Comparison of Celsior and UW Solution in Experimental Pancreas Preservation

Journal of Surgical Research 105, 173–180 (2002) doi:10.1006/jsre.2002.6408

Comparison of Celsior and UW Solution in Experimental Pancreas Preservation Dirk Uhlmann,* Barbara Armann,* Stefan Ludwig,* Evelyn Escher,* Uta-Carolin Pietsch,† Andrea Tannapfel,‡ Daniel Teupser,§ Johann Hauss,* and Helmut Witzigmann* *2nd Department of Surgery, †Department of Anesthesiology, ‡Institute of Pathology, and §Institute of Laboratory Medicine, Clinical Chemistry and Molecular Diagnostics, University of Leipzig, 04103 Leipzig, Germany Submitted for publication October 4, 2001; published online July 8, 2002

Background. The University of Wisconsin solution (UW) is the gold standard for pancreas preservation. Celsior (CEL) was formulated specifically for heart preservation. Recently, experimental and clinical experience has been reported on the application of CEL to abdominal organs. In this animal study, pancreas preservation with CEL was compared with that in UW solution. Patients and Materials. Heterotopic, allogeneic pancreaticoduodenal transplantation was performed in female Go¨ttingen Minipigs (n ⴝ 12 donors, n ⴝ 12 recipients). The grafts were flushed and stored for 6 h at 4°C in UW or CEL. The recipients were randomized into two groups receiving either UW (n ⴝ 6)- or CEL (n ⴝ 6)-preserved grafts with a follow-up of 5 days. Blood flow (laser Doppler), partial oxygen tension, histological changes, endothelin-1 (plasma, immunohistochemistry), lipase, amylase, trypsinogen activation peptide, and C-reactive protein (CRP) were measured. Results. Partial oxygen tension was lower in the CEL group (P < 0.05). However, blood flow did not differ between UW- and CEL-preserved organs. The histomorphologic analysis of the pancreatic grafts revealed significantly less edema in the UW-preserved organs. Serum levels of amylase, lipase, CRP, and TAP taken from the central venous blood were comparable in the two groups, except for higher amylase values 36 h after reperfusion in the CEL group compared to the UW group (P < 0.05). Likewise, TAP taken from the portal venous effluent of the graft was found to be higher in the CEL group than in UW (P < 0.05). Endothelin-1 serum levels rose significantly during reperfusion without differences between the two groups. ET-1 immunohistochemistry revealed increased local ET-1 during reperfusion in all grafts. However, the ET-1 immunostaining in the CEL group

was more pronounced than that in the UW group (P < 0.05). Conclusions. Our results suggest that CEL solution is not as effective in preventing pancreatic ischemia/ reperfusion damage as the standard UW solution in experimental pancreas transplantation. Increased ET-1 immunostaining and reduced p tiO 2 in the CEL group indicate increased microcirculatory damage in the CEL group. © 2002 Elsevier Science (USA)

INTRODUCTION

Simultaneous pancreas/kidney transplantation is accepted as standard therapy in the treatment of patients with type I diabetes mellitus and end stage renal disease by the American Diabetes Association [1]. The introduction of the University of Wisconsin solution (UW) into the clinical pancreas and kidney programs in 1987 has allowed marked prolongation of cold ischemic time without any obvious deleterious effects on immediate pancreas or kidney function [2]. To date, no changes were made in the formulation of the commercial solution despite experimental evidence that some of the components of this complex solution were of no demonstrable benefit in terms of kidney, liver, or pancreas preservation [3]. In particular, the omission of hydroxyethyl starch [4] and reversing the Na:K ratio [3] appears to have no detrimental effect. It improves the flushing rate, presumably by reducing the viscosity and the vasoconstrictive property of UW. However, the rationale for a universal preservation solution has prevailed and no new formulations based on UW have been put on the market. The application of UW to clinical heart preservation is less widespread [5] due to a concern about the high concentration of potassium

173

0022-4804/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.

174

JOURNAL OF SURGICAL RESEARCH: VOL. 105, NO. 2, JUNE 15, 2002

and the lack of calcium and despite experimental [6] and clinical [7] evidence of effective heart preservation. Celsior (CEL) was formulated specifically for heart preservation [8] and its composition reflects the putative shortcomings of UW: hydroxyethyl starch was omitted, the potassium concentration was decreased, calcium was included, and the magnesium concentration was increased. Histidine was added to improve buffering, offsetting the removal of phosphate buffer as well as scavenging for toxic free radicals. Mannitol replaced raffinose. It has been suggested that CEL is equivalent or superior to UW for cardiac preservation [9]. Also in rat lung [10] and pig liver [11] preservation models and in a prospective multicenter randomized study of kidney preservation [12] Celsior provides results equal to or better than those with UW solution. Recently, a preliminary report has suggested that CEL is also effective for pancreas preservation. In a pancreas segemental autotransplantation model in pigs, Celsior was found to be an effective alternative to the established UW preservation solution [13]. For these reasons, in this study, porcine pancreas allografts, preserved by cold flushing and storage in either UW or CEL solution, were compared with regard to microcirculatory disorders, histologic changes, and biochemical alterations after reperfusion. MATERIALS AND METHODS

Experimental Protocol The investigation was performed in accordance with the German legislation on protection of animals and the experiments were approved by the Committee on Animal Care (Regierungspra¨ sidium Leipzig, Germany; No. 02/00). Twenty-four female Go¨ ttingen Minipigs (Ellegaard Go¨ ttingen Minipigs ApS, Denmark) weighing 20 –25 kg were used as either donors (n ⫽ 12) or recipients (n ⫽ 12). The recipients were randomized into two groups: in group 1 (n ⫽ 6) the pancreatic grafts were preserved in UW solution (Viaspan, DuPont Pharma GmbH, Bad Homburg, Germany), and in group 2 (n ⫽ 6) in Celsior solution (Sangstat Medical, Menlo Park, CA). On day 5 after surgery, the animals underwent relaparotomy to obtain tissue specimens for histologic evaluation of graft injury and were subsequently sacrificed.

Surgical Procedure and Immunosuppression Azaperone (15 mg/kg iv; Stresnil, Cilag-Janssen, Neuss, Germany), atropine (0.2 mg/kg iv; Atropinsulfat, Braun, Melsungen, Germany), and ketanest (3 mg/kg iv; Ketamin, Ratiopharm, Ulm, Germany) were administered for premedication. The anesthesia was induced with thiopental (8 mg/kg iv; Trapanal, Byk Gulden, Konstanz, Germany) and maintained with isoflurane (Forene, Abbott, Wiesbaden, Germany) and fentanyl (Fentanyl, Janssen, Neuss, Germany). Ventilation was performed with a oxygen/air mixture (respirator “Julian,” Draegerwerk AG, Luebeck, Germany) with a F iO 2 of 40%. A composite graft, including the whole pancreas and part of the duodenum, was harvested from the donor pig. Immediately after harvesting, the graft was perfused with either UW or CEL solution via the aortic patch until clear fluid returned from the portal vein. A maximum perfusion pressure of 100 mm Hg was used. The graft was prepared in cold saline solution at ⫹4°C and stored for 6 h in either

UW or CEL solution at 4°C. Transplantation of the pancreas was essentially performed according to the technique described by Ka¨ lle´ n et al. [14]. Immunosuppression consisted of 4 mg/kg/day of cyclosporin A (Sandimmun, Novartis, Nuernberg, Germany) and 1 mg/ kg/day of prednisolone (Prednisolut, Jenapharm, Jena, Germany), Beginning on the first postoperative day. The cyclosporin A dose was adjusted to achieve a blood trough concentration of 150 –200 ng/ml.

Partial Oxygen Tension in the Tissue Continuous measurement of the partial oxygen tension in the pancreatic tissue (p tiO 2) was performed by implantation a Clarketype electrode (Licox, Gesellschaft fu¨ r medizinische Sondentechnik mbH, Kiel-Mielkendorf, Germany) into the corpus of the pancreas. Tissue p tiO 2 was measured in the donor before organ harvesting for 30 min and in the recipient for 1 h beginning immediately before the onset of reperfusion.

Laser Doppler Flow Measurement Blood flow was measured by placing the Doppler flow probe (DP 1) at three different points of the pancreas and duodenum: (a) the tail of the pancreas, (b) the head of the pancreas, and (c) the duodenum. The measurements were performed in the donor before organ harvesting and in the recipient 30 and 90 min after reperfusion and were monitored on a Moor Instruments DRT4 Monitor (Moor Instruments, Devon, United Kingdom). The principle of the probe is that light generated by a laser diode (780-nm wavelength with maximal emission energy of 1.0 mW) penetrates the tissue, where it is reflected by circulating blood cells. Analog laser Doppler flow signals were digitalized and processed on a personal computer with the DRTSOFT V2.9 software (Moor Instruments). Blood flow was recorded for at least 30 s after a stable signal was obtained. Postsampling data processing included mean blood flow (given in arbitrary units, aU) and pulse wave analysis with integral under the curve. For integral estimation, the mean of the pulse waves within the 30-s sampling period was calculated.

Histology Specimens were taken for light microscopy from the corpus of the donor pancreas before organ procurement and from the recipient before reperfusion, 1 h after reperfusion, and on the fifth postoperative day. Development of interstitial edema was quantitatively assessed by planimetric analysis (area measurement, given in a percentage of the complete area under investigation) using a computer-assisted image analysis system (CapImage; Zeintl, Heidelberg, Germany) [15]. Leukocyte infiltration into tissue was analyzed histomorphologically by counting the number of polymorphonuclear cells in 50 high-power fields (HPF). Thereafter, the numbers were scored using a classification system from 0 (no infiltration) to 3 (severe). Likewise, acinar necrosis and hemorrhagic injury were documented using a scoring system from 0 (no pathologic changes) to 3 (severe necrosis and severe erythrocyte infiltration, respectively). Scoring was performed in a blinded fashion.

Immunohistochemical Analysis of Endothelin-1 For immumohistochemical analysis of endothelin-1, the material was routinely fixed in 4% formaldehyde solution and embedded in paraffin. Sections (4 ␮m thick) were cut, dewaxed in xylene, and then rehydrated. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in PBS for 20 min. After a short rinse with phosphate-buffered saline, the sections were boiled in citrate buffer for 15 min in a microwave (600 Wt). After cooling, the sections were covered with normal goat serum for 20 min and then incubated with the primary antibody against endothelin-1 (DPC Biermann, Bad Nauheim, Germany). Thereafter, the sections were washed with phosphate-buffered saline, incubated with biotinylated goat antirab-

UHLMANN ET AL.: CEL AND UW SOLUTIONS IN PANCREAS PRESERVATION

175

bit immunglobulin G (BioGenex, Hamburg, Germany) for 30 min, and covered with peroxidase-conjugated streptavidin (BioGenex, Hamburg, Germany). The peroxidase reaction was allowed to proceed for 5 min, with 3-amino-9-ethylcarbazole solution as substrate. Slides were counterstained with Meyer’s hemalaun and finally mounted. The intensity of intracytoplasmatic Endothelin -1 (ET-1) immunostaining in pancreatic tissues was graded semiquantitatively from 0 (no staining) to 3 (maximal intensity of staining). The number of positive cells was counted and scored semiquantitatively from 0 to 4 (0 ⫽ 0; 1 ⫽ 1–30; 2 ⫽ 31– 60; 3 ⫽ 61–90; 4 ⫽ ⬎ 90 stained cells in 1 of 10 observed areas). Negative controls were included in each batch replacing the primary antibody. The slides were examined and scored independently by three investigators (B.A., A.T., D.U.) who were blinded to clinical and pathological information.

Serum Lipase and Amylase and C-Reactive Protein (CRP) Serum lipase, amylase, and CRP were measured on an automatic analyser (Hitachi 917) by photometry in the recipient before surgery (baseline), 30 min and 1, 2, 6, and 12 h after reperfusion, and on postoperative days 2 and 5.

Plasma Endothelin-1 ET-1 levels from central venous blood of the recipient were determined in EDTA-plasma before surgery (baseline), before reperfusion, and 45 min and 2, 6, and 12 h after reperfusion using the QuantiGlo ET-1 Chemiluminescent Immunoassay (R&D Systems, Nivelles, Belgium), a 5-h solid phase ELISA. It contains synthetic human ET-1 and antibodies raised against the synthetic factor.

Trypsinogen Activation Peptide TAP values were determined with a quantitative competitive enzyme-linked immunosorbent assay [16, 17], based on the competition between free and immobilized peptide binding to an affinitypurified rabbit antibody to TAP (TAPKIT, Biotrin Int. Ltd., Dublin, Ireland). TAP was measured in the recipient from central venous blood before surgery (baseline) and 45 min and 12 h after reperfusion and from the portal venous blood of the graft 45 min after reperfusion.

Malondialdehyde (MDA) MDA was determined using the thiobarbituric acid method by photometry in the recipient before surgery (baseline), 30 min and 1, 2, 6, and 12 h after reperfusion, and on postoperative days 2 and 5.

Statistical analysis Data were expressed as means ⫾ SEM. Mean were compared using the Mann–Whitney U test. P values ⬍0.05 were considered significant.

RESULTS

Tissue Oxygenation (Fig. 1) The mean p tiO 2 levels of the donor groups were not significantly different (UW, 53.8 ⫾ 11.7 mm Hg, and CEL, 49.6 ⫾ 10.5 mm Hg). In the recipients, the levels dropped at the end of cold ischemic time to 0 –1 mm Hg in both groups. During reperfusion, the mean p tiO 2 level in the UW group increased 10 min after reperfu-

FIG. 1. p tiO 2, i.e., partial oxygen tension in the pancreatic grafts before ischemia in the donor and 10, 30, and 60 min after reperfusion in the recipients. Organ preservation with either UW (n ⫽ 6) or CEL (n ⫽ 6) for 6 h. Values are means ⫾ SEM, *P ⬍ 0.05 vs CEL preservation.

sion to 47.4 ⫾ 9.4 mm Hg but decreased again 30 min after reperfusion to 30.1 ⫾ 4.2 mm Hg. This decrease indicates a secondary hypoxic period. Thereafter, a slight decrease was found 60 min after reperfusion, with a subsequent plateau at 22.8 ⫾ 4.6 mm Hg. In the CEL group, the corresponding p tiO 2 values were found to be significantly lower 10 (39.9 ⫾ 10.9 mm Hg), 30 (19.2 ⫾ 5.5 mm Hg), and 60 (14.5 ⫾ 3.8 mm Hg) min after reperfusion compared to those of the UW group (P ⬍ 0.05). However, the mean p tiO 2 levels after reperfusion remained below the preischemic levels in both groups at all times. Laser Doppler Flow Measurement (Table 1) Integral under the curve. In the two donor groups, the integrals under the curve were highest in the duodenum, lower in the head, and lowest in the tail of the pancreas. The values were not significantly different between the UW and CEL group at any places measured. However, the integral showed significantly higher levels in the duodenum than in the head and tail in each group (P ⬍ 0.05). The values in the head of the pancreas did not differ from those in the tail in either group. At 30 min after reperfusion, the values at all estimated locations were significantly lower than those of the corresponding areas of the donors in both the UW and CEL recipient groups (P ⬍ 0.05). No significant differences were seen between the groups at this time. At 90 min after reperfusion, the values remained at a stable level in both groups. A significantly higher value was measured in the UW group than in the CEL group only in the duodenum.

176

JOURNAL OF SURGICAL RESEARCH: VOL. 105, NO. 2, JUNE 15, 2002

morphologic analysis of pancreatic grafts 1 h after reperfusion and at the fifth postoperative day revealed significantly less edema in the UW-preserved grafts (P ⬍ 0.05). No significant differences between the two groups were found with respect to leukocytic tissue infiltration, acinar necrosis, and hemorrhagic lesions. The histomorphologic alterations were found to be almost homogeneously distributed over the entire pancreas, i.e., the head, corpus, and tail of the gland. Immunohistochemical Analysis Immunohistochemical analysis of the donor pancreas before surgery revealed no or only a focal presence of ET-1 immunoreactivity in either group. At the end of the cold ischemic time, we observed a faint intracytoplasmatic immunoreactivity (grade 1) in vascular endothelial cells, in endocrine cells of Langerhans islets, and in exocrine acinar cells in the UW group. There was a slightly more pronounced immunoreactivity (grade 2) in these cells in the CEL group. Furthermore, the staining intensity varied from grade 1 (UW) to grade 2 (CEL). One hour after reperfusion a significantly stronger immunoreactivity in CELpreserved tissue (intensity, 3; number of cells, 3) compared to that in the UW group (intensity, 2, number of cells, 1) was observed. On the fifth postoperative day, no or only a very few positive cells were seen in either group. Lipase, Amylase, CRP

FIG. 2. Histological lesions of pancreatic tissue: (A) UW group, (B) CEL group. There was significantly less interstitial edema in the UW group. Hemateoxylin and eosin; ⫻200.

Serum lipase and amylase were within the normal range in both recipient groups before surgery (lipase, UW 0.15 ⫾ 0.04 ␮mol/L, CEL 0.21 ⫾ 0.04 ␮mol/L; amylase, UW 36.7 ⫾ 4.6 ␮mol/L, CEL 33.6 ⫾ 7.6 ␮mol/ L). After reperfusion, serum lipase and amylase dis-

Mean blood flow. Likewise the mean blood values (aU) in the donor revealed no significant differences between the UW and CEL groups, with highest levels in the duodenum and lowest in the tail of the pancreas; 30 min after reperfusion, the values were decreased at all measurement points in both groups compared to the preischemic levels. The decrease was found to be significant only in the duodenum of each group (P ⬍ 0.05). A further decrease of blood flow was seen 90 min after reperfusion in both groups. On comparison of the UW and CEL groups 30 and 90 min after reperfusion, a significant difference in blood flow was observed only in the duodenum in favor of the UW solution (P ⬍ 0.05). Histology (Table 2) Before organ procurement, there was no evidence of morphological damage in either donor group. Histo-

FIG. 3. Immunohistochemical demonstration of endothelin protein in acinar cells 1 h after reperfusion; ⫻400.

177

UHLMANN ET AL.: CEL AND UW SOLUTIONS IN PANCREAS PRESERVATION

TABLE 1 Laser Doppler Flow Measurement Duodenum UW Integral under the curve Donor Recipient 30 min after reperfusion 90 min after reperfusion Mean flow values (aU) Donor Recipient 30 min after reperfusion 90 min after reperfusion

Head of the pancreas

CEL

2809.3 ⫾

2636.6 ⫾

1239.4 ⫾ 1423.9 ⫾

P

UW

CEL

Tail of the pancreas P

UW

CEL

P

ns

1834.4 ⫾

1836.5 ⫾

ns

1455.4 ⫾

1544.1 ⫾

ns

1226.9 ⫾ 973.4 ⫾ 143.1

ns ⬍0.05

1147.5 ⫾ 1118.2 ⫾

1104.9 ⫾ 922.5 ⫾ 250.8

ns ns

1017.7 ⫾ 905.6 ⫾ 112.7

950.1 ⫾ 258.4 1005.9 ⫾

ns ns

193.2 ⫾ 26.3

165.1 ⫾ 33.0

ns

154.3 ⫾ 33.6

122.4 ⫾ 28.7

ns

95.1 ⫾ 15.0

86.5 ⫾ 18.5

ns

148.5 ⫾ 21.3 128.4 ⫾ 35.0

98.2 ⫾ 21.8 57.0 ⫾ 9.9

⬍0.05 ⬍0.05

93.8 ⫾ 15.4 72.5 ⫾ 13.3

105.4 ⫾ 19.2 71.4 ⫾ 12.9

ns ns

89.5 ⫾ 27.0 83.7 ⫾ 13.4

69.2 ⫾ 21.5 77.9 ⫾ 16.6

ns ns

Note. Mean blood flow values and integral under the curve levels from the donor and from the recipient 30 and 90 min after reperfusion. Values are means ⫾ SEM.

played a marked increase, with a maximum at 36 h in both groups (lipase, UW 17.1 ⫾ 5.0 ␮mol/L, CEL 26.8 ⫾ 7.8 ␮mol/L; amylase, UW 421.1 ⫾ 85.8 ␮mol/L, CEL 621.2 ⫾ 74.0 ␮mol/L). At this time, lipase values were comparable between the UW and CEL groups, whereas amylase levels revealed a significantly lower increase in the UW group (P ⬍ 0.05). On the second and fifth postoperative days, these parameters had normalized to baseline levels in both groups. CRP values (normal range ⬍ 0.5 mg/L) showed a first slight increase 12 h after reperfusion (UW 6.3 ⫾ 1.1 mg/L, CEL 7.5 ⫾ 2.1 mg/L) and remained elevated until the fifth postoperative day (UW 11.1 ⫾ 2.5 mg/L, CEL 13.1 ⫾ 2.3 mg/L). CRP values did not differ between the UW and CEL group at any time. Plasma Endothelin-1 Basal plasma endothelin-1 levels of 0.82 ⫾ 0.2 pg/ml were measured in the central venous blood of the recipients in both groups. Before reperfusion, the levels rose significantly to 1.76 ⫾ 0.63 pg/ml in the UW and 1.94 ⫾ 0.69 pg/ml in the CEL group. Reperfusion re-

sulted in a further increase in ET-1, with maximum levels 2 h after reperfuion (UW, 3.48 ⫾ 1.09 pg/ml; CEL, 3.78 ⫾ 0.96 pg/ml). ET-1 values did not differ between the two groups at any time of measurement. Trypsinogen Activation Peptide Plasma levels of TAP in the central venous blood of the recipients increased from 2.4 ⫾ 0.4 ng/ml (baseline) to 3.9 ⫾ 1.6 ng/ml (45 min after reperfusion) and to 17.8 ⫾ 11.9 ng/ml (12 h after reperfusion) in the UW group. The corresponding levels in the CEL group were 2.8 ⫾ 0.5, 4.6 ⫾ 2.2, and 19.2 ⫾ 13.0 ng/ml. There was no significant difference between the two groups. Serum levels of TAP taken from the portal venous effluent of the pancreas graft 45 min after reperfusion were lower in the UW-preserved organs (UW, 26.3 ⫾ 6.8 ng/ml, vs CEL 46.7 ⫾ 7.8 ng/ml; P ⬍ 0.05). Malondialdehyde Before surgery, MDA levels of 0.54 ⫾ 0.08 and 0.52 ⫾ 1.0 ␮mol/L were measured in the UW and CEL groups

TABLE 2 Histology of Pancreatic Grafts before and after Reperfusion After reperfusion Before reperfusion

Edema Leukocytes Acinar necrosis Hemorrhage

a

1h

5 days

UW

CEL

P

UW

CEL

P

UW

CEL

P

7.1 ⫾ 2.4 1.2 ⫾ 0.6 0.0 ⫾ 0.0 0.5 ⫾ 0.5

13.8 ⫾ 5.8 1.6 ⫾ 0.7 0.2 ⫾ 0.2 0.9 ⫾ 0.6

⬍0.05 ns ns ns

17.6 ⫾ 2.3 1.0 ⫾ 0.6 0.5 ⫾ 0.5 0.7 ⫾ 0.5

47.4 ⫾ 5.7 1.8 ⫾ 0.5 0.8 ⫾ 0.4 1.2 ⫾ 0.7

⬍0.05 ns ns ns

22.5 ⫾ 3.1 2.0 ⫾ 0.6 1.3 ⫾ 0.7 0.7 ⫾ 0.5

80.3 ⫾ 9.2 2.8 ⫾ 0.5 2.2 ⫾ 0.7 1.5 ⫾ 0.5

⬍0.05 ns ns ns

Note. Values of edema are represented as percentages of the complete area under investigation. Scores for necrosis, leukocyte infiltration, and hemorrhage range from 0 (no pathologic changes) to 3 (severe changes). ns, not significant. Values are means ⫾ SEM. a Assessment at the end of cold ischemic time.

178

JOURNAL OF SURGICAL RESEARCH: VOL. 105, NO. 2, JUNE 15, 2002

respectively. After reperfusion, MDA levels increased to a maximum at 6 h in both groups (UW, 0.69 ⫾ 0.09 ␮mol/L, CEL, 0.75 ⫾ 0.12 ␮mol/L, P not significant). DISCUSSION

The pancreas is highly susceptible to edematous swelling during preservation and reperfusion. This may cause impairment of the microcirculation and consequent acute pancreatitis after reperfusion. Severe graft pancreatitis is an infrequent but potentially fatal complication. The University of Wisconsin solution was mainly developed to lessen these adverse effects of ischemia and hypothermia on pancreas grafts [18] and resulted in considerable improvement of clinical results [2]. The UW solution can be regarded as the “gold standard” of pancreas preservation. The solution is viscous (4.8 cP at 1°C) and has an electrolyte composition close to that of intracellular fluid, with high potassium (135 mmol/L) and low sodium (30 mmol/L) levels at a pH of 7.4. The pH is maintained by using phosphate as a buffer. Magnesium, sulfate, and lactobionate are added as calcium chelators to stabilize the membrane potential. Hydroxyethyl starch (HES), lactobionate, and raffinose are nondiffusible, preventing the development of cellular edema during hypothermia. Adenosine contributes to regeneration of ATP. The use of allopurinol blocks both the breakdown of adenosine nucleotides and the generation of oxygen radicals. CEL preservation solution was formulated specifically for heart preservation [8]. This solution has been shown to improve heart preservation in isolated rat heart preparations and heterotopic rabbit heart transplants compared to standard cardioplegia and preservation solutions [8, 19]. Initial clinical experience confirmed that CEL is an effective solution for heart preservation [20]. CEL has not been found to be as effective as a colloid-free UW solution for long-term liver preservation in a rat model of transplantation [21]. A retrospective clinical study comparing UW with CEL preservation in liver transplantation comprising only a few patients showed no differences between these solutions [22]. In a pig renal autotransplantation model, CEL proved to be comparable to UW solution [23]. In a pancreas autotransplantation model in pigs, Celsior was found to be an alternative to the established UW preservation solution [13]. The same group compared CEL with the standard solution UW in this segmental autotransplantation model [24]. They found no differences between the two solutions. However, in this model of segmental autotransplantation there is no correlation to the clinical situation. The follow-up period was only 2 h and the exocrine secretion was drained into the peritoneal cavity, which would induce inflammation of the surrounding organs. A follow-up of

2 h does not allow evaluation of ischemia/reperfusion injury. The clinical standard today is whole pancreas transplantation with a duodenal segment. This study also provided no information on the influence of immunosuppression in pancreas transplantation. A marked impairment of the microcirculation in the early reperfusion period after pancreatic ischemia has been reported in animal models [25, 26] and in human pancreas grafts [27]. Edematous graft swelling after reperfusion plays a central role in the development of microcirculatory disturbances and subsequent graft pancreatitis. The main consideration in pancreas preservation is to avoid edema. Capillary perfusion failure was confirmed in our study by a secondary decrease of the pancreatic p tiO 2 (30 min after reperfusion) after an initial peak in both groups. However, tissue oxygenation was significantly decreased in the CEL group compared to preservation with UW solution during the entire period of p tiO 2 measurement. In contrast to tissue oxygenation, microvascular perfusion, assessed by blood flow and the integral under the curve using laser Doppler flowmetry, does not differ in CEL- and UW-preserved grafts. This may be because reperfusion of the supplying vessels and precapillary sphincters are not influenced by the two preservation solutions or because they are influenced in the same way. The blood flow tended to be higher in the head than in the tail of the pancreas, in accordance with the studies by Schilling et al. [28] and Studley et al. [29]. Of the biochemical parameters (lipase, amylase, CRP, TAP) measured in this study, only TAP in the portal effluent of the graft showed a significant difference in favor of UW solution. TAP is a useful parameter for assessing the severity of clinical and experimental pancreatitis [17]. Evaluation of the tissue changes showed that preservation with UW solution significantly reduced pancreatic tissue edema compared to CEL preservation. Endothelins are known to play an important role in preservation and reperfusion injury of the heart [30], lung [31], liver [32], kidney [33], and pancreas [34]. In our study, immunohistochemical staining revealed a more marked ET-1 expression in the CEL group. The immunohistological findings were not reflected by the plasma ET-1 levels. One can speculate that increased local ET-1 release is in part responsible for mirocirculatory disturbances followed by decreased p tiO 2 and increased edema formation, which are found to be more pronounced in the CEL group. Expression of endothelins and the corresponding receptors has been demonstrated for the exocrine pancreas [35]. Furthermore, it has been reported that ET-1 down-regulates pancreatic blood flow under physiological conditions [36]. Plusczyk et al. have shown that ET-1 mediates a deterioration of the pancreatic micro-

UHLMANN ET AL.: CEL AND UW SOLUTIONS IN PANCREAS PRESERVATION

circulation, which is similar to the microcirculatory failure found in sodium taurocholate-induced experimental pancreatitis. This is associated with severe edema of the pancreas and focal acinar cell necrosis, which are distinct characteristics of severe acute pancreatitis [37]. Microcirculatory impairment with subsequent ischemia is the determining factor for the escalation of acinar cell necrosis in the pathogenesis of acute pancreatitis [38]. There are some theoretical reasons why UW solution is more effective with regard to tissue edema and tissue oxygenation than CEL solution. A critical difference in the composition of CEL and UW with respect to edema development is the substitution of mannitol for raffinose. In the original development of UW using canine pancreas slices, it was shown that there was less edema in the presence of raffinose than mannitol [18]. In the same investigation, cold storage of the intact canine pancreas in a lactobionate–raffinose solution showed little tissue edema after isolated normothermic reperfusion [18]. Lactobionate is an anion that replaces chloride, to which to the cell is impermeable. Theoretically, lactobionate reduces cell swelling [39]. It is also present in CEL, albeit in a lower concentration than in UW, where it has also been suggested that it exerts a protective role as a free radical scavenger, as well as having impermeant properties [40]. Furthermore, the higher concentration of cell-permeable chloride ions in CEL vs UW solution (41.5 vs 10 mmol/L) may cause an increase in osmotic pressure and cell swelling. Ploeg showed that the addition of a colloid (hydroxyethyl starch) plays a crucial role in the development of edema [41]. It is not entirely clear why a colloid would be more important in pancreas preservation than in lung, kidney, or liver. UW has a number of pharmacological additives, whereas only reduced glutathione is added to CEL. There is some evidence that optimal pancreas preservation requires some of these additives. The pancreas cold preservation time we have chosen (6 h) is not comparable to the internationally reported cold ischemic times of 14 to 16 h. However, the aim of the middle German organ sharing network is to reach cold ischemic times of about 6 to 8 h. For this reason we used this time in our investigation A possible disadvantage of this study may be the fact that the recipient pancreas remained in situ. We made this choice for the following reasons: The major outcome following pancreas transplantation must be euglycemia. In three pigs not mentioned in this study, we performed pancreatectomy in the recipients. Technically, this operation is possible, and euglycemia was attained in two of the three animals. The third died of abdominal infection. Pancreatectomy entails a prolonged operation time of 90 to 120 min and an addi-

179

tional major trauma. The aim of this study was to investigate the influence of two different preservation solutions on pancreas ischemia/reperfusion injury with respect to histomorphological changes, microcirculatory parameters, and biochemical parameters. These parameters indirectly represent the functional status of the grafts and would be greatly influenced by an additional pancreatectomy. Taking into account all these considerations, we decided to leave the recipients’ pancreas in situ. In summary, our data suggest that CEL is not as effective as UW in preventing pancreatic ischemia/ reperfusion injury. ACKNOWLEDGMENTS We thank the team of the Medical Experimental Center at the University of Leipzig for professional assistance and Dr. J. Kratzsch (University of Leipzig) for determining plasma endothelin levels. This study was supported by a grant from the Else Kro¨ ner-Fresenius Stiftung and by a junior research fund of the Medical Faculty at the University of Leipzig.

REFERENCES 1. 2.

3.

4.

5.

6.

7.

8.

9.

American Diabetes Association. Pancreas transplantation for patients with diabetes mellitus. Diabetes Care 23: 112, 2000. Belzer, F. O., D’Alessandro, A. M., Hoffmann, R. M., Knechtle, S. J., Reed, A., Pirsch, J. D., Kalayoglu, M., and Sollinger, H. W. The use of UW solution in clinical transplantation. A 4-year experience. Ann. Surg. 215: 579 –583, 1992. Moen, J., Claesson, K., Pienaar, H., Lindell, S., Ploeg, R. J., McAnulty, J. F., Vreugdenhil, P., Southard, J. H., and Belzer, F. O. Preservation of dog liver, kidney, and pancreas using the Belzer-UW solution with a high-sodium and low-potassium content. Transplantation 47: 940, 1989. Howden, B. O., Jablonski, P., Thomas, A. C., Walls, K., Biguzas, M., Scott, D. F., Grossman, H., and Marshall, V. C. Liver preservation with UW solution. I. Evidence that hydroxyethyl starch is not essential. Transplantation 49: 869, 1990. Demmy, T. L., Biddle, J. S., Bennett, L. E., Walls, J. T., Schmaltz, R. A., and Curtis, J. J. Organ preservation solutions in heart transplantation-patterns of usage and related survival. Transplantation 63: 262, 1997. Aziz, S., Tada, Y., Jaffery, S., Mori, Y., Reichenbach, D. D., Gronka, R., Kushmerick, M., and Verrier, E. D. University of Wisconsin solution provides superior myocardial preservation compared with Stanford cardioplegic solution. J. Heart Lung Transplant. 13: 1099, 1994. Stringham, J. C., Love, R. B., Welter, D., Canver, C. C., and Mentzer, R. M. Impact of University of Wisconsin solution on clinical heart transplantation. A comparison with Stanford solution for extended preservation. Circulation 98(Suppl.): II 157, 1998. Menasche, P., Pradier, F., Grousset, C., Peynet, J., Mouas, C., Bloch, G., and Piwnica, A. Improved recovery of heart transplants with a specific kit of preservation solution. J. Thorac. Cardiovasc. Surg. 105: 353, 1993. Mohara, J., Morishita, Y., Takahashi, T., Oshima, K., Yamagishi, T., Takeyoshi, I., and Matsumoto, K. A comparative study of Celsior and University of Wisconsin solutions based on 12-hr preservation followed by transplantation in canine models. J. Heart Lung Transplant. 18: 1202, 1999.

180 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

JOURNAL OF SURGICAL RESEARCH: VOL. 105, NO. 2, JUNE 15, 2002 Roberts, R. F., Nishanian, G. P., Carey, J. N., Sakamaki, Y., Starnes, V. A., and Barr, M. L. A comparison of the new preservation solution Celsior to Euro-Collins and University of Wisconsin solutions in lung reperfusion injury. Transplantation 67: 152, 1999. Audet, M., Alexandre, E., Mustun, A., David, P., Chenard-Neu, M. P., Tiollier, J., Jaeck, D., Cinqualbre, J., Wolf, P., and Boudjema, K. Comparative evaluation of Celsior solution versus Viaspan in a pig liver transplantation model. Transplantation 71: 1731, 2001. Faenza, A., Catena, F., Nardo, B., Montalti, R., Capocasale, E., Busi, N., Boggi, U., Vistoli, F., Di Naro, A., Albertazzi, A., Mosca, F., and Cavallari, A. Kidney preservation with university of Wisconsin and Celsior solution: A prospective multicenter randomized study. Transplantation 72: 1274, 2001. Baldan, N., Parise, P., Furian, L., Savio, M. L., Valente, M. L., Calabrese, F., Venturini, R., Girotto, A., and Rigotti, P. Swine pancreas preservation with Celsior solution. Transplant. Proc. 32: 29, 2000. Kallen, R., Borgstrom, A., and Falt, K. Pancreatic enzymes in serum and urine as indicators of pancreatic allograft rejection in the pig. Transplantation 48: 376, 1989. Menger, M. D., Bonkhoff, H., and Vollmar, B. Ischemiareperfusion-induced pancreatic microvascular injury. An intravital fluorescence microscopic study in rats. Dig. Dis. Sci. 41: 823, 1996. Gudgeon, A. M., Heath, D. I., Hurley, P., Jehanli, A., Patel, G., Wilson, C., Shenkin, A., Austen, B. M., Imrie, C. W., and Hermon-Taylor, J. Trypsinogen activation peptides assay in the early prediction of severity of acute pancreatitis. Lancet 335: 4 – 8, 1990. Neoptolemos, J. P., Kemppainen, E. A., Mayer, J. M., Fitzpatrick, J. M., Raraty, M. G., Slavin, J., Beger, H. G., Hietaranta, A. J., and Puolakkainen, P. A. Early prediction of severity in acute pancreatitis by urinary trypsinogen activation peptide: A multicentre study. Lancet 355: 1955–1960, 2000. Wahlberg, J. A., Southard, J. H., and Belzer, F. O. Development of a cold storage solution for pancreas preservation. Cryobiology 23: 477, 1986. Menasche, P., Termignon, J. L., Pradier, F., Grousset, C., Mouas, C., Alberici, G., Weiss, M., Piwnica, A., and Bloch, G. Experimental evaluation of Celsior, a new heart preservation solution. Eur. J. Cardiothorac. Surg. 8: 207, 1994. Llosa, J. C., Rodriguez-Lambert, J. L., Naya, J. L., Gosalbez, F., and Valle, J. M. Celsior, a novel cardioplegic solution for arrest and storage in heart transplantation. Transplant. Proc. 32: 2589, 2000. Howden, B. O., and Jablonski P. Liver preservation: A comparison of celsior to colloid-free University of Wisconsin solution. Transplantation 70: 1140, 2000. Maggi, U., Caccamo, L., Gatti, S., Paone, G., Reggiani, P., Rossi, G., Latham, L., Vannelli, A., Melada, E., Brambilla, R., Damilano, I., Trezza, P., and Fassati, L. R. Celsior solution and clinical liver transplantation. Transplant. Proc. 32: 36, 2000. Baldan, N., Toffano, M., Cadrobbi, R., Codello, L., Calabrese, F., Bacelle, L., and Rigotti, P. Kidney preservation in pigs using celsior, a new organ preservation solution. Transplant. Proc. 29: 3539, 1997. Baldan, N., Rigotti, P., Furian, L., Valente, M. L., Calabrese, F., DiFilippo, L., Parise, P., Sarzo, G., Frison, L., and Ancona, E. Pancreas preservation with Celsior solution in a pig autotransplantation model: Comparative study with University of Wisconsin solution. Transplant. Proc. 33: 873, 2001. Mayer, H., Thies, J. C., Schmidt, J., Ryschich, E., Gebhard, M. M., Herfarth, C., and Klar, E. Characterization and reduc-

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39. 40.

41.

tion of ischemia/reperfusion-injury after experimental pancreas transplantation. J. Gastrointest. Surg. 3: 162, 1999. Benz, S., Schnabel, R., Morgenroth, K., Weber, H., and Hopt, U. T. The ischemia/reperfusion injury of the pancreas: A new animal model. J. Surg. Res. 75: 109, 1998. Benz, S., Pfeffer, F., Adam, U., Schareck, W., and Hopt, U. T. Impairment of pancreatic microcirculation in the early reperfusion period during simultaneous pancreas-kidney transplantation. Transplant. Int. 11(Suppl. 1): S433, 1998. Schilling, M. K., Redaelli, C., Reber, P. U., Friess, H., Signer, C., Stoupis, C., and Bu¨ chler, M. W. Microcirculation in chronic alcoholic pancreatitis: A laser Doppler flow study. Pancreas 19: 21, 1999. Studley, J. G., Mathie, R. T., and Blumgart, L. H. Regional pancreatic perfusion: An experimental study in the dog. Acta Chir. Scand. 152: 373, 1986. Szabo, G., Fazekas, L., Bahrle, S., MacDonald, D., Stumpf, N., Vahl, C. F., Hagl, S. Endothelin-A and -B antagonists protect myocardial and endothelial function after ischemia/reperfusion in a rat heart transplantation model. Cardiovasc. Res. 39: 683, 1998. Minamoto, K., Pinsky, D. J., Fujita, T., and Naka, Y. Timing of nitric oxide donor supplementation determines endothelin-1 regulation and quality of lung preservation for transplantation. Am. J. Respir. Cell. Mol. Biol. 26: 14, 2002. Kraus, T., Golling, M., Mehrabi, A., Fernandes, L., Angelescu, M., Gebhard, M. M., Herfarth, C., and Klar, E. Endothelin-1 and big-endothelin concentrations are elevated in liver graft tissue during cold storage and reperfusion. Eur. Surg. Res. 33: 1, 2001. Inman, S. R., Burns, T. E., Plott, W. K., Pomilee, R. A., Antonelli, J. A., and Lewis, R. M. Addition of an endothelin receptor antagonist to modified Belzer’s perfusion preservation solution mitigates the adverse effect of preretrieval warm ischemic injury on posttransplant glomerular filtration rate. Transplant. Proc. 33: 2975, 2001. Uhlmann, D., Ludwig, S., Escher, E., Armann, B., Gabel, G., Teupser, D., Tannapfel, A., Hauss, J., Witzigmann, H. Protective effect of a selective endothelin 1a receptor antagonist (BSF 208075) on graft pancreatitis in pig pancreas transplantation. Transplant. Proc. 33: 3732, 2001. Hildebrand, P., Mrozinski, J. E., Mantey, S. A., Patto, R. J., and Jensen, R. T. Pancreatic acini possess endothelin receptors whose internalization is regulated by PLC-activating agents. Am. J. Physiol. 264: G984, 1993. Takaori, K., Inoue, K., Kogire, M., Higashide, S., Tun, T., Aung, T., Doi, R., Fujii, N., and Tobe, T. Effects of endothelin on microcirculation of the pancreas. Life Sci. 51: 615, 1992. Plusczyk, T., Bersal, B., Westermann, S., Menger, M., and Feifel, G. ET-1 induces pancreatitis-like microvascular deterioration and acinar cell injury. J. Surg. Res. 85: 301, 1999. Plusczyk, T., Bauer, M., Marzi, I., Harbauer, G., and Feifel, G. Comparative effects of secretin (SEC) and cholecystokininoctapeptide (CCK-8) on pancreatic microcirculation. Dig. Dis. Sci. 40: 1199, 1995. Belzer, F. O., and Southard, J. H. Principles of solid-organ preservation by cold storage. Transplantation 45: 673, 1988. Menasche, P., Hricak, B., Pradier, F., Cheav, S. L., Grousset, C., Mouas, C., Alberici, G., Bloch, G., and Piwnica, A. Efficacy of lactobionate-enriched cardioplegic solution in preserving compliance of cold-stored heart transplants. J. Heart Lung Transplant. 12: 1053, 1993. Ploeg, R. J. Importance of a colloid in Canine Pancreas Preservation. ’s-Gravenhage: Pasmans Offsetdrukkerij, 1991.