Hypothermic Machine Perfusion versus Cold Storage on the rescuing of livers
from non-heart beating donor rats.
Matías E. Carnevale1*, Cecilia L. Balaban1, Edgardo E. Guibert1, Hebe Bottai2 and
Joaquin V. Rodriguez1
1
Centro Binacional (Argentina-Italia) de Investigaciones en Criobiología Clínica y
Aplicada (CAIC) - Universidad Nacional de Rosario. Avda. Arijón 28 bis, Rosario
(S2011BXN), Argentina. 2Estadística, Fac. Cs. Bioq. y Farm. UNR
E-mail: jrodrig@fbioyf.unr.edu.ar
Abstract: The aim of this work was to compare the preservation efficiency of grafts excised from Non
Heart Beating Donors that have suffered 45 min of warm ischemia, employing Cold Storage (CS) and
Hypothermic Machine Perfusion (HMP) methods. We have developed a BES–Gluconate–Polyethylene
Glycol based solution (BGP-HMP), a new solution for HMP of livers for transplants. After 24 h of HMP
or CS, these livers were reperfused at 37C with Krebs-Henseleit Dextran solution. In both procedures,
portal pressure and flow were measured and the intrahepatic resistance (IR) was calculated. Perfusate
pH oscillations and enzymes (LDH, AST, and ALT) activities in perfusate were evaluated during
normothermic reperfusion. Also, the O2 consumption of the liver, glucose production and the bile flow
were measured during the normothermic reperfusion period. Results: the portal flow (PF) and IR in both
groups (n=5) showed statistical differences (p<0.05). HMP with BGP-HMP solution showed higher
values of PF and lesser IR than CS with HTK solution. Enzyme release after 90 min of reperfusion didn’t
show statistical differences between both groups. With regard to bile flow and O 2 consumption, preserved
livers by both process and solutions were able to produce bile, but the preserved livers with HMP were
able to uptake more O2 than livers preserved by CS.
Keywords: hypothermic machine perfusion, liver, cold storage
1.
INTRODUCTION
Hepatic transplantation has become the only safety solution to various
irreversible liver problems. According to this, cold storage technique (CS) was an
important factor to achieve this success. Brain death donors are currently used in liver
transplantation. However, a growing discrepancy between demand for organs and their
availability from brain dead donors has led to a re-examination of non-heart beating
donors (NHBD) to expand the pool of organs suitable for transplant. These organs may
have a decreased survival after being transplanted because they suffered a period of
warm ischemia. For the above reasons it is necessary to develop new strategies for the
rescuing of these organs. There are experimental as well as clinical reports indicating
that Hypothermic Machine Perfusion (HMP) of marginal donor livers is superior to cold
static preservation. Consequently, this method is an alternative to overcome the present
shortage of donors by expansion of the existing donor pool and possibly lengthening of
the storage time (Van der Plaats A. et al, 2004).
HMP is a dynamic technique where a continuous flow of preservation solution
perfuses the organ during cold storage step and keeps a residual metabolism, largely
dependent on energy generation. This, in mammalian systems is synonymous with a
need for oxygen supply for aerobic metabolism delivered by vascular perfusion (Fuller
B.J., Lee C.Y., 2007). Besides, the HMP system allows applying pharmacological
maneuvers on the hypothermically perfused organs. However, there still are doubts as to
which is the best perfusion preservation solution to use. For these reasons, we have
developed a BES–Gluconate–Polyethylene Glycol based solution (BGP-HMP), a new
solution for HMP of livers for transplants (Pascucci F. 2012, Carnevale ME, 2012).
So, the aim of this work was to compare the preservation efficiency of grafts
excised from NHBD that have suffered 45 min of warm ischemia, employing CS and
HMP methods.
2.
MATERIAL AND METHODS
2.1 Animals
Male Wistar rats weighing 250-300 g were used in all experiments. The rats were
allowed access to a standard laboratory diet and water ad libitum freely prior to the
experiment and received care in compliance with international regulations. The ethical
Committee of School of Biochemistry and Pharmaceutical Sciences, National
University of Rosario approved animal protocols.
2.2 Solutions
The composition of Bretschneider solution (HTK – Brestchneider, Franz Kholer
Chemie GMBH, Germany) was described in (Pokorny H., 2004). The BGP-HMP
solution (Pascucci F., 2012) was prepared in our laboratory, it has the following
composition (mM): 100 Na+ gluconate, 7 K+ gluconate, 20 Sucrose, 30 BES, 2.5
KH2PO4, 0.03 Polyethylene Glycol 35 kDA, 5 MgSO4, 3 Glutathione, 5 Adenosine, 15
Glycine, 0.25 mg/mL Streptomycin and 10 (U/mL) Penicillin G. Osm: 297  4
mOsm/Kg H2O, pH= 7.40, Na+ = 120  2 and K+ = 10  1 mEq/L. BES: [N, N-bis (2hydroxyethyl)-2-aminoethanesulfonic acid]. The HTK solution was bubbled with
Medicinal Synthetic Air at the selected temperatures at a gas flow of 600 mL/min
during 20 min before use. In contrast, the BGP-HMP solution was saturated with room
air.
2.3 The hypothermic perfusion system
A photograph of the recirculating perfusion system is shown below. It consists in a
reservoir that contains the perfusion solution, a bubbler to deliver air and a sample port,
a masterflex peristaltic pump, a flowmeter, a nylon filter and a bubble trap that connects
the inflow fluid with a three-way stopcock used to communicate the liver inflow and the
hydrostatic manometer. The liver was set up floating in a thermostatic chamber that
driven the perfusate fluid exiting from the liver to the reservoir. The perfusion portal
pressure could be measured with a hydrostatic manometer. A thermocouple and a pH
glass electrode could also be inserted into the thermostatic chamber to measure the fluid
temperature and pH during device operation. They are all assembled in a thermostated
chamber in which the temperature could be regulated at 5.0  0.5 or 10.0  0.5 C.
2.4 Non-heart beating donation rat model, HMP and Cold storage
Male Wistar rats, weighing between 300 and 320 g, were anaesthetized by
intraperitoneal injection of chloral hydrate (Parafarm, Argentina) (500 mg/Kg). The
abdomen was opened by midline incision and the liver freed from all ligamentous
attachments. In brief, the bile duct was cannulated with a PE-50 catheter (Intramedic,
USA) and the portal vein was cannulated with a 14 G catheter (Abbocath) but perfusion
was stalled until later. Cardiac arrest was induced by intravenous injection of
concentrated potassium chloride solution (2 M) (García Valdecasas J.C., 1998). Also,
150 U of heparin were injected into the femoral vein. 45 minutes later, perfusion with
buffer Krebs-Henseleit started and the suprahepatic inferior vena cava was cannulated
with steel tubing (internal diameter 3 mm). Finally, the liver was removed and aconnected on a recirculating perfusion system maintained at 5 ºC and perfused with 250
mL of BGP-HMP solution or b- flushed with 20 mL of cold Bretschneider solution
(HTK) and then stored (0-4°C) in the same solution up to 24 h. HMP was performed up
to 24 h at a constant pressure of 40 mmH2O (equivalent to 25% of the normothermic
portal pressure) and a flow of 0.23 mL/min.gliver. The perfusion solution was air
equilibrated during the HMP. After 24 h of HMP or CS, these livers were reperfused at
37C with Krebs-Henseleit Dextran solution (Balaban C.L., 2011). In both procedures,
portal pressure and flow were measured and the intrahepatic resistance (IR) was
calculated. Perfusate pH oscillations and enzymes (LDH, AST, and ALT) activities in
perfusate were evaluated during normothermic reperfusion. Bile was collected in
preweighed tubes every 15 min and bile flow was estimated gravimetrically assuming
the bile density to be equal to water and expressed as L.min-1.g liver-1.
The ability of the liver to extract oxygen was also measured as a function of
perfusate flow. Oxygen consumption was determined at 30 min intervals from samples
of liver inflow and outflow perfusates using an YSI model 5300 Biological Oxygen
Monitor (Yellow Springs, Ohio, USA), equipped with a Clark–type sensor (YSI 5331
oxygen probe, Yellow Springs, Ohio, USA). At the end of perfusion, the livers were
collected, weighed and cut in small blocks (4 mm thickness) for histological studies.
The livers were fixed in 10 % formaldehyde in PBS (pH=7.40) and histologically
processed for paraffin embedding. Slices of 5 µm thick were cut, stained with H&E and
analyzed with bright light microscopy. Also, a biopsy was frozen for glycogen content
determinations.
2.5 Liver Glycogen content after reperfusion
The technique of preservation and the composition of the preservation solution
might contribute in different way to restore the energy during reperfusion. Liver
glycogen was determined in biopsy specimens taken after perfusion. Glycogen content
was calculated from the amount of glucose released by treatment of homogenized tissue
with α-amyloglucosidase following the determination of free glucose content (Carr R.S,
1984). Glycogen content was expressed as mg glycogen/g.liver tissue and determined at
the end of the perfusion period.
2.6 Histology
Hematoxilyn and Eosin staining was used to judge the morphological integrity of
the parenchyma. We have analyzed the morphology of the sinusoid such as sinusoidal
dilatation and the amount of cell damages. Two operators independently, but working
concurrently, examined by light microscopy at least fifteen microscopic fields of three
different biopsies. The fields were chosen randomly within the liver parenchyma from
each experimental group and the images were captured by a Cannon Power Shot A 620
digital camera. Images were analyzed with Image J software (NIH) using the pointcounting method with grids that contained 80 points, placed over the images (Pizarro M.
D., 2009). Every point was analyzed and the field under the point was counted, looking
for the occurrence of sinusoidal dilatation, vacuolization and sinusoidal endothelial cell
detachment, and expressed as percentage of each observation.
2.7 Calculations
(1)
(2)
(3)
(4)
(5)
IR (mm Hg min g liver / mL) = portal pressure (10.3 mmHg) / portal flow (mL /
min.g liver).
LDH (U/L.g liver)= [LDH] / liver weight
AST (U/L.g liver)= [AST] / liver weight
ALT (U/L.g liver)= [AST] / liver weight
Oxygen consumption (µmol O2 / min. g liver) = (Cin–Cout) / portal flow (mL /
min. g liver) where Cin and Cout are the oxygen concentration in the inflow and
outflow, respectively. CO2 (µmol O2 / mL) = pO2 (kPa). SO2 37°C (µmol O2 /
mL.kPa). Where SO237°C is the oxygen solubility in water at 37°C. SO237°C =
0.01056 µmol O2 / mL.kPa (Gnaiger E., 2004).
2.8 Statistical analysis
Mean ± SD are presented. Statistical significance of the differences between
values was assessed by Analysis of variance (ANOVA). Student t test was applied
for Glycogen data analysis. Also, for the morphometric analysis ANOVA test was
performed. A P value less than 0.05 was considered statistically significant.
3.
RESULTS AND DISCUSSION
3.1 Perfusion flow and intrahepatic resistance
Fig. 1 shows the evolution of perfusion flow and intrahepatic resistance during
90 min of normothermic reperfusion. Intrahepatic vascular resistance was calculated as
a function of perfusion flow at a constant perfusion pressure of 10.3 mmHg. As
indicated in Fig.1A, there were statistical differences between the portal flow of HMP
livers respect to CS livers during the experiment (p<0.05). Fig 1B, shows the evolution
of IR during the perfusion period, a statistical difference was found between HMP
group and CS at 75 and 90 minutes.
A
B
Figure 1A. Perfusion flow and 1B. Intrahepatic resistance during 90 min of reperfusion. Livers were
perfused for 90 min at constant pressure, and the intrahepatic resistance was calculated and expressed as
(mmHg.min.g liver/mL) in 5 separate experiments in each analyzed group. (*p<0.05 compared with CS)
3.2 Bile flow production
Bile production was impaired in the non-heart beating donation rat model,
showed in Fig. 2. Bile flow was diminished in the HMP group respect to the CS group
(*p<0.05). Additionally the mean bile flow in the HMP group was not different along
the 90 min of perfusion. This phenomenon could be due to the liver washing of the bile
salts during the 24 hours of HMP.
Figure 2. Bile flow production. The bile was collected in pre-weighed tubes every 15 min and the bile
was estimated gravimetrically and expressed as μL/min.g.liver in 5 separate experiments in each analyzed
group. (*p<0.05 compared with HMP group)
3.3 Oxygen consumption rate
Figure 3 shows how HMP and CS affect metabolic activity of livers in terms of
O2 consumption. HMP livers have shown statistically significant higher rates of
consumption respect to the CS livers.
Figure 3. Oxygen consumption. Oxygen delivery and consumption was determined at 30 min intervals
from samples of liver inflow and outflow perfusate and expressed as μmol O2/min.g liver in 5 separate
experiments in each analyzed group (*p= 0.0339).
3.4 LDH, AST and ALT release during reperfusion
The accumulation of LDH, AST and ALT enzymes in the perfusate during 90
min of perfusion, indicates the level of cell injury. After 24 hs of CS or HMP the liver
generates a high release of LDH, AST and ALT during the reperfusion experiments.
Statistical differences was not shown between both groups. After 90 min from the begin
of perfusion the LDH concentration was: CS: 25.01 ± 9.31; HMP: 26.74 ± 11.78 (U/L.g
liver), AST concentration was: CS: 4.66 ± 0.67; HMP: 5.07 ± 2.57 and ALT
concentration was: CS: 1.98 ± 0.31; HMP: 2.25 ± 1.38, (U/L.g liver)
3.5 Glycogen content after cold preservation and 90 min of reperfusion
Simple cold storage of livers for transplantation activates glycolysis due to lack
of oxygen. Energy derived from glycolysis may be critical for cell survival and liver cell
death may occur once glycolysis is inhibited in the liver due to accumulation of end
products or lack of substrates (glycogen) (Quintana A.B., 2005). In figure 4 it is
possible to appreciate the variation in glycogen content after CS and HMP.
Paradoxically glycogen depletion was critical during reperfusion after 24 hs of HMP, in
comparison with CS (*p<0.0343). The glycogen depletion probably was due to the
continuous hypothermic perfusion with an oxygenated solution: HMP-BGP solution,
which it does not contain energetic substrates as glucose to replenish the glycogen
reserves.
Figure 4. Glycogen
content. Liver
glycogen was
determined in biopsy
specimens taken
after reperfusion and
it was expressed as
mg glycogen/g.liver
tissue in 5 separate
experiments in each
analyzed group.
3.6 Liver histology
The histological observations were semi quantified by a point-counting method
consisting on 80 point grids superposed over the images, points that were in contact
with injured morphology were counted and divided by total number of points on the
grid (n=50 for each group). The results are expressed in figure 5.
Sinusoidal dilatation: the widening of sinusoids was more evident for group HMP, than
CS group. HMP had a slightly grade of sinusoidal dilatation (p = 0.06). See Fig 5A.
Vacuolization: an elevated incidence of vacuoles in the hepatocyte cytoplasm as a sign
of cell damage was not present in both groups (data not shown).
Endothelial cell injury: both preserved groups of livers presented differences in
endothelial cell injury levels (Fig. 5B). CS group shows a statistically difference from
HMP group (p=0.03).
5A
5B
Figure 5. Hematoxilyn/Eosin Staining pictures. 40X zoom images from H&E stained biopsies. Green
arrows show dilatated sinusoids (5A) and endothelial cell injury (5B). Morphometric analysis.
Percentage values account for frequency of injured morphology event in each microscopic field.
CONCLUSION
In conclusion, the livers preserved with both preservations methods showed the
same membrane integrity (with respect to enzymes release) and they are able to keep an
active metabolism. However, livers preserved by HMP showed better hemodynamic
response and were able to enhance O2 consumption capacity. However, both systems of
preservation for livers could be suitable to be applied in different circumstances to get a
better use of livers from NHBD to increase the pool of available organs.
ACKNOWLEDGEMENTS
This work was funded by grant PIP-1208 from CONICET, grant - Prot
19096/PT Regione Autonoma Friuli-Venezia Giulia, Italy and grant 1BIO176 from
UNR.
REFERENCES
Balaban C.L., Rodriguez J.V. and Guibert E.E.. (2011) "Delivery of the bioactive gas H2S during cold
preservation of rat liver. Effects on hepatic function in an ex-vivo model". Artificial Organs, 35, 508-515
Carnevale M. E., Balaban C.L., Guibert E.E., Bottai H., Rodríguez J. V.. (2012) "Evaluation of a new
solution for Hypothermic Machine Perfusion (HMP) of the liver. II - A study in the perfused rat liver in
vitro". Cryobiology (in press).
Carr R.S., Neff J.M.. (1984) “Quantitative semi-automated enzymatic assay for tissue Glycogen”. Comp.
Biochem. Physiol, 77, 447–449.
Fuller BJ, Lee CY (2007): Hypothermic perfusion preservation: the future of organ preservation
revisited? Cryobiology 54:129–145.
García Valdecasas J.C., Tabet J., Valero R., Taurá P., Rull R., García F, Montserrat E., González F., Ordi
J., Beltran J., López Boado M., Deulofeu R., Angás J., Cifuentes A., Visa J. (1998) "Liver conditioning
after cardiac arrest: the use of normothermic recirculation in an experimental animal model". Transpl
Int, 11, 424-432.
Gnaiger E. Oxygen solubility in experimental media. Mitochondrial Physiology Network (2001–2004)
6:1–6. Available at: http://www.oroboros.at/index.php?oxygen-solubility.
Pascucci F., Carnevale M.E., Balaban C.L., Mamprin M.E., Guibert E.E, Rodríguez J.V.. (2012)
"Evaluation of a new solution for Hypothermic Machine Perfusion (HMP) of the liver. I - Composition
and physicochemical parameters". Cryobiology (in press).
Pizarro M.D., Rodriguez J.V., Mamprin M.E., Fuller B.J., Mann B.E., Motterlini R., Guibert E.E.. (2009)
"Protective effects of a carbon monoxide-releasing molecule (CORM-3) during hepatic cold
preservation". Cryobiology 58, 248-55.
Pokorny H., Masoul-Rockenschaub S., Langer F. et al. (2004) "Histidine-tryptophan-ketoglutarate
solution for organ preservation in human liver transplantation-a prospective multi-centre observation
study". Transplant. Int. 17, 256-260.
Quintana A.B.;. Guibert E. E; Rodríguez J.V.. (2005) “Effect of cold preservation/reperfusion on
glycogen content of liver. Concise review”. Annals of Hepatology, 4(1), 25-31.
Rahmatullah M. and Boyde T.R. (1980), “Improvements in the determination of urea using diacetyl
monoxime; methods with and without deproteinisation”, Clin.Chim.Acta 107 3-9.
Rodríguez J.V., Guibert E.E., Quintana A.B., Scandizzi A.L., Almada L.L. (1999) “Role of sodium
nitroprusside in the improvement of rat liver preservation in University of Wisconsin solution: a study in
the isolated perfused liver model”. J. Surg. Res. , 87, 201–208
Van der Plaats A, ’t Hart NA, Verkerke GJ, Leuvenink HGD, Ploeg RJ, Rakhorst G (2004): Hypothermic
machine preservation in liver transplantation revisited: concepts and criteria in the new millennium.
Ann Biomed Eng 32:623–631.