Ischemic-type biliary lesions (ITBL) are reported to. Prevention of Ischemic-Type Biliary Lesions by Arterial Back-Table Pressure Perfusion

Transcription

1 Prevention of Ischemic-Type Biliary Lesions by Arterial Back-Table Pressure Perfusion Christian Moench, * Kerstin Moench, Ansgar W. Lohse, Jochen Thies, * and Gerd Otto * Ischemic-type biliary lesions (ITBLs) lead to considerable morbidity after orthotopic liver transplantation (OLT). The exact pathogenesis is unknown. We tested the hypothesis that insufficient perfusion of biliary arterial vessels might be responsible for ITBLs. This could be prevented by improved perfusion techniques. Since February 2000, we performed a controlled study using arterial back-table pressure perfusion (AP) to achieve reliable perfusion of the biliary-tract capillary system, which may be impaired by the high viscosity of University of Wisconsin solution. We retrospectively analyzed 190 OLTs performed between September 1997 and July 2002 with regard to ITBLs. One hundred thirty-one grafts were preserved by in situ standard perfusion (SP), including portal perfusion, whereas in 59 cases, additional AP was performed. Donor-related factors, recipient age, indication for OLT, OLT technique, immunosuppression, and ischemia time were similar in both groups. In the SP group, 21 of 131 patients (16%) developed ITBLs. Only 1 of 59 patients with grafts receiving AP developed ITBLs. This difference was highly significant (P.004). Peak aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels within the first 3 days were significantly lower in the AP group (AST, P.016; ALT, P.007). Multivariate analysis showed a significant influence of AP (P.010) and donor age (P.003) on the development of ITBLs. AP is an easy and reliable method to prevent ITBLs in OLT. It therefore should be used as the standard technique in liver procurement. (Liver Transpl 2003;9: ) Ischemic-type biliary lesions (ITBL) are reported to occur in up to 26% of all patients undergoing orthotopic liver transplantation (OLT). 1,2 In many instances, sequelae, namely cholestasis and cholangitis, lead to repeated interventional or endoscopic treatment, reoperation, and, in many cases, even to re-olt. Features of ITBLs are bile-duct stenoses, dilatations, and cast formation. Pathomorphologically, these alterations are caused by epithelial and muscular necrosis of the biliary system. Periductal connective tissue usually is remarkably well preserved. This leads to the assumption that merely the bile duct itself is damaged by the responsible mechanism. By international definition, ITBLs are nonischemic and nonimmunologic lesions; arterial thrombosis or stenosis, blood group type A, type B, type O (ABO) incompatibility, and chronic rejection must be ruled out before the diagnosis is established. Features of ITBLs have been categorized repeatedly. With regard to therapeutic consequences, our own classification distinguishes between two major types: in type A lesions, the complete biliary system is affected, and in type B lesions, only the major extrahepatic bile ducts are involved. 1 Biliary cast or biliary sludge formation has already been described in other studies. 3-5 However, the lesions were properly characterized to be a particular entity by the Mayo group 6 and briefly thereafter by Li et al 7 and Kadmon et al. 8 Review of the literature showed the percentage of patients with ITBLs ranges between 15% and 25%. The greatest rate (26%) is reported by the Mayo group. The underlying cause of ITBLs remains unclear despite numerous studies. The rate of ITBLs increased with the introduction of University of Wisconsin (UW) solution in comparison to the rate seen with the use of Collins solution. 7 We hypothesized that the high viscosity of UW solution might lead to inadequate perfusion of the biliary-tree small arteries and therefore cause insufficient preservation of bile ducts and their epithelium. Accordingly, ITBLs have been reported to occur with a very low probability if high pressure is used for aortal perfusion. 9 It may easily be feasible to exert pressure on the perfusate bag. However, the resulting pressure or even flow in the hepatic artery itself may barely be controlled. To achieve controlled arterial perfusion of the graft, we performed additional arterial back-table pressure perfusion (AP). Methods Experimental Procedures One hundred ninety patients who underwent OLT between September 1997 and July 2002 and had a graft that survived From the *Department of Transplantation and Hepatobiliary Surgery and Medical Department, Johannes Gutenberg University Mainz; and the Deutsche Stiftung Organtransplantation, Region Mitte, Germany. Address reprint requests to Christian Moench, MD, Department of Transplantation and Hepatobiliary Surgery, Johannes Gutenberg University Mainz, Langenbeckstrasse 1, Mainz, Germany. Telephone: ; FAX: ; christian.moench@planet-interkom.de Copyright 2003 by the American Association for the Study of Liver Diseases /03/ $30.00/0 doi: /jlts Liver Transplantation, Vol 9, No 3 (March), 2003: pp

2 286 Moench et al Table 1. Influence of Graft and Recipient Factors on Occurrence of ITBLs With ITBL (n 22) Without ITBL (n 168) P Donor age (yr) Donor GGT (U/L) Recipient age (yr) Blood group compatibility (%) Cold ischemia time (min) Warm ischemia time (min) Acute rejection (%) CMV reactivation (%) AST 72 hr (U/L) 835 1,058 1,327 2, ALT 72 hr (U/L) ,200 1, Alkaline phosphatase 72 hr (U/L) GGT 72 hr (U/L) Bilirubin 72 hr (mg/dl) Hospital stay (d) ICU stay (d) Quick day 14 (%) Creatinine day 14 (mg/dl) for longer than 1 month were included in this retrospective analysis. All grafts were preserved with UW solution. In 131 grafts, standard perfusion (SP) was used. In 59 grafts, AP was used between February 2000 and July 2002 (44 OLTs performed in our center; 15 OLTs, at remote centers). Organs in the SP group were recruited from September 1997 until July 2002 (n 63, September 1997 to February 2000; n 68, February 2000 to July 2002). Until February 2000, all organs transplanted in our center (n 63) were preserved by SP (retrieved by our own team and remote teams). Since February 2000, all organs procured by a remote team and transplanted in our center (n 68) were preserved by SP, whereas all organs retrieved by our own team were preserved by AP. Therefore, a comparison between AP- and SP-preserved grafts during the same period was possible. No other changes in harvesting or transplantation technique were made. Data were collected by questionnaire for the 15 grafts procured by our team using AP and transplanted in remote centers. In detail, the procurement procedure was performed using standardized methods. 10,11 The aorta and caval vein were clamped subdiaphragmatically. Four liters of UW solution containing penicillin, insulin, and dexamethasone was used for aortal perfusion. A pressure of 200 to 300 mm Hg was exerted on the fluid bag by a pressure bag. At the end of aortal perfusion, a cannula was placed into the superior mesenteric vein through a lateral infrapancreatic incision, and 1 L of UW solution was used for portal in situ perfusion with a gravity flow of 50 cm H 2 O (SP). Portal perfusion was performed sequentially after aortic perfusion to avoid major bleeding during the procurement surgery. During the perfusion procedure, in situ cooling was provided by cold Ringer s solution, and sterile ice was poured into the abdominal cavity. Bile ducts also were flushed with UW solution. In addition to SP, AP has been performed through the celiac trunk or common hepatic artery in all livers harvested by our own explantation team since February Pressure was adjusted to 150 mm Hg. A pressure bag with a pressure gauge was used to adjust to the correct pressure. Pressure of 150 mm Hg was selected to slightly exceed physiological blood pressure. In addition, there is a high risk for damaging the graft artery (intima dissection) if greater pressure is used. A total of 300 ml of UW solution was used (AP). In case of an aberrant right hepatic artery, both vessels (the celiac trunk and aberrant artery) received back-table pressure perfusion with 300 ml of UW solution each. Back-table portal vein perfusion was not considered necessary because it has not been a problem to control in situ portal venous perfusion. Development of ITBLs was the study end point. Diagnosis of ITBLs was based on endoscopic retrograde cholangiography in every case. Clinical symptoms of biliary obstruction and/or increased alkaline phosphatase and -glutamyltransferase (GGT) levels led to the diagnostic procedure. In every case, anastomotic bile-duct complications, arterial thrombosis, and chronic rejection were excluded. Several factors (Table 1) that could be related to ITBL formation were included in univariate analysis comparing SP and AP. Statistical analyses were performed using SPSS (SSPS Inc, Chicago, IL) and Microsoft Office (Microsoft Corp, Redmond, WA). P less than.05 is considered significant. Chisquared and Student s t-test were used to compare the different average values, and the Kaplan-Meier method and logrank test were used to compare survival rates. Results ITBLs occurred in 22 of 190 grafts (11.6%) preserved by UW solution (Table 2). AP exerted a highly signifi-

3 Prevention of Ischemic-Type Biliary Lesions 287 Table 2. ITBLs After SP and AP Between September 1997 and July 2002 SP (n 131) AP (n 59) Total (n 190) With ITBLs 21 (16)* 1 (1.7)* 22 (11.6) Without ITBLs 110 (84) 58 (98.3) 168 (88.4) NOTE. Values expressed as number (percent). *P.004, Pearson s Chi-squared. Figure 1. Patient survival with and without AP. cant influence on the prevention of ITBLs. In organs procured by SP, 21 of 131 patients (16%) experienced ITBLs, whereas ITBLs occurred in only 1 of 59 patients with organs procured using AP (P.004, Pearson s Chi-squared). Significant results were the same if only grafts preserved by SP and AP transplanted between February 2000 and July 2002 were investigated. During this period, we saw ITBLs in 10 of 68 patients (14.7%) with grafts preserved with SP in comparison to only 1 of 59 patients (1.7%) with grafts preserved with AP (Table 3; P.009, Pearson s Chi-squared). Indications for OLT were alcoholic cirrhosis in 21%, viral hepatitis in 28% (hepatitis C virus, 19%; hepatitis B virus, 9%), hepatocellular carcinoma in 14%, autoimmune liver disease in 15% (primary sclerosing cholangitis, 8%; primary biliary cirrhosis, 2%; autoimmune hepatitis, 5%), graft dysfunction in 6%, acute liver failure in 6%, and others in 10%. Average donor age was years. Cold ischemia time was minutes, and warm ischemia time (piggyback technique, sequential reperfusion) was minutes (warm ischemia time was available in only 74% of cases). Average recipient age was years. Acute rejection occurred in 33%. The immunosuppressive protocol included tacrolimus in combination with steroids in 72%. In 28%, triple therapy containing cyclosporine A, azathioprine, and steroids was used. The two different immunosuppressive protocols were distributed equally in the SP and AP groups. Cytomegalovirus (CMV) infections occurred in 22% and were treated with ganciclovir. Average intensive care unit (ICU) stay was 7 12 days, included in the total hospital stay of days. At day 14 after OLT, average prothrombin time value (Quick) was 87% 14%, and average serum creatinine level was mg/dl. Survival rates by the Kaplan-Meier method are shown in Figures 1 and 2. AP was associated with a slight survival benefit(p.307, log rank). Patients with ITBLs had reduced survival compared with those without ITBLs (P.307, log rank). Graft and recipient characteristics (donor age; donor GGT level; recipient age; blood group compatibility; cold and warm ischemia time; acute rejection rate; CMV infection rate; peak alkaline phosphatase, Table 3. ITBLs After SP and AP Between February 2000 and July 2002 After Starting AP SP (n 68) AP (n 59) Total (n 127) With ITBLs 10 (14.7)* 1 (1.7)* 11 (8.7) Without ITBLs 58 (85.3) 58 (98.3) 116 (91.3) NOTE. Values expressed as number (percent). *P.009, Pearson s Chi-squared. Figure 2. Patient survival after OLT with and without ITBLs.

4 288 Moench et al Table 4. Graft and Recipient Data in SP and AP SP (n 131) AP (n 59) P Donor age (yr) Donor GGT (U/L) Recipient age (yr) Blood group compatibility (%) Cold ischemia time (min) Warm ischemia time (min) Acute rejection (%) CMV reactivation (%) AST 72 hr (U/L) 1,510 2, ALT 72 hr (U/L) 1,348 1, Alkaline phosphatase 72 hr (U/L) GGT 72 hr (U/L) Bilirubin 72 hr (mg/dl) Hospital stay (d) ICU stay (d) Quick day 14 (%) Creatinine day 14 (mg/dl) GGT, and bilirubin levels during the first 3 postoperative days; duration of hospital and ICU stays; quick day 14; and creatinine level day 14) were similar in both groups (AP and SP; Table 4). AP led to less graft damage, reflected by significantly lower peak aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels within the first 3 days after OLT (P.01). Recipients of grafts were older in the AP group (P.041). Table 1 lists results of univariate analysis comparing graft and recipient characteristics in patients with and without ITBLs. In patients with ITBLs, donor age was significantly older (P.003) and GGT levels were significantly lower (P.029). There was no major difference regarding the other factors. Multivariate analysis was performed to determine influencing factors on ITBLs. AP (P.010), as well as donor age (P.003), had a significant influence on the development of ITBLs. Rejection (P.791), CMV rates (P.218), recipient age (P.703), cold (P.839) and warm (P.230) ischemia time, and year of OLT (P.144) were not associated with ITBLs. Discussion Nonanastomotic bile-duct strictures after OLT occur in patients with hepatic artery thrombosis and in ABOincompatible grafts. 12,13 They may be distinguished from chronic rejection and recurrence of primary sclerosing cholangitis. Therefore, bile duct damage after OLT can be divided into ischemic biliary lesions (IBLs), bile duct lesions caused by immunologic problems, and ITBLs. Multiple factors have been claimed to be responsible for damage to the biliary tree. 9 Among them are type of biliary reconstruction, primary liver disease, CMV infection, repeated acute graft rejection, positive lymphocyte cross-match, poor HLA match, long cold ischemia time, initial high transaminase levels reflecting preservation damage, old donor age, and experience of the procuring surgeon. Immunologic factors were supposed to be involved, but it is not clear whether they cause the damage or are induced by it. 14 Unequivocally, the incidence of ITBLs is significantly less in grafts with short cold ischemia times. 2,3,7 ITBLs were almost completely absent in grafts preserved using the less viscous Collins solution. 7 This led to the assumption that the bile-duct capillary system may be incompletely perfused during preservation using UW solution. Accordingly, a remarkably low rate of ITBLs has been reported after the introduction of high-pressure aortal perfusion. 9 Despite this innovation, ITBLs have remained a serious problem in numerous transplant centers. Aiming to preserve the integrity of microvascular structures, we decided to perform controlled perfusion of the hepatic artery after graft retrieval. Even if aortal pressure perfusion is performed, the resulting flow of perfusate in the hepatic artery is unpredictable.

5 Prevention of Ischemic-Type Biliary Lesions 289 The number of grafts preserved with AP is limited in this study. Nevertheless, results have been surprising. The incidence of ITBLs in the SP group was 16% (Table 2), a rate also reported by other centers. The rate of ITBLs after AP was 1.7%. AP leads to a significantly lower initial increase in AST and ALT levels within the first 3 days. Although the lower bilirubin, alkaline phosphatase, and GGT levels within the first 10 days are not significant (Table 4), they may reflect favorable preservation of bile ducts by AP. We observed an influence of donor age on ITBLs. Worse condition of the arterial system in older patients may impair regular perfusion of bile-duct capillaries. Age therefore may be a cofactor. Interestingly, mean age of donors in the AP group was slightly older, emphasizing the role of AP as the preventive factor for ITBLs. In this context, it is noteworthy that the graft of the patient with ITBLs in the AP group was harvested from a marginal donor aged 68 years with poor circulatory conditions before retrieval (systolic blood pressure, 40 mm Hg for 30 minutes). Thus, clogging of small arterial vessels supplying the biliary tree may already have occurred before AP. Regarding these results, diffuse bile-duct damage after OLT, to date called ITBLs, could be subclassified in a new way: a large part of ITBLs might be a classified as a kind of IBL. In this sense, IBLs can occur as either macroangiopathic (arterial occlusion, stenosis, thrombosis) or microangiopathic biliary lesions (inadequate perfusion of small vessels during preservation). Cases of diffuse bile-duct injury with adequate perfusion during procurement should continue to be classified as ITBLs. In our series, since AP was introduced in 2000, this form of ITBL had a very low frequency of 1.7%. Ongoing investigations must determine whether other factors (e.g., immunologic factors) influence ITBLs. In conclusion, controlled AP leads to a considerable reduction in bile-duct damage after preservation with UW solution. The improvement is apparent as early as within the first few days after OLT. Considering the advantages of this perfusion technique compared with the ordinary procedure, a multicenter randomized study should be discussed. We suggest AP be used as the standard technique for liver procurement. References 1. Theilmann L, Kuppers B, Kadmon M, Roeren T, Notheisen H, Stiehl A, et al. Biliary tract strictures after orthotopic liver transplantation: Diagnosis and management. Endoscopy 1994;26: Sanchez-Urdazpal L, Gores G, Ward E, Maus JP, Wahlstrom HE, Moore SB, et al. Ischemic-type biliary complications after orthotopic liver transplantation. Hepatology 1992;16: Starzl T, Putnam C, Hansbrough J, Porter K, Reid H. Biliary complications after liver transplantation: With special reference to the biliary cast syndrome and techniques of secondary duct repair. Surgery 1977;81: Iwatsuki S, Shaw BJ, Starzl T. Biliary tract complications in liver transplantation under cyclosporine-steroid therapy. Transplant Proc 1983;15: McMaster P, Herbertson B, Cusick C, Calne R, Williams R. Biliary sludging following liver transplantation in man. Transplantation 1978;25: Sanchez-Urdazpal L, Gores G, Ward E, Maus TP, Buckel EG, Steers JL, et al. Diagnostic features and clinical outcome of ischemic-type biliary complications after orthotopic liver transplantation. Hepatology 1993;17: Li S, Stratta R, Langnas A, Wood R, Marujo W, Shaw BJ. Diffuse biliary tract injury after orthotopic liver transplantation. Am J Surg 1992;164: Kadmon M, Bleyl J, Küppers B, Otto G, Herfarth C. Biliary complications after prolonged UW-preservation of liver allografts. Transplant Proc 1993;25: Langrehr J, Schneller A, Neuhaus R, Vogl T, Hintze R, Neuhaus P. Etiologic factors and incidence of ischemic type biliary lesions (ITBL) after liver transplantation. Langenbecks Arch Surg 1998; 115(suppl):S1560-S Gubernatis G. Technique of organ procurement and preservation of liver and pancreas. Baillieres Clin Gastroenterol 1989;3: Strazl T, Hakala T, Shaw B, Hardesty RL, Rosenthal TJ, Griffith BP, et al. A flexible procedure for multiple cadaveric organ procourement. Surg Gynecol Obstet 1984;158: Gugenheim J, Samuel D, Reynes M, Bismuth H. Liver transplantation across ABO blood group barriers. Lancet 1990;336: Zajko A, Campbell W, Logsdon G, Bron KM, Tzakis A, Esquivel CO, et al. Cholangiographic findings in hepatic artery occlusion after liver transplantation. AJR Am J Roentgenol 1987;149: Golling M, Zipperle S, Weimer R, Otto G, Herfarth C, Opelz G, et al. Chronic liver immunologic factors in ischemic type biliary lesions (ITBL) Reduced TH1 and increased TH2 response. Langenbecks Arch Surg 1998;115(suppl):S1557- S1559.