Langenbecks Arch Surg DOI 10.1007/s00423-014-1161-2
REVIEW ARTICLE
Machine perfusion in solid organ transplantation: where is the benefit? Helge Bruns & Peter Schemmer
Received: 16 July 2013 / Accepted: 1 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Machine perfusion (MP) in solid organ transplantation has been a topic of variable importance for decades. At the dawn of organ transplantation, MP was one of the standard techniques for preservation; today’s gold standard for organ preservation for transplantation is cold storage (CS). The outcome after transplantation of solid organs has tremendously improved over the last five decades. MP has been continuously under investigation and may be an option for organ preservation in selected cases; however, there is only little evidence from clinical trials that can be used to advocate for MP as a routine organ preservation method. Methods This article reviews the current knowledge on MP in the field of solid organ transplantation with special focus on findings from clinical trials. Conclusion Especially in heart and lung transplantation, MP seems to be a promising tool to improve postoperative outcome, but a general evidence-based recommendation for or against an application of MP cannot be given due to the lack of the highest level of clinical evidence. Gold standards such as CS should not be left behind without good reason. Randomized clinical trials are desperately needed to further improve outcome and for better understanding of the underlying pathophysiological mechanisms.
Keywords Machine perfusion . Organ preservation . Organ transplantation
H. Bruns : P. Schemmer (*) Department of General and Transplant Surgery, Ruprecht Karls University, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany e-mail: [email protected]
Introduction In 1812, Le Gallois published his ideas on the principles of organ perfusion that can be transferred to machine perfusion (MP) [1]. However, the first concepts for mechanical devices for MP were published in 1935 with the Carrel–Lindbergh machine [2, 3]. While this device was not designed for organ preservation, this was the first publication on a modern oxygenated, normothermic pulsatile perfusion system. At the dawn of transplantation, MP was the gold standard for organ preservation since this technique seemed to be mostly physiological [4–6]. With the development of modern preservation solutions, cold storage (CS) became superior to MP regarding the outcome after transplantation [7]. The early 1980s can be considered to be the turning point in organ preservation towards CS. In 1982, a retrospective analysis for kidney preservation identified a better 1-year survival after CS with Collins solution in comparison to MP with the Belzer machine (53 vs 40 %, respectively) [8]. Today, the 1-year survival after CS with UW- or HTK-solution has further improved to 85 % [9, 10]. It is not fully clear whether MP with optimized devices can keep up with CS with optimized preservation and storage solutions [11–13]. While MP potentially optimizes grafts for transplantation, there are several methodical drawbacks. One of the major concerns in MP is that organ perfusion might lead to mechanical stress and thus could lead to further damage of the explanted organs. Increased pressure during perfusion has long been identified as a negative predictor of postoperative graft function, and MP can be used to monitor the flow characteristics and reevaluate organs before transplantation [14]. MP might be an option to select organs with good or bad function after transplantation [15–19]; it is not fully clear whether MP could really improve postoperative organ function in general. Currently, MP plays only a minor role in solid organ preservation but has continuously been investigated in both
Langenbecks Arch Surg Table 1 Clinical studies and trials reviewed in this publication Study Kidney Retrospective analysis: 1 year survival after MP with Belzer (n=565) and Waters (n=1,441) machine versus CS (Collins; n=282) [8] Retrospective analysis: CTS data; risk ratio for failure after CS (n=85.944) and MP (n=2.202) [12] Retrospective analysis: USRDS data; transplantation of EDC kidneys after MP (n= 1.114) and CS (n=4.726) [35] Randomized controlled trial: MP (n=336) versus CS (n=336), kidneys taken from the same donor and randomized; primary endpoint: DGF [20, 25] Randomized controlled trial: MP (n=45) vs CS (n=45), kidneys taken from same donor and randomized; primary endpoint: DGF [27] Liver Randomized controlled trial: transplantation after MP (Medtronic PBS device with Vasosol-R; n=20) vs CS (UW; n=20) [23] Heart Phase I trial: PROTECT: transplantation after MP using Organ Care System and warm oxygenated blood (n=20) [30] Phase I trial: PROCEED: transplantation after MP using Organ Care System and warm oxygenated blood (n=15) [31] Lung Clinical feasibility study and retrospective analysis; reevaluation of high risk lungs after 4 h MP (n=23), comparison to regular lung transplants (n=116), primary endpoint: primary graft dysfunction after 72 h [65]
Findings
Criticism
Decreased or equal to 1 year survival (40 and Retrospective analysis; poor survival in any of the groups; graft allocation criteria to groups unclear 48 versus 53 %) after MP with Belzer and Waters machine Increased risk ratio for failure after MP Retrospective analysis; kidneys used for MP may be marginal grafts Retrospective analysis; impact of preservation/ Decreased rate of DGF after MP, no perfusion solutions not analyzed difference in long-term graft survival after 2 and 3 years DGF defined as the subjective need for dialysis Decreased DGF (21 vs 27 %) after MP; increased graft survival after 1 (94 vs 90 %) and 3 years (91 vs 87 %) after MP No difference in DGF; no difference in renal DGF defined as the subjective need for dialysis function (eGFR) and patient and graft survival after 3 and 12 months Clinical relevance of decreased transaminases No significant difference (trends toward unclear; effect on transaminases might be related decreased PDF and hospital stay after MP); to different preservation/perfusion solutions used decreased postoperative transaminases in the trial after MP 30-day survival=100 %; acute rejection in 10 % and transient left ventricular dysfunction in 10 % 30-day survival=93 %
Phase I trial (no controls included in the study)
Phase I trial (no controls included in the study)
Phase I trial (no controls included in the study) 20 high-risk lungs were transplanted; no difference for primary graft dysfunction compared to regular lung transplants after CS
CTS collaborative transplant study, CS cold storage, MP machine perfusion, USRDS United States Renal Data System, DGF delayed graft function, EDC extended donor criteria, eGFR estimated glomerular filtration rate, PDF primary dysfunction, UW University of Wisconsin, PROTECT prospective multicenter European trial to evaluate the safety and performance of the Organ Care System for heart transplants, PROCEED prospective multicenter safety and effectiveness evaluation of the Organ Care System Device for heart use
preclinical studies and in clinical trials [20–31]. Multiple perfusion systems and options (normothermic, hypothermic, and oxygenated) exist [32]. Most of the existing clinical trials have different approaches and address different problems, thus making reasonable, valid meta-analyses of these data impossible (Table 1).
MP in solid organ transplantation Kidney MP might be best investigated in kidney transplantation; unfortunately, these findings cannot be generalized to the whole field of solid organ transplantation [33]. Multiple trials on MP for kidney preservation after organ donation from cardiac deceased (CDD) and brain dead donors (BDD) do exist [20, 25, 27]. The largest retrospective analyses failed to
identify beneficial effects of MP. However, all of these large retrospective analyses are either too old or biased. In the older analyses, abandoned techniques for MP or CS are compared with an unacceptable overall survival even in the best subgroups. Moreover, these analyses lack important data concerning kidney donors that might have an influence on graft survival and graft function [7, 8, 12]. Some publications exist that expect an improved function of organs with extended donor criteria (EDC) after MP [34]. Thus, it may be possible that a high proportion of kidney grafts that underwent MP in these retrospective analyses were marginal organs that have an already increased probability of delayed graft function (DGF) per se. If these organs are compared to cold-stored kidneys which might have had optimal qualification, differences found between groups might be related to differences between organs and not to different methods of organ
Langenbecks Arch Surg
preservation. To fully eliminate this bias, randomized controlled trials are needed, especially those that randomly assign one of a donors kidneys to MP and the other one to CS. Retrospective analyses systematically underestimate the effect of EDC in kidney transplantation and thus might be biased and cannot be reliable in this point [12]. Moreover, all different types of MP are mixed together in these analyses. In general, studies such as the retrospective analysis of the CTS data published by Opelz and Dohler in 2007 help to identify today’s standards and outcome after transplantation but should not be used as a scientific tool to advocate for or against MP [12]. Concerning kidney transplantation, two randomized clinical trials (RCTs) have been published, and findings from these studies might have an impact on MP in kidney transplantation or at least on the development of further study protocols [20, 25, 27]. In these clinical trials, one of the donor’s kidneys was randomized to MP and the other one to CS; the primary endpoint of both studies was DGF, which was defined as the need to undergo dialysis during the first 7 days after kidney transplantation. While Moers et al. identified a decreased rate of DGF after MP of 21 versus 27 %, and graft survival was 94 versus 90 % and 91 versus 87 % after MP and CS after 1 and 3 years, respectively, the difference was only available for those kidneys that showed DGF; concerning those kidneys without DGF, there was no difference in graft survival [20, 25]. The subgroup analysis of this study did not identify advantages in the EDC group or after donation after cardiocirculatory death. An analysis of perfusate from donor kidneys from this study with EDC that underwent MP was correlated to DGF after transplantation, and lipid peroxidation products were identified as a marker for DGF [17]. Thus, it can be proposed that MP could be a tool to identify those patients at risk for DGF. Nonetheless, it is not fully clarified yet weather DGF does necessarily lead to decreased graft survival. In a retrospective analysis that focused on costs of MP, Buchanan et al. identified a decreased rate of DGF after MP but no difference in long-term graft survival [35]. In their RCT, Watson et al. have investigated the effect of MP on kidney donation after cardiac death and failed to identify any significant difference or even a trend in terms of DGF, eGFR after 3 and 12 months, and patient and graft survival [27]. In both trials published by Moers et al. and Watson et al., DGF was defined as the need for dialysis in the first week after transplantation; this is necessarily one of the main criticisms concerning both trials: While this primary endpoint clearly is clinically relevant, related to kidney function, can easily be monitored, and is one of the most relevant cost-drivers after kidney transplantation, the need for dialysis is a subjective surrogate parameter for kidney function and depends on the physician in charge; standard laboratory values that serve as
triggers for postoperative dialysis may differ between transplant centers. This end point of both studies is therefore biased. Using the perfusion device employed in both clinical trials, flow parameters can be monitored easily during perfusion of the organs [16, 19], but it remains unclear whether perfusion itself causes mechanical shear stress and might lead to further damage of some of the organs. Thus, modern approaches also need to focus on optimization of flow characteristics during perfusion. Pressure, temperature, and pulsatile frequency have to be adapted to the specific organs’ needs [36]. On first sight, MP seems to be a rather costly procedure compared to CS, but at least in kidney MP, good evidence exists that potentially, MP could even decrease total costs after transplantation, if the right patients are selected. MP decreases DGF, and DGF leads to decreased long-term graft survival [37]. Depending on reimbursement and observational period, this effect might disappear after some years [35]. Given, that the need for dialysis in the early postoperative phase decreased and long-term outcome is improved, total costs are decreased as well. Taken together, MP might be an attractive option in kidney transplantation, but until now, there is no clear evidence which patients or organs benefit most from this kind of kidney preservation. Further clinical trials are needed to identify the optimal candidates for MP. It remains unclear whether the same effect can be achieved using optimized storage solutions, be it in kidney transplantation or concerning any other solid organ [38–43]. Liver While clinical trials exist, there is only one clinical trial with published data on MP in liver transplantation [23]. Guarrera et al. found a beneficial effect on laboratory parameters after MP in terms of decreased postoperative serum transaminases, but there was no difference in graft and patient survival. In their study, 20 liver transplantations after MP using a modified Medtronic PBS device and Vasosol-R solution were compared to 20 liver transplantations after CS with the University of Wisconsin (UW) [23]. There was a trend toward decreased primary dysfunction and a decreased hospital stay after MP, but the clinical relevance of this finding remains unclear. Guarrera et al. did not find any difference between both groups in terms of survival. Moreover, the observed effect might be related to the different preservation solutions used in both groups [24, 44]. Currently, clinical trials would not only need to focus on the direct effect of MP itself but also on MP as a tool to manipulate the perfused organ ex vivo. MP could be used for pharmacological conditioning of the perfused organ, to deliver substrates and/or to remove metabolic waste products [45]. There is some evidence from preclinical studies that MP might be a tool to improve marginal liver grafts [46, 47]. Especially oxygenated MP seems to have a beneficial effect in these organs [48–50]. In porcine models, MP was identified as a tool
Langenbecks Arch Surg
to increase the donor pool [51–53]. It is not clear yet whether these findings can be transferred to the clinical situation [46]. Especially concerning marginal livers from EDC, further optimization of perfusion and storage solutions, but also introduction of pharmacological conditioning, might be an option to improve outcome as much or even better than MP [38, 43, 54–56]. Pancreas No clinical trial has been performed on MP of the pancreas yet. In 1983, preclinical studies in a canine model demonstrated an increased failure rate after MP (30 and 40 % after 24 and 48 h after MP vs 0 and 0 % after 24 and 48 h after CS) [57]. It may be worth to perform further preclinical studies in this organ, but in the authors’ opinion, it is still too early for clinical trials on MP of the pancreas. MP has been discussed as a helpful tool to improve islet yield especially after prolonged ischemia time [58–60]. In a porcine model, insulin content in islets was increased after MP compared to CS [59, 60]. Leeser et al. found an increased islet viability after MP compared to CS in human pancreata (86 vs74 %, respectively) but a decreased total islet yield compared to pancreata with
REVIEW ARTICLE
Machine perfusion in solid organ transplantation: where is the benefit? Helge Bruns & Peter Schemmer
Received: 16 July 2013 / Accepted: 1 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Machine perfusion (MP) in solid organ transplantation has been a topic of variable importance for decades. At the dawn of organ transplantation, MP was one of the standard techniques for preservation; today’s gold standard for organ preservation for transplantation is cold storage (CS). The outcome after transplantation of solid organs has tremendously improved over the last five decades. MP has been continuously under investigation and may be an option for organ preservation in selected cases; however, there is only little evidence from clinical trials that can be used to advocate for MP as a routine organ preservation method. Methods This article reviews the current knowledge on MP in the field of solid organ transplantation with special focus on findings from clinical trials. Conclusion Especially in heart and lung transplantation, MP seems to be a promising tool to improve postoperative outcome, but a general evidence-based recommendation for or against an application of MP cannot be given due to the lack of the highest level of clinical evidence. Gold standards such as CS should not be left behind without good reason. Randomized clinical trials are desperately needed to further improve outcome and for better understanding of the underlying pathophysiological mechanisms.
Keywords Machine perfusion . Organ preservation . Organ transplantation
H. Bruns : P. Schemmer (*) Department of General and Transplant Surgery, Ruprecht Karls University, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany e-mail: [email protected]
Introduction In 1812, Le Gallois published his ideas on the principles of organ perfusion that can be transferred to machine perfusion (MP) [1]. However, the first concepts for mechanical devices for MP were published in 1935 with the Carrel–Lindbergh machine [2, 3]. While this device was not designed for organ preservation, this was the first publication on a modern oxygenated, normothermic pulsatile perfusion system. At the dawn of transplantation, MP was the gold standard for organ preservation since this technique seemed to be mostly physiological [4–6]. With the development of modern preservation solutions, cold storage (CS) became superior to MP regarding the outcome after transplantation [7]. The early 1980s can be considered to be the turning point in organ preservation towards CS. In 1982, a retrospective analysis for kidney preservation identified a better 1-year survival after CS with Collins solution in comparison to MP with the Belzer machine (53 vs 40 %, respectively) [8]. Today, the 1-year survival after CS with UW- or HTK-solution has further improved to 85 % [9, 10]. It is not fully clear whether MP with optimized devices can keep up with CS with optimized preservation and storage solutions [11–13]. While MP potentially optimizes grafts for transplantation, there are several methodical drawbacks. One of the major concerns in MP is that organ perfusion might lead to mechanical stress and thus could lead to further damage of the explanted organs. Increased pressure during perfusion has long been identified as a negative predictor of postoperative graft function, and MP can be used to monitor the flow characteristics and reevaluate organs before transplantation [14]. MP might be an option to select organs with good or bad function after transplantation [15–19]; it is not fully clear whether MP could really improve postoperative organ function in general. Currently, MP plays only a minor role in solid organ preservation but has continuously been investigated in both
Langenbecks Arch Surg Table 1 Clinical studies and trials reviewed in this publication Study Kidney Retrospective analysis: 1 year survival after MP with Belzer (n=565) and Waters (n=1,441) machine versus CS (Collins; n=282) [8] Retrospective analysis: CTS data; risk ratio for failure after CS (n=85.944) and MP (n=2.202) [12] Retrospective analysis: USRDS data; transplantation of EDC kidneys after MP (n= 1.114) and CS (n=4.726) [35] Randomized controlled trial: MP (n=336) versus CS (n=336), kidneys taken from the same donor and randomized; primary endpoint: DGF [20, 25] Randomized controlled trial: MP (n=45) vs CS (n=45), kidneys taken from same donor and randomized; primary endpoint: DGF [27] Liver Randomized controlled trial: transplantation after MP (Medtronic PBS device with Vasosol-R; n=20) vs CS (UW; n=20) [23] Heart Phase I trial: PROTECT: transplantation after MP using Organ Care System and warm oxygenated blood (n=20) [30] Phase I trial: PROCEED: transplantation after MP using Organ Care System and warm oxygenated blood (n=15) [31] Lung Clinical feasibility study and retrospective analysis; reevaluation of high risk lungs after 4 h MP (n=23), comparison to regular lung transplants (n=116), primary endpoint: primary graft dysfunction after 72 h [65]
Findings
Criticism
Decreased or equal to 1 year survival (40 and Retrospective analysis; poor survival in any of the groups; graft allocation criteria to groups unclear 48 versus 53 %) after MP with Belzer and Waters machine Increased risk ratio for failure after MP Retrospective analysis; kidneys used for MP may be marginal grafts Retrospective analysis; impact of preservation/ Decreased rate of DGF after MP, no perfusion solutions not analyzed difference in long-term graft survival after 2 and 3 years DGF defined as the subjective need for dialysis Decreased DGF (21 vs 27 %) after MP; increased graft survival after 1 (94 vs 90 %) and 3 years (91 vs 87 %) after MP No difference in DGF; no difference in renal DGF defined as the subjective need for dialysis function (eGFR) and patient and graft survival after 3 and 12 months Clinical relevance of decreased transaminases No significant difference (trends toward unclear; effect on transaminases might be related decreased PDF and hospital stay after MP); to different preservation/perfusion solutions used decreased postoperative transaminases in the trial after MP 30-day survival=100 %; acute rejection in 10 % and transient left ventricular dysfunction in 10 % 30-day survival=93 %
Phase I trial (no controls included in the study)
Phase I trial (no controls included in the study)
Phase I trial (no controls included in the study) 20 high-risk lungs were transplanted; no difference for primary graft dysfunction compared to regular lung transplants after CS
CTS collaborative transplant study, CS cold storage, MP machine perfusion, USRDS United States Renal Data System, DGF delayed graft function, EDC extended donor criteria, eGFR estimated glomerular filtration rate, PDF primary dysfunction, UW University of Wisconsin, PROTECT prospective multicenter European trial to evaluate the safety and performance of the Organ Care System for heart transplants, PROCEED prospective multicenter safety and effectiveness evaluation of the Organ Care System Device for heart use
preclinical studies and in clinical trials [20–31]. Multiple perfusion systems and options (normothermic, hypothermic, and oxygenated) exist [32]. Most of the existing clinical trials have different approaches and address different problems, thus making reasonable, valid meta-analyses of these data impossible (Table 1).
MP in solid organ transplantation Kidney MP might be best investigated in kidney transplantation; unfortunately, these findings cannot be generalized to the whole field of solid organ transplantation [33]. Multiple trials on MP for kidney preservation after organ donation from cardiac deceased (CDD) and brain dead donors (BDD) do exist [20, 25, 27]. The largest retrospective analyses failed to
identify beneficial effects of MP. However, all of these large retrospective analyses are either too old or biased. In the older analyses, abandoned techniques for MP or CS are compared with an unacceptable overall survival even in the best subgroups. Moreover, these analyses lack important data concerning kidney donors that might have an influence on graft survival and graft function [7, 8, 12]. Some publications exist that expect an improved function of organs with extended donor criteria (EDC) after MP [34]. Thus, it may be possible that a high proportion of kidney grafts that underwent MP in these retrospective analyses were marginal organs that have an already increased probability of delayed graft function (DGF) per se. If these organs are compared to cold-stored kidneys which might have had optimal qualification, differences found between groups might be related to differences between organs and not to different methods of organ
Langenbecks Arch Surg
preservation. To fully eliminate this bias, randomized controlled trials are needed, especially those that randomly assign one of a donors kidneys to MP and the other one to CS. Retrospective analyses systematically underestimate the effect of EDC in kidney transplantation and thus might be biased and cannot be reliable in this point [12]. Moreover, all different types of MP are mixed together in these analyses. In general, studies such as the retrospective analysis of the CTS data published by Opelz and Dohler in 2007 help to identify today’s standards and outcome after transplantation but should not be used as a scientific tool to advocate for or against MP [12]. Concerning kidney transplantation, two randomized clinical trials (RCTs) have been published, and findings from these studies might have an impact on MP in kidney transplantation or at least on the development of further study protocols [20, 25, 27]. In these clinical trials, one of the donor’s kidneys was randomized to MP and the other one to CS; the primary endpoint of both studies was DGF, which was defined as the need to undergo dialysis during the first 7 days after kidney transplantation. While Moers et al. identified a decreased rate of DGF after MP of 21 versus 27 %, and graft survival was 94 versus 90 % and 91 versus 87 % after MP and CS after 1 and 3 years, respectively, the difference was only available for those kidneys that showed DGF; concerning those kidneys without DGF, there was no difference in graft survival [20, 25]. The subgroup analysis of this study did not identify advantages in the EDC group or after donation after cardiocirculatory death. An analysis of perfusate from donor kidneys from this study with EDC that underwent MP was correlated to DGF after transplantation, and lipid peroxidation products were identified as a marker for DGF [17]. Thus, it can be proposed that MP could be a tool to identify those patients at risk for DGF. Nonetheless, it is not fully clarified yet weather DGF does necessarily lead to decreased graft survival. In a retrospective analysis that focused on costs of MP, Buchanan et al. identified a decreased rate of DGF after MP but no difference in long-term graft survival [35]. In their RCT, Watson et al. have investigated the effect of MP on kidney donation after cardiac death and failed to identify any significant difference or even a trend in terms of DGF, eGFR after 3 and 12 months, and patient and graft survival [27]. In both trials published by Moers et al. and Watson et al., DGF was defined as the need for dialysis in the first week after transplantation; this is necessarily one of the main criticisms concerning both trials: While this primary endpoint clearly is clinically relevant, related to kidney function, can easily be monitored, and is one of the most relevant cost-drivers after kidney transplantation, the need for dialysis is a subjective surrogate parameter for kidney function and depends on the physician in charge; standard laboratory values that serve as
triggers for postoperative dialysis may differ between transplant centers. This end point of both studies is therefore biased. Using the perfusion device employed in both clinical trials, flow parameters can be monitored easily during perfusion of the organs [16, 19], but it remains unclear whether perfusion itself causes mechanical shear stress and might lead to further damage of some of the organs. Thus, modern approaches also need to focus on optimization of flow characteristics during perfusion. Pressure, temperature, and pulsatile frequency have to be adapted to the specific organs’ needs [36]. On first sight, MP seems to be a rather costly procedure compared to CS, but at least in kidney MP, good evidence exists that potentially, MP could even decrease total costs after transplantation, if the right patients are selected. MP decreases DGF, and DGF leads to decreased long-term graft survival [37]. Depending on reimbursement and observational period, this effect might disappear after some years [35]. Given, that the need for dialysis in the early postoperative phase decreased and long-term outcome is improved, total costs are decreased as well. Taken together, MP might be an attractive option in kidney transplantation, but until now, there is no clear evidence which patients or organs benefit most from this kind of kidney preservation. Further clinical trials are needed to identify the optimal candidates for MP. It remains unclear whether the same effect can be achieved using optimized storage solutions, be it in kidney transplantation or concerning any other solid organ [38–43]. Liver While clinical trials exist, there is only one clinical trial with published data on MP in liver transplantation [23]. Guarrera et al. found a beneficial effect on laboratory parameters after MP in terms of decreased postoperative serum transaminases, but there was no difference in graft and patient survival. In their study, 20 liver transplantations after MP using a modified Medtronic PBS device and Vasosol-R solution were compared to 20 liver transplantations after CS with the University of Wisconsin (UW) [23]. There was a trend toward decreased primary dysfunction and a decreased hospital stay after MP, but the clinical relevance of this finding remains unclear. Guarrera et al. did not find any difference between both groups in terms of survival. Moreover, the observed effect might be related to the different preservation solutions used in both groups [24, 44]. Currently, clinical trials would not only need to focus on the direct effect of MP itself but also on MP as a tool to manipulate the perfused organ ex vivo. MP could be used for pharmacological conditioning of the perfused organ, to deliver substrates and/or to remove metabolic waste products [45]. There is some evidence from preclinical studies that MP might be a tool to improve marginal liver grafts [46, 47]. Especially oxygenated MP seems to have a beneficial effect in these organs [48–50]. In porcine models, MP was identified as a tool
Langenbecks Arch Surg
to increase the donor pool [51–53]. It is not clear yet whether these findings can be transferred to the clinical situation [46]. Especially concerning marginal livers from EDC, further optimization of perfusion and storage solutions, but also introduction of pharmacological conditioning, might be an option to improve outcome as much or even better than MP [38, 43, 54–56]. Pancreas No clinical trial has been performed on MP of the pancreas yet. In 1983, preclinical studies in a canine model demonstrated an increased failure rate after MP (30 and 40 % after 24 and 48 h after MP vs 0 and 0 % after 24 and 48 h after CS) [57]. It may be worth to perform further preclinical studies in this organ, but in the authors’ opinion, it is still too early for clinical trials on MP of the pancreas. MP has been discussed as a helpful tool to improve islet yield especially after prolonged ischemia time [58–60]. In a porcine model, insulin content in islets was increased after MP compared to CS [59, 60]. Leeser et al. found an increased islet viability after MP compared to CS in human pancreata (86 vs74 %, respectively) but a decreased total islet yield compared to pancreata with
