Biliary Complications Adversely Affect Patient and Graft Survival After Liver Retransplantation

Transcription

1 LIVER TRANSPLANTATION 19: , 2013 ORIGINAL ARTICLE Biliary Complications Adversely Affect Patient and Graft Survival After Liver Retransplantation C. Kristian Enestvedt, 1 Saloni Malik, 1 Peter P. Reese, 2,3 Alexander Maskin, 4 Peter S. Yoo, 5 Sameh A. Fayek, 6 Peter Abt, 1 Kim M. Olthoff, 1 and Abraham Shaked 1 1 Division of Transplantation, Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, PA; 2 Renal Division and 3 Department of Biostatistics and Epidemiology, Penn Perelman School of Medicine, Philadelphia, PA; 4 Section of Transplantation, Department of Surgery, University of Nebraska Medical Center, Omaha, NE; 5 Transplant and Immunology, Department of Surgery, Yale School of Medicine, New Haven, CT; and 6 Section of Transplant Surgery, Department of Surgery, Rush University Medical Center, Chicago, IL Inferior outcomes are consistently observed for recipients of liver retransplantation (re-lt) versus recipients of primary transplants. Few studies have examined the incidence and impact of biliary complications (BCs) on outcomes after re-lt. The aim of this study was to compare patient and graft survival for re-lt recipients with BCs (BC 1 ) and re-lt recipients without BCs (BC 2 ). Additional aims were to determine the impact of biliary reconstruction on the incidence of BCs and to identify risk factors for BCs after re-lt. A single-center, retrospective analysis of all re-lt recipients over a decade was performed. Univariate analyses were performed, and survival was compared with the log-rank method. A multivariate Cox regression analysis was performed to determine independent predictors of death and graft failure. The BC rate was 20.9% (n 5 23) for 110 re-lt cases. The average follow-up was 55 months. The survival rates for BC 2 recipients at 3 months and 1, 3, and 5 years were 95.3%, 91.7%, 85.4%, and 80.9%, respectively, whereas BC 1 patients had survival rates of 64.3%, 49.7%, 34.8%, and 29.8%, respectively (P < 0.001, log-rank). The graft survival rates at 3 months and 1, 3, and 5 years were 92.0%, 88.5%, 82.4%, and 78.0%, respectively, for the BC 2 group and 60.9%, 43.5%, 30.4%, and 26.1%, respectively, for the BC 1 group (P < 0.001, log-rank). BCs, a length of stay 12 days, and donor age were strongly associated with death and graft failure in a regression analysis, whereas retransplant indications other than chronic rejection and recurrent disease also affected graft failure. In conclusion, BCs significantly affected both patient and graft survival, with an increased risk of death and graft loss among BC 1 recipients. Early recognition, appropriate interventions, and preventative measures for BCs are critical in the clinical management of re-lt recipients. Liver Transpl 19: , VC 2013 AASLD. Received January 14, 2013; accepted June 1, Liver retransplantation (re-lt) has accounted for approximately 7% to 10% of transplants performed annually over the past decade in the United States. According to the most recently available data from the Scientific Registry of Transplant Recipients, the retransplant cohort accounted for 464 of 6101 liver transplantation (LT; 8%) procedures in It has been well established that both graft survival and patient survival are inferior for multitransplant patients versus primary LT patients. A further examination of data from the Scientific Registry of Transplant Recipients for the same period reveals that the graft survival rates at 1, 3, and 5 years were 82.6%, 74.2%, and 56.0%, respectively, for re-lt and 92.4%, 86.1%, and 69.9%, respectively, for primary grafts. In addition, the annual death rate per 1000 patient years has been 2 to 3 times higher for recipients of multiple LT procedures versus recipients of primary transplants over the past 9 years. Numerous authors have looked at causal relationships between complications and mortality in the re-lt population. 2-7 To date, however, few authors have examined the Abbreviations: BC, biliary complication; CI, confidence interval; HR, hazard ratio; LT, liver transplantation; MELD, Model for End- Stage Liver Disease; re-lt, liver retransplantation. Address reprint requests to C. Kristian Enestvedt, M.D., Penn Transplant Institute, 3400 Spruce Street, 2 Dulles, Philadelphia, PA kenestvedt@hotmail.com DOI /lt View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI /lt. Published on behalf of the American Association for the Study of Liver Diseases VC 2013 American Association for the Study of Liver Diseases.

2 966 ENESTVEDT ET AL. LIVER TRANSPLANTATION, September 2013 incidence of biliary complications (BCs) after re-lt and the effects of these complications on patient and graft survival. 8,9 BCs are frequent causes of morbidity after both deceased and living donor transplants and occur in 9% to 37% of primary deceased donor LT cases Numerous factors have been implicated: they range from the more overt (profound ductal ischemia in the setting of hepatic artery thrombosis) to the subtler forms of ischemic injury related to periductal dissection and other technical factors (eg, T-tube placement and type of reconstruction) to patient factors (most notably primary sclerosing cholangitis as an indication for LT) Bile duct complications include intrahepatic and extrahepatic strictures or both, bile leaks, and ischemic cholangiopathy. The last is one of the most morbid biliary complications and is seen most commonly in the setting of donation after cardiac death transplantation. 15,17 In fact, the use of donation after cardiac death organs is a wellrecognized risk factor for BCs after primary LT. 16,18,19 Given the inferior graft and patient survival after re- LT, the known challenges posed by BCs for primary transplants, and the paucity of data regarding the incidence of this clinical problem, we examined our experience with BCs in the re-lt setting. The primary aims of this study were to examine the incidence of BCs after re-lt and to determine the impact of early BCs on patient and graft survival. The secondary aims were to determine whether the type of biliary reconstruction affected the incidence of BCs and to identify risk factors for BCs after re-lt. PATIENTS AND METHODS A retrospective review of all re-lt procedures performed at a single center from January 2000 to June 2010 was conducted with a prospectively maintained database. Only recipients of whole liver cadaveric transplants were included. Data were obtained from both electronic medical records and hospital charts. Covariates included recipient demographics, pretransplant information, intraoperative data, detailed donor demographics, and postoperative outcomes. Both calculated and exception Model for End-Stage Liver Disease (MELD) scores were included in the recipient analysis. Indications for both primary LT and re-lt were evaluated. In accordance with the model published by Feng et al., 20 the donor risk index was calculated for each donor/recipient pair. Each reported BC was reviewed in detail for the type of complication (leak, stricture, or both), its location, and the type and timing of the intervention. In all BC cases, hepatic artery patency was evaluated either radiologically (duplex ultrasound, magnetic resonance imaging, computed tomography angiography, or conventional angiography) or during surgical exploration. The reconstruction technique (duct-to-duct or Rouxen-Y) was also coded for primary LT and re-lt. The statistical analysis was performed with SPSS Statistics 20 (IBM Corp., 2010). Continuous variables were compared between groups with the t test or the rank-sum test as appropriate. Categorical variables were compared between groups with the chi-square test or Fisher s exact test as appropriate. Patient survival was compared between groups with and without BCs with Kaplan-Meier survival curves via the logrank method. Multivariate Cox regression analysis was used to identify predictors of death and graft failure after re-lt. BCs were treated as a time-varying covariate to account for variations in the time of presentation of these complications. Only variables with a P value < 0.05 in the univariate analysis were entered into the Cox regression model. The follow-up duration was calculated from the date of retransplantation to the date of death, the date of the last documented follow-up at the completion of data collection (February 2012), or the date of subsequent retransplantation (if required). Two-sided tests of hypotheses were conducted, and a P value < 0.05 was the criterion for statistical significance. This study was approved by the institutional review board at the Hospital of the University of Pennsylvania. RESULTS During the study period, adult LT was performed 1344 times; 110 of these procedures (8.2%) were re- LT performed for 104 recipients. Six third transplants were included, with each re-lt representing a distinct event for analysis. The minimum follow-up was 10 months, and the average follow-up was 55 months. The recipient demographics are presented in Table 1. The median age at retransplantation was 49 years (range years). The majority of the recipients were male and Caucasian. The average listed MELD score at retransplantation was The median number of days from relisting to retransplantation was 737. The indication for primary transplantation was most often hepatitis C virus related cirrhosis, which was followed by primary sclerosing cholangitis and hepatocellular carcinoma. Indications for retransplantation included chronic rejection (31.8%), hepatic artery thrombosis (22.7%), primary nonfunction (16.4%), recurrent hepatitis C virus (10.0%), other recurrent disease (10.9%), and other causes (n 5 9 or 8.2%). The mean cold ischemia time for re-lt was 5.8 hours, and the mean warm ischemia time was 56 minutes. A T-tube was used in 16 recipient operations (14.5%). The median number of units of intraoperatively administered packed red blood cells was 7. Eighty-five of the 110 recipients (77.3%) had duct-toduct biliary reconstruction, and 25 (22.7%) had Rouxen-Y reconstruction. For all patients, the reconstruction type was unchanged from primary LT to retransplantation. BCs occurred in 20.9% of the retransplants: 65.2% of these complications (n 5 15) were anastomotic leaks, and 34.8% (n 5 8) were anastomotic strictures [13% (n 5 3) were both anastomotic leaks and strictures]. The BC rate did not differ with the reconstruction technique (P ). Recipient characteristics,

3 LIVER TRANSPLANTATION, Vol. 19, No. 9, 2013 ENESTVEDT ET AL. 967 TABLE 1. Recipient Demographics for the Groups With and Without BCs BC 1 Group BC 2 Group Total (n 5 110) (n 5 23) (n 5 87) P Value Age (years)* 49.0 ( ) 49.2 ( ) 48.9 ( ) 0.92 Male sex [n (%)] 72 (65.5) 15 (65.2) 57 (65.5) >0.99 Race [n (%)] 0.68 White 79 (71.8) 15 (65.2) 64 (73.6) Black 22 (20.0) 5 (21.7) 17 (19.5) Hispanic 7 (6.4) 2 (8.7) 5 (5.7) Asian 2 (1.8) 1 (4.3) 1 (1.1) Interval to re-lt (days)* 737 (2-8426) 77 (2-4770) 879 (2-8426) 0.16 Listed MELD score at re-lt Calculated MELD score at re-lt Etiology of disease for primary LT [n (%)] 0.60 Hepatitis C virus 32 (29.1) 9 (39.1) 23 (26.4) Hepatitis C virus 1 hepatocellular carcinoma 13 (11.8) 4 (17.4) 9 (10.3) Hepatitis B virus 2 (1.8) 0 (0) 2 (2.3) Alcohol 9 (8.2) 1 (4.3) 8 (9.2) Primary sclerosing cholangitis 14 (12.7) 2 (8.7) 12 (13.8) Primary biliary cirrhosis 10 (9.1) 3 (13.0) 7 (8.0) Nonalcoholic steatohepatitis 2 (1.8) 1 (4.3) 1 (1.1) Fulminant hepatic failure 4 (3.6) 1 (4.3) 3 (3.4) Cryptogenic 6 (5.5) 0 (0) 6 (6.9) Other 18 (16.4) 2 (8.7) 16 (18.4) Indication for re-lt [n (%)] 0.60 Chronic rejection 35 (31.8) 5 (21.7) 30 (34.5) Hepatic artery thrombosis 25 (22.7) 5 (21.7) 20 (23.0) Primary nonfunction 18 (16.4) 6 (26.1) 12 (13.8) Recurrent hepatitis C virus 11 (10.0) 2 (8.7) 9 (10.3) Other recurrent disease 12 (10.9) 2 (8.7) 10 (11.5) Other/unknown 9 (8.2) 3 (13.0) 6 (6.9) Operative/hospitalization data for re-lt Cold ischemia time (hours) Warm ischemia time (minutes) Duct reconstruction method [n (%)] 0.21 Duct-to-duct 85 (77.3) 20 (87.0) 65 (74.7) Roux-en-Y 25 (22.7) 3 (13.0) 22 (25.3) T-tube use [n (%)] 16 (14.5) 7 (30.4) 9 (10.3) 0.04 Red blood cells (U)* 7 (0-61) 9.0 (2-28) 6.0 (0-61) 0.11 Length of stay (days)* 10 (3-145) 24.1 (6-145) 13.3 (3-121) 0.03 Postoperative hepatic artery thrombosis [n (%)] 3 (2.7) 2 (8.7) 1 (1.1) 0.12 Postoperative primary nonfunction [n (%)] 14 (12.7) 3 (13.0) 11 (12.6) >0.99 NOTE: All categorical P values were derived with the chi-square test unless otherwise noted. *The data are presented as medians and ranges. The data are presented as means and standard deviations. Primary biliary cirrhosis or primary sclerosing cholangitis. Fisher s exact test. including age, sex, race, MELD score, interval to retransplantation, and indications for primary transplantation and retransplantation, did not differ between patients with BCs (BC 1 ) and patients without BCs (BC 2 ; Table 1). Notably, BC 1 patients had a significantly longer length of stay than BC 2 patients (24 versus 13 days, P ). A T-tube was used at re-lt in a statistically significant proportion of BC 1 patients versus BC 2 patients (30.4% versus 10.3%, P ). Donor characteristics were compared between patients who experienced BCs and those who did not (Table 2). The mean donor age for re-lt recipients was 34 years (range years). The cause of donor death was head trauma for 45.5% (n 5 50), cerebrovascular accident for 36.4% (n 5 40), and anoxia for 18.2% (n 5 20). Most livers were procured locally (66.4%) or from regional centers (31.8%), and few (1.8%) were allocated nationally. Only 1 recipient (0.9%) received a donation after cardiac death organ, and no split livers were used. The donor race was Caucasian for 72.7% of the recipients (n 5 80), black for 15.5% (n 5 17), and Hispanic for 11.8% (n 5 13). Patients with BCs were less likely to have received livers from Caucasian donors than patients without

4 968 ENESTVEDT ET AL. LIVER TRANSPLANTATION, September 2013 TABLE 2. Donor Demographics BC 1 Group BC 2 Group Donor Characteristic Total (n 5 110) (n 5 23) (n 5 87) P Value Age (years)* 33.5 (14-81) 35.0 (16-81) 33.0 (14-74) >0.99 Height (cm)* ( ) ( ) ( ) Cause of death [n (%)] 0.41 Anoxia 20 (18.2) 6 (26.1) 14 (16.1) Cerebrovascular/stroke 40 (36.4) 9 (39.1) 31 (35.6) Trauma 50 (45.5) 8 (34.8) 42 (48.3) Sharing [n (%)] 0.56 Local 73 (66.4) 14 (60.9) 59 (67.8) Regional 35 (31.8) 9 (39.1) 26 (29.9) National 2 (1.8) 0 (0) 2 (2.3) Race [n (%)] < Caucasian 80 (72.7) 10 (43.5) 70 (80.5) Black 17 (15.5) 5 (21.7) 12 (13.8) Hispanic 13 (11.8) 8 (34.8) 5 (5.7) Donation after cardiac death [n (%)] 1 (0.9) 0 (0) 2 (2.3) >0.99 Split/partial liver [n (%)] 0 (0) 0 (0) 0 (0) Not applicable Donor risk index score *The data are presented as medians and ranges. The data are presented as means and standard deviations. Figure 1. Patient survival was significantly higher for the BCcohort when compared to the BC; group (P < 0.001, log rank). BCs, and they were more likely to have received livers from Hispanic donors than patients without BCs. However, the overall donor risk index scores were not significantly different between the 2 groups (P ), and the mean donor risk index for all re-lt recipients was low ( ). The overall survival rates for all 110 retransplant recipients were 88.8%, 83.1%, 74.9%, and 70% at 3 months and 1, 3, and 5 years, respectively. The survival rates for patients without BCs were 95.3%, 91.7%, 85.4%, and 80.9% at 3 months, 1 year, 3 years, and 5 years, respectively, whereas patients with BCs had survival rates of 64.3%, 49.7%, 34.8%, and 29.8%, respectively (P < 0.001, log-rank; Fig. 1). The BC type (stricture versus leak) did not affect survival (P ). The overall graft survival rates after retransplantation were 85.5%, 79.1%, 71.3%, and 66.6% at 3 months and 1, 3, and 5 years, respectively. The graft survival rates at 3 months and 1, 3, and 5 years were 92.0%, 88.5%, 82.4%, and 78.0%, respectively, for patients without BCs and 60.9%, 43.5%, 30.4%, and 26.1%, respectively, for patients with BCs (P < 0.001, log-rank; Fig. 2). The analysis of factors associated with patient and graft survival evaluated all of the variables included in Table 1 plus donor age. Only those factors identified by the univariate analysis with a P value < 0.05 were included in the multivariate model. The presence of T- tubes, although statistically significant in the univariate analysis, was excluded from the final model because of confounding with other variables. The relationship of continuous variables to survival outcomes was also explored graphically. This approach revealed that the length of stay had a nonlinear relationship with outcomes. In multivariate analyses, the length of stay was, therefore, categorized as a binary variable divided at the median (12 days). In a subsequent Cox regression analysis, the presence of a BC [hazard ratio (HR) , confidence interval (CI) , P < 0.01], a length of stay 12 days (HR ,

5 LIVER TRANSPLANTATION, Vol. 19, No. 9, 2013 ENESTVEDT ET AL. 969 CI , P ), and increasing donor age (HR , CI , P < 0.01) were strongly associated with death (Table 3). For graft failure, BCs (HR , CI , P < 0.01), donor age (HR , CI , P < 0.01), a length of stay 12 days (HR , CI , P ), and indications for re-lt other than chronic rejection and recurrent disease (HR , CI , P ) were strongly associated in a regression analysis (Table 4). Among the patients with BCs, 52.2% (n 5 12) received treatment for their complications within the first 30 days after the operation, and 47.8% (n 5 11) received treatment 31 days to 1 year after transplantation. All BCs were identified within 93 days of retransplantation. The time to treatment was not associated with survival (P ). Five of the 23 patients with BCs (21.7%) were treated operatively, with 3 of these undergoing endoscopic retrograde cholangiopancreatography with stent placement treatment before reoperation. Another 15 patients underwent endoscopic Figure 2. Graft survival was signficantly higher for the BCcohort when compared to the BC; group (P < 0.001, log rank). retrograde cholangiopancreatography and stent placement without an operative intervention, and 4 patients received another treatment (including biloma drainage and percutaneous transhepatic cholangiography). There was no difference in survival between patients who were treated operatively and patients who were treated nonoperatively (P ). DISCUSSION In the present study, we have shown a significant association between mortality and BCs after re-lt. For re-lt recipients with BCs, the survival rate at 5 years was less than half the rate for recipients without these complications (29.8% versus 80.9%). BCs not only affected patient survival but also significantly decreased graft survival for re-lt recipients. Furthermore, the presence of BCs had the strongest association with both patient death and graft failure in the regression analysis. Why is there such a profound risk for mortality with BCs after re-lt? Although we do not have a clear answer to this critical question, there may be several contributing factors. First, the strong association between BCs and survival might indicate that these complications are markers of underlying allograft dysfunction. They are also a source of potential or recurring infection, particularly when they are related to overt cholangitis. Unfortunately, we do not have adequate information on the causes of death for the entire cohort; therefore, we cannot accurately determine the impact of sepsis-related events on survival in this group. The link between causes of death, sepsis-related events, and post re-lt BCs is clearly an area in which more research is warranted. Furthermore, these patients spend more time in the hospital and undergo multiple other interventions, increasing the risk for additional complications. Not surprisingly, we have shown a significant association between the length of stay and the outcomes of death and graft failure. BCs contribute to significant morbidity after primary LT and re-lt. In the current study, the BC rate was approximately 21%, which is well within the range reported in numerous series for primary LT. Fewer data exist for re-lt, but rates ranging from 7% to 27% have been published. 9,21 In fact, it has been TABLE 3. Cox Regression: Predictors of Death After Re-LT Univariate Multivariate Analysis Variable Analysis: P Value HR CI P Value BCs < <0.001 Donor age (years) < <0.001 Length of stay 12 days < Indication for re-lt 0.01 Chronic rejection Reference Reference Reference Recurrent disease Other

6 970 ENESTVEDT ET AL. LIVER TRANSPLANTATION, September 2013 TABLE 4. Cox Regression: Predictors of Graft Failure Univariate Multivariate Analysis Variable Analysis: P Value HR CI P Value BCs < <0.001 Donor age (years) < <0.001 Length of stay 12 days Indication for re-lt <0.01 Chronic rejection Reference Reference Reference Recurrent disease Other suggested by some that BC rates are rising because sicker patients are undergoing transplantation more frequently in the MELD era. 21 When the factors associated with the presence of BCs after re-lt were evaluated, several previously identified risk factors were implicated. Previously identified as a risk for bile duct complications, hepatic artery thrombosis is a common complication among BC 1 re-lt recipients. It occurred in only 3 of the patients in this series overall, but it represented nearly 9% of those patients with bile duct problems. It is somewhat surprising that there was not a statistically significant difference in the number of RBCs transfused intraoperatively for patients in the BC 1 group. This variable is often an indicator of the complexity of the re-lt operation in this group and has been shown to affect patient and graft survival by other authors. 22 BC 1 patients had a significantly longer hospital stay, and this was consistent with increased postoperative morbidity. A comparison of patients with BCs and patients without BCs showed no differences in the listed or calculated MELD scores, etiologies for primary LT and re-lt, times to retransplantation, or ischemia times. Notably, the presence of T-tubes was statistically higher among recipients with BCs. Finally, donor factors, including the donor risk index composite, did not affect the risk of BCs after re-lt. The 1 donation after cardiac death liver used did not result in bile duct related morbidity. According to the results of the current study, outcomes are clearly worse for patients with BCs after re- LT. However, the appropriate management strategy for these patients remains unclear, and numerous approaches have been advocated When we compared outcomes for patients who received less invasive interventions (including endoscopic and percutaneous therapy) and patients who had surgical revisions/interventions, there was no difference in survival between the groups. Does this indicate that the timing to surgical interventions affected these outcomes? For those with high-volume leaks, our practice is to intervene early and to favor operative interventions. These types of leaks are nearly always identified in the first few days after re-lt, and expeditious re-exploration with revision of the anastomosis is warranted. For those patients with low-volume leaks, careful observation followed by endoscopic retrograde cholangiopancreatography is the usual strategy. In this study, when either percutaneous or endoscopic therapy was used for BCs, the choice of therapy did not affect mortality. All patients in this series received the same type of biliary reconstruction upon retransplantation that was employed for primary LT. Indeed, the type of reconstruction did not affect the presence of BCs or graft or patient survival. On the basis of this cohort analysis, there does not appear to be a rationale for changing the reconstruction technique at retransplant unless particular anatomic issues are encountered. Unfortunately, few authors have evaluated the incidence and risk factors for BCs at re-lt. In the only recent publication addressing this phenomenon in the re-lt setting, Sibulesky et al. 9 examined the impact that the type of biliary reconstruction had on the incidence of complications. Comparing the outcomes for duct-to-duct and Roux-en-Y reconstruction, the authors found no difference in the incidence of any BCs 30 days (7% versus 10%) and 1 year after re-lt (11% versus 10%, P ). Additionally, there was no difference in either the death rate or graft survival according to the type of reconstruction. At 5 years, both techniques led to identical survival (62%) and equivalent rates of graft loss (45% for duct-to-duct reconstruction and 37% for Roux-en-Y reconstruction, P ). However, the authors evaluated neither patient nor graft survival according to the presence of BCs. The present study has several limitations. First, it has the inherent limitations typical of a retrospective database analysis (eg, unmeasured confounders). Second, these results are based on a single-center experience. Although this has the benefit of making the reconstruction techniques and management strategies for BCs fairly uniform, it also limits generalizability. Future studies examining this problem at multiple centers may help to further identify clinical practice patterns that affect the incidence of BCs and subsequent outcomes. The focus of the current study was primarily early events complicating survival. It is possible that late BCs and particularly late strictures may have been excluded from the analysis. Also noteworthy is the fact that T-tubes were used only in the early part of this series. Although there has been

7 LIVER TRANSPLANTATION, Vol. 19, No. 9, 2013 ENESTVEDT ET AL. 971 ongoing controversy regarding the use of T-tubes in primary LT, including some recent randomized controlled trials re-examining their use, we have abandoned this technique at our center almost entirely We have noted a strong association between T-tube use and BCs after re-lt, which is supported by the present study. Indeed, when patients with T-tubes were excluded from the analysis, overall patient survival and graft survival improved among the patients with BCs, and there was a lower BC rate overall (data not shown). Finally, although this study represents one of the largest examinations of outcomes for re-lt related to BCs, it does suffer from its small sample size, which limits our ability to detect clinically meaningful associations between clinical exposures and important outcomes after re-lt. In conclusion, early BCs after LT adversely affect both patient and graft survival. These results are independent of the reconstruction technique and the types of postoperative management strategies employed once these complications are recognized. Transplant surgeons must be acutely aware of the potential for BCs and their particularly detrimental nature in the retransplant setting, give careful consideration to techniques aimed at their prevention, and intervene appropriately once they occur. Further study is warranted to determine which risk factors play the most significant role in the development of these complications and how they might most appropriately be mitigated. REFERENCES 1. Organ Procurement and Transplantation Network (OPTN) and Scientific Registry of Transplant Recipients (SRTR). OPTN/SRTR 2010 Annual Data Report. Rockville, MD: Department of Health and Human Services, Health Resources and Services Administration, Healthcare Systems Bureau, Division of Transplantation; Jain A, Orloff M, Abt P, Kashyap R, Mohanka R, Lansing K, et al. Survival outcome after hepatic retransplantation for hepatitis C virus-positive and -negative recipients. Transplant Proc 2005;37: Chen GH, Fu BS, Cai CJ, Lu MQ, Yang Y, Yi SH, et al. A single-center experience of retransplantation for liver transplant recipients with a failing graft. Transplant Proc 2008;40: Perez-Saborido B, Meneu-Dıaz JC, de los Galanes SJ, Barra V, Fundora Y, Abradelo M, et al. Short- and longterm overall results of liver retransplantation: "Doce De Octubre" Hospital experience. Transplant Proc 2009;41: Postma R, Haagsma EB, Peeters PM, van den Berg AP, Slooff MJ. Retransplantation of the liver in adults: outcome and predictive factors for survival. Transpl Int 2004;17: Reese PP, Yeh H, Thomasson AM, Shults J, Markmann JF. Transplant center volume and outcomes after liver retransplantation. Am J Transplant 2009;9: Yamauchi Y, Wettergren A, Yamashita Y, Mikami K, Hyodo M, Larsen PN, et al. Retransplantation of the liver in adults: long-term outcome and prognostic modeling. Hepatogastroenterology 2007;54: Zhu ZJ, Rao W, Sun JS, Cai JZ, Deng YL, Zheng H, et al. Liver retransplantation for ischemic-type biliary lesions after orthotopic liver transplantation: a clinical report of 66 cases. Hepatobiliary Pancreat Dis Int 2008;7: Sibulesky L, Heckman MG, Perry DK, Taner CB, Willingham DL, Nguyen JH. A single-center experience with biliary reconstruction in retransplantation: duct-toduct or Roux-en-Y choledochojejunostomy. Liver Transpl 2011;17: Gastaca M. Biliary complications after orthotopic liver transplantation: a review of incidence and risk factors. Transplant Proc 2012;44: Sundaram V, Jones DT, Shah NH, de Vera ME, Fontes P, Marsh JW, et al. Posttransplant biliary complications in the pre- and post-model for End-Stage Liver Disease era. Liver Transpl 2011;17: Pascher A, Neuhaus P. Biliary complications after deceased-donor orthotopic liver transplantation. J Hepatobiliary Pancreat Surg 2006;13: Qian YB, Liu CL, Lo CM, Fan ST. Risk factors for biliary complications after liver transplantation. Arch Surg 2004;139: Pirenne J, Van Gelder F, Coosemans W, Aerts R, Gunson B, Koshiba T, et al. Type of donor aortic preservation solution and not cold ischemia time is a major determinant of biliary strictures after liver transplantation. Liver Transpl 2001;7: Maheshwari A, Maley W, Li Z, Thuluvath PJ. Biliary complications and outcomes of liver transplantation from donors after cardiac death. Liver Transpl 2007;13: Foley DP, Fernandez LA, Leverson G, Anderson M, Mezrich J, Sollinger HW, D Alessandro A. Biliary complications after liver transplantation from donation after cardiac death donors: an analysis of risk factors and long-term outcomes from a single center. Ann Surg 2011;253: Chan EY, Olson LC, Kisthard JA, Perkins JD, Bakthavatsalam R, Halldorson JB, et al. Ischemic cholangiopathy following liver transplantation from donation after cardiac death donors. Liver Transpl 2008;14: Abt P, Crawford M, Desai N, Markmann J, Olthoff K, Shaked A. Liver transplantation from controlled nonheart-beating donors: an increased incidence of biliary complications. Transplantation 2003;75: Foley DP, Fernandez LA, Leverson G, Chin LT, Krieger N, Cooper JT, et al. Donation after cardiac death: the University of Wisconsin experience with liver transplantation. Ann Surg 2005;242: Feng S, Goodrich NP, Bragg-Gresham JL, Dykstra DM, Punch JD, DebRoy MA, et al. Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant 2006;6: Welling TH, Heidt DG, Englesbe MJ, Magee JC, Sung RS, Campbell DA, et al. Biliary complications following liver transplantation in the Model for End-Stage Liver Disease era: effect of donor, recipient, and technical factors. Liver Transpl 2008;14: Hong JC, Kaldas FM, Kositamongkol P, Petrowsky H, Farmer DG, Markovic D, et al. Predictive index for longterm survival after retransplantation of the liver in adult recipients: analysis of a 26-year experience in a single center. Ann Surg 2011;254: Sauer P, Chahoud F, Gotthardt D, Stremmel W, Weiss KH, B uchler M, et al. Temporary placement of fully covered self-expandable metal stents in biliary complications after liver transplantation. Endoscopy 2012;44: Park JS, Kim MH, Lee SK, Seo DW, Lee SS, Han J, et al. Efficacy of endoscopic and percutaneous treatments for biliary complications after cadaveric and living donor liver transplantation. Gastrointest Endosc 2003;57:78-85.

8 972 ENESTVEDT ET AL. LIVER TRANSPLANTATION, September Melcher ML, Freise CE, Ascher NL, Roberts JP. Outcomes of surgical repair of bile leaks and strictures after adult-to-adult living donor liver transplant. Clin Transplant 2010;24:E230-E Llado L, Fabregat J, Baliellas C, Gonzalez-Castillo A, Ramos E, Gonzalez-Vilatarsana E, et al. Surgical treatment of biliary tract complications after liver transplantation. Transplant Proc 2012;44: Gonzalez MR, Cascales PA, Abellan I, Pons JA, Miras M, Capel A, et al. The evolution of therapeutic strategies for biliary tract complications after liver transplantation over a period of 20 years. Transplant Proc 2012;44: Benıtez Cantero JM, Costan Rodero G, Montero Alvarez JL, Ayllon Teran MD, Naveas Polo C, Fraga Rivas E, et al. Biliary complications after liver transplantation using side-to-side choledochocholedochostomy reconstruction with or without T-tube. Transplant Proc 2012; 44: Paes-Barbosa FC, Massarollo PC, Bernardo WM, Ferreira FG, Barbosa FK, Raslan M, Szutan LA. Systematic review and meta-analysis of biliary reconstruction techniques in orthotopic deceased donor liver transplantation. J Hepatobiliary Pancreat Sci 2011;18: Weiss S, Schmidt SC, Ulrich F, Pascher A, Schumacher G, Stockmann M, et al. Biliary reconstruction using a side-to-side choledochocholedochostomy with or without T-tube in deceased donor liver transplantation: a prospective randomized trial. Ann Surg 2009;250: