ABSTRACT
Background
Normothermic machine perfusion (NMP) is increasingly used to assess liver grafts before transplantation. While viability assessment of the hepatocellular compartment has been widely adopted, viability assessment of the biliary compartment remains controversial due to the lack of reliable markers. Glutathione (GSH) plays a key role in bile formation and may serve as a marker for biliary viability. Thus, the present study aimed to investigate the suitability of GSH as a potential marker for biliary viability assessment during NMP.
Methods
Between 2018 and 2021, livers undergoing NMP were included in the study. Bile samples were collected at 1 h, 6 h, and at the end of NMP, and then analyzed using enzyme‐linked immunosorbent assay (ELISA). Results were correlated with perfusion parameters and clinical outcomes.
Results
56 livers underwent NMP during the study period, of which 41 were successfully transplanted (73.2%). Within 1 year after transplantation, 19 patients (46.3%) developed biliary complications. 1‐year patient survival was 80.5% and 1‐year graft survival (death‐censored) was 94.6%. The GSH in transplanted livers showed a significant increase over time (Hour 1: 25.9 ± 31.2 μM vs. End: 108.7 ± 95.3 μM, p < 0.001) and significant inverse correlation with the occurrence of biliary complications after transplantation (p < 0.05). GSH concentration at the first hour measurement was significantly lower in livers that were deemed non‐transplantable (11.2 μM vs. 25.9 μM, p = 0.006).
Conclusions
Our data show an increase of GSH in bile during liver ex situ NMP. We found a negative correlation between GSH concentration and the development of biliary complications, suggesting GSH as a new marker for biliary viability assessment.
Keywords: biliary complications, liver preservation, liver regeneration, normothermic machine perfusion, viability assessment
This study identifies biliary glutathione concentration during normothermic machine perfusion as a potential biomarker for biliary viability. Increasing GSH levels correlate with fewer biliary complications post‐transplant, supporting its role in assessing graft quality and improving decision‐making in liver transplantation.
Abbreviations
- AP
alkaline phosphatase
- AST
alanine aminotransferase
- ATP
adenosine triphosphate
- BAR score
balance of risk score
- CIT
cold ischemia time
- CRP
C‐reactive protein
- DCD
donors after circulatory death
- DRI
donor risk index
- EAD
early allograft dysfunction
- ECD
extended criteria donors
- ELISA
enzyme‐linked immunosorbent assay
- ERCP
endoscopic retrograde cholangiopancreatography
- gGT
gamma‐glutamyltransferase
- GSH
glutathione
- H2O2
hydrogen peroxide
- IRI
ischemia reperfusion injury
- LDH
lactate dehydrogenase
- LDRI
Liver Donor Risk Index
- L‐GrAFT risk score
Liver Graft Assessment for Transplantation score
- MRP2
multidrug resistance‐associated protein 2
- NMP
normothermic machine perfusion
- PNF
primary nonfunction
- PTCD
percutaneous transhepatic cholangiography
- ROS
reactive oxygen species
- TR rats
transport deficient rats
1. Introduction
Liver transplantation remains the only curative therapeutic option for patients suffering from acute and chronic liver failure [1]. Yet, due to the severe organ shortage, patients still die while awaiting a liver transplantation [2, 3]. In view of this ongoing scarcity, novel strategies have been developed to allow for a greater acceptance of organs previously deemed unsuitable for transplantation, such as liver grafts retrieved from extended criteria donors (ECD) [4] and donors after circulatory death (DCD) [5, 6].
In recent years, ex situ normothermic machine perfusion (NMP) has emerged as a method for maintaining liver grafts in an environment that closely mimics their natural physiological state. Current liver ex situ NMP strategies allow for increased preservation times and thus provide a platform for viability assessment. This is especially important considering the increased utilization of ECD liver grafts, which are known to have a higher risk for post‐operative complications such as primary nonfunction (PNF) [7] and the development of biliary complications [8]. Viability criteria to evaluate hepatocellular function in marginal organs during NMP include lactate clearance, maintenance of physiologic pHs, bile production, and transaminase levels [9, 10]. These markers have been widely accepted in clinical practice. In contrast, viability assessment of the biliary compartment in the setting of ex situ liver NMP remains controversial. To date, biliary viability assessment is based on markers such as biliary bicarbonate, pH, and glucose levels. These markers are thought to reflect the functional integrity of the biliary epithelium [11, 12].
Ischemia during the transplant procedure causes profound damage to the biliary epithelium [13], which subsequently needs to regenerate. Insufficient biliary regeneration is thought to lead to the development of biliary complications. However, the exact pathogenesis of biliary complications remains incompletely understood, and multiple mechanisms such as peribiliary gland injury [14], cellular senescence [15] and altered (more toxic) bile composition [16] are likely to contribute.
It has been described that antioxidants alleviate the effects of ischemic injury [17]. The antioxidant glutathione (GSH) is prevalent in many tissues, but particularly in the liver, which is a primary site for GSH synthesis. GSH is one of the key determinants of acid‐independent bile flow and serves as the osmotic driving force in the formation of acid‐independent bile [18]. GSH is a tripeptide synthesized in the cytosol by the consumption of two adenosine triphosphate (ATP) and is involved in a plethora of different cellular processes associated with cell survival. Approximately half of the GSH synthesized in the liver is released into the blood plasma; the other half is released into bile through the hepatic canalicular membrane [19]. In the liver, GSH levels are maintained through a balanced GSH synthesis and turnover. Changes in GSH levels, for example, due to increased turnover or impaired synthesis/transport, are associated with a wide variety of liver diseases [20]. Ballatori et al. demonstrated that an increase or decrease in GSH excretion is accompanied by a simultaneous change in bile flow [18, 21].
It has been published that biliary GSH efflux is impaired following warm ischemia. In a murine model, biliary excretion was impaired, and bile flow decreased after cold ischemia [22]. In addition, it plays an important role in guarding cells against oxidative injury by reducing hydrogen peroxide (H2O2) and scavenging reactive oxygen species (ROS) [20], which makes it an interesting biomarker, especially in regard to ischemia reperfusion injury (IRI). The aim of this study was to analyze GSH concentration and the changes over time in bile of normothermic perfused liver grafts and its role as a possible biomarker in the assessment of the biliary compartment and predictor of biliary complications after liver transplantation. To our knowledge, this is the first study investigating biliary glutathione as a potential viability marker during human liver NMP.
2. Materials and Methods
2.1. Normothermic Machine Perfusion
We used a standardized normothermic machine perfusion (NMP) protocol for liver grafts, which was implemented at the Medical University of Innsbruck in 2018. The protocol follows the back‐to‐base principle and includes real‐time perfusate analysis over the perfusion period [23]. All liver grafts were primarily accepted with the intention to perform NMP. Indications for NMP were as follows: (1) donor factors: marginal organ quality, findings in the donor requiring workup, etc.; (2) logistics: facilitating scheduling, avoiding nighttime procedures, limited theater capacity, and/or (3) recipient factors: complex surgical cases [23].
The perfusion system is a closed system containing approximately 1250–1340 mL of perfusate (composed of 3 units of red blood cells of 250–280 mL each and 500 mL gelofusine). Before the start of the perfusion, bolus doses of cefuroxime, heparin, and calcium‐gluconate were added to the perfusion circuit. Via syringe pumps, bile salts, heparin, insulin, and epoprostenol are delivered during the perfusion process according to previously published protocols [24, 25, 26].
After a perfusion period of up to 24 h, the decision whether to transplant a liver graft was based on previously established parameters, such as a rapid lactate clearance and maintenance of physiological pH levels. Exceptionally high levels of lactate dehydrogenase, alanine aminotransferase, and aspartate aminotransferase measured in the perfusate are considered risk factors and may contribute to the decision to discard the respective liver graft [9, 11].
The perfusion protocol as well as its clinical application was approved by the ethics committee of the Medical University of Innsbruck (protocol #1175/2018).
2.2. Patient Data
Recipients of a liver graft included in the study were adults ≥ 18 years of age, listed for a first or a retransplantation. Patient records were retrospectively analyzed. Data was documented in a prospectively maintained database. Data collected included the following: Donor and recipients demographics, cause of end‐stage liver failure, Balance of Risk score (BAR score), Liver Donor Risk Index (LDRI), L‐GrAFT risk score, ischemia times, preservation times, 1‐, 2‐, 3‐year patients' survival, and biliary complication rate (Table 1).
TABLE 1.
Recipients, donor, and preservation characteristics. Clinical outcome and complications.
| Recipients | (N = 41) |
| Recipient age in years (mean, SD) | 54.7 ± 15.5 |
| Recipient BMI, kg/m2 (mean, SD) | 25.4 ± 4.5 |
| Recipient female gender (n, %) | 11 (26.8%) |
| Prior Tx (n, %) | 7 (17%) |
| Balance of Risk Score (BAR Score) (mean, SD) | 7.2 ± 3.7 |
| Balance of Risk (BAR) Score > 8 (n, %) | 11 (26.9%) |
| Donor | Overall cohort (N = 56) |
| Donor age in years (mean, SD) | 58.2 ± 17 |
| Donor BMI in kg/m2 (mean, SD) | 27.4 ± 4.7 |
| Donor female gender (n, %) | 28 (50%) |
| Extended criteria donor (ECD) (n, %) | 48 (85.7%) |
| Donation after determination of death by circulatory criteria (DCD) (n, %) | 14 (25%) |
| Liver donor risk index (DRI) (mean, SD) | 1.9 ± 0.5 |
| Liver DRI > 2 (n, %) | 18 (32.1%) |
| Preservation times | Overall cohort (N = 56) |
| Cold ischemia time (h) (mean, SD) | 6.5 ± 2.8 |
| Normothermic machine perfusion time in hours (mean, SD) | 16.5 ± 6.2 |
| Overall preservation time in hours (mean, SD) | 23.0 ± 6.6 |
| Outcome parameters (n, %) | (N = 41) |
| Biliary complications (< 1 year post transplant) | 19 (46.3%) |
| Anastomotic Biliary Strictures | 10 (24.4) |
| Anastomotic Biliary Leakages | 6 (14.6%) |
| Intrahepatic post‐transplant cholangiopathy | 1 (2.4%) |
| Hilar post‐transplant cholangiopathy | 2 (4.8%) |
| Early allograft dysfunction (EAD) | 14 (34.1%) |
| L‐GrAFT risk score | −0.4 ± 1.2 |
| 1‐year patient survival | 33 (80.5%) |
| 2‐year‐patient survival | 29 (77.8%) |
| 3‐year patient survival | 26 (74.9%) |
| Causes of end stage liver disease (recipients) | (N = 41) |
| Non‐alcoholic steatohepatitis | 6 (14.6%) |
| Alcoholic steatohepatitis | 9 (22%) |
| Primary sclerosing cholangitis | 4 (9.8%) |
| Secondary sclerosing cholangitis | 0 (0%) |
| Hepatitis B | 2 (4.9%) |
| Hepatitis C | 3 (7.3%) |
| Haemochromatosis | 1 (2.4%) |
| Ischemic cholangiopathy | 4 (9.8%) |
| Primary non‐function | 1 (2.4%) |
| Others | 11 (26.8%) |
Note: Values are presented as means with standard deviation or absolute numbers and percentages in parentheses. EAD was defined according to the Olthoff criteria: The presence of at least one of the following criteria (1) bilirubin > 10 mg/dL on Day 7, (2) international normalized ration ≥ 1.6 on Day 7, and/or (3) ALT or AST > 2000 IU/L within the first 7 days after transplantation. ECD criteria: Donor age > 65 years, ICU stay with ventilation > 7 days, BMI > 30 kg/m2, steatotic liver > 40%, serum sodium > 165 mmol/L, ALT > 105 U/L, AST > 90 U/L, serum bilirubin > 3 mg/dL, donation after cardiocirculatory death. Primary non‐function (PNF) was defined as peak AST ≥ 3000 IU/L plus at least one of the following criteria: INR ≥ 2.5, serum lactate ≥ 4 mmol/L, and total bilirubin ≥ 10 mg/dL.
2.3. Bile Sample Analysis
During NMP, bile samples were collected one and 6 h following the start of perfusion, as well as at the end of the perfusion. Samples were analyzed immediately for bile pH, bile bicarbonate, and bile glucose (Table S1) using an ABL800 FLEX analyzer (Radiometer MP Austria GmbH, Innsbruck, Austria). Additional bile samples were stored at −80°C and later analyzed for GSH concentration using the Glutathione Colorimetric Detection Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA).
2.4. Hepatocellular Injury and Function—Perfusate Parameters and Transplant Characteristics
Perfusate samples were taken at predefined time points at the start of NMP, as well as 1, 2, 4, 6, 12, 18, and 24 h after the beginning of normothermic perfusion. Samples were analyzed immediately for perfusate pH, perfusate lactate, and perfusate glucose (Table S1) using an ABL800 FLEX analyzer (Radiometer MP Austria GmbH, Innsbruck, Austria). Additional perfusate samples were analyzed for alkaline phosphatase (AP), bilirubin, C‐reactive protein (CRP), gamma‐glutamyltransferase (gGT), alanine aminotransferase (AST), lactate dehydrogenase (LDH), sodium, as well as differential blood count (Table S1) (all Roche Diagnostics GmbH, Mannheim, Germany) according to our institutional protocol [23].
3. Statistics
Statistical analysis was performed using GraphPad Prism 9, the R software environment (Version 4.3.1) [27] with the “survminer” [28] package for plotting survival curves (Version 4.3.1) and IBM SPSS Statistics Version 29 (IBM Corporation, Armonk, NY, USA). For the analysis of means, a t‐test was performed because of its proven robustness even if the assumption of normality is violated [29, 30]. For identifying associations/relationships between categorical variables, Pearson's chi‐squared test or Fisher's exact test was utilized. To assess the significance of perfusate and bile parameters, a binomial logistic regression regarding biliary complications 1 year after transplantation was performed. The linearity of the variables was checked with the Box–Tidwell method [31]. p values < 0.05 were considered statistically significant.
4. Results
4.1. Donor, Recipients, and Preservation Characteristics
Between February 2018 and December 2021, 56 liver allografts were perfused on an OrganOx metra device (UK, Oxford) and included in the study. The mean duration of NMP was 16.5 ± 6.2 h and was based on logistics or the need for further assessment of the liver grafts. The minimum normothermic perfusion time was 5.0 h and the maximum 27.8 h, which was due to an unstable recipient during the induction of anesthesia. After NMP, 41 livers were transplanted, whereas 15 had to be discarded. Livers were discarded if one of the following reasons occurred: (1) no bile production, (2) rising lactate or non‐clearance of lactate, (3) no maintenance of pH levels or high amounts of bicarbonate > 100 mL to maintain physiologic pH levels, (4) AST > 20 000 IU/L in the perfusate.
Of all liver grafts included in this study, 48 (85.7%) were from extended criteria donors (ECD) [4] and 14 (25.0%) from donors after circulatory death (DCD). Total mean ± SD preservation time was 23.0 ± 6.6 h, mean cold ischemia time (CIT) 6.5 ± 2.8 h. Mean recipient age was 54.7 ± 15.5 years and mean donor age 58.2 ± 17. After transplantation, 14 (34.1%) patients developed an early allograft dysfunction (EAD), defined as the presence of at least one of the following criteria: (1) bilirubin > 10 mg/dL on day 7, (2) international normalized ratio ≥ 1.6 on day 7, and/or (3) ALT or AST > 2000 IU/L within the first 7 days after transplantation [32]. The mean balance of risk (BAR) [33] score was 7.2 ± 3.7 and 11 (26.8%) recipients had a BAR score higher than 8. Mean donor risk index (DRI) [34] of all 56 livers was 1.9 ± 0.5 and 13 (31.7%) out of the transplanted livers had a DRI higher than 2.0. Sixteen (39.0%) out of 41 livers had mild steatosis. In the EAD group, 8 out of 14 (57.1%) livers had steatosis and, in the no‐EAD group, 8 out of 27 (29.6%) (p = 0.105). One year after transplantation, 19 (46.3%) patients developed a biliary complication [35] including anastomotic biliary strictures in 10 (24.4%), anastomotic biliary leakage in 6 (14.6%), intrahepatic post‐transplant cholangiopathy in 1 (2.4%), and hilar post‐transplant cholangiopathy in 2 (4.8%) patients. In 5 out of the 41 transplanted patients (12.2%), arterial complications were reported (Table S2).
Total mean ± SD preservation time of transplanted livers was 22.2 ± 6.6 h, NMP duration was 16.0 ± 6.0 h, and mean cold ischemia time (CIT) was 6.2 ± 2.4 h.
Ninety‐day readmission rate was 29.3%. In six cases (14.6%), 90‐day mortality occurred with functioning graft due to sepsis (aspergillosis, candidiasis). One patient died with graft dysfunction due to severe invasive aspergillosis on postoperative day 29 (BAR score 9, EAD, graft deterioration in the second week post‐transplant) (patient and recipient demographics, preservation times, outcome parameters, and causes of end‐stage liver disease of the recipients are listed in Table 1).
4.2. GSH Concentration
The GSH concentration in bile of transplanted livers showed a significant increase over time (Hour 1: 25.9 ± 31.2 μM vs. End: 108.7 ± 95.3 μM, p < 0.001). In contrast, the livers that were not transplanted after NMP did not show this significant increase (Hour 1: 11.2 ± 6.0 μM vs. End: 85.0 ± 134.8 μM, p = 0.060). GSH content at the first hour measurement was significantly lower in livers that were deemed non‐transplantable with 11.2 μM in comparison to 25.9 μM (p = 0.006) (Figure 1).
FIGURE 1.
Comparison of biliary glutathione (GSH) content between transplanted and non‐transplanted livers. At the first hour of normothermic machine perfusion, GSH concentration was significantly lower in liver grafts deemed unsuitable for transplantation (n = 15) (11.2 μM) compared to those successfully transplanted (n = 41) (25.9 μM; p = 0.006, unpaired t‐test). Values represent mean concentrations. [Color figure can be viewed at wileyonlinelibrary.com]
GSH concentration at Hour 6 correlated significantly with biliary pH (p = 0.025) and bile bicarbonate concentration (p = 0.011). There was no correlation between GSH content at 6 h and lactate in perfusate (p = 0.385) or bile (p = 0.713).
After 6 h and at the end of NMP, the GSH concentration was significantly lower in patients who suffered from a post‐transplant cholangiopathy in the first year after transplantation (56.0 ± 45.0 vs. 126.1 ± 104.8 μM, p = 0.013) (77.2 ± 74.1 vs. 135.8 ± 104.5 μM, p = 0.048) (Figure 2).
FIGURE 2.
Biliary glutathione (GSH) content in transplanted livers with and without post‐transplant cholangiopathy. At 6 h and at the end of normothermic machine perfusion (NMP), GSH concentrations were significantly lower in grafts that developed post‐transplant cholangiopathy (n = 19) within the first year compared to those without complications (n = 22) (6 h: 56.0 ± 45.0 vs. 126.1 ± 104.8 μM, p = 0.013; end of NMP: 77.2 ± 74.1 vs. 135.8 ± 104.5 μM, p = 0.048; unpaired t‐test). Values represent mean ± SD. [Color figure can be viewed at wileyonlinelibrary.com]
If taking the upper bound value of the 95% confidence interval for the mean of the post‐transplant cholangiopathy cohort (78.4 μM) as a threshold value, the chance of suffering from a post‐transplant cholangiopathy is 6.10 times higher if the GSH concentration is below 78.4 μM after 6 h of NMP (p = 0.022) (Figure 3).
FIGURE 3.
Predicted probability of post‐transplant cholangiopathy based on biliary glutathione (GSH) levels. Binomial logistic regression analysis showing the predicted probability of developing biliary complications within 1 year after transplantation, with 95% confidence intervals. Using the upper bound value of the 95% confidence interval for the mean GSH concentration in the cholangiopathy group (n = 19) at 6 h (78.4 μM) as a threshold, livers with GSH levels below this value had a 6.10‐fold increased risk of post‐transplant cholangiopathy (p = 0.022, Fisher's exact test). [Color figure can be viewed at wileyonlinelibrary.com]
Changes of GSH concentration in the first 6 h of NMP were lower in patients who developed a biliary complication in the first year (39.0 ± 67.1 vs. 97.9 ± 106.1 μM, p = 0.044).
One‐year graft survival showed no significant association with GSH levels at hour six, GSH levels at the end of NMP, or change in GSH levels (p = 0.534/0.881/0.834). One‐year overall survival showed no significant association with GSH levels at Hour 6, GSH levels at the end of NMP, and change in GSH levels (p = 0.942/0.587/0.586).
4.3. Confounder Analysis
To check for possible confounders for the GSH results, a subgroup analysis of the 1‐year biliary complication (n = 19) and no‐biliary complication cohort (n = 22) was performed.
All demographic data between cohorts were comparable except donor type and the rate of EAD, with more DCD donors in the post‐transplant cholangiopathy cohort (31.6 vs. 4.5%, p = 0.036), but a lower EAD rate (10.5 vs. 54.5%, p = 0.004) (Table 2).
TABLE 2.
Comparison of recipient and donor characteristics in patients with and without biliary complications 1 year after transplantation.
| Variable | Biliary complication | No biliary complication | p |
|---|---|---|---|
| (1 year, N = 19) | (1 year, N = 22) | ||
| Recipient sex | |||
| Female (n, %) | 5 (26.3) | 6 (27.3) | 1.000 |
| Male (n, %) | 14 (73.7) | 16 (72.7) | |
| Blood group | |||
| O (n, %) | 8 (42.1) | 7 (35.0) | 0.280 |
| A (n, %) | 7 (36.8) | 10 (50.0) | |
| B (n, %) | 4 (21.1) | 1 (5.0) | |
| AB (n, %) | 0 | 2 (10.0) | |
| Indication | |||
| NASH (n, %) | 4 (21.1) | 2 (9.1) | 0.698 |
| ASH (n, %) | 4 (21.1) | 5 (22.7) | |
| PSC (n, %) | 1 (5.3) | 3 (13.6) | |
| Hepatitis (n, %) | 1 (5.3) | 4 (18.2) | |
| Ischemic cholangiopathy (n, %) | 2 (10.5) | 2 (9.1) | |
| Other (n, %) | 7 (36.8) | 6 (27.3) | |
| Malignancy (n, %) | 5 (26.3) | 7 (31.8) | 0.744 |
| No Malignancy (n, %) | 14 (73.7) | 15 (68.2) | |
| Milan criteria | |||
| Not Applicable (n, %) | 14 (73.7) | 16 (72.7) | 0.684 |
| Inside (n, %) | 4 (21.1) | 3 (13.6) | |
| Outside (n, %) | 1 (5.3) | 3 (13.6) | |
| Portal vein thrombosis (n, %) | 1 (5.3) | 3 (13.6) | 0.610 |
| No portal vein thrombosis (n, %) | 18 (94.7) | 19 (86.4) | |
| TIPS (n, %) | 3 (15.8) | 4 (18.2) | 1.000 |
| No TIPS (n, %) | 16 (84.2) | 18 (81.8) | |
| Donor Sex | |||
| Female (n, %) | 12 (63.2) | 11 (50.0) | 0.531 |
| Male (n, %) | 7 (36.8) | 11 (50.0) | |
| Donor type | |||
| DBD (n, %) | 13 (68.4) | 21 (95.5) | 0.036 |
| DCD (n, %) | 6 (31.6) | 1 (4.5) | |
| ECD (n, %) | 16 (84.2) | 17 (77.3) | 0.703 |
| No ECD (n, %) | 3 (15.8) | 5 (22.7) | |
| Mild steatosis (n, %) | 8 (42.1) | 8 (36.4) | 0.757 |
| No steatosis (n, %) | 11 (57.9) | 14 (63.6) | |
| Primary function (n, %) | 17 (89.5) | 10 (45.5) | 0.004 |
| EAD (n, %) | 2 (10.5) | 12 (54.5) | |
| Age recipient (mean, SD) | 56.6 (±13.4) | 53.0 (±17.2) | 0.472 |
| Age donor (mean, SD) | 57.8 (±15.8) | 62.3 (±16.7) | 0.380 |
| BMI recipient (mean, SD) | 24.7 (±3.6) | 25.9 (±5.1) | 0.415 |
| BMI donor (mean, SD) | 26.3 (±3.5) | 26.9 (±4.0) | 0.616 |
| CIT (mean, SD) | 332.8 (±137.1) | 408.1 (±143.8) | 0.096 |
| NMP time (mean, SD) | 913.6 (±345.6) | 1001.0 (±371.6) | 0.443 |
| Total preservation time (mean, SD) | 1246.4 (±401.2) | 1409.1 (±387.0) | 0.195 |
| Second CIT time (mean, SD) | 20.2 (±28.5) | 11.7 (±6.7) | 0.222 |
| Total preservation time adjusted (mean, SD) | 1266.6 (±395.9) | 1354.8 (±489.9) | 0.534 |
Note: Values are presented as means with standard deviation or absolute numbers and percentages in parentheses. Bold values indicate significant p values. For the analyses of means a t‐test was performed, for the analysis of categorical variables Pearsons's chi‐squared test or Fisher's exact test was utilized. p values < 0.05 were considered statistically significant.
4.4. Secondary Performance Parameters
For distinction of possible parameters predicting biliary complications after 1 year post‐transplant, additional parameters were analyzed (Table S2).
The level of bicarbonate in the perfusate at the end of NMP was significantly higher in the complication cohort (15.84 mmol/L vs. 12.25 mmol/L, p = 0.024). The level of lactate in the perfusate at 6 h and at the end of NMP was significantly lower in the complication cohort (3.68 mg/dL vs. 6.06 mg/dL, p = 0.039). Both values did not correlate with GSH levels at the end of NMP (Spearman correlation coefficients: −0.81/0.058, p = 0.632 [0.734]). No significant associations were observed between other analyzed parameters and the incidence of biliary complications.
5. Discussion
In this study, we demonstrate the antioxidant GSH as a potential new biomarker for the biliary viability assessment during NMP of livers. So far, no previous studies have investigated biliary GSH concentration as a viability marker in human ex situ liver perfusion.
Our initial results showed that the GSH content in the bile of NMP livers, which were successfully transplanted after NMP, increased significantly during the perfusion period. In contrast, this increase was not observed in non‐transplantable liver grafts. Especially at the first hour following the start of NMP, the rise of GSH was significantly higher in transplantable liver grafts compared to those deemed unsuitable for transplantation. These findings could be a first indication that livers considered unsuitable for transplantation have impaired GSH synthesis and/or secretion, possibly due to previous damage or impaired function. This is supported by the findings of Koeppel et al. [22]. They demonstrated that a lack of GSH secretion in TR rats (transport deficient rats), due to a deficiency of multidrug resistance‐associated protein 2 (MRP2), associated with reduced biliary GSH excretion during reperfusion leads to an impairment in bile formation. Notably, the expression and tissue localization of canalicular MRP2, which mediates GSH excretion, is rapidly reduced in response to a wide range of liver injuries [36] including oxidative stress [37].
Furthermore, we found a negative correlation between GSH concentration in bile during NMP and the occurrence of post‐transplant biliary complications. A GSH value of 78.4 μM was identified as a threshold, below which the likelihood of developing biliary complications increases by a factor of 6.10, suggesting GSH as a marker of biliary viability. This indicates that GSH may serve as a biomarker for cholangiocyte assessment during NMP.
The antioxidant properties of GSH [38] make it an interesting candidate for therapeutic interventions. Enhancing GSH levels during NMP could possibly lead to an improved performance of marginal liver grafts during NMP and even make them accessible for transplantation. This is in line with the findings from Clarke et al. [39]. They showed that N‐acetylcysteine (NAC), a precursor of GSH, has been used during NMP as a therapeutic intervention to limit the formation and accumulation of methemoglobin during NMP, and as a result, allows a significant extension of perfusion duration.
Although our study primarily focuses on GSH levels during NMP, other parameters (perfusion parameters, donor and recipient demographics, transplant outcomes such as patient and graft survival after 3 years) were analyzed to rule out any confounders of our main findings and long‐term implications. There was no correlation between GSH levels and patient/graft survival after 1, 2, and 3 years. Higher GSH levels could potentially protect against bile duct injury and subsequent cholangiopathy, making it a crucial early biomarker for assessing cholangiocyte function during NMP.
In conclusion, our study demonstrates that higher GSH levels in bile during NMP are associated with better liver perfusion performance and a reduced risk of post‐transplant cholangiopathy. GSH serves not only as an early biomarker for cholangiocyte viability but also as a potential target for therapeutic intervention to enhance transplant outcomes. These findings add to the growing body of evidence supporting the protective role of GSH as an antioxidant in liver perfusion and offer promising opportunities for future organ‐repair strategies. Further research is essential to validate these findings and explore the therapeutic potential of GSH during machine perfusion to improve liver transplant outcome.
6. Limitations
While our study provides valuable insights, it is important to acknowledge its limitations. This is a single‐center study and the defined GSH threshold value of 78.4 μM is based on our investigated cohort, as no further data or literature is available at present. This approach might lead to overfitting and requires follow‐up validation in an independent cohort and the need for external validation. The sample size was rather small, due to the nature of the investigated patient cohort.
As this is a retrospective study, prospective measurement of GSH could provide real‐time insight into biliary function during NMP. The assay used can be performed within 45–90 min, which is sufficient time for a timely decision on organ acceptance during NMP. However, it should be noted that on‐site laboratory capacity and infrastructure must be available. This is a potentially limiting factor in clinical practice.
Additionally, the non‐consecutive inclusion of livers due to logistical reasons may introduce a selection bias. The timing of the last GSH measurement varied, as it was taken just before NMP termination, which could affect the consistency of the results. Furthermore, the high incidence of post‐transplant cholangiopathy observed in this study may be partly explained by the substantial proportion of DCD livers, which are known to have a higher risk for biliary complications [40]. Nevertheless, we could demonstrate significant findings, which could be expanded to future research approaches to validate these results across a broader population.
Author Contributions
Study concept and design (C.B., A.W.), acquisition of experimental data (C.B., A.W.), acquisition of clinical data (C.B., H.E., F.P., B.C., F.M., R.O., T.R., D.Ö., S.S., A.W.), analysis of data (C.B., F.P.), verification of data (C.B., A.W.), interpretation of data (C.B., F.P., H.E., A.W.), drafting of the manuscript (C.B., A.W., F.P., H.E.), critical revision of the manuscript for important intellectual content (C.B., H.E., F.P., B.C., F.M., R.O., T.R., D.Ö., S.S., A.W.). All authors read and approved the final manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Bile parameters, perfusate parameters and differential blood count of the perfusate and the correlation with biliary complication 1 year after liver transplantation.
Table S2: Treatment, Time of diagnosis and arterial complications in the biliary complication cohort (N = 19).
Acknowledgments
Open Access funding provided by Medizinische Universitat Innsbruck/KEMÖ.
Funding: The work was funded by the “MUI‐START Förderung” (project number 2018‐01‐009) granted to A.W.
References
- 1. Asrani S. K., Devarbhavi H., Eaton J., and Kamath P. S., “Burden of Liver Diseases in the World,” Journal of Hepatology 70, no. 1 (2019): 151–171. [DOI] [PubMed] [Google Scholar]
- 2. Eurotransplant , “Eurotransplant Yearly Statistics Overview 2021 [Internet],” (2022), https://statistics.eurotransplant.org/index.php?search_type=overview&search_text=9023.
- 3. Jochmans I., van Rosmalen M., Pirenne J., and Samuel U., “Adult Liver Allocation in Eurotransplant,” Transplantation 101, no. 7 (2017): 1542–1550. [DOI] [PubMed] [Google Scholar]
- 4. Eurotransplant Foundation , “Eurotransplant Manual Chapter 9: The Donor [Internet],” (2017), https://www.eurotransplant.org/wp‐content/uploads/2022/11/H9‐The‐Donor‐08‐11‐2022.pdf.
- 5. Smith M., Dominguez‐Gil B., Greer D. M., Manara A. R., and Souter M. J., “Organ Donation After Circulatory Death: Current Status and Future Potential,” Intensive Care Medicine 45, no. 3 (2019): 310–321. [DOI] [PubMed] [Google Scholar]
- 6. Mihaylov P., Mangus R., Ekser B., et al., “Expanding the Donor Pool With the Use of Extended Criteria Donation After Circulatory Death Livers,” Liver Transplantation 25, no. 8 (2019): 1198–1208. [DOI] [PubMed] [Google Scholar]
- 7. Hartog H., Hann A., and Perera M. T. P. R., “Primary Nonfunction of the Liver Allograft,” Transplantation 106, no. 1 (2022): 117–128. [DOI] [PubMed] [Google Scholar]
- 8. O'Neill S., Roebuck A., Khoo E., Wigmore S. J., and Harrison E. M., “A Meta‐Analysis and Meta‐Regression of Outcomes Including Biliary Complications in Donation After Cardiac Death Liver Transplantation,” Transplant International 27, no. 11 (2014): 1159–1174. [DOI] [PubMed] [Google Scholar]
- 9. Watson C. J. E., Kosmoliaptsis V., Randle L. V., et al., “Normothermic Perfusion in the Assessment and Preservation of Declined Livers Before Transplantation,” Transplantation 101, no. 5 (2017): 1084–1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sutton M. E., 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 9, no. 11 (2014): e110642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Watson C. J. E. and Jochmans I., “From ‘Gut Feeling’ to Objectivity: Machine Preservation of the Liver as a Tool to Assess Organ Viability,” Current Transplantation Reports 5, no. 1 (2018): 72–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ghinolfi D., Rreka E., De Tata V., et al., “Pilot, Open, Randomized, Prospective Trial for Normothermic Machine Perfusion Evaluation in Liver Transplantation From Older Donors,” Liver Transplantation 25, no. 3 (2019): 436–449. [DOI] [PubMed] [Google Scholar]
- 13. Brunner S. M., Junger H., Ruemmele P., et al., “Bile Duct Damage After Cold Storage of Deceased Donor Livers Predicts Biliary Complications After Liver Transplantation,” Journal of Hepatology 58, no. 6 (2013): 1133–1139. [DOI] [PubMed] [Google Scholar]
- 14. op den Dries S., Westerkamp A. C., Karimian N., et al., “Injury to Peribiliary Glands and Vascular Plexus Before Liver Transplantation Predicts Formation of Non‐Anastomotic Biliary Strictures,” Journal of Hepatology 60, no. 6 (2014): 1172–1179. [DOI] [PubMed] [Google Scholar]
- 15. Ferreira‐Gonzalez S., Man T. Y., Esser H., et al., “Senolytic Treatment Preserves Biliary Regenerative Capacity Lost Through Cellular Senescence During Cold Storage,” Science Translational Medicine 14, no. 674 (2022): eabj4375. [DOI] [PubMed] [Google Scholar]
- 16. Buis C. I., Geuken E., Visser D. S., et al., “Altered Bile Composition After Liver Transplantation Is Associated With the Development of Nonanastomotic Biliary Strictures,” Journal of Hepatology 50, no. 1 (2009): 69–79. [DOI] [PubMed] [Google Scholar]
- 17. Orellana‐Urzúa S., Briones‐Valdivieso C., Chichiarelli S., Saso L., and Rodrigo R., “Potential Role of Natural Antioxidants in Countering Reperfusion Injury in Acute Myocardial Infarction and Ischemic Stroke,” Antioxidants 12, no. 9 (2023): 1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ballatori N. and Truong A. T., “Glutathione as a Primary Osmotic Driving Force in Hepatic Bile Formation,” American Journal of Physiology. Gastrointestinal and Liver Physiology 263, no. 5 (1992): G617–G624. [DOI] [PubMed] [Google Scholar]
- 19. Ballatori N. and Rebbeor J., “Roles of MRP2 and oatp1 in Hepatocellular Export of Reduced Glutathione,” Seminars in Liver Disease 18, no. 4 (1998): 377–387. [DOI] [PubMed] [Google Scholar]
- 20. Yuan L. and Kaplowitz N., “Glutathione in Liver Diseases and Hepatotoxicity,” Molecular Aspects of Medicine 30, no. 1–2 (2009): 29–41. [DOI] [PubMed] [Google Scholar]
- 21. Ballatori N. and Truong A. T., “Relation Between Biliary Glutathione Excretion and Bile Acid‐Independent Bile Flow,” American Journal of Physiology. Gastrointestinal and Liver Physiology 256, no. 1 (1989): G22–G30. [DOI] [PubMed] [Google Scholar]
- 22. Koeppel T. A., Trauner M., Albert M., Arrese M., Rios‐Velez L., and Boyer J. L., “Role of Glutathione in Hepatic Bile Formation During Reperfusion After Cold Ischemia of the Rat Liver,” Journal of Hepatology 28, no. 5 (1998): 812–819. [DOI] [PubMed] [Google Scholar]
- 23. Cardini B., Oberhuber R., Fodor M., et al., “Clinical Implementation of Prolonged Liver Preservation and Monitoring Through Normothermic Machine Perfusion in Liver Transplantation,” Transplantation 104 (2020): 1917–1928. [DOI] [PubMed] [Google Scholar]
- 24. Ravikumar R., Jassem W., Mergental H., et al., “Liver Transplantation After Ex Vivo Normothermic Machine Preservation: A Phase 1 (First‐in‐Man) Clinical Trial,” American Journal of Transplantation 16, no. 6 (2016): 1779–1787. [DOI] [PubMed] [Google Scholar]
- 25. Nasralla D., Coussios C. C., Mergental H., et al., “A Randomized Trial of Normothermic Preservation in Liver Transplantation,” Nature 557, no. 7703 (2018): 50–56. [DOI] [PubMed] [Google Scholar]
- 26. Ceresa C. D. L., Nasralla D., Watson C. J. E., et al., “Transient Cold Storage Prior to Normothermic Liver Perfusion May Facilitate Adoption of a Novel Technology,” Liver Transplantation 25, no. 10 (2019): 1503–1513. [DOI] [PubMed] [Google Scholar]
- 27. R Core Team , R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2021), https://www.R‐project.org. [Google Scholar]
- 28. Kassambara A., Kosinski M., Biecek P., and Fabian S., “survminer: Drawing Survival Curves Using ‘ggplot2’ [Internet],” (2021), https://cran.rproject.org/web/packages/survminer/survminer.pdf.
- 29. Wilcox R. R., Introduction to Robust Estimation and Hypothesis Testing, 3rd ed. (Elsevier Science & Technology Books, 2013), http://www.sciencedirect.com/science/book/9780123869838. [Google Scholar]
- 30. Pagano R. R., Understanding Statistics in the Behavioral Sciences, 9th ed. (Wadsworth Publishing Co Inc, 2010). [Google Scholar]
- 31. Box G. E. P. and Tidwell P. W., “Transformation of the Independent Variables,” Technometrics 4, no. 4 (1962): 531–550. [Google Scholar]
- 32. Olthoff K. M., 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 Transplantation 16, no. 8 (2010): 943–949. [DOI] [PubMed] [Google Scholar]
- 33. Martínez J. A., Pacheco S., Bachler J. P., et al., “Accuracy of the BAR Score in the Prediction of Survival After Liver Transplantation,” Annals of Hepatology 18, no. 2 (2019): 386–392. [DOI] [PubMed] [Google Scholar]
- 34. Feng S., Goodrich N. P., Bragg‐Gresham J. L., et al., “Characteristics Associated With Liver Graft Failure: The Concept of a Donor Risk Index,” American Journal of Transplantation 6, no. 4 (2006): 783–790. [DOI] [PubMed] [Google Scholar]
- 35. Esser H., D. Jong, I , Roos F. M., et al., “Defining Biliary Complications After Liver Transplantation—Consensus Guidelines From the BileducTx Meeting,” in Preparation. [DOI] [PMC free article] [PubMed]
- 36. Geier A., Wagner M., Dietrich C. G., and Trauner M., “Principles of Hepatic Organic Anion Transporter Regulation During Cholestasis, Inflammation and Liver Regeneration,” Biochimica et Biophysica Acta 1773, no. 3 (2007): 283–308. [DOI] [PubMed] [Google Scholar]
- 37. Sekine S., Yano K., Saeki J., Hashimoto N., Fuwa T., and Horie T., “Oxidative Stress Is a Triggering Factor for LPS‐Induced Mrp2 Internalization in the Cryopreserved Rat and Human Liver Slices,” Biochemical and Biophysical Research Communications 399, no. 2 (2010): 279–285. [DOI] [PubMed] [Google Scholar]
- 38. Averill‐Bates D. A., “The Antioxidant Glutathione,” Vitamins and Hormones 121 (2023): 109–141. [DOI] [PubMed] [Google Scholar]
- 39. Clarke G., Mao J., Fan Y., et al., “N‐Acetylcysteine: A Novel Approach to Methaemoglobinaemia in Normothermic Liver Machine Perfusion,” Scientific Reports 13, no. 1 (2023): 19022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Meier R. P. H., Kelly Y., Braun H., et al., “Comparison of Biliary Complications Rates After Brain Death, Donation After Circulatory Death, and Living‐Donor Liver Transplantation: A Single‐Center Cohort Study,” Transplant International 35 (2022): 10855. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Bile parameters, perfusate parameters and differential blood count of the perfusate and the correlation with biliary complication 1 year after liver transplantation.
Table S2: Treatment, Time of diagnosis and arterial complications in the biliary complication cohort (N = 19).