AIM: To evaluate whether urokinase perfusion of non-heart-beating cadaveric donor livers reduces the incidence of intrahepatic ischemic-type biliary lesions (IITBLs).

METHODS: A prospective study was conducted to investigate potential microthrombosis in biliary microcirculation when non-heart-beating cadaveric livers were under warm or cold ischemic conditions. The experimental group included 140 patients who underwent liver transplantation during the period of January 2006 to December 2007, and survived for more than 1 year. The control group included 220 patients who received liver transplantation between July 1999 and December 2005 and survived for more than 1 year. In the experimental group, the arterial system of the donor liver was perfused twice with urokinase during cold perfusion and after trimming of the donor liver. The incidence of IITBLs was compared between the two groups.

RESULTS: In the control group, the incidence of IITBLs was 5.9% (13/220 cases) after 3-11 mo of transplantation. In the experimental group, two recipients (1.4%) developed IITBLs at 3 and 6 mo after transplantation, respectively. The difference in the incidence between the two groups was statistically significant (P < 0.05).

CONCLUSION: Double perfusion of cadaveric livers from non-heart-beating donors with urokinase may reduce the incidence of IITBLs.

Intrahepatic ischemic-type biliary lesions (IITBLs) after orthotopic liver transplantation are the most frequent cause of non-anastomotic biliary strictures in liver grafts[1–4]. They affect 2%-19% of patients after liver transplantation and have become a leading cause of liver re-transplantation in China[56]. Several risk factors for IITBLs have been identified, including ischemia-related injury, immunologically induced injury, and cytotoxic injury[7–9]. Although IITBLs have a multifactorial origin, ischemia-reperfusion injury and hepatic arterial thrombosis are considered to be the major causes[10–12]. In recent years, however, impaired biliary microcirculation has led to an increasing concern. Improving microcirculation can increase oxygen supply and might prevent biliary injuries. Urokinase is a proteolytic enzyme produced by the kidney, which is found in the urine. Urokinase acts on the endogenous fibrinolytic system by converting plasminogen to plasmin, which, in return, degrades fibrin clots. Since January 2006, our center has used a procedure based on the therapeutic principle of urokinase to prevent potential microthrombosis in biliary microcirculation. We have used non-heart-beating cadaveric donor grafts in which warm or cold ischemic insult was induced and which were perfused with urokinase; this procedure has produced good results.

In the present study, we evaluated prospectively 140 liver transplantation patients who received grafts with urokinase perfusion.

The incidence of IITBLs was 5.9% in patients who received grafts without urokinase perfusion (13/220 patients) 3-11 mo (mean, 5.1 mo) after transplantation, and 1.4% in the patients who received grafts with urokinase perfusion (2/140 patients) at 3 and 6 mo after transplantation. The difference in the incidence between the two groups was statistically significant (P < 0.05). The main symptom noted in the 15 IITBL patients in both groups was progressive hyperbilirubinemia. Cholangitis was observed as a complication in three of the 15 patients. The results of a biochemical assay showed that the level of total bilirubin increased progressively. The level of direct-reacting bilirubin increased the most, followed by the level of biliary enzymes such as alkaline phosphatase and γ-glutamyltransferase. A slight increase in the levels of aspartate aminotransferase and alanine aminotransferase levels was detected.

Of the 13 IITBL patients who received grafts without urokinase perfusion, four were diagnosed with IITBLs by T-tube cholangiography; two by percutaneous transhepatic cholangiography; four by endoscopic retrograde cholangiography; and three by magnetic resonance cholangiography (MRCP). In the two IITBL patients who received grafts with urokinase perfusion, IITBLs were diagnosed by MRCP. The imaging of the 15 IITBL patients showed varying degrees of diffuse stricture and segmental dilatation of the intrahepatic bile duct, with withered-branch-like changes in the biliary tree (Figure 1). The stricture or embolic lesion of the hepatic artery and its major branches was excluded by Doppler ultrasonography, and acute or chronic rejections were excluded by liver biopsy.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Percutaneous transhepatic cholangiography shows intrahepatic ITBL. Variant degrees of diffuse stricture and segmental dilatation of the intrahepatic bile duct in the biliary tree.

Eight of 13 IITBL patients who received grafts without urokinase perfusion required liver re-transplantation, and the other five patients died from liver failure while still on the waiting list. The two IITBL patients who received grafts with urokinase perfusion underwent liver re-transplantation. The injured grafts were removed from the 10 re-transplantation patients and observed pathologically. Gross findings showed patchy necrosis of the biliary endomembrane, thickened and sclerosed duct wall, and biliary sludge in the narrowed lumen. Microscopic findings showed severe cholestasis of hepatocytes, spotty or patchy necrosis, infiltration of inflammatory cells and mild to moderate proliferation of fibrous tissue in the portal area, necrosis and exfoliation of a great number of endothelial cells in small biliary lumens, formation of bile thrombus, and fibrosis and luminal stenosis of some bile ducts (Figure 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Cholestasis of hepatocytes, infiltration of inflammatory cells and proliferation of fibrous tissue in the portal area, fibrosis and luminal stenosis of bile ducts. (HE stain, × 200).

IITBLs have become a leading cause of liver re-transplantation in China[1314]. This is closely associated with the fact that non-heart-beating cadaveric livers constitute a major source of donor livers. Several studies conducted in China and other countries have suggested that cadaveric donor grafts with warm ischemic injury leads to a higher incidence of IITBLs than grafts from brain-dead donors. Abt et al[15] have reported that the ratio of IITBLs in biliary complications was 66.6% with non-heart-beating cadaveric livers and 19.2% with livers from brain-dead donors. Nakamura et al[16] have reported that the incidence of IITBLs was only 1.4% in recipients grafted with brain-dead donor livers. Zhang et al[17] have reported that in 235 patients transplanted with cadaveric livers and 36 transplanted with living-donor livers, the incidence of biliary complications was 19.1% and 5.6%, respectively. The incidence of IITBLs was 7.29% and 0%, respectively, and the difference was statistically significant. Researchers in China and other countries have agreed that the difference was mainly caused by warm ischemia.

What kinds of pathological or physiological changes associated with IITBLs may occur in a donor liver during warm ischemia? Generally, the initiation of intrinsic coagulation requires stoppage or slowing of blood flow, high blood viscosity, and injury of vascular endothelial cells. After these three requirements are met, most non-heart-beating cadaveric donors are usually in a hypercoagulable state during warm ischemia. Thus, blood coagulation or microthrombi may develop rapidly in the arterioles and peribiliary capillary plexus of the donor liver within a short period. The blood clots or microthrombi are not washed out easily by subsequent cold perfusion under normal pressure, which results in inadequate perfusion of the peribiliary capillary plexus. Even after blood circulation in the donor liver returns to normal, the arterioles and peribiliary capillary plexus still lack perfusion with arterial blood. This may lead to ischemia and hypoxia of the bile duct wall; formation of biliary sludge caused by degeneration, necrosis or even exfoliation of a great number of endothelial cells; and luminal stenosis caused by gradual fibrosis of the muscular layers. By periodically sampling livers from rats dying suddenly from cardiac arrest for light microscopy, we observed blood coagulation in most arterioles in the portal area in hepatic tissue samples collected after 25 min of cardiac arrest. For the reasons described above, we adopted the strategy of double perfusion with urokinase, which aimed to achieve maximum dissolution of microthrombi, ensure better effects of cold perfusion and reperfusion, and prevent the development of IITBLs. In the present study, we found that the difference between the two groups was significant, which indicated that urokinase perfusion can effectively reduce the incidence of IITBLs.

The in vitro use of urokinase may not have any effect on the coagulation function of patients and is not dose-limited. Thus, urokinase usually can be administered in vitro at a higher concentration than when injected intravenously. For better blood circulation in arterioles and the peribiliary capillary plexus, a higher concentration of urokinase could be used to reperfuse the liver after trimming. A urokinase-free preservation solution should be used to perfuse the artery to wash out excess urokinase before completing the whole perfusion process. The activity and function of urokinase at 0-4°C have not been reported yet. In separate tests, a significant thrombolytic effect was observed when fresh blood clots in tubes at 4°C were immersed in urokinase at the above-mentioned concentrations.

The results of the present study confirm that: (1) liver transplantation from non-heart-beating cadaveric donors may lead to a higher incidence of IITBLs and a higher rate of re-transplantation; and (2) double perfusion of cadaveric livers from non-heart-beating donors with urokinase may reduce the incidence of IITBLs.