Response to endoscopic therapy for biliary anastomotic strictures in deceased versus living donor liver transplantation

Vancouver, Canada

Response to endoscopic therapy for biliary anastomotic strictures in deceased versus living donor liver transplantation

Calvin HY Chan, Fergal Donnellan, Michael F Byrne, Alan Coss, Mazhar Haque, Holly Wiesenger, Charles H Scudamore, Urs P Steinbrecher, Alan A Weiss and Eric M Yoshida

Vancouver, Canada

BACKGROUND:Endoscopic therapy has been successful in the management of biliary complications after both deceased donor liver transplantation (DDLT) and living donor liver transplantation (LDLT). LDLT is thought to be associated with higher rates of biliary complications, but there are few studies comparing the success of endoscopic management of anastomotic strictures between the two groups. This study aims to compare our experience in the endoscopic management of anastomotic strictures in DDLT versus LDLT.

METHODS:This is a retrospective database review of all liver transplant patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) after liver transplantation. The frequency of anastomotic stricture and the time to develop and to resolve anastomotic stricture were compared between DDLT and LDLT. The response of anastomotic stricture to endoscopic therapy was also analyzed.

RESULTS:A total of 362 patients underwent liver transplantation between 2003 and 2011, with 125 requiring ERCP to manage biliary complications. Thirty-three (9.9%) cases of DDLT and 8 (27.6%) of LDLT (P=0.01) were found to have anastomotic stricture. When comparing DDLT and LDLT, there was no difference in the mean time to the development of anastomotic strictures (98±17 vs 172±65 days,P=0.11), likelihood of response to ERCP [22 (66.7%) vs 6 (75.0%),P=0.69],mean time to the resolution of anastomotic strictures (268±77 vs 125±37 days,P=0.34), and the number of ERCPs required to achieve resolution (3.9±0.4 vs 4.7±0.9,P=0.38).

CONCLUSIONS:Endoscopic therapy is effective in the majority of biliary complications relating to liver transplantation. Anastomotic strictures occur more frequently in LDLT compared with DDLT, with equivalent endoscopic treatment response and outcomes for both groups.

(Hepatobiliary Pancreat Dis Int 2013;12:488-493)

endoscopic retrograde cholangiopancreatography;liver transplantation; biliary tract surgical procedures; biliary tree; liver failure

Introduction

Biliary strictures remain a common cause of morbidity and mortality after liver transplantation, with reported incidence rates between 4% to 16%.[1-7]The majority of biliary strictures develop at the anastomosis,[6]with ischemia thought to be the primary etiological factor. Endoscopic therapy in the form of endoscopic retrograde cholangiopancreatography (ERCP) with dilatation and stenting is now the primary treatment modality for the management of biliary anastomotic strictures, with reported response rates varying from 64% to 91%.[6-9]Biliary complications are common in patients who undergo living donor liver transplantation (LDLT), with reported incidence of up to 30%,[10,11]although this incidence is declining with improved surgical techniques. To date, there are few reports on direct comparisons of the success of endoscopic managementof anastomotic stricture between LDLT and deceased donor liver transplantation (DDLT) recipient groups. The present study was undertaken to compare a single center's experience with the endoscopic management of anastomotic stricture in DDLT compared with LDLT.

Methods

We retrospectively reviewed the database of the Liver Transplantation Unit of Vancouver General Hospital between May 2003 and August 2011. Patients who had received a liver transplantation and who underwent a post-transplantation ERCP were identif i ed. A computerised patient record system was reviewed to obtain information on patient demographics and on transplantation and ERCP related events. A database was constructed to compare ERCP indications, stricture characteristics, and response to endoscopic therapy.

ERCP was performed by one of four endoscopists (BMF, SCH, SUP, and WAA), all with high volume tertiary hospital experience. For many of the procedures a senior endoscopy fellow would be involved with the case under direct supervision by one of the four endoscopists. A Pentax duodenoscope (ED-3230K, Pentax America, Montvale, NJ, USA) was used for ERCPs at the institution until 2010, when Olympus duodenoscopes (JF-180, Olympus America, Melville, NY, USA) were subsequently used. The choice of accessories and decision on type of endoscopic intervention was at the endoscopist's discretion. Generally, the endoscopist would perform either bougie or balloon dilatation of the stricture if the stenosis was felt to be high grade and stent insertion across the stricture anticipated to be diff i cult. The number of stents inserted was dependent on the anticipated response rate to stenting, with a more aggressive approach the later the onset of stricture developed from the time of transplantation. The timing of repeat ERCP was also at the endoscopist's discretion, but the procedure was usually performed at 3-month intervals until the resolution of stricture. A minimum of 12 months of endoscopic therapy was attempted before the patient was deemed to have a failed endoscopic therapy. An anastomotic stricture was def i ned as the presence of visible stenosis at the anastomosis on cholangiogram at ERCP, accompanied by biochemical evidence of biliary obstruction (persistently raised or rising liver biochemistry), without another cause identif i able to account for jaundice. Successful endoscopic therapy was def i ned as the resolution of stenosis as assessed by the endoscopist at ERCP, typically by the ability to traverse a biliary extraction balloon across the anastomosis, accompanied with improvement in serum bilirubin, alkaline phosphatase, and gamma glutamyltransferase, after the removal of the biliary stent. Stricture recurrence was def i ned as the clinical and cholangiographic recurrence of stricture after initial success.

Data were expressed as mean±standard error. Student'sttest was used to compare continuous variables and Fisher's exact test was used to compare categorical variables. Statistical analysis was performed with the GraphPad Quickcalcs computerized statistical software package. APvalue less than 0.05 was considered statistically signif i cant.

Results

ERCP in liver transplant recipients

A total of 362 patients who had received liver transplantation were identif i ed during the study period. Three hundred thirty-three patients (333/362, 92.0%) received DDLT. All patients suspected to have biliary complications after transplantation proceeded to ERCP. Of the series, 125 (34.5%) patients underwent ERCP after transplantation. In the 125 patients, 111 (88.8%) underwent DDLT and 14 (11.2%) underwent LDLT, respectively. All patients undergoing ERCP had duct-toduct anastomosis. Biliary cannulation was successful in all patients, with three DDLT patients requiring precut sphincterotomy. All but two patients undergoing LDLT and three of the DDLT patients receiving ERCP had an intraoperative biliary stent (5Fr or 8Fr pediatric feeding tube) inserted. Patient characteristics are summarized in Table 1. HCV was the most common cause for liver failure in both DDLT and LDLT patients requiring ERCP (45.9% vs 50.0%). Indications for transplantation in patients undergoing ERCP are summarized in Table 2.

Anastomotic strictures: baseline characteristics

Forty-one (11.3%) patients (33 of DDLT and 8 of LDLT patients) developed an anastomotic stricture. In patients with anastomotic strictures, HCV was again the most common indication for transplantation (48.5% vs 50.0%). LDLT patients were younger than DDLT patients (45.25±6.39 vs 54.38±1.11 years,P=0.02) (Table 3). Three of the LDLT donor livers were left lobes. Four (50.0%) LDLT and 10 (30.3%) DDLT patients required repeat transplantation before ERCP (11 of post-operative bleeding, 2 of portal vein thrombosis, and 1 of sepsis). In the DDLT patients, one was found to have concomitant biliary stones on initial ERCP and two developed stone disease after ERCP. Three patients (9.1%) were initially diagnosed with a bile leak on index ERCP,and subsequently developed an anastomotic stricture, compared with six patients (75.0%) in LDLT (P=0.0004).

Table 1.The baseline characteristics of all patients

Table 2.Indications for ERCP after transplantation (n, %)

Anastomotic strictures: response to endoscopic therapy

The mean follow-up period was 28.1±4.9 months for DDLT patients (range 2 to 82 months), and 18± 8.7 months for LDLT patients (range 3 to 63 months). The mean time to stricture onset was not statistically different between the two groups, although there was a trend towards a more delayed onset with LDLT (98 ± 17 vs 172±65 days,P=0.11). Stricture dilatation was performed in 3 (37.5%) of the LDLT patients compared to 13 (39.4%) of the DDLT patients. In all but one DDLT patient a balloon dilator was used. In the LDLT group, one (12.5%) patient had a maximum of three stents inserted, two (25.0%) patients had a maximum of 2 stents inserted, four (50.0%) patients had 1 stent inserted, and one (12.5%) patient did not have astent inserted (The patient was subjected to surgery predominantly for bile leak). In the DDLT group, fi ve (15.2%) patients had a maximum 3 stents, nine (27.3%) patients had 2 stents, and seventeen (51.5%) patients had 1 stent, and two (6.1%) patients had no stents inserted (One stricture was not able to be traversed, and the other underwent a dilatation alone.) The mean number of stents required was not different between the two groups (1.4±0.9 stents for LDLT vs 1.5±0.8 for DDLT,P=0.835). Two patients had recurrence of the stricture as evidenced by increased liver function tests after a prior ERCP failed to demonstrate a stricture. Both patients showed the recurrence of the stricture within 3 months after ERCP and stent removal and both of them received DDLT. They were re-stented and the stricture was eventually resolved after endoscopic therapy.

Table 3.Characteristic of patients with biliary anastomotic strictures

On an intention to treat basis, 66.7% of patients with anastomotic strictures responded to endoscopic therapy. If the patients currently undertaking active endoscopic treatment were excluded, the response rate was 81.5% for DDLT, 85.7% for LDLT, and 82.4% for overall. Response to endoscopic therapy was similar between DDLT and LDLT patients. In all, 66.7% and 75.0% of DDLT and LDLT patients had a successful response to endoscopic therapy (P=0.69). There was no statistical difference between the time to resolution and the number of ERCPs required to resolution (Table 3). There were 4 complications in the DDLT group (1 mildbleeding after sphincterotomy, 2 stent migration, and 1 mild ERCP pancreatitis) and none in the LDLT group.

In the DDLT group, six patients are still undergoing endoscopic therapy. In patients who failed to have endoscopic therapy, four patients underwent surgery (1 underwent surgical bypass, 2 stricture resection, and 1 stricturoplasty) and one patient required a retransplantation. In the LDLT group, one patient underwent surgical bypass and one is still undergoing endoscopic management. None of the patients who required def i nitive surgical therapy had immediate posttransplant complications requiring repeat surgery, and all of these strictures occurred more than 30 days after transplantation.

Discussion

Whilst ERCP is now a well-established fi rst line modality for the treatment of biliary anastomotic strictures, data on the response to endoscopic therapy for LDLT patients are somewhat limited. LDLT usually involves anastomosis of the small donor intrahepatic duct, with a larger donor to recipient duct discrepancy, and hence a higher propensity to stricture formation.[12-14]The reported stricture rates of 10% and 28% in both the DDLT and LDLT patients in our study are comparable to the published ones. There was no statistical difference in the time to stricture formation between the two groups of patients.

Endoscopic therapy was successful in over 68% of patients in our study, with similar response rates in the DDLT and LDLT patients. The response rates increased to over 82% if the patients who received active treatment were excluded from the present analysis. The response rates were consistent with those in the DDLT patients but higher than those in the LDLT patients. The only other study that has directly compared endoscopic therapy for anastomotic stricture between the two groups demonstrated successful resolution in 77.8% (21 out of 27) of the DDLT patients and only 20.0% (2 out of 10) of the LDLT patients.[15]Reasons for the failure of treatment included an inability to traverse the stricture with a wire (n=6), and the lack of placement of a biliary stent after dilatation (n=2). Both patients who received dilatation and stenting had a resolution of stricture. Furthermore, some studies demonstrated successful endoscopic management of post-LDLT strictures in only 42.5% of 113 patients[16]and 36.9% of 147 patients,[17]respectively. Interestingly, failure to traverse the stricture occurred in 23% and 44%, respectively.

Whilst the favorable response rate in our study may be related to the relatively small sample size of the LDLT patients leading to a type II error, lack of direct comparisons with DDLT patients in the two previously mentioned series limits the interpretation of their analysis. However, we believe that the high rate of endoscopic success in our study is related to the placement of an intraoperative biliary stent. This stent facilitates easier biliary cannulation and guidewire stricture traversal, which has contributed to lower success rates in the other series.

The most common cause of initial endoscopic failure is an inability to traverse the guidewire across the stricture.[1]T tubes have been proven to increase the rates of bile leak and cholangitis,[18,19]while there are conf l icting studies on the role of intraoperative stents in the rate of bile leak and structuring.[20-23]Few studies have assessed the effects of an intraoperative stent on subsequent biliary access and initial endoscopic success. Improving initial biliary access amplif i es the overall response rate in our patients compared with that in the published series. Chang et al[16]reported that wire passage by ERCP was not successful in 26 of 113 patients, and Kim et al[17]reported that 65 of 147 patients had no initial endoscopic success because of an inability to pass a guidewire across the stricture. Gómez et al[15]demonstrated a signif i cant difference in response rates between DDLT and LDLT (78% vs 20%), almost entirely explained by an initial inability to traverse a guidewire across the stricture in 60% of LDLT patients.

It has been hypothesized that the duct diameter discrepancy and often acute angulation between the living donor and recipient ducts are signif i cant factors in lowering initial success of endoscopic therapy. Once overcomed with the initial ERCP, response rates should theoretically be comparable to DDLT. This study, by directly comparing the LDLT and DDLT patients, reinforces the strategy of ERCP as an acceptable fi rst line management of anastomotic strictures in both DDLT and LDLT populations.

Less than 40% of DDLT and LDLT patients underwent dilatation before stent insertion. A study described dilatation of 24Fr, followed by stent insertion.[8]The ability to traverse the stricture and place multiple stents with increasing diameter and number reduces the need for dilatation before stent insertion. As most patients had an 8Fr stentin situbefore ERCP, the need for dilatation was alleviated. Endoscopic treatment was not different in the DDLT and LDLT groups. The mean number of stents required was not statistically different between the two groups.

The recurrence rate in our patients was very low, with two DDLT patients and none of LDLT patients. The recurrence rate in LDLT patients was found to be upto 30%,[24]and the low recurrence rate may be related to the prolonged duration of stent therapy. The shorter follow-up and relatively small sample size of LDLT in our study may be limiting factors for this analysis.

Bile leak has been described as an risk factor of anastomotic stricture,[1,25]and our study indicates that it is particularly relevant in LDLT. Local ischemia and inf l ammation are possible pathophysiological factors associated with both stricture and leak, but the close association with LDLT has not been reported in the past. Bile leaks, however, are more common in LDLT,[14]and this may be an indirect surrogate for a technically more challenging operative anastomosis. Further studies are required to clarify whether there is a real etiological variance in the formation of strictures in DDLT versus LDLT patients.

Limitations of this study include retrospective design and a small number of LDLT patients. Since the rate of complications can be underestimated in retrospective studies, bias is likely to affect both groups equally. The relevance of an intraoperative stent to the overall success of endoscopic therapy should ideally be conf i rmed with a prospective controlled study. However, stricture rate and response to endotherapy in DDLT patients are comparable to the published data, supporting the generalisability of our results.

In conclusion, anastomotic biliary strictures remain a signif i cant potential complication in both DDLT and LDLT recipients. They occur more commonly in LDLT, and response to endoscopic therapy appears to be effective, with the majority of patients responding to endotherapy. This study supports the need for further prospective studies comparing DDLT and LDLT characteristics, and to validate whether the routine use of intraoperative biliary stents will improve overall endoscopic success, so that optimal endoscopic therapy can be tailored to manage post-transplant biliary complications in this growing subgroup of patients.

Contributors:DF, BMF and YEM proposed the study. CCHY, CA, HM and WH performed research. CCHY and DF wrote the fi rst draft. CCHY, CA, HM, WH, SCH, SUP and WAA collected and analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. DF is the guarantor.Funding:None.

Ethical approval:No needed.

Competing interest:No benef i ts in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Verdonk RC, Buis CI, Porte RJ, van der Jagt EJ, Limburg AJ, van den Berg AP, et al. Anastomotic biliary strictures after liver transplantation: causes and consequences. Liver Transpl 2006;12:726-735.

2 Pfau PR, Kochman ML, Lewis JD, Long WB, Lucey MR, Olthoff K, et al. Endoscopic management of postoperative biliary complications in orthotopic liver transplantation. Gastrointest Endosc 2000;52:55-63.

3 Thuluvath PJ, Atassi T, Lee J. An endoscopic approach to biliary complications following orthotopic liver transplantation. Liver Int 2003;23:156-162.

4 Thethy S, Thomson BNj, Pleass H, Wigmore SJ, Madhavan K, Akyol M, et al. Management of biliary tract complications after orthotopic liver transplantation. Clin Transplant 2004; 18:647-653.

5 Verdonk RC, Buis CI, Porte RJ, Haagsma EB. Biliary complications after liver transplantation: a review. Scand J Gastroenterol Suppl 2006:89-101.

6 Graziadei IW, Schwaighofer H, Koch R, Nachbaur K, Koenigsrainer A, Margreiter R, et al. Long-term outcome of endoscopic treatment of biliary strictures after liver transplantation. Liver Transpl 2006;12:718-725.

7 Rerknimitr R, Sherman S, Fogel EL, Kalayci C, Lumeng L, Chalasani N, et al. Biliary tract complications after orthotopic liver transplantation with choledochocholedochos tomy anastomosis: endoscopic fi ndings and results of therapy. Gastrointest Endosc 2002;55:224-231.

8 Zoepf T, Maldonado-Lopez EJ, Hilgard P, Malago M, Broelsch CE, Treichel U, et al. Balloon dilatation vs. balloon dilatation plus bile duct endoprostheses for treatment of anastomotic biliary strictures after liver transplantation. Liver Transpl 2006;12:88-94.

9 Pasha SF, Harrison ME, Das A, Nguyen CC, Vargas HE, Balan V, et al. Endoscopic treatment of anastomotic biliary strictures after deceased donor liver transplantation: outcomes after maximal stent therapy. Gastrointest Endosc 2007;66:44-51.

10 Takatsuki M, Eguchi S, Kawashita Y, Kanematsu T. Biliary complications in recipients of living-donor liver transplantation. J Hepatobiliary Pancreat Surg 2006;13:497-501.

11 Wang SF, Huang ZY, Chen XP. Biliary complications after living donor liver transplantation. Liver Transpl 2011;17:1127-1136.

12 Sharma S, Gurakar A, Jabbour N. Biliary strictures following liver transplantation: past, present and preventive strategies. Liver Transpl 2008;14:759-769.

13 Thuluvath PJ, Pfau PR, Kimmey MB, Ginsberg GG. Biliary complications after liver transplantation: the role of endoscopy. Endoscopy 2005;37:857-863.

14 Freise CE, Gillespie BW, Koffron AJ, Lok AS, Pruett TL, Emond JC, et al. Recipient morbidity after living and deceased donor liver transplantation: fi ndings from the A2ALL Retrospective Cohort Study. Am J Transplant 2008;8:2569-2579.

15 Gómez CM, Dumonceau JM, Marcolongo M, de Santiba?es E, Ciardullo M, Pekolj J, et al. Endoscopic management of biliary complications after adult living-donor versus deceased-donor liver transplantation. Transplantation 2009; 88:1280-1285.

16 Chang JH, Lee IS, Choi JY, Yoon SK, Kim DG, You YK, et al. Biliary stricture after adult right-lobe living-donor liver transplantation with duct-to-duct anastomosis: long-term outcome and its related factors after endoscopic treatment. Gut Liver 2010;4:226-233.

17 Kim TH, Lee SK, Han JH, Park do H, Lee SS, Seo DW, et al. The role of endoscopic retrograde cholangiography for biliary stricture after adult living donor liver transplantation: technical aspect and outcome. Scand J Gastroenterol 2011;46:188-196.

18 Amador A, Charco R, Marti J, Alvarez G, Ferrer J, Mans E, et al. Cost/eff i cacy clinical trial about the use of T-tube in cadaveric donor liver transplant: preliminary results. Transplant Proc 2005;37:1129-1130.

19 Scatton O, Meunier B, Cherqui D, Boillot O, Sauvanet A, Boudjema K, et al. Randomized trial of choledochochole dochostomy with or without a T tube in orthotopic liver transplantation. Ann Surg 2001;233:432-437.

20 Kusano T, Randall HB, Roberts JP, Ascher NL. The use of stents for duct-to-duct anastomoses of biliary reconstruction in orthotopic liver transplantation. Hepatogastroenterology 2005;52:695-699.

21 Bawa SM, Mathew A, Krishnan H, Minford E, Talbot D, Mirza DF, et al. Biliary reconstruction with or without an internal biliary stent in orthotopic liver transplantation: a prospective randomised trial. Transpl Int 1998;11:S245-247.

22 Johnson MW, Thompson P, Meehan A, Odell P, Salm MJ, Gerber DA, et al. Internal biliary stenting in orthotopic liver transplantation. Liver Transpl 2000;6:356-361.

23 Tranchart H, Zalinski S, Sepulveda A, Chirica M, Prat F, Soubrane O, et al. Removable intraductal stenting in duct-toduct biliary reconstruction in liver transplantation. Transpl Int 2012;25:19-24.

24 Seo JK, Ryu JK, Lee SH, Park JK, Yang KY, Kim YT, et al. Endoscopic treatment for biliary stricture after adult living donor liver transplantation. Liver Transpl 2009;15:369-380.

25 Ostroff JW. Post-transplant biliary problems. Gastrointest Endosc Clin N Am 2001;11:163-183.

Received August 8, 2012

Accepted after revision February 6, 2013

AuthorAff i liations:Division of Gastroenterology (Chan CHY, Donnellan F, Byrne MF, Coss A, Haque M, Wiesenger H, Steinbrecher UP, Weiss AA and Yoshida EM), and Department of Surgery (Scudamore CH), Vancouver General Hospital, University of British Columbia, Vancouver, BC, V5Z 1M9, Canada

Fergal Donnellan, MD, Division of Gastroenterology, Vancouver General Hospital, 5th Floor, 2775 Laurel Street, Vancouver, BC, V5Z 1M9, Canada (Tel: 604-875-5244; Fax: 604-875-5447; Email: fdonnellan77@hotmail.com)

? 2013, Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872(13)60077-6