Machine perfusion is receiving increasing attention as a potentially better alternative to traditional static cold preservation of donor livers for transplantation.1-4 Compared to static cold storage (SCS), machine perfusion may provide better graft preservation and less reperfusion injury, resulting in better outcomes after transplantation.5-8 When performed at normal core body temperature, machine perfusion mimics physiological circumstances thus limiting cold ischemia-induced injury and allowing for functional assessment of a donor liver.6,9,10 Ex situ normothermic machine perfusion (NMP) can be performed either for the entire preservation period between procurement and implantation, or after a donor liver has undergone a period of traditional SCS for example, during transportation to the transplant centre (end-ischemic machine perfusion).9,11,12 The first type of ex situ NMP is referred to as “normothermic preservation machine perfusion”13 and the first clinical series of this novel type of organ preservation was recently reported.14 The latter type of NMP, referred to as “post-SCS” NMP, is performed at the transplant centre and provides the possibility of testing the function and viability of the organ before transplantation.13 Successful transplantation of donor livers initially declined for transplantation but underwent end-ischemic NMP and viability testing has been recently described.15-17
When ex situ NMP is performed after a period of cold ischemic preservation, a donor liver is subjected to rewarming and reoxygenation which may lead to reperfusion injury18; similar to what is generally observed in vivo during transplantation. The only difference however is that the perfusion fluid during NMP usually does not contain leucocytes and platelets. This therefore eliminates the detrimental effects of these blood components in the ischemia-reperfusion (I/R) injury cascade. Perfusion solutions based on a combination of red blood cells and fresh frozen plasma or a colloid solution are hence generally used by most research groups.
The liver is principally responsible for the synthesis of anti- and pro-coagulant proteins along with components of the fibrinolytic system which play an essential role in the regulation of coagulation and fibrinolysis. One of the features of reperfusion of a donor liver during transplantation is the activation of these 2 systems. During ex situ NMP, donor livers resume normal metabolic functions, including the production of bile, proteins, urea and the clearance of lactate.15,19 Perioperative haemostatic disorders may occur during liver transplantation which could result in an increased risk of the development of bleeding problems in the recipient.20,21 Although the changes in blood coagulation and fibrinolysis after graft reperfusion during liver transplantation have been described in great detail,22,23 little is known about activation of coagulation and fibrinolysis during end-ischemic ex situ NMP. To prevent fibrin formation during NMP as a result of activation of the coagulation cascade, most groups have been adding heparin to the perfusion fluid, however it remains unknown whether coagulation activation still occurs or whether end-ischemic NMP results in the activation of fibrinolysis. The aim of this study is to therefore determine whether activation of coagulation and/or fibrinolysis occurs during end-ischemic NMP of human donor livers and whether this could be used as a marker for graft I/R injury and/or function.
MATERIALS AND METHODS
Twelve human donor livers that were declined for transplantation by all 3 transplant centres in the Netherlands, as well as other centres within the Euro transplant region, were included in this study. Ten livers were obtained from controlled donation after circulatory death (DCD) donors and 2 livers from donation after brain death (DBD) donors. The retrieval and preparation procedure of these livers has been previously described.11,24 Livers were retrieved using a standard surgical technique of in situ cooling and flush-out with ice cold preservation fluid (University of Wisconsin or histidine-tryptophan-ketoglutarate solution). The surgical procedure was not started until after a 5 minute “no touch” period after declaration of cardiac arrest and circulatory death in case of a DCD donor. In case of DBD liver procurement, the administration of 25 000 units of heparin was given intravenously before cross clamping. The same dose of heparin was added to the preservation solution in case of DCD liver procurement. Livers were subsequently packed and stored on ice and transported to our center. In all 12 cases, permission for the use of these livers for research purposes was requested from and granted by the relatives. The study protocol was approved by the medical ethical committee of the University Medical Center Groningen and the Nederlandse Transplantatie Stichting, the national organisation responsible for the coordination and regulation of organ donation in the Netherlands.
End-Ischemic Ex Situ NMP
Upon arrival at our centre, livers were prepared and perfused at 37°C using a pressure controlled liver perfusion device (Organ Assist, Groningen, The Netherlands) which perfused through both the hepatic artery and portal vein, as described previously.11,24 Livers were perfused for 6 hours with a perfusion solution based on red blood cells and heparinized human plasma fortified with nutrients, calcium, trace elements, and antibiotics. Total volume of perfusion fluid was 2120 mL to which 20 000 IE of heparin were added. Samples of the perfusion fluid were collected before the liver was connected to the perfusion machine (baseline) and at 30-minute intervals during the 6 hours of NMP. All samples were centrifuged (2700 rpm for 5 minutes at 4°C) to remove erythrocytes and plasma was collected, snap-frozen, and stored at −80°C until analysis.
After NMP and subsequent evaluation of the liver function, the livers were divided into 2 groups depending on their functionality. The livers with a high cumulative bile production (≥30 g during 6 hours of NMP) were considered to be “good functioning” livers whereas those with a low bile output (<30 g during 6 hours of NMP) were considered to be “poor functioning” livers.11,24 Levels of alanine aminotransferase (ALT) and lactate in the perfusate (established markers of hepatocellular I/R injury) were measured at regular intervals using a standard biochemical method.
Assessment of Coagulation and Fibrinolysis Activation
Activation of coagulation leads to the conversion of the zymogen prothrombin to the serine protease thrombin, which releases prothrombin fragment 1 + 2 as an activation peptide. To determine whether activation of coagulation occurred during NMP, plasma levels of prothrombin fragment 1 and 2 (F1 + 2) were determined using the Enzygnost F1 + 2 ELISA kit (Siemens Healthcare Diagnostics, The Hague, The Netherlands).
Activation of fibrinolysis was assessed by measuring the concentrations of tissue plasminogen activator (tPA) antigen, plasminogen activator inhibitor-1 (PAI-1) antigen and plasmin-antiplasmin (PAP) complexes, using an IMUBIND tPA ELISA kit, (Sekisui (USA) via Werfen, Breda, Netherlands), Quantikine Human Serpin E1/PAI-1 ELISA kit (Duoset DY1786 R&D systems, Abingdom, UK) and TECHNOZYM PAP complex ELISA kit (Technoclone, Vienna, Austria), respectively. All ELISAs were performed according to the manufacturers’ instructions. In addition, concentration of D-dimers in the perfusion fluid was measured using an automated latex enhanced immunoassay (D-dimer HS 500, ACL 300 TOP, Instrumentation Laboratory, Breda, The Netherlands). D-dimer is a fibrin degradation product, a small protein fragment that is released after crossed linked fibrin is degraded by fibrinolysis.
Biopsies of the liver parenchyma were taken before machine perfusion and immediately stored in formalin. Paraffin-embedded slides of the liver biopsies were stained using Maurits, Scarlet and Blue (MSB) stain, a trichome staining technique particularly used for the selective demonstration of fibrin.25 A strong bright red stain was representative of fibrin deposition. The slides were subsequently analysed by light microscopy (at a magnification of ×40, ×100, and ×200) to determine whether microthrombi or fibrin depositions were present in the liver microcirculation. All histological analyses were supervised by an experienced hepatopathologist (A.S.H.G.).
Statistical analyses were performed using SPSS version 20 for Windows (SPSS Inc., Chicago, IL, USA). Continuous variables were presented as medians and interquartile range (IQR)] and categorical variables were presented as total numbers and percentages. The Mann-Whitney U test was used for comparison of continuous variables between groups and the Wilcoxon signed rank test for comparison within a group. Categorical variables were compared using the Fischer exact test. Correlations between continuous variables were determined by the Pearson correlation coefficient or by linear regression analysis, as appropriate. A p-value less than 0.05 was considered statistically significant.
Characteristics of the twelve donor livers that underwent end-ischemic NMP are presented in Table 1. We performed NMP for 6 hours because based on the experience we have gathered whilst conducting machine perfusion (studies performed by colleagues' op den Dries et al and Sutton et al from our research group11,24), we concluded that 6 hours were sufficient to draw credible conclusions on graft viability and function. During the 6 hours of NMP, as Table 2 illustrates, 6 livers produced 30 g or more of bile,11 which contained significantly higher bilirubin (measure of bile quality). These livers also exhibited higher lactate clearance rates and were thus classified as “good functioning” livers. The remaining 6 livers produced less than 30 g bile, with significantly less bilirubin, and lower lactate clearance, were grouped as “poor functioning” livers.
No Activation of Coagulation During End-Ischemic NMP
An overview of baseline values and changes in parameters of coagulation and fibrinolysis activation during NMP of all twelve livers is presented in Table 3. At baseline, before the liver was connected to the perfusion device, very small amounts of prothrombin fragment F1 + 2 (good functioning livers - median 278 pmol/L; IQR, 120-556 and poor functioning livers - median 127 pmol/L; IQR, 127-648) were detected in the perfusion fluid. During 6 hours of NMP, F1 + 2 concentrations remained stable and there was no significant difference in the delta increase or decrease of F1 + 2 during 6 hours of NMP between the groups of livers with good or poor function (P = 0.86) (Figure 1A).
Activation of Fibrinolysis During End-Ischemic NMP
The concentration of D-dimer and PAP complexes in the perfusion fluid of both good and poor functioning livers increased more than 20-fold soon after the start of NMP (Table 3). Concentrations of tPA remained relatively stable during NMP. However, when comparing the change in tPA concentration during NMP in the group of livers with good or poor function, a significant difference was noted. While tPA in the perfusion fluid decreased during the 6 hours of NMP in the good functioning livers, an increase was noted in livers with poor function (Figure 1B). Similarly, the increase in D-dimer and PAP complexes during NMP was significantly higher in the group with poor functioning livers, compared to good functioning livers (Figures 1C and D). PAI-1 concentrations were low for both groups during the first hours of NMP, but a sharp increase was noted during the second part of the 6 hours of NMP (Table 3). Although the increase in PAI-1 was more pronounced in livers that displayed good function during NMP, compared to poor functioning livers, this difference did not reach statistical significance (data not shown).
Correlation between Activation of Fibrinolysis and I/R Injury
We next examined whether activation of fibrinolysis during NMP correlated with the degree of I/R injury. ALT levels in the perfusion fluid were used as an established marker of hepatocellular I/R injury. Significant correlations were observed between tPA and ALT concentrations measured at all time points, and between tPA and D-dimer (Figures 2A and B). In accordance with this, a significant correlation was observed between ALT and D-dimer concentrations (Figure 3A). Moreover, livers that displayed good function during NMP were found to have lower levels of ALT and D-dimer in the perfusion fluid, compared to poorly functioning livers. In fact, very high D-dimer levels were only found in the group of poorly functioning livers (Figure 3A). Altogether, these data suggest that ex situ end-ischemic NMP is associated with activation of fibrinolysis, the degree of which is linked to the severity of I/R injury.
As could be expected from the lack of change in prothrombin F1 + 2 concentrations, there was no correlation between this marker of coagulation activation and ALT levels in the perfusion fluid (Figure 3B).
Histological Detection of Fibrin in Donor Livers
Since we found no evidence of coagulation activation during end-ischemic NMP, yet a significant release of D-dimers into the perfusate soon after the start of NMP, it is likely that the D-dimers released were derived from lysis of (micro)clots or fibrin depositions already present in the liver grafts before ex situ NMP. To investigate the presence of (micro)thrombi or fibrin depositions in the liver microcirculation, the liver biopsies obtained before NMP underwent MSB trichome staining. A total of 292 portal venous structures and 213 central venous structures were examined in biopsy slides of the twelve liver grafts. In 4 (1.35%) of these 292 portal tracts and in none of the 213 central venous tracts, MSB staining revealed signs of intravascular fibrin. In the 4 portal venous tracts with positive fibrin staining, only small amounts of free floating fibrin were detected, without evidence of vessel-occluding microthrombi (Figures 4A and B). There was no significant difference in the presence of fibrin in portal venous tracts of good or poor functioning livers.
Ex vivo NMP has been proposed as an alternative preservation method to traditional SCS as well as a novel method to resuscitate extended criteria donors livers and to assess their function and viability before transplantation.11,15,18,24 When used as a selection tool to identify donor livers that are transplantable after initially being declined for transplantation based on high donor risk factors, NMP could potentially contribute to an increase of the number of donor livers for transplantation.24,26
If ex situ NMP is performed after a period of traditional cold ischemic preservation, the donor liver is subject to re-warming and re-oxygenation which potentially leads to I/R injury. Proteolysis and activation of fibrinolysis have been identified as key events in clinical reperfusion of cold stored donor livers.20,23 The main finding of this study is that end-ischemic ex vivo NMP of donor livers results in activation of fibrinolysis, but not of coagulation. Markers of fibrinolysis activation correlate significantly with markers of hepatocellular I/R injury during NMP. High concentration of D-dimers in the perfusion fluid soon after the start of NMP can be considered a marker of severe I/R injury and a predictor of poor liver graft function.
D-dimer is a fibrin degradation product, a small protein fragment that is released during fibrinolysis after crossed linked fibrin is degraded. It is named so because it contains 2 cross linked D fragments of the fibrin protein.27 D-dimers are a specific marker for degradation of cross linked fibrin. The sharp increase in D-dimers in the perfusion fluid during end-ischemic NMP in the absence of an increase in prothombin fragment F1 + 2 suggests that this D-dimer increase is a result of degradation of preexisting rather than newly formed fibrin in the donor livers. Prothrombin fragment F1 + 2 is a reliable indicator of coagulation activation. Activation of coagulation leads to the conversion of the zymogen prothrombin to the serine protease thrombin, which releases prothrombin fragment F1 + 2 as an activation peptide. During NMP, prothrombin fragment F1 + 2 remained stable and levels did not change significantly compared with baseline values. Although the increase in D-dimers levels after the start of NMP is most likely explained by the breakdown of preexisting fibrin in the livers, histological evidence of fibrin in the hepatic microvasculature was found only sporadically. After microscopic evaluation of 292 portal and 213 central venous branches in parenchymal biopsies, traces of fibrin were found in only 4 of the 292 (1.4%) portal venous branches. Moreover, no signs of vessel-occluding microthrombi were seen. This observation is in accordance with a recent clinical study on histological analyses of bile duct biopsies of 128 human donor livers before transplantation in which the incidence of microthrombi in the peribiliary vascular plexus after SCS was 2.7%.8,28 A calculation of the estimated amount of fibrin that must have been present in livers, based on molecular weight of D-dimers and fibrin is approximately 14 mg per liver (equal to the size of a pinhead). Based on an average liver weight of 2000 g, this would amount to an approximate value of 0.0007% of the total liver weight. This information adds an interesting new perspective to the discussion about whether thrombolytic therapy of (DCD) liver grafts before transplantation is potentially beneficial. Thrombolytic therapy of DCD livers grafts with plasminogen activators such as recombinant tPA or streptokinase has been proposed as a method to reduce the incidence of ischemic cholangiopathy after DCD liver transplantation.29,30 On the other hand, several studies have indicated that hypoxia due to circulatory arrest (as occurs in a dying person) is associated with a pronounced stimulation of the fibrinolytic system because of the release of endogenous plasminogen activators.31,32 In accordance with this, it was recently reported that evidence of clinically relevant endogenous hyper-fibrinolysis can be found in DCD donors (Maastricht type II) which therefore argued against the need for additional thrombolytic treatment of DCD donors or their livers.33 Our study demonstrated that end-ischemic NMP also results in activation of endogenous fibrinolysis (with no particular difference noted in regard to whether or not the donor liver is derived from a brain or circulatory death donor). This may be a clinically relevant aspect of NMP that contributes to the improved preservation of a donor liver during this type of machine perfusion. Activation of fibrinolysis was most pronounced in livers of poor quality as illustrated by a high release of ALT and low bile production during NMP. In fact, high levels of D-dimers in the perfusion fluid correlated with liver function and ALT levels, reflecting I/R injury.
The observation that end-ischemic NMP of suboptimal quality donor livers is associated with activation of fibrinolysis is very much in line with data from clinical studies that have identified proteolysis and fibrinolysis as key components of graft reperfusion and I/R injury.21,23,34 Hyper-fibrinolysis has been proposed as an explanation for the higher risk of blood loss in recipients after reperfusion of extended criteria donor livers.35,36 However, clinical studies determining whether end-ischemic NMP of donor livers of suboptimal quality before transplantation leads to a reduction hyper-fibrinolysis in recipients after transplantation are still lacking.
In contrast to the activation of fibrinolysis, we found no evidence of activation of coagulation during NMP. When the perfusion fluid used for NMP is based on human plasma (as was the case in this study) or whole blood, the addition of an anticoagulant to the perfusion fluid is necessary. In fact, even when plasma is replaced by a (synthetic) colloid solution, which does not contain coagulation proteins, investigators generally still add heparin to the perfusion fluid.12,15,37 Given the restoration of metabolic function of the liver during ex situ NMP, coagulation activation and subsequent fibrin formation may occur. However, the half-life of most proteins involved in the coagulation cascades is relatively long and large amounts of de novo production during a few hours of NMP are not to be expected. Nevertheless, we would certainly advise to include heparin or another potent anticoagulant drug to the perfusion fluid during liver machine perfusion. Additionally an important detail to take into consideration in regard to maintaining a favourable haemostatic balance of these livers during NMP is the donor type. As mentioned previously, depending on the donor type (DBD or DCD), a hypercoagulable or hyperfibrinolytic state may be seen. This difference, however, was not observed in this study as data from DBD livers were comparable to the DCD livers. This may have been attributed to the small sample size in which a conclusive comparison between the DBD and DCD livers could not be made.
In conclusion, end-ischemic ex situ NMP results in activation of fibrinolysis, but not of coagulation. Markers of fibrinolysis activation correlate significantly with markers of I/R injury and high concentrations of D-dimer early after start of NMP can be considered a marker of severe I/R injury and a predictor of poor liver graft function.
The authors are grateful to all transplant coordinators in the Netherlands for identifying potentially discarded livers and for obtaining informed consent from the relatives of the donors.
1. Pavel MC, Fondevila Campo C, Calatayud Mizrahi D, et al. Normothermic perfusion machine in liver transplant with cardiac death donor grafts. Cir Esp
2. Brockmann J, Reddy S, Coussios C, et al. Normothermic perfusion: a new paradigm for organ preservation. Ann Surg
3. Reddy S, Greenwood J, Maniakin N, et al. Non–heart-beating donor porcine livers: the adverse effect of cooling. Liver Transpl
4. Vogel T, Brockmann JG, Friend PJ. Ex-vivo normothermic liver perfusion: an update. Curr Opin Organ Transplant
5. Xu H, Berendsen T, Kim K, et al. Excorporeal normothermic machine perfusion resuscitates pig DCD livers with extended warm ischemia. J Surg Res
6. Banan B, Watson R, Xu M, et al. Development of a normothermic extracorporeal liver perfusion system toward improving viability and function of human extended criteria donor livers. Liver Transpl
7. Fondevila C, Hessheimer AJ, Maathuis MH, et al. Superior preservation of DCD livers with continuous normothermic perfusion. Ann Surg
8. op den Dries S, Westerkamp AC, Karimian N, et al. Injury to peribiliary glands and vascular plexus before liver transplantation predicts formation of non-anastomotic biliary strictures. J Hepatol
9. Imber CJ, St Peter SD, Lopez de Cenarruzabeitia I, et al. Advantages of normothermic perfusion over cold storage in liver preservation. Transplantation
10. Henry SD, Guarrera JV. Protective effects of hypothermic ex vivo perfusion on ischemia/reperfusion injury and transplant outcomes. Transplant Rev (Orlando)
11. op den Dries S, Karimian N, Sutton ME, et al. Ex vivo normothermic machine perfusion and viability testing of discarded human donor livers. Am J Transplant
12. St Peter SD, Imber CJ, Lopez I, et al. Extended preservation of non-heart-beating donor livers with normothermic machine perfusion. Br J Surg
13. Karangwa SA, Dutkowski P, Fontes P, et al. Machine perfusion of donor livers for transplantation: a proposal for standardized nomenclature and reporting guidelines. Am J Transplant
14. Ravikumar R, Jassem W, Mergental H, et al. Liver transplantation after ex vivo normothermic machine preservation: a phase 1 (first-in-man) clinical trial. Am J Transplant
15. Mergental H, Perera MT, Laing RW, et al. Transplantation of declined liver allografts following normothermic ex-situ evaluation. Am J Transplant
16. Watson CJ, Kosmoliaptsis V, Randle LV, et al. Preimplant normothermic liver perfusion of a suboptimal liver donated after circulatory death. Am J Transplant
17. Perera T, Mergental H, Stephenson B, et al. First human liver transplantation using a marginal allograft resuscitated by normothermic machine perfusion. Liver Transpl
18. Ravikumar R, Leuvenink H, Friend PJ. Normothermic liver preservation: a new paradigm? Transpl Int
19. Boehnert MU, Yeung JC, Bazerbachi F, et al. Normothermic acellular ex vivo liver perfusion reduces liver and bile duct injury of pig livers retrieved after cardiac death. Am J Transplant
20. de Boer MT, Molenaar IQ, Hendriks HG, et al. Minimizing blood loss in liver transplantation: progress through research and evolution of techniques. Dig Surg
21. Dzik WH, Arkin CF, Jenkins RL, et al. Fibrinolysis during liver transplantation in humans: role of tissue-type plasminogen activator. Blood
22. Porte RJ. Coagulation and fibrinolysis in orthotopic liver transplantation: current views and insights. Semin Thromb Hemost
23. Porte RJ, Bontempo FA, Knot EA, et al. Systemic effects of tissue plasminogen activator-associated fibrinolysis and its relation to thrombin generation in orthotopic liver transplantation. Transplantation
24. Sutton ME, op den Dries S, Karimian N, et al. Criteria for viability assessment of discarded human donor livers during ex vivo normothermic machine perfusion. PLoS One
25. Fisseler-Eckhoff A, Muller KM. Lendrum (-MSB) staining for fibrin identification in sealed skin grafts. Pathol Res Pract
26. Broomhead RH, Patel S, Fernando B, et al. Resource implications of expanding the use of donation after circulatory determination of death in liver transplantation. Liver Transpl
27. Adam SS, Key NS, Greenberg CS. D-dimer antigen: current concepts and future prospects. Blood
28. Burlage LC, Karangwa SA, Lisman T, et al. Thrombolytic protocol minimizes ischemic-type biliary complications in liver transplantation from donation after circulatory death donors. Liver Transpl
29. Hashimoto K, Eghtesad B, Gunasekaran G, et al. Use of tissue plasminogen activator in liver transplantation from donation after cardiac death donors. Am J Transplant
30. Seal JB, Bohorquez H, Reichman T, et al. Thrombolytic protocol minimizes ischemic-type biliary complications in liver transplantation from donation after circulatory death donors. Liver Transpl
31. Gando S, Sawamura A, Hayakawa M. Trauma, shock, and disseminated intravascular coagulation: lessons from the classical literature. Ann Surg
32. Porte RJ, Clavien PA. Preflush with plasminogen activator in non-heart-beating donors: is it worth it? Transplantation
33. Vendrell M, Hessheimer AJ, Ruiz A, et al. Coagulation profiles of unexpected DCDD donors do not indicate a role for exogenous fibrinolysis. Am J Transplant
34. Calmus Y, Cynober L, Dousset B, et al. Evidence for the detrimental role of proteolysis during liver preservation in humans. Gastroenterology
35. Hartmann M, Szalai C, Saner FH. Hemostasis in liver transplantation: pathophysiology, monitoring, and treatment. World J Gastroenterol
36. Kang YG, Martin DJ, Marquez J, et al. Intraoperative changes in blood coagulation and thrombelastographic monitoring in liver transplantation. Anesth Analg
37. Reddy SP, Bhattacharjya S, Maniakin N, et al. Preservation of porcine non-heart-beating donor livers by sequential cold storage and warm perfusion. Transplantation