|Home | About | Journals | Submit | Contact Us | Français|
Over the last decade, two major advances have made liver transplantation a widely applied and successful treatment of endstage liver disease. First, improvements in immunosuppression made possible by cyclosporine (1) and the recently introduced, even more powerful, immunosuppressant FK 506 (2) have reduced graft loss due to rejection. Second, a breakthrough in preservation of the liver for transplantation was recently achieved by the University of Wisconsin cold storage solution (UW)* (3). Thus, it is now possible to safely preserve a human liver for up to 24 hr (4). This has simplified the logistics in liver transplantation and also contributed to an improved quality of the organs that are transplanted. In spite of this, there is still a need for methods to further improve both the duration and the quality of preservation.
Numerous studies have demonstrated that cell and organ injury from ischemia can be reduced by avoiding the occurrence of elevated intracellular Ca2+ levels, e.g., by the use of Ca2+ antagonists (5-8). Previously, a protective effect was found when lidoflazine, a Ca2+ antagonist, was added to UW solution for preservation of the rat kidney (9). In the present study, we have investigated the effect of lidoflazine on liver preservation with UW solution, using the isolated perfused rat liver model.
The experiments were performed on male Lewis rats, weighing 240–340 g (Harlan, Indianapolis, IN). The animals had free access to standard pellet diet and tap water, and were not fasted before surgery. Anesthesia was with inhalational methoxyflu-rane (Metofane; Pitman-Moore, Inc., Washington Crossing, NJ).
Heparin, 300 IU, was given i.v. 10 min before harvesting. Ligatures were placed around the inferior vena cava, cranially to the renal vein and artery, and around the portal vein. For bile collection, polyethylene tubing (PE-10,1.D. 0.28 mm, O.D. 0.61 mm; Clay Adams, Parsippany, NJ) was inserted into the bile duct. Immediately after cannulation of the portal vein with a 16-gauge catheter (Critikon, Inc., Tampa, FL), this catheter was connected to an extension set and a 60-ml syringe (Abbott Laboratories, North Chicago, IL) filled with ice-cold preservation solution and immediately flushed in situ. The vena cava was transected distally to the ligature. Using this procedure, the warm ischemia time was always shorter than 20 sec. The liver was flushed with 30 ml at a hydrostatic pressure of 15–20 cm H20. During flushing, a 14-gauge catheter (Deseret Medical Inc., Sandy, UT) was inserted into the infrahepatic vena cava; the suprahepatic vena cava was transsected and the hepatectomy was completed. Each liver was weighed and stored in 100 ml of preservation solution for 72 hr, kept surrounded by ice.
The perfusions were performed in a thermostatically controlled Plexiglass cabinet (65×32×60 cm; Fischer Scientific, Pittsburgh, PA). Each liver was continuously perfused through the portal vein with 200 ml of recirculating buffer at 38°C for 90 min, using a Masterflex pump controller (Cole-Parmer Instrument, Chicago, IL) calibrated to maintain a flow rate of 3.5 ml/min/g. The perfusate was oxygenated while passing through oxygen-permeable tubing (T57111115-6,1.D. 1.5 mm; American Scientist, Warrendale, PA) inside a plastic container with a 95% O2 and 5% CO2 mixture at a flow rate of 6 ml/min, giving an oxygen tension of 480–550 mmHg on the arterial side of the system. Air bubbles were avoided by connecting a disposable Nylon filter (blood set 64; Abbott Laboratories) in line between the oxygenator and the inflow. The portal vein pressure was continuously monitored with an in-line manometer (Abbott Laboratories).
Krebs-Hanseleit bicarbonate solution containing 2% albumin (Sigma, St. Louis, MO) and 5 mM glucose (Abbott Laboratories) was used as the perfusate. The initial pH was adjusted to 7.38–7.42, but was not further adjusted during the perfusion, since the pH had earlier been found to decrease by not more than 0.10–0.15 (unpublished data). The preserved livers were weighed, then flushed with 3 ml of lactated Ringer's solution (Baxter Healthcare Corp., Deerfield, IL) at room temperature through the portal vein and immediately connected to the perfusion apparatus. The bile was collected in preweighed test tubes placed outside the cabinet to avoid dilution of the bile by condensed water, and the amount of bile produced was established by weighing the bile samples, assuming the bile density to be equal to water. Perfusate samples were taken every 30 min for measurement of aspartate aminotransferase (ASAT), alanine aminotransferase (ALAT), lactate dehydrogenase (LDH), and purine nucleotide phosphorylase (PNP). Perfusate samples were also taken for immediate measurements of pH, pO2, and pCO2 from the inflow as well as the outflow using a pH-blood gas meter (ABL 2-acid-base laboratory; Radiometer, Copenhagen, Denmark). After the perfusion, the liver was again weighed and the bile volume was measured. The release of the hepatocellular enzymes was determined with colorimetric methods (Sigma), using a Technicon RA-500 analyzer. Levels of PNP in the perfusate were measured, using the method of Hoffeeetal. (10).
Six different doses of lidoflazine (4-[4,4-bis(4-fluorophenyl)butyl]-N-(2,6-dimethylphenyl)-l-piperazineacetamide; Kabi Pharmacia, La Jolla, CA) were dissolved in UW solution (DuPont, Wilmington, DE) on the day of the harvest. The doses tested were 0.078 mg/L (1.6×10−7 M), 0.312 mg/L (6.3×10−7 M), 1.25 mg/L (2.5×l0−6 M), 5 mg/L (l.0×10−5 M), 20 mg/L (4.1×10−6 M), and 80 mg/L (1.6×10−4 M). The same solution was used for both flush-out and cold storage. In a control group, plain UW solution was used. Six experiments were performed in each group.
Results were calculated as mean ± SD. Statistical comparisons were performed using the Wilcoxon rank sum test. A P-value less than 0.05 was considered statistically significant.
As seen in Figure 1, in which the amount of ASAT released into the perfusate is presented in control livers with 0, 24, 48, and 72 hr of cold ischemia time, a clear-cut difference between the 48-hr and 72-hr preserved livers was seen. Since a certain amount of damage is required in a model, in which quantification of improved preservation is a major goal, 72 hr was chosen for the experiments. The other enzymes tested (ALAT, LDH, and PNP) behaved in a similar fashion in these control experiments, and the differences reached the same levels of significance (data not shown).
The levels of enzymes in the 72-hr preserved livers, released into the perfusate during reperfusion, are given in Figure 2. For simplicity, only the 90-min values are presented. The 30-min and 60-min values were proportionally lower than the corresponding 90-min values, and the statistical comparisons in the 30- and 60-min values reached the same levels of significance as these. The four lowest doses of iidoflazine (0.078–5 mg/L) were all effective in preventing enzyme release, as compared with when plain UW was used. The ASAT, LDH, and PNP 90-min values from these groups were all significantly lower than the values from the control group. When comparing the ALAT values, the 0.312- and 5-mg/L groups had significantly lower values. However, the two highest doses (20 and 80 mg/L, groups 5 and 6) were not significantly different in UW for any enzyme tested.
In contrast to our experiences from the isolated perfused rabbit liver (11–15), measurement of the bile production was technically difficult, since biliary sludge would sometimes obstruct the thin catheter in the groups of preserved livers, or kinking of the catheter would occur, which was not easily detected because of the low volumes produced. Therefore, a number of experiments in each group had to be excluded, and no valid conclusions could be drawn from this parameter only. Instead, the results of the enzyme analyses were regarded as more reliable indicators of liver cell injury after preservation in this rat model. However, as seen in previous studies (11,15), bile production in control livers padually decreased with increasing ischemia time (1.1±3.4 [n=6], 0.59±0.22 [n-5], 0.27±0.15 [n=3], and 0.19±0.06 [n=4] ml/90 min in the 0-, 24- 48-, and 72-hr preserved livers, respectively. Bile production after 72 hr appeared to improve with addition of at least moderate to higher doses of lidoflazine (0.17±0.06, [n=4], 0.21±0.04 [n=3], 0.25±0.09 [n=4], 0.25±0.06 [n=3], 0.25±0.08 [n=5], and 0.25±0.05 [n=4] ml/90 min for the 0.078-, 0.312-, 1.25-, 5-, 20-, and 80 mg/L groups, respectively, but due to the low number of evaluable experiments, no statistically significant differences could be calculated. The bile in 72-hr preserved livers was pale as compared with fresh or 24-hr preserved livers. The portal vein pressure was 11.0±2.2 cm H2O for all groups without any significant differences. One experiment in the 1.25-mg/L group was excluded because of air embolism. There was no difference in liver weights after flush-out between the groups. During preservation, all livers decreased in weight, which is a well-known phenomenon with UW (13). The addition of lidoflazine did not affect the liver weight during cold storage.
As indicated by the enzyme data, lidoflazine significantly improved the capacity of UW to preserve rat livers. Since the concentrations of the hepatocellular enzymes (ASAT, ALAT, and LDH) and the endothelial enzyme (PNP) were lower for the four lowest doses of lidoflazine tested, compared with UW alone, the present data suggest that lidoflazine improves the preservation of hepatocytes as well as liver endothelial cells. Preservation of the endothelial cells is of special importance, since injury to these cells can cause an aggravation of hepatocellular injury during reperfusion. Furthermore, the endothelium is the first target of allogenic blood after transplantation, and injury here may cause an increase in the immunogenicity of the liver (16).
It was found that lidoflazine increased the flush-out rate during harvesting, for all doses that also reduced the release of enzymes, except for the lowest dose (6.2±1.0, 6.8±0.4, 8.1±0.5 [P<0.01], 7.9±0.8 [(P<0.05], 9.1±1.2 [P<0.01], 6.8±0.4, and 6.8±1.3 ml/min for the 0, 0.078, 0.312, 1.25, 5, 20, and 80-mg lidoflazine/L groups, respectively; levels of significance compared with the control group are within brackets). This effect is probably mediated by vasodilation. Theoretically, vasodilation may be beneficial by contributing to a more rapid cooling of the organ and by improving the distribution of the preservation solution within the parenchyma. However, in a small animal, such as the rat, both mechanisms are probably of little importance, since the duration of the harvesting procedure is short, and the small organ can quickly achieve a low temperature, once surrounded by ice. Also, in cold storage of rat organs with a relatively large surface area/volume ratio, simple diffusion may be sufficient to supply a satisfactory distribution of the preservation solution. Thus, the present data suggest that the capacity of lidoflazine to improve liver preservation in this model is not caused mainly by dilatation of the vascular bed. However, for the preservation of larger organs such as the human liver, vasodilation may be an additional beneficial effect in a dose-related manner.
The two highest doses of lidoflazine used here had no ameliorating effect on the release of liver enzymes, nor did they increase the flush-out rate. Nevertheless, lidoflazine appears to have a wide therapeutic window, as it was effective in a concentration 10−7 M–10−5 M. It cannot be ruled out that an even lower dose than 10−7 could be effective, although this was never tested.
The results from this investigation are in agreement with those of a previous study, in which lidoflazine was found to improve the preservation of rat kidneys (9).
We conclude that lidoflazine improves the quality of liver preservation with UW solution in this in vitro model. However, further evaluation of this drug in a transplant model is required before the clinical usefulness can be determined.
Lidoflazine was generously donated by Kabi Pharmacia, La Jolla, CA. The authors gratefully acknowledge Dr. Rene Duquesnoy for general support, and James Snyder for measuring the PNP levels.
1This work was supported by grants from the Department of Veteran Affairs, the Swedish-American Foundation, and by Project Grant DK29961 from the National Institutes of Health.
*Abbreviations: ALAT, alanine aminotransferase; ASAT, aspartate aminotransferase; LDH, lactate dehydrogenase; PNP, purine nucleotide phosphorylase; UW, University of Wisconsin solution.