying studies. Also, only kidneys that were able to be placed on pump were ultimately perfused via end-HMPo2. Extreme anatomical variants or the mere fact of not achieving a perfect perfusion circuit were reasons to not place graft on the perfusion machine. Immunosuppression was documented from all kidney transplant recipients and was similar in all trial sites. The immunosuppression regimen was consistent and comparable within the past decade, minimizing the likelihood as being attributable to the overall outcome of the trial.
Second, HMP has repeatedly been shown to decrease the incidence of DGF and improve graft survival up to years after transplant,23-26 especially in donor kidneys 65 years and older,27 when HMP was applied throughout the entire preservation period, ie, donor kidneys were perfused immediately after procurement and until transplant at the recipient center. End-HMPo2 is a strategy that hopes to facilitate logistics during organ procurement and transportation, using valuable time on arrival at the recipient center to recondition a statically cold-stored organ, while avoiding prolongation of cold ischemia time.28 However, the exact time when to start the reconditioning, and thus estimating the exact balance between the duration of kidneys being cold stored and then machine perfused with (or without) oxygen required to maintain the positive effect on transplant outcomes, is difficult to determine. Some experimental data suggest a beneficial effect of end-ischemic (nonoxygenated) machine perfusion of as little as 1 hour.29 Other data in a porcine model by Hosgood et al30 did not find functional improvement when kidneys were subjected to 4 hours of HMP following 14 hours of SCS vs 18 hours of SCS alone without HMP. In a 2017 study that reported paired analysis of ECD kidneys, (nonoxygenated) end-HMP with a mean preservation time of 6.15 hours proved to be an independent factor for the prevention of DGF, which in turn was the strongest risk factor for 1-year graft failure.16 In our current trial, a mandated minimum machine preservation time of 2 hours and a mean preservation time of 4.67 hours using end-HMPo2 did not improve clinical outcomes after transplant, suggesting that either a longer period of oxygenated HMP or an earlier supply of oxygen and/or HMP is required to maintain a clinically relevant improved outcome. Further in-depth analysis of our cohort regarding the balance of SCS and HMP did not reveal any possible improvement in graft survival, not even in kidneys that were perfused for the longest time after a relatively shorter period of SCS. The accurate timing of HMP administration either at the beginning (preconditioning) or at the end (reconditioning) can be discussed. In parallel to this trial in ECD kidney transplant, another prospective randomized clinical trial by Consortium for Organ Preservation in Europe has directly compared continuous oxygenated vs nonoxygenated HMP in paired donor kidneys from the moment of procurement until implantation.31 This study found that prolonged oxygenated HMP provides significantly better results in donation after circulatory death kidneys compared with standard HMP, showing a lower rate of graft failure, reduction in incidence of acute rejection, and better estimated glomerular filtration rate at 12 months after kidney transplant.32
In concordance with those results, recent experimental analyses have shown that oxygenation during HMP increases kidney flow during HMP preservation and suggest that an early application of oxygenated HMP may be a more effective method than end-HMPo2.33,34 In addition, it remains unclear whether an initial potentially more damaging period of SCS can be overcome by offering a subsequent shorter or longer duration of (oxygenated) HMP. Another aspect of HMP is the possible assessment of perfusion characteristics, such as kidney resistance, that has been found to predict DGF, when assessed at the end of HMP.35 In our study, we have observed a decrease of kidney resistance and subsequent increase in kidney artery flow during the perfusion period (data not shown), and the presence or absence of any correlation with clinical outcome has to be investigated.
While cold storage is currently the most widely applied technique in organ preservation, a brief period of normothermic machine perfusion has been increasingly tested in an experimental setting. By choosing normothermic conditions, especially in higher-risk donors, it is thought to avoid cold ischemic injury and allow a better assessment of organ viability.36 While a first trial in a donation after circulatory death setting comparing SCS alone with 1-hour end–normothermic machine perfusion immediately prior to transplant after SCS is underway and results have to be evaluated (ISRCTN15821205), current experimental data using oxygenated end-HMP after prior HMP (and without any SCS) have shown that this combination can improve early graft function.33 First-in-man data on controlled rewarming of a cold-stored kidney graft have recently shown good results in a clinical setting.37
This international clinical trial aimed to increase insight on the question of relevance and if so, on duration and initiation of oxygenated HMP when following SCS preservation in kidney transplant, which is widely perceived as beneficial. Although this study is statistically underpowered owing to the improved graft survival rates achieved today in this high-risk group of ECD kidneys, we have failed to find any clinically beneficial effect by hypothermically machine perfusing donor kidneys for a brief period including oxygenation in the recipient center after a prior prolonged period of SCS preservation. Our current data do not support the use of a noncontinuous, brief period of HMPo2 placed at the final stage of organ preservation in ECD kidneys, which appears to be clinically ineffective while generating additional cost.
The baseline assumption of 80.2% 1-year graft survival in ECD kidneys has been exceeded by far in the control group of the present study. This study is statistically underpowered owing to the improved graft survival rates achieved today in comparison with clinical trial data used for the statistical analysis plan.
Reconditioning of higher-risk ECD kidneys from donors after brain death using short-term oxygenated HMP immediately prior to transplant after a period of SCS does not lead to improved graft survival or graft function when compared with simple SCS alone.
Corresponding Author: Peri Husen, MD, Department of General, Visceral and Transplantation Surgery, University Hospital Essen, Hufelandstr 55, 45122 Essen, Germany (firstname.lastname@example.org).
Accepted for Publication: January 19, 2021.
Published Online: April 21, 2021. doi:10.1001/jamasurg.2021.0949
Author Contributions: Drs Husen, Paul, and Ploeg had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Ploeg and Paul contributed equally.
Concept and design: Husen, Jochmans, Knight, Ablorsu, Pratschke, Mathe, Leuvenink, Minor, Ploeg.
Acquisition, analysis, or interpretation of data: Husen, Boffa, Jochmans, Krikke, Davies, Mazilescu, Brat, Knight, Wettstein, Cseprekal, Banga, Bellini, Szabo, Darius, Quiroga, Mourad, Pratschke, Papalois, Mathe, Pirenne, Ploeg, Paul.
Drafting of the manuscript: Husen, Davies, Mazilescu, Cseprekal, Ablorsu, Darius, Mathe, Leuvenink, Ploeg, Paul.
Critical revision of the manuscript for important intellectual content: Husen, Boffa, Jochmans, Krikke, Brat, Knight, Wettstein, Banga, Bellini, Szabo, Darius, Quiroga, Mourad, Pratschke, Papalois, Mathe, Minor, Pirenne, Ploeg, Paul.
Statistical analysis: Husen, Davies.
Obtained funding: Husen, Ploeg.
Administrative, technical, or material support: Husen, Boffa, Mazilescu, Brat, Knight, Wettstein, Cseprekal, Banga, Bellini, Szabo, Ablorsu, Darius, Quiroga, Mourad, Pratschke, Papalois, Mathe, Minor, Ploeg, Paul.
Supervision: Husen, Jochmans, Knight, Darius, Quiroga, Pratschke, Papalois, Mathe, Leuvenink, Ploeg, Paul.
Conflict of Interest Disclosures: Dr Husen reports grants from European Union 7th Framework Programme during the conduct of the study. Dr Boffa reports other support from Astellas Pharma outside the submitted work. Dr Jochmans reports grants from European Union 7th Framework Programme during the conduct of the study and other support from European Society for Organ Transplantation and European Association for the Study of the Liver outside the submitted work. Dr Knight reports personal fees from OrganOx for clinical trial design outside the submitted work. Dr Leuvenink reports grants from European Union 7th Framework Programme during the conduct of the study. Dr Minor reports grants from the European Union during the conduct of the study. No other disclosures were reported.
Funding/Support: The trial was funded by the European Union 7th Framework Programme (Theme Health.2012.1.4-1, grant agreement 305934). Perfusion devices and disposables were obtained from Organ Assist. MedAssist provided logistical support in terms of delivery and collection of devices, disposables, and samples.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Data Sharing Statement: See Supplement 3.
Additional Contributions: This trial was conducted by the members of the Consortium for Organ Preservation in Europe. We thank the European Commission for their funding through the Seventh Framework Programme. We also thank Timothy Boland (trial management); Ally Bradley (trial management); Virginia Chiocchia, MSc (statistical analysis); Katherine Corr (trial management) (Nuffield Department of Surgical Sciences, University of Oxford, UK); H. Sijbrand Hofker, MD (local trial coordination) (Department of Surgery, University Medical Center Groningen, the Netherlands); Undine Gerlach, MD (local trial coordination) (Department of Surgery, Charité, Berlin, Germany); Halil Karadag, MD (sample collection) (Department of General, Visceral and Transplantation Surgery, University Hospital Essen, Germany); Martin Kuizenga (assistance in machine preservation) (OrganAssist, the Netherlands); Rajeev Kumar, PhD (database management); Margaux Laspeyres, MA (trial management) (Nuffield Department of Surgical Sciences, University of Oxford, UK); Sarah Mertens (data collection and local coordination) (Transplant Research Group, Laboratory of Abdominal Transplantation, Department of Microbiology, Immunology and Transplantation, KU Leuven, Leuven, Belgium); Bhumika Patel (biobank management) (Nuffield Department of Surgical Sciences, University of Oxford, UK); and Marianne Thijssen-Grooten (logistics support) (MedAssist, the Netherlands) for their support. We are indebted to the members of the data monitoring committee: Christopher J. E. Watson, MD (chair) (University of Cambridge, UK); Josep M Grinyó, MD (University of Barcelona, Spain); Gabriel C Oniscu, MD (Royal Infirmary of Edinburgh, UK); and Susan Charman, BSc, Dip Ed, MSc (London, UK), as well as to all our patients, the donors and their families. These individuals were not compensated.