Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2018 Aug;18(8):2005-2020.
doi: 10.1111/ajt.14687. Epub 2018 Mar 14.

Observations on the Ex Situ Perfusion of Livers for Transplantation

Affiliations
Free PMC article
Observational Study

Observations on the Ex Situ Perfusion of Livers for Transplantation

Christopher J E Watson et al. Am J Transplant. .
Free PMC article

Abstract

Normothermic ex situ liver perfusion might allow viability assessment of livers before transplantation. Perfusion characteristics were studied in 47 liver perfusions, of which 22 resulted in transplants. Hepatocellular damage was reflected in the perfusate transaminase concentrations, which correlated with posttransplant peak transaminase levels. Lactate clearance occurred within 3 hours in 46 of 47 perfusions, and glucose rose initially during perfusion in 44. Three livers required higher levels of bicarbonate support to maintain physiological pH, including one developing primary nonfunction. Bile production did not correlate with viability or cholangiopathy, but bile pH, measured in 16 of the 22 transplanted livers, identified three livers that developed cholangiopathy (peak pH < 7.4) from those that did not (pH > 7.5). In the 11 research livers where it could be studied, bile pH > 7.5 discriminated between the 6 livers exhibiting >50% circumferential stromal necrosis of septal bile ducts and 4 without necrosis; one liver with 25-50% necrosis had a maximum pH 7.46. Liver viability during normothermic perfusion can be assessed using a combination of transaminase release, glucose metabolism, lactate clearance, and maintenance of acid-base balance. Evaluation of bile pH may offer a valuable insight into bile duct integrity and risk of posttransplant ischemic cholangiopathy.

Keywords: clinical research/practice; liver allograft function/dysfunction; liver transplantation/hepatology; metabolism/metabolite; organ perfusion and preservation.

Figures

Figure 1
Figure 1
Vascular resistance and flows during perfusions. Lines represent trends in mean resistance (A and B) or flows (C and D) in portal vein (A and C) and hepatic artery (B and D); dots represent individual perfusions. Once perfusion temperature was ≥35°C, HA and PV pressures were fixed at 60 mmHg and 8 to 10 mmHg, respectively. HA and PV resistances both fall to a steady state by 60 minutes of perfusion (HA 60 median resistance 235 and 254 mmHg/L/min for transplanted and research livers respectively, and PV median 60 minute resistance 9.95 and 9.3 mmHg/L/min, respectively). Vascular flows are dependent on resistance and perfusion pressure, and at 60 minutes median HA flows were 250 and 236 mls/min for transplanted and research livers, respectively, and PV flows were 0.81 and 0.85L/min, respectively.( HA, hepatic artery; PV, portal vein
Figure 2
Figure 2
Graphs showing the change in perfusate ALT (A and B), AST (C and D) concentrations throughout the course of NESLiP of transplanted (B and D) and nontransplanted (A and C) livers. Figure 2 (A to D) shows the perfusate ALT and AST rising following the start of NESLiP. ALT and AST release from the livers had abated by two hours in most livers. In B and D, the orange line is the liver T6 that developed primary nonfunction. In A and C, three livers show continued transaminase release beyond 2 hours suggesting ongoing hepatocellular damage; from highest to lowest peak ALT and AST, these were livers R8, R11, and R1. ALT, alanine transaminasel; AST, aspartate transaminase; NESLiP, normothermic ex situ liver perfusion
Figure 3
Figure 3
Relationship between the ALT in the perfusate at 2 hours and (A) the ALT concentration in the preservation effluent and (B) the peak ALT in the first 7 days posttransplant. (A) 28 livers were flushed with Hartmann's solution before undergoing NESLiP from which transaminases were measured. There is a strong correlation between the concentration of ALT in the effluent Hartmann's solution after it has flushed through the liver and the perfusate ALT at 2 hours (Pearson r = .776, P < .001, linear regression equation y = 1.122x); the symbol x indicates transplanted livers while + indicates a liver that was not transplanted. (B) There is a correlation between the ALT after 2 hours of NESLiP and the peak ALT in the first 7 days posttransplant (Pearson r = .733, P = .0001, linear regression equation, y = 0.35x+280). ALT, alanine transaminase; NESLiP, normothermic ex situ liver perfusion
Figure 4
Figure 4
Lactate concentrations for livers at the start of NESLiP. The lactate concentration in the perfusate fell to below 2.5 mmol/L (dotted red line) in all but five livers by 2 hours, and all but one liver by 3 hours. Liver R10 was a DCD donor with a nonperfused replaced left hepatic artery and consequently a poorly perfused left lobe, which was presumably producing lactate while the well perfused right lobe was metabolising it. R13 and R15 were markedly steatotic livers. Liver T6 is highlighted in orange and was the only liver to suffer primary nonfunction following transplantation. Liver T9 had an intraparenchymal contusion of the left lobe following trauma. Supplementary Figure S1 shows the absolute change in lactate over time for each liver. NESLiP, normothermic ex situ liver perfusion
Figure 5
Figure 5
Glucose concentrations during the course of normothermic perfusion in (A) transplanted livers and (B) nontransplanted livers. Perfusate glucose concentrations rose for the first 1 to 2 hours during NESLiP, and then in all but one perfusions glucose fell at varying rates but each suggestive of zero order kinetics. Dashed lines represent livers where the initial perfusate glucose concentration was low and glucose boluses were given to probe liver function; the heavy line with open circle symbols in (A) is the glucose profile for the liver that suffered primary nonfunction. NESLiP, normothermic ex situ liver perfusion
Figure 6
Figure 6
Response of two livers with an initially “low” perfusate glucose to a glucose challenge to confirm functional capacity. (A) shows the glucose concentration in the perfusate for liver T10. The glucose only rose to 8.9 mmol/L by 90 minutes. A glucose infusion was begun at 190 minutes, delivering 7.5 g (41.6 mmol), following which glucose fell. (B) shows the glucose profile of liver R20. After 90 minutes, during which a peak glucose of 9.5 mmol/L was achieved, a bolus of 2.5 g (13.9 mmol) was added. The glucose rose to 16.6 mmol/L and fell exhibiting zero order kinetics
Figure 7
Figure 7
Bile flow during normothermic perfusion. Livers produced bile at a variable rate and the rate did not correlate with bile duct damage or posttransplant outcome. Each line represents a single liver perfusion, colored according to whether they were transplanted and developed cholangiopathy or not, or for research livers, whether there was grade 3 or 4 stromal necrosis of the major septal ducts on histology or not. Research livers early in the program did not have detailed histology on major septal ducts and are omitted
Figure 8
Figure 8
Bile duct histology illustrating normal intrahepatic ducts and ducts with moderate/severe (grade 3/4) stromal necrosis. (A) is a normal septal duct (H&E ×20) from liver R24 showing preserved epithelium, intact peribiliary glands ([B] H&E ×100), and endothelial staining in the viable stroma ([C], CD31 stain, ×100). This is grade 1 stromal necrosis (<25% of circumference affected). (D) and (E) are from liver R22; (D) (H&E ×100) shows an intact epithelial layer with preservation of some peribiliary glands but an acellular necrotic stroma indicating widespread necrosis; (E) (CD31 stain, 100x) shows relatively little CD31 staining signifying loss of endothelial cells within the stroma. (F) is from liver R14 (H&E ×40) showing patchy loss of surface epithelium, preservation of some peribiliary glands and stromal necrosis affecting between 50% and 75% of the circumference of the duct (Grade 3). (G) is from liver R21 (H&E ×40) showing loss of surface epithelium, degenerate peribiliary glands and complete stromal necrosis in an effectively dead duct (Grade 4). H&E, hematoxylin and eosin
Figure 9
Figure 9
Change in pH of bile during perfusion. Figure illustrating the change in biliary pH over time in NESLiP livers. Transplanted livers which did not develop cholangiopathy are indicated in pink, those that did are in mauve. Also shown are the biliary pH measurements of livers not transplanted. These are colored such that those with minimal evidence of stromal necrosis of major septal bile ducts (grade 1) on histology are in purple, grade 2 (25‐50% circumferential necrosis) in blue, and those with greater degrees of stromal necrosis, typically affecting >50% of the circumference, are in orange. pH appears to rise initially after the start of perfusion, and then plateau above pH 7.5 in livers with viable bile ducts with the exception of liver T20 in which the pH fell back to under 7.5 (dashed pink line). )NESLiP, normothermic ex situ liver perfusion
Figure 10
Figure 10
(A) The glucose concentration in bile during perfusion and (B) the difference between biliary glucose and perfusate glucose concentrations. (A) Glucose is reabsorbed from bile by cholangiocytes, so normal bile has a low glucose. The presence of glucose was associated with cholangiopathy or duct injury. Since perfusate glucose concentrations were often high, impaired cholangiocyte function often resulted in a very high glucose reflecting the high levels in the perfusate at the time. (B) depicts the difference between perfusate and bile glucose showing that most healthy ducts exhibited a large glucose gradient while diseased ducts did not
Figure 11
Figure 11
Biliary bicarbonate concentration. The bicarbonate concentration in bile increases early after perfusion as bile starts to be produced, but appears to be less discriminatory than the bile pH

Similar articles

See all similar articles

Cited by 25 articles

See all "Cited by" articles

References

    1. Watson CJ, Kosmoliaptsis V, Randle LV, et al. Preimplant normothermic liver perfusion of a suboptimal liver donated after circulatory death. Am J Transplant. 2016;16(1):353‐357. - PubMed
    1. Watson CJ, Randle LV, Kosmoliaptsis V, Gibbs P, Allison M, Butler AJ. 26‐hour storage of a declined liver before successful transplantation using ex vivo normothermic perfusion. Ann Surg. 2017;265(1):e1‐e2. - PubMed
    1. Bral M, Gala‐Lopez B, Bigam D, et al. Preliminary single‐center Canadian experience of human normothermic ex vivo liver perfusion: results of a clinical trial. Am J Transplant. 2017;17(4):1071‐1080. - PubMed
    1. 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. 2016;16(6):1779‐1787. - PubMed
    1. Mergental H, Perera MT, Laing RW, et al. Transplantation of declined liver allografts following normothermic ex‐situ evaluation. Am J Transplant. 2016;16:3235‐3245. - PubMed

Publication types

Feedback