Liver transplantation (LT) has emerged as standard of care for treatment of irreversible acute and chronic liver failure. Despite excellent outcomes, the limited number of available organs leaves a large number of patient with established and novel indications underserved. The organ shortage and the changing demographics of deceased donors resulted in acceptance of an increasing number of livers of extended criteria donors (ECDs).1 A number of studies showed a higher susceptibility of these organs to prolonged ischemic time, a higher rate of primary nonfunction (PNF), early allograft dysfunction, and poor long-term outcome.2-4
In the light of these developments, normothermic machine perfusion (NMP) has evolved as a promising approach to reduce and counteract the detrimental effects of ischemia—especially in high-risk organs.5-8 The introduction of this new technology has several implications on the field, including the evaluation of graft function and injury before transplantation, possibly resulting in lower organ discard rates.9-11
Safe prolongation of preservation time has the potential to ease the everyday clinical logistic challenges and to facilitate LT without pressure of time. In the light of the current global threat of the SARS-CoV2 pandemic, expansion of preservation times for the sake of donor and recipient testing adds an immediate potential benefit.
Although the knowledge in machine perfusion is rapidly growing, the experience outside of clinical trials is still very limited. Considering that NMP is now rapidly being adopted by an increasing number of centers, guidance for safe introduction and streamlining in data recording and decision-making is warranted. Since NMP is not solely used for preservation during transportation but also overnight storage, monitoring, and evaluation, standard procedures need to be established for safe implementation of this technology into clinical routine. Several factors, such as task sharing, monitoring, response to technical failure, benchmarking for decision-making, and actual net benefit of the technology require closer attention. We choose a poststatic cold storage NMP approach since this approach facilitates clinical adoption. Working from our own experience, we herein describe standard operating procedures (SOPs) for NMP in LT and summarize the findings of functional testing and prolongation of total preservation time. We provide guidance for establishing NMP, including organ monitoring and care in the clinical routine.12
MATERIALS AND METHODS
NMP was introduced at our center in February 2018. An NMP device (OrganOx Metra, Oxford, United Kingdom) was set up and on-site test runs under the instruction of a company representative were undertaken. Two training sessions for setup and handling of the device were held on the days of implementation. The on-site training of the perfusionists and the surgical team was done with support of the company. Further details are described below and in supporting information. Owing to the complexity of the entire process, a decision was made to initiate NMP in a back-to-base concept before NMP in remote hospitals. While the device has a european conformity mark, it was considered novel from a clinical application perspective. The use in the clinical phase II/III trial was somewhat different from a sustainable clinical routine in the sense that a dedicated person was performing NMP in the trial, whereas the entire team would require training when established as a clinical routine. Considering that approximately 15 surgeons are involved in transplantation at our center and that intensive care medical personnel, perfusionists, and nurses require specific knowledge and skills in order to participate in this routine, such training was established with a focus on the individual tasks. Because of the novelty of the procedure, a specific consent form was implemented at our center. Out of 106 patients waitlisted and transplanted since the beginning of the NMP program, 77.4% gave their informed consent and 1.9% declined the preservation method. In 20.7%, patients were waitlisted before availability of NMP and NMP was not required when transplantation was performed.
Technical Training and Establishing Center-specific Protocols
A key step to establish NMP in clinical routine is the intradisciplinary and interdisciplinary approach with clear definitions of tasks and responsibilities. Table 1 summarizes the responsibilities of the different stakeholders in the process.
Involvement of the transplant coordinators, anesthetists, perfusionists, nurses, laboratory, and blood bank staff is essential to enable a straightforward application of NMP in the clinical routine. Over 20 in-depth training sessions for transplant coordinators, anesthesiologists, perfusionists, nurses, laboratory, and blood bank staff were performed (Table 2). More task-specific training with all members of these individual disciplines was conducted by the NMP-program lead. This involved either a hands-on training, for example, sample collection, application of specific substrates, blood samplings, glucose settings as well as troubleshooting for the anesthesiology as well as nurse staff. A more conceptual introduction was offered for the staff of the central laboratory, blood bank, and transplant coordinators. Table 2 summarizes the specific training including the approximate duration of each session until NMP was fully established in 24/7 clinical routine. Furthermore, a working group on a social media platform was established as a 24/7 troubleshooting service. The group included a technical expert from the NMP company and also served for immediate information sharing within the group. Data on the platform were purely technical in order not to breach confidentiality.
Indications for NMP
In reference to a concept developed previously by 1 of the authors (C.J.E.W.), NMP was applied for the following indications and recommend this algorithm as an SOP for decision-making and documentation:
- (I) Door-related indications: The beneficial effects of NMP may be most clinically relevant in extended criteria organs (donor age > 65, elevated risk for transmission of a disease, eg, hepatitis b/c; donor body mass index [BMI] > 30 kg/m2; total bilirubin > 3 mg/dL; serum sodium > 165 mmol/L; hospitalization in intensive care unit [ICU] > 7 d; hepatic steatosis > 40%, acute hemodynamic deterioration with risk of organ loss13) including donation after circulatory determination of death (DCD) livers. Especially in combination with expected prolonged ischemia times, NMP can facilitate better preservation of the graft and allows determination of graft performance during NMP.5,14,15
- (II) Logistic-related indications: NMP has been shown to be feasible and safe although mechanisms of action are still under investigation.16,17 An extended preservation time has a significant implication on the logistics of the surgical procedure. It transforms transplantation from an emergency into a scheduled procedure and allows time to, for example, better prepare the recipient with a complex medical condition for the surgical procedure.5 In cases of parallel organ transplantations or overlap with other urgent surgeries, NMP reduces the need for parallel surgeries and is advantageous in the light of human resource management and working hour restrictions.
- (III) Recipient-related indications: In cases of surgically highly complex recipients or high-risk patients, NMP is extremely helpful in order to relieve time pressure and reallocate the organ in case hepatectomy is considered impossible. Although this seems trivial, it proved extremely valuable in some cases in our experience.
Logistics, Back-table Preparation, and Monitoring
Once the decision had been made as to whether a liver is suitable for NMP according to the above-mentioned criteria, the transplant coordinators established a donor organ identity case in our digital clinical data system (see also Table 1). Donor data sheets, organ labels, SOPs for NMP setup, and a perfusion protocol indicating the time points and samples for laboratory analysis (Table 3) were sent to the operating room (OR) and strictly remained with the NMP device during the entire period.
In case of a very short hepatic artery, an extension graft may be necessary in order to ensure sufficient length of the artery for transplantation. In our experience, especially prolonged NMP may lead to a dry hepatic artery and eventually result in a need to shorten the vessel. In order to avoid this, a moistened gauze was placed on the artery.
For liver NMP monitoring, the NMP device together with the data set was transferred to the ICU and handed over by the surgeon on duty with a sign out to the anesthetist and nurse in charge. Management of nutrition, buffering with sodium bicarbonate, glucose control, perfusate sampling for laboratory analysis (according to protocol, Table 3), and blood gas (BG) analysis was performed on-site. Maintenance of physiological pH values between 7.2 and 7.4 throughout NMP was a target. The amount of sodium bicarbonate required to achieve the pH target was recorded but not considered a parameter for decision-making at this point. The results of BG analyses as well as laboratory analyses are automatically transferred to the digital clinical data system and reviewed during rounds at the ICU. Although the selection of parameters recorded during NMP are considered to portray the quality of the organ, only a few parameters can be considered to serve decision-making at this point.
Blood Gas Analyses and Laboratory Parameters
Selected BG parameters were continuously monitored through the built in BG-analyzer. For a more comprehensive assessment of perfusion parameters, a center protocol was established (Table 3; Table S1, SDC, https://links.lww.com/TP/B937). At present, there is little evidence about the predictive value of individual perfusion parameters. Lactate degradation, aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels, and dynamics as well as bile pH have been correlated with initial graft function.7 With patient consent for application of this innovative technology, a panel of parameters beyond the above mentioned was assessed as part of the routine.
Decision for Transplantation and Management of NMP in the OR
Laboratory parameters assessed during NMP are displayed in Table S1, SDC, https://links.lww.com/TP/B937. Lactate decrease, pH maintenance, and glucose consumption have been established as meaningful parameters to assess organ quality during NMP.6,9 A maintenance of physiological pH values without sodium bicarbonate supplementation after 2 h of NMP as well as a rapid decrease and maintenance of lactate to physiological levels are considered key factors indicating good organ function. Further to this, exceptionally high AST, ALT, and lactate dehydrogenase levels and a sharp incline of these parameters are considered warning signals. Other values such as bile output and bile pH are valued as indicators for bile duct viability and function. After the organ has been classified as suitable for transplantation, the surgical procedure is initiated. Once hepatectomy is close to completion, the NMP device is transferred to the OR. At the end of the procedure, the NMP minimal data set is transferred to a database (Table S2, SDC, https://links.lww.com/TP/B937).
This study was approved by the ethics committee of the Medical University of Innsbruck (No. 1175/2018). The need for written informed consent was waived by the ethics committee for this retrospective, descriptive study.
NMP was introduced in the clinical routine at the Medical University of Innsbruck in February 2018. It was since applied in 34 donor livers: 25 of these were eventually transplanted and 9 organs were discarded because of poor performance during NMP (in addition to donor risk factors) (Table 4 shows donor risk factors and final reasons for decline). The main reasons to opt for NMP in the 25 transplanted cases were marginal donor organ quality in 9 patients, logistics in 6 patients (multiple parallel organ transplantation such as heart, lung, or combined kidney pancreas transplantations), the combination of marginal donor quality and logistics in 5 patients, the combination of logistics and surgically complex recipients in 2 cases as well as the combination of the marginal organ quality, and logistics as well as surgically complex recipient in 3 cases (2 case descriptions are included in the SDC, section 2, SDC, https://links.lww.com/TP/B937) (Figure 1). As per the center threshold, 9 lower organ quality livers would not have been accepted for transplantation, if NMP was not in place to allow for further assessment of the liver (Table 5 shows donor risk factors and reasons to opt for NMP). Extension of NMP duration up to 24 h allowed to omit nighttime procedures and parallel transplantations. During this initial experience of NMP some minor events and technical hurdles occurred; however, we did not encounter graft losses or injuries due to NMP. Technical issues and ways to overcome these errors are described in the supporting material (SDC, https://links.lww.com/TP/B937). Especially, the clear definition and distribution of responsibilities allowed for a rapid and efficient troubleshooting.
Practical Issues (SOPs) in Starting a Liver NMP Program
There are 3 conceptual key challenges with the establishment of a liver NMP program in the clinical routine:
- (1) The calibration and setup of the device is not formally specifically embedded in any job description and successful application requires good will, engagement, and teamwork between perfusionists and surgeons. Senior surgeons find themselves mixing drugs as adjuncts and pulling syringes in the middle of the night in the early phase. Interteam and intrateam reservations need to be sorted and a sense of joint effort established until the novel tasks are eventually formally included in training and job descriptions.
- (2) The holding place for the device during preservation and assessment requires presence of attentive personnel and equipment for perfusate (and tissue) sampling. We currently position the NMP device in the ICU and use the future recipients’ space since it fulfills these requirements. The allocation of such resources, however, is only a transient solution and eventually, dedicated rooms with dedicated personnel are required.
- (3) The involvement of numerous disciplines holds the risk of miscommunication and loss of information at the interfaces. Clear communication, complete documentation (see Table 2), and clear determination of responsibilities (see Table 1) are required. In addition, the multidisciplinary character requires uniform support by the hospital leadership and the representatives of all disciplines involved since additional work time and attention is needed especially during the early phase of NMP.
- (4) The decision-making is not always straightforward. In some cases, the parameters assessed display a mixed picture and true determination of the quality of the parenchyma and the bile ducts remains an approximation. Some of the laboratory values are far outside the common norm and the determination of the significance remains incomplete. For example, potassium values of up to 13 mmol/L. The difference between organs is significant, and the relevance of this and many other findings remains to be established. Handling this level of uncertainty can be challenging.
Donor and Recipient Demographics of the First 25 Transplanted Cases
Preservation times and donor demographics of the 25 transplanted organs are summarized in Table 6. Median donor age was 57 (range, 18–80), BMI was 28 (range, 22–33), and the liver donor risk index was 1.9 (range, 1–3.5). Twenty-three of 25 donors met ECD criteria (92%). Thirteen donors were female (52%). Causes of death were cerebrovascular accident in 12, circulatory arrest not otherwise specified in 3, trauma in 5, and hypoxia in 5 donors. Twenty-one organs were retrieved from DBD and 4 after DCD. Prior cardiac arrest was recorded in 15 donors with a mean return of spontaneous circulation of 42 min (±38 min). The mean cold ischemia time until start of NMP was 397 min (±144 min), mean NMP time was 816 min (±386.0 min) resulting in a total preservation time of 1213 min (±426 min). Total preservation time exceeded 24 h in 8 cases and 30 h in 2 cases. Figure 2 shows mean courses of AST (a), ALT (b), and lactate dehydrogenase (c) during NMP of livers eventually transplanted and nontransplanted.
The median recipient age was 64 y (range, 33–71 y), the median model of end stage liver disease was 14 (range, 8–40 y). The BMI was 25.8 (range, 15.5–35.4 y). Twenty-three recipients were male and 2 were female. Indications for LT were nonalcoholic liver cirrhosis (n = 2), alcoholic liver cirrhosis (n = 9), secondary sclerosing cholangiopathy (n = 1), cirrhosis due to chronic hepatitis B (n = 2), and hepatitis C (n = 1) and cryptogenic cirrhosis (n = 4). Furthermore, 5 patients underwent second LT (causes for retransplantation: hepatitis B recurrence in 1, ischemic-type biliary lesion in 3, PNF in 1 patient) and 1 patient underwent third LT due to recurrence of primary sclerosing cholangiopathy.
A hepatocellular carcinoma was present in 9 patients. Two patients were accepted for high-urgency LT. Table 7 summarizes recipient demographics.
Graft and patient survival were 88% (22 out of 25 patients) at 20 mo (mean follow-up of 8.6 ± 5.9 mo). Causes of death were fungal sepsis in 1 patient on day 57 posttransplantation and multiorgan failure 246 d posttransplant due to a rapidly progressing posttransplant lymphoproliferative disease in a second patient and multiorgan failure due to septis at day 43 in the third patient. Liver function was normal in all 3 cases until death or immediately before death hence death-censored graft survival is 100%.
While primary poor function according to Olthoff18 was observed in 5 patients (20%), no PNF occurred. Mean bilirubin serum levels and mean international normalized ratio values on day 7 posttransplant were 2.7 mg/dL (±3.0 mg/dL) and 1.1 (±0.1), respectively. Furthermore, mean bilirubin serum levels were 1.9 mg/dL (±4.4 mg/dL) and international normalized ratio remained stable at 1.1 (±0.3) in the last follow-up.
The 90-d readmission rate was 28% (7 out of 25 patients). The reasons for readmission were anastomotic stricture (n = 3), acute rejection, pneumonia, atrial fibrillation, and fungal infection (1 patient each).
Early biliary complications (<30 d posttransplantation) were observed in 5 patients (20%). Anastomotic leakage was found in 3 and anastomotic strictures in 2 patients. All were successfully treated surgically or endoscopically. Late biliary complications (>30 d posttransplantation) occurred in 6 patients (24%). Five patients were successfully treated by endoscopic retrograde cholangio-pancreatography for anastomotic strictures. One patient underwent resection of the left lateral segments due to an intrahepatic abscess following anastomotic stricture and ascending cholangitis. No alterations in the right hepatic biliary tree were observed in this patient. No ischemic-type biliary lesion was observed in our patient cohort. Biliary complications are resumed in Table 8.
We herein present our experience with the initiation of a liver NMP program and describe major logistic and operational as well as administrative challenges, that need to be overcome in order to successfully establish a 24/7 service in the clinical routine. In summary, we find that liver NMP is extremely helpful in several aspects of LT. As often with new technologies (consider also robotic assisted surgery), the advantage is immediately identifiable in some aspects (functional assessment, increased flexibility) but less measurable in some other aspects (lowering discard rate, optimizing the surgical conditions in complex recipients). In our clinical-routine experience, the relative advantage of NMP seems to go beyond better preservation as investigated in the randomized controlled trial by Nasralla et al.7 The time reserve gained is an advantage that we feel is highly underestimated in the current discussion. The focus of the mechanistic aspects and the bioenergetics events as they are altered by machine perfusion under different conditions is very relevant and the critical discussion regarding the endpoints in any machine perfusion trials is meaningful,19 but a controversy over mechanisms may overlook the actual and immediate benefit in a clinical routine and result in withholding a technology with advantage for our patients instead of collectively advancing with the development of this novel field. We have completely omitted nighttime LT and can adopt the call schedule accordingly. We feel that higher safety and flexibility are relevant parameters for human resource management and education and training in LT.
Our back-to-base approach made a safe introduction of NMP in a larger team possible. Immediate advantages are the familiar environment and the opportunity to teach NMP to a larger group. Together with a center-specific protocol and a clear definition of roles and tasks, this allowed for a rapid development of a routine. Considering that the preservation periods in LT are now extended to over 30 h, a multidisciplinary approach and clear definition of roles, responsibilities, interface, sign in, sign out, and decision-making are warranted. We feel that we have established a framework in this regard, which could be useful for others. Responsibilities for care of the organ during NMP warrant clarification by a local protocol and the competent national authority. At present, the legal interpretation regarding who owns an organ after donation refers to organs in the state of cold storage. The circumstances of warm perfusion and eventually and foreseeably organ treatment are conditions that will require reconsideration of the legal framework and a more precise definition of ownership and responsibility. Since NMP is still novel, we have decided to develop a specific consent form for patients to indicate, that they have been informed about the novelty of the procedure and the limited experience and therefore limited knowledge regarding the long-term outcome.
In the process of establishing normothermic liver perfusion as a standard for marginal livers, logistic demands, and complex recipient operations, we have identified several challenges. We believe that the technology has advanced to a stage in which safe use in humans can be achieved under well-controlled circumstances and by a well-trained and dedicated team. NMP is changing from an experimental procedure to a routine clinical application. Means for immediate troubleshooting are essential since blood perfusion of the organ leaves little room for machine malfunction. A 24-h service, remote life data read, and possible remote machine control are considerations for a backup safety procedure in case of malfunction or human error. The fact that the human organ is kept “viable” and active during this period has a number of risks and challenges that need to be addressed in order to facilitate the safe and wide spread use of NMP. We herein provide a report on the establishment of NMP as a clinical routine and identify several important aspects. SOPs for setup, management, data collection, and decision-making may serve as basis for others to use or adopt. We believe that machine perfusion may emerge to replace or complement cold storage and strongly suggest to standardize procedures and data collection in order for this technology to advance in a meaningful way.
Significant efforts are needed to adopt such a novel technology in reference to all existing guidelines and national regulations. A safe implementation of a novel technology relies mainly on the prudence of the team and the key coordinators of the project. The specific tasks emerging with NMP will eventually need to be implemented into the respective job descriptions. The respective training will need to be formalized and certified. The fact that we are handling life-saving organs leaves no room for mistakes. We have identified eminent challenges in the clinical implementation of this technology and share our experience in order to facilitate implementation of this and similar novel technologies in transplantation.
For wider spread routine clinical application of NMP, reimbursement needs to be addressed. Registry data and real world experience in addition to clinical trials may eventually provide sufficient accumulated evidence in order for caregivers to agree reimbursement. Since the technology is expensive in comparison to cold static storage, the advantage for the individual patient needs to be specified.
We herein describe 3 indications for NMP. Although NMP for ECD organs may result in a higher number of transplanted organs, the eventual discard of some of these organs puts an additional economic burden on the procedure. Logistics and complex recipients may seem trivial as indications in comparison, but our experience in these cases was extremely positive. Although the number of biliary complications as a key outcome parameter may seem high with approximately 30%, the majority was related to anastomotic strictures and were successfully treated. Biliary complication rates of up to 40% have been reported after static cold storage.20 Considering, that the ECD rate was over 90%, we feel that the results are overall reasonable. There are no indications, that NMP would improve liver or bile duct viability. Parallel nighttime transplantations were omitted entirely since introduction of NMP. LT was performed as a daytime procedure at our center in almost all NMP cases.
The decision-making toward an eventual transplantation or not requires further attention. Current knowledge does not allow us to determine a cutoff or a summative score that would indicate suitability for transplantation. Instead, the collective data and interpretation of the individual parameters in addition to previously established parameters such as the liver donor risk index eventually describe a picture of the condition of a liver. The final decision to transplant or discard an organ remains an educated assumption at this point. A joint platform for data collection and correlation with the outcome would be needed in order to advance the understanding of the relevance of the individual parameters. Unfortunately, this is pending and the current registries and consortia need time to adopt to this novel technology.
In summary, establishing a normothermic perfusion in the clinical setting is achievable in an interdisciplinary approach, wherein a strict partition of tasks is clarified and an appropriate training of different disciplines is performed.
We are in debt to the staff members of the teams involved in all aspects of NMP for their assistance and to the Tirol Kliniken as well as the Medical University of Innsbruck for supporting the NMP program.
1. Nemes B, Gámán G, Polak WG, et al. Extended criteria donors in liver transplantation part I: reviewing the impact of determining factors. Expert Rev Gastroenterol Hepatol. 2016; 10:827–839. doi:10.1586/17474124.2016.1149061
2. Ploeg RJ, D’Alessandro AM, Knechtle SJ, et al. Risk factors for primary dysfunction after liver transplantation–a multivariate analysis. Transplantation. 1993; 55:807–813. doi:10.1097/00007890-199304000-00024
3. Feng S, Goodrich NP, Bragg-Gresham JL, et al. Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant. 2006; 6:783–790. doi:10.1111/j.1600-6143.2006.01242.x
4. Hashimoto K, Miller C. The use of marginal grafts in liver transplantation. J Hepatobiliary Pancreat Surg. 2008; 15:92–101. doi:10.1007/s00534-007-1300-z
5. Ceresa CDL, Nasralla D, Jassem W. Normothermic machine preservation of the liver: state of the art. Curr Transplant Rep. 2018; 5:104–110. doi:10.1007/s40472-018-0186-9
6. Watson CJE, Kosmoliaptsis V, Randle LV, et al. Normothermic perfusion in the assessment and preservation of declined livers before transplantation: hyperoxia and vasoplegia-important lessons from the first 12 cases. Transplantation. 2017; 101:1084–1098. doi:10.1097/TP.0000000000001661
7. Nasralla D, Coussios CC, Mergental H, et al.; Consortium for Organ Preservation in Europe. A randomized trial of normothermic preservation in liver transplantation. Nature. 2018; 557:50–56. doi:10.1038/s41586-018-0047-9
8. Boteon YL, Attard J, Boteon APCS, et al. Manipulation of lipid metabolism during normothermic machine perfusion: effect of defatting therapies on donor liver functional recovery. Liver Transpl. 2019; 25:1007–1022. doi:10.1002/lt.25439
9. Watson CJE, Jochmans I. From “gut feeling” to objectivity: machine preservation of the liver as a tool to assess organ viability. Curr Transplant Rep. 2018; 5:72–81. doi:10.1007/s40472-018-0178-9
10. Mergental H, Perera MT, Laing RW, et al. Transplantation of declined liver allografts following normothermic ex-situ evaluation. Am J Transplant. 2016; 16:3235–3245. doi:10.1111/ajt.13875
11. Ghinolfi D, Pezzati D, Rreka E, et al. Nonagenarian grafts for liver transplantation. Liver Transpl. 2019; 25:1439–1444. doi:10.1002/lt.25580
12. Ceresa CDL, Nasralla D, Watson CJE, et al. Transient cold storage prior to normothermic liver perfusion may facilitate adoption of a novel technology. Liver Transpl. 2019; 25:1503–1513. doi:10.1002/lt.25584
13. Fischer-Fröhlich CL, Lauchart W. Expanded criteria liver donors (ECD): effect of cumulative risks. Ann Transplant. 2006; 11:38–42
14. Brockmann J, Reddy S, Coussios C, et al. Normothermic perfusion: a new paradigm for organ preservation. Ann Surg. 2009; 250:1–6. doi:10.1097/SLA.0b013e3181a63c10
15. Ravikumar R, Leuvenink H, Friend PJ. Normothermic liver preservation: a new paradigm? Transpl Int. 2015; 28:690–699. doi:10.1111/tri.12576
16. Watson CJ, Randle LV, Kosmoliaptsis V, et al. 26-hour storage of a declined liver before successful transplantation using ex vivo normothermic perfusion. Ann Surg. 2017; 265:e1–e2. doi:10.1097/SLA.0000000000001834
17. Laing RW, Mergental H, Mirza DF. Normothermic ex-situ liver preservation: the new gold standard. Curr Opin Organ Transplant. 2017; 22:274–280. doi:10.1097/MOT.0000000000000414
18. Olthoff KM, Kulik L, Samstein B, et al. Validation of a current definition of early allograft dysfunction in liver transplant recipients and analysis of risk factors. Liver Transpl. 2010; 16:943–949. doi:10.1002/lt.22091
19. Dutkowski P, Guarrera JV, de Jonge J, et al. Evolving trends in machine perfusion for liver transplantation. Gastroenterology. 2019; 156:1542–1547. doi:10.1053/j.gastro.2018.12.037
20. Testa G, Malagò M, Broelseh CE. Complications of biliary tract in liver transplantation. World J Surg. 2001; 25:1296–1299. doi:10.1007/s00268-001-0113-5