Original Clinical ScienceçLiver
Ischemic Postconditioning of the Liver Graft in Adult Liver Transplantation Luana Ricca, MD,1,2,3 Antoinette Lemoine, PharmD, PhD,3,4 François Cauchy, MD,5 Jocelyne Hamelin, PharmD, PhD,3,4 Mylène Sebagh, MD, PhD,2,6 Davide Degli Esposti, PhD,4 Chady Salloum, MD,7 Eric Vibert, MD, PhD,5,6 Genoveffa Balducci, MD,1 and Daniel Azoulay, MD, PhD7,8
Background. Ischemia-reperfusion (I/R) injury is the main cause of graft failure in liver transplantation (LT). Ischemic
postconditioning (IPo) has shown to be beneficial against I/R injury. Our objective was to compare the results of LT with or without IPo. Methods. One hundred patients undergoing LT alternatively received IPo or not. At the time of arterial reperfusion, IPo consisted of three 1-minute arterial occlusions, interspersed with 1-minute reperfusion pauses. The primary endpoint was postoperative aspartate aminotransferase (AST) peak value; early graft dysfunction and histological I/R injury were secondary endpoints. Results. Median postoperative AST peak values was similar in both groups (426 vs 463 IU/L, P = 0.21); no difference was found in other postoperative liver function tests. In the IPo group, fewer grafts presented severe histological I/R injury (12% vs 28%; P = 0.029). Ischemic postconditioning did not induce changes in cellular apoptosis but triggered autophagy in periportal areas. Independent predictors of severe I/R injury were IPo (odds ratio, 0.20; P = 0.008) and arterial warm ischemia duration (odds ratio, 1.05; P = 0.008). Early graft dysfunction rate was similar in both groups (20% versus 26%, P = 0.47) and was associated with severe histological I/R injury and longer cold ischemia. Morbidity, mortality, and 1-year graft and patient survival were similar in both groups. Conclusions. Ischemic postconditioning did not influence postoperative AST peak values or other liver function tests. However, our results showed a better tolerance to I/R injury on histological findings of grafts receiving IPo. Future studies are necessary to optimize the IPo protocol in LT, to clarify its clinical impact, and to deepen the molecular understanding. (Transplantation 2015;99: 1633–1643)
L
iver transplantation (LT) involves the devascularization of the liver graft with prolonged deprivation of tissue oxygen, which converts cellular metabolism to anaerobic pathways. Reperfusion and the subsequent restoration of oxygen delivery can lead to graft injury, known as ischemiareperfusion (I/R) injury, which is the principal cause of liver graft failure.1-3 Several strategies have been developed to limit this injury, including discarding grafts with severe steatosis,4,5 minimizing ischemia time,6 matching graft quality to the recipient status,7 providing nitric oxide inhalation at the revascularization of
the liver graft.8 However, organ shortages constituted a major incentive to use donor organs under extended criteria, such as organs from elderly or cardiac death donors or from those affected by significant steatosis or prolonged cold ischemia. These grafts are inevitably prone to a higher rate of I/R injury.9 Thus, the development of other strategies to reduce I/R injury is of critical importance. Protective measures against I/R in transplantation situations are called preconditioning, conditioning, and postconditioning, indicating their application before inflow occlusion in the donor, during the ischemia or after graft revascularization in the recipient, respectively. A number of clinical studies have demonstrated reductions in postoperative serum aminotransferase levels (hepatocyte
Received 18 July 2014. Revision requested 29 September 2014. Accepted 5 January 2015. 1
Dipartimento di Scienze Medico-Chirurgiche e Medicina Traslazionale, Università degli Studi di Roma La Sapienza, Rome, Italy. 2
AP-HP Hôpital Paul Brousse-Centre Hépatobiliaire, Villejuif, France.
3
INSERM UMR-S 1004, Université Paris Sud, Villejuif, France.
4
AP-HP Hôpital Paul Brousse, Biochimie et Oncogénétique, Villejuif, France.
5
AP-HP Hôpital Paul Brousse, Centre Hépatobiliaire, Villejuif, France.
6
INSERM UMR-S 785, Université Paris Sud, Villejuif, France.
7 AP-HP Hôpital Henri Mondor, Service de Chirurgie Digestive Hépatobiliaire, Creteil, France. 8
INSERM U 955, Université Paris Est, Créteil, France.
The work of Luana Ricca, PhD student of the Department of Medical and surgical Sciences and Translational Medicine of La Sapienza University of Rome, was supported by a grant of La Sapienza University of Rome, Italy.
■
August 2015
■
Correspondence: Daniel Azoulay, Service de Chirurgie Digestive Hépatobiliaire 51 Avenue du Maréchal de Lattre de Tassigny, 94010 Creteil, France. (daniel. [email protected]). Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal's Web site (www.transplantjournal.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0041-1337/15/9908-1633 DOI: 10.1097/TP.0000000000000685
The authors declare no conflicts of interest.
Transplantation
L.R. participated in the performance of the research, in data analysis and in the writing of the paper. A.L. participated in the performance of the research and in the writing of the paper. F.C. participated in data analysis and in the writing of the paper. J.H. participated in the performance of the research. M.S. and D.D.E. contributed analytic tools. C.S. and E.V. participated in the performance of the research. G.B. contributed analytic tools and participated in data analysis. D.A. participated in research design, in the performance of the research and in the writing of the paper.
Volume 99
■
Number 8
www.transplantjournal.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
1633
1634
■
August 2015
■
Volume 99
■
Number 8
www.transplantjournal.com
death markers) and inflammatory infiltrates after ischemic preconditioning of the graft in the donor.10-15 Nevertheless, most of these studies did not show any clinical benefit in terms of patient or graft survival.9 Hence, the routine use of ischemic preconditioning in LT is not currently in use. Graft postconditioning in the recipient seems to be a more appealing strategy because it can be applied selectively in patients at greater risks for I/R injury due to donor-, procurement-, and/or recipient-related factors. Postconditioning focuses on the early events of reperfusion as the key therapeutic window. After prolonged ischemia, ischemic postconditioning (IPo) consists in repetitive very short cycles of ischemia interspersed with intermittent reperfusion applied at the onset of revascularization. To date, the beneficial effects of IPo has been proven in several preclinical studies in different organs16 and confirmed in human clinical studies on heart I/R.17,18 In the context of LT, the protective role of IPo has been documented in animal models.19-24 Different protocols for short repetitive periods of arterial occlusion and reperfusion after prolonged ischemia have been tested (3 to 6 cycles of 15 to 60 seconds), but so far, no standard procedure has been established.17-24 The aim of the present study was to evaluate the role of IPo in human LT and its effects on I/R injury and liver graft function. In an attempt to standardize the procedure in the setting of LT, we decided to apply IPo only on arterial reperfusion on the basis of previous clinical experience.17,18 Postoperative serum transaminase levels, graft function, I/R injury at pathological examination, mortality, morbidity, and patient and graft survival were compared between patients receiving IPo or not.
groups to be clinically relevant. Assuming a standard deviation of 550 IU/L for nontreated patients and 500 IU/L for patients receiving postconditioning, with a statistical power of 80% and an α error of 5% (2-sided), it was necessary to include at least 98 patients. All patients older than 18 years receiving a cadaveric donor graft were eligible. The exclusion criteria were the following: urgent LT, serum transaminase levels higher than 1000 IU/l before LT and graft cold ischemia longer than 10 hours. All patients meeting these criteria and transplanted by the surgeons participating in this study (D.A., L.R., C.S., and E.V.) were included prospectively. To minimize selection bias, IPo was performed in an alternate fashion, based on the protocol applied for the first clinical application of ischemic preconditioning in human.25 During pretransplant evaluation, all patients gave their written consent to participate in research protocol and for the experimental use of blood samples and graft biopsy specimens. The technique of liver procurement and transplantation are detailed in Methods (SDC, http://links.lww.com/TP/B139). Biologists, intensive care unit clinicians, and pathologists were not informed whether the graft had been subjected to IPo or not.
Transplantation
MATERIALS AND METHODS Study Design and Population
This study was designed as an observational study to assess the role of IPo against I/R injury of the liver graft. As the optimal protocol is still to define, IPo was applied on arterial reperfusion on the basis of available literature. Ischemic postconditioning consisted of three 1-minute episodes of arterial occlusion interspersed with three 1-minute pauses of inflow immediately after arterial reperfusion (Figure 1). The primary endpoint was the peak level of aspartate aminotransferase (AST) during the first 10 postoperative days because this variable is the measure of liver damage usually adopted in the literature.10-12,15,22,24 Secondary endpoints were the occurrence of early graft dysfunction (EGD) and the severity of I/R injury as determined from histological findings. Mortality, morbidity, and 1-year patient and graft survival were also studied. The sample size was calculated considering a difference of 300 IU/L in postoperative AST peak levels between the
FIGURE 1. IPo protocol.
Pathology
Both baseline (before ischemia) and reperfusion biopsies were fixed in formaldehyde, paraffin-embedded, and stained with standard hematoxylin-eosin stain. All of the biopsies were reviewed by the same (blinded) pathologists (M.S.) to ensure a homogeneous interpretation. Specimens were considered adequate if more than 3 adjacent lobules were available. Grafts were evaluated, on both baseline and reperfusion biopsy, for the presence of steatosis and necrosis; steatosis was distinguished as microvesicular or macrovesicular and classified as absent, mild (
Ischemic Postconditioning of the Liver Graft in Adult Liver Transplantation Luana Ricca, MD,1,2,3 Antoinette Lemoine, PharmD, PhD,3,4 François Cauchy, MD,5 Jocelyne Hamelin, PharmD, PhD,3,4 Mylène Sebagh, MD, PhD,2,6 Davide Degli Esposti, PhD,4 Chady Salloum, MD,7 Eric Vibert, MD, PhD,5,6 Genoveffa Balducci, MD,1 and Daniel Azoulay, MD, PhD7,8
Background. Ischemia-reperfusion (I/R) injury is the main cause of graft failure in liver transplantation (LT). Ischemic
postconditioning (IPo) has shown to be beneficial against I/R injury. Our objective was to compare the results of LT with or without IPo. Methods. One hundred patients undergoing LT alternatively received IPo or not. At the time of arterial reperfusion, IPo consisted of three 1-minute arterial occlusions, interspersed with 1-minute reperfusion pauses. The primary endpoint was postoperative aspartate aminotransferase (AST) peak value; early graft dysfunction and histological I/R injury were secondary endpoints. Results. Median postoperative AST peak values was similar in both groups (426 vs 463 IU/L, P = 0.21); no difference was found in other postoperative liver function tests. In the IPo group, fewer grafts presented severe histological I/R injury (12% vs 28%; P = 0.029). Ischemic postconditioning did not induce changes in cellular apoptosis but triggered autophagy in periportal areas. Independent predictors of severe I/R injury were IPo (odds ratio, 0.20; P = 0.008) and arterial warm ischemia duration (odds ratio, 1.05; P = 0.008). Early graft dysfunction rate was similar in both groups (20% versus 26%, P = 0.47) and was associated with severe histological I/R injury and longer cold ischemia. Morbidity, mortality, and 1-year graft and patient survival were similar in both groups. Conclusions. Ischemic postconditioning did not influence postoperative AST peak values or other liver function tests. However, our results showed a better tolerance to I/R injury on histological findings of grafts receiving IPo. Future studies are necessary to optimize the IPo protocol in LT, to clarify its clinical impact, and to deepen the molecular understanding. (Transplantation 2015;99: 1633–1643)
L
iver transplantation (LT) involves the devascularization of the liver graft with prolonged deprivation of tissue oxygen, which converts cellular metabolism to anaerobic pathways. Reperfusion and the subsequent restoration of oxygen delivery can lead to graft injury, known as ischemiareperfusion (I/R) injury, which is the principal cause of liver graft failure.1-3 Several strategies have been developed to limit this injury, including discarding grafts with severe steatosis,4,5 minimizing ischemia time,6 matching graft quality to the recipient status,7 providing nitric oxide inhalation at the revascularization of
the liver graft.8 However, organ shortages constituted a major incentive to use donor organs under extended criteria, such as organs from elderly or cardiac death donors or from those affected by significant steatosis or prolonged cold ischemia. These grafts are inevitably prone to a higher rate of I/R injury.9 Thus, the development of other strategies to reduce I/R injury is of critical importance. Protective measures against I/R in transplantation situations are called preconditioning, conditioning, and postconditioning, indicating their application before inflow occlusion in the donor, during the ischemia or after graft revascularization in the recipient, respectively. A number of clinical studies have demonstrated reductions in postoperative serum aminotransferase levels (hepatocyte
Received 18 July 2014. Revision requested 29 September 2014. Accepted 5 January 2015. 1
Dipartimento di Scienze Medico-Chirurgiche e Medicina Traslazionale, Università degli Studi di Roma La Sapienza, Rome, Italy. 2
AP-HP Hôpital Paul Brousse-Centre Hépatobiliaire, Villejuif, France.
3
INSERM UMR-S 1004, Université Paris Sud, Villejuif, France.
4
AP-HP Hôpital Paul Brousse, Biochimie et Oncogénétique, Villejuif, France.
5
AP-HP Hôpital Paul Brousse, Centre Hépatobiliaire, Villejuif, France.
6
INSERM UMR-S 785, Université Paris Sud, Villejuif, France.
7 AP-HP Hôpital Henri Mondor, Service de Chirurgie Digestive Hépatobiliaire, Creteil, France. 8
INSERM U 955, Université Paris Est, Créteil, France.
The work of Luana Ricca, PhD student of the Department of Medical and surgical Sciences and Translational Medicine of La Sapienza University of Rome, was supported by a grant of La Sapienza University of Rome, Italy.
■
August 2015
■
Correspondence: Daniel Azoulay, Service de Chirurgie Digestive Hépatobiliaire 51 Avenue du Maréchal de Lattre de Tassigny, 94010 Creteil, France. (daniel. [email protected]). Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal's Web site (www.transplantjournal.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0041-1337/15/9908-1633 DOI: 10.1097/TP.0000000000000685
The authors declare no conflicts of interest.
Transplantation
L.R. participated in the performance of the research, in data analysis and in the writing of the paper. A.L. participated in the performance of the research and in the writing of the paper. F.C. participated in data analysis and in the writing of the paper. J.H. participated in the performance of the research. M.S. and D.D.E. contributed analytic tools. C.S. and E.V. participated in the performance of the research. G.B. contributed analytic tools and participated in data analysis. D.A. participated in research design, in the performance of the research and in the writing of the paper.
Volume 99
■
Number 8
www.transplantjournal.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
1633
1634
■
August 2015
■
Volume 99
■
Number 8
www.transplantjournal.com
death markers) and inflammatory infiltrates after ischemic preconditioning of the graft in the donor.10-15 Nevertheless, most of these studies did not show any clinical benefit in terms of patient or graft survival.9 Hence, the routine use of ischemic preconditioning in LT is not currently in use. Graft postconditioning in the recipient seems to be a more appealing strategy because it can be applied selectively in patients at greater risks for I/R injury due to donor-, procurement-, and/or recipient-related factors. Postconditioning focuses on the early events of reperfusion as the key therapeutic window. After prolonged ischemia, ischemic postconditioning (IPo) consists in repetitive very short cycles of ischemia interspersed with intermittent reperfusion applied at the onset of revascularization. To date, the beneficial effects of IPo has been proven in several preclinical studies in different organs16 and confirmed in human clinical studies on heart I/R.17,18 In the context of LT, the protective role of IPo has been documented in animal models.19-24 Different protocols for short repetitive periods of arterial occlusion and reperfusion after prolonged ischemia have been tested (3 to 6 cycles of 15 to 60 seconds), but so far, no standard procedure has been established.17-24 The aim of the present study was to evaluate the role of IPo in human LT and its effects on I/R injury and liver graft function. In an attempt to standardize the procedure in the setting of LT, we decided to apply IPo only on arterial reperfusion on the basis of previous clinical experience.17,18 Postoperative serum transaminase levels, graft function, I/R injury at pathological examination, mortality, morbidity, and patient and graft survival were compared between patients receiving IPo or not.
groups to be clinically relevant. Assuming a standard deviation of 550 IU/L for nontreated patients and 500 IU/L for patients receiving postconditioning, with a statistical power of 80% and an α error of 5% (2-sided), it was necessary to include at least 98 patients. All patients older than 18 years receiving a cadaveric donor graft were eligible. The exclusion criteria were the following: urgent LT, serum transaminase levels higher than 1000 IU/l before LT and graft cold ischemia longer than 10 hours. All patients meeting these criteria and transplanted by the surgeons participating in this study (D.A., L.R., C.S., and E.V.) were included prospectively. To minimize selection bias, IPo was performed in an alternate fashion, based on the protocol applied for the first clinical application of ischemic preconditioning in human.25 During pretransplant evaluation, all patients gave their written consent to participate in research protocol and for the experimental use of blood samples and graft biopsy specimens. The technique of liver procurement and transplantation are detailed in Methods (SDC, http://links.lww.com/TP/B139). Biologists, intensive care unit clinicians, and pathologists were not informed whether the graft had been subjected to IPo or not.
Transplantation
MATERIALS AND METHODS Study Design and Population
This study was designed as an observational study to assess the role of IPo against I/R injury of the liver graft. As the optimal protocol is still to define, IPo was applied on arterial reperfusion on the basis of available literature. Ischemic postconditioning consisted of three 1-minute episodes of arterial occlusion interspersed with three 1-minute pauses of inflow immediately after arterial reperfusion (Figure 1). The primary endpoint was the peak level of aspartate aminotransferase (AST) during the first 10 postoperative days because this variable is the measure of liver damage usually adopted in the literature.10-12,15,22,24 Secondary endpoints were the occurrence of early graft dysfunction (EGD) and the severity of I/R injury as determined from histological findings. Mortality, morbidity, and 1-year patient and graft survival were also studied. The sample size was calculated considering a difference of 300 IU/L in postoperative AST peak levels between the
FIGURE 1. IPo protocol.
Pathology
Both baseline (before ischemia) and reperfusion biopsies were fixed in formaldehyde, paraffin-embedded, and stained with standard hematoxylin-eosin stain. All of the biopsies were reviewed by the same (blinded) pathologists (M.S.) to ensure a homogeneous interpretation. Specimens were considered adequate if more than 3 adjacent lobules were available. Grafts were evaluated, on both baseline and reperfusion biopsy, for the presence of steatosis and necrosis; steatosis was distinguished as microvesicular or macrovesicular and classified as absent, mild (
