UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Complications in hepato-pancreato-biliary surgeryMultidisciplinary and interdisciplinary approachSchreuder, A.M.

Link to publication

Creative Commons License (see https://creativecommons.org/use-remix/cc-licenses):Other

Citation for published version (APA):Schreuder, A. M. (2019). Complications in hepato-pancreato-biliary surgery: Multidisciplinary andinterdisciplinary approach.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 14 Aug 2020

3

Endoscopy 2018; 50:577-587

A. Marthe Schreuder, Klaske A.C. Booij, Philip R. de Reuver, Otto M. van Delden, Krijn P. van Lienden, Marc G. Besselink,

Olivier R. Busch, Dirk J. Gouma, Erik A.J. Rauws, Thomas M. van Gulik

CHAPTER 3Percutaneous-endoscopic rendezvous procedure for

the management of bile duct injuries after cholecystectomy:

short- and long-term outcomes

Part I | Chapter 3

54

Abstract

Background and study aimsBile Duct Injury (BDI) remains a daunting complication of laparoscopic cholecystectomy. In patients with complex BDI, a percutaneous-endoscopic rendezvous (RV) procedure may be required to establish bile duct continuity. The aim of this study is to assess short-term and long-term outcomes of the RV.

Patients and methodsAll consecutive patients with BDI referred to our tertiary referral center between 1995 and 2016 were analysed. RV procedure was performed when endoscopic or radiologic intervention failed and when deemed feasible by a dedicated multidisciplinary team including hepato-pancreato-biliary surgeons, gastrointestinal endoscopists and interventional radiologists. Classification of BDI, technical success of RV, procedure-related adverse events and outcomes were assessed.

ResultsAmong a total of 812 patients, RV was performed in 47 (5.8%), of which 31 (66%) were diagnosed with complete transection of the bile duct (Amsterdam type D/Strasberg type E injury). Primary success rate of RV was 94% (44/47 patients). Overall morbidity was 18% (10 patients). No life-threatening adverse events or 90-day mortality occurred. After a median follow-up of 40 months (IQR 23-54) RV was the final successful treatment in 26/47 patients (55%). In 14/47 patients (30%) RV acted as a bridge to surgery while hepaticojejunostomy (HJ) was chosen either primarily or secondarily to treat refractory or relapsing stenosis.

ConclusionsIn experienced hands, RV is safe with a long-term success rate of 55%. When endoscopic or transhepatic interventions fail to restore bile duct continuity in patients with BDI, RV should be considered either as definitive treatment or as a bridge to elective surgery.

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

55

Introduction

Bile duct injury (BDI) remains a major complication of laparoscopic cholecystectomy, and occur in 0.3 to 1.4% of patients [1-3]. BDI ranges in severity from cystic duct leakage to complete transection (and even resection) of the common bile duct (CBD) or major bile ducts, sometimes combined with vascular injury [4,5]. Recognition and early treatment of BDI is of utmost importance, since inadequate management can have severe effects and will lead to deterioration of the patients’ general condition, resulting in biliary peritonitis, sepsis, multiple organ failure and even death [4,6-8]. The majority of BDIs or strictures can be treated successfully by endoscopic or radiological intervention [7,9-11]. In case of complete transection of a major bile duct, reconstructive surgery by means of a hepaticojejunostomy (HJ) is generally the treatment of choice [7,11]. However, morbidity and mortality rates of this surgical treatment of BDI are considerable, especially when performed early after cholecystectomy [12-14]. Long-term recurrent stenosis occurs in 13 to 30% of patients [6,13-15]. When endoscopic and radiologic treatments fail in patients with complex BDI, an escape procedure before surgery is provided by the percutaneous-endoscopic rendezvous (RV) procedure, with both approaches combined in one session [16,17]. The RV procedure has been described for different indications. It may be useful in case of failed cannulation or canalization [18], and stone extraction [19]. In patients with BDI the RV has been described in small case series as safe and effective both for management of bile duct disruption and stenosis [16,17]. Sufficient data regarding long-term outcomes of the RV procedure for the management of BDI are lacking. The aim of this study is to analyse the short-term and long-term outcomes of RV in patients with BDI.

Methods

PatientsBetween January 1995 and December 2016, all consecutive patients who were referred to the departments of surgery, interventional radiology and gastroenterology of the Academic Medical Center (AMC) in Amsterdam for the management of a BDI after cholecystectomy were included into a prospective database. From this database, all patients in whom an RV procedure was attempted after referral to the AMC were included in the present study. Patients’ medical charts were reviewed to analyse operation reports and clinical data including indication for initial cholecystectomy, type of operation and diagnostic and therapeutic interventions that were attempted prior to the RV, both from the referring center and our tertiary referral center. Initial

Part I | Chapter 3

56

type of injury was classified according to the Amsterdam, Strasberg and Bismuth classifications [20-22]. We evaluated the indication for an RV procedure, the approach of percutaneous transhepatic biliary drainage (PTBD), use of dilatation, number of stents or biliary drainage catheters inserted and the number of stent or drain revisions.

Treatment protocol The standard endoscopic and radiologic management by endoscopic retrograde cholangiopancreatography (ERCP) and PTBD for BDI at our institution have been published previously [9,10]. In patients in whom endoscopic and/or percutaneous treatment failed, further treatment course was determined by a dedicated multidisciplinary team, including hepato-pancreato-biliary surgeons, gastrointestinal endoscopists and interventional radiologists, either deciding upon direct surgical reconstruction or upon an RV procedure. All RV procedures were performed by interventional radiologists and endoscopists with more than 10 years of experience in biliary intervention.The RV procedure entails a collaboration between the endoscopist and interventional radiologist. RV can be carried out extraluminally, e.g. in the subhepatic space, or intraluminally, within the bile duct or in the duodenum. The procedure is explained in figure 1 and in our supplementary video (video 1). The procedure is generally performed under moderate sedation using a combination of midazolam, fentanyl and local anaesthetics or deep sedation using a combination of propofol, opioids and local anaesthetics. During and after the procedure, the patient’s vital signs are monitored and O2 is administered if necessary. After intravenous administration of broad-spectrum prophylactic antibiotics, a diagnostic cholangiogram is obtained either by PTBD or by ERCP and the location and extent of injury is determined. The intrahepatic biliary system was percutaneously approached from the left (subxiphoid approach) or right side (sub- or intercostal approach) or by bilateral approach. Following ultrasound-guided puncture of the intrahepatic biliary system with a 22G Chiba needle, an introducer set (Neff-set, Cook Medical Europe Ltd, Limerick, Ireland) was used to place an 8F vascular access sheath for diagnostic cholangiography and intra-procedural drainage. Subsequently, a hydrophilic guidewire is passed to the leak or stenosis, either using the percutaneous route or endoscopically (figure 1b). In the opposite direction, a loop or dormia basket is manoeuvred near the guidewire in an attempt to snare the wire and externalize it (figure 1c-e). This ‘rendezvous’ of wires may take place in the subhepatic space, intraductally or in the duodenum.After passing the defect the wire is used as a guide for the insertion of either an internally draining PTBD or endoscopic stent (Amsterdam-type straight polyethylene) bridging the defect (figure 1g). If possible, multiple stents were placed at the same time. In case of a tight stenosis, either catheter dilatation or balloon dilatation was performed in order

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

57

Video 1. Rendezvous procedure for bile duct injury. Online content viewable at: https://doi.org/10.1055/s-0043-123935

to facilitate stent placement (figure 1f ).

After a successful RV procedure, successive treatment consisted of elective endoscopic stent revisions every 3 months, until bile duct patency was restored (absence of leakage and/or stenosis). Patency of the bile duct was assessed by the endoscopist based on cholangiographic appearance, adequate drainage of contrast into the duodenum and absence of any significant resistance when passing an extraction balloon through the area of stenosis. Before the year 2002, common practice was to insert one or two stents at revision. After 2002, progressive stenting by means of inserting as many stents as possible was implemented in our institution [9]. Old stents were always removed before inserting new stents. In case of surgical repair, a HJ was constructed using a Roux-en-Y jejunal limb. The surgical technique as performed in our institution has been published previously [12].

Outcome and follow-upOutcome parameters that were assessed included primary, secondary and long term success rate of the RV treatment, procedure related adverse events, number of stent revisions, duration of stent treatment and adverse events occurring during the period of stent treatment.The RV procedure is considered a primary success (“technically” successful) if the biliary gap was passed by a guidewire and subsequent internal biliary drainage could be achieved, either by means of one or more endoscopically placed stents or by an internal PTBD. Secondary success of RV treatment is achieved when after a period of stenting the stents

Part I | Chapter 3

58

A

E

B

F

C

G

D

H

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

59

I J

Figure 2. Example of a successful RV procedure in a 20-year old female with an Amsterdam type D/Strasberg type E3 BDI. A) Cholangiography showing a complete transection of the CBD (bold arrow) with leakage of contrast into the subhepatic space, forming a biloma. A PTBD is in place in the right system (arrow) B) Transhepatically a guidewire (arrow) is passed into the subhepatic space. C) A second guidewire is passed into the subhepatic space endoscopically (bold arrow) D) A percutaneous inserted snare (arrow) approaches the endoscopic wire (bold arrow). E) After snaring the endoscopic wire, this wire is exteriorized percutaneously. F) Dilatation of stricture of the proximal CBD. G) Two stents are placed in the CBD endoscopically, bridging the defect. H) contrast given through the PTBD drains towards the duodenum, with some leakage of contrast to the subhepatic space. I) and J) Percutaneous cholangiography 4 weeks later shows adequate drainage of contrast through the endoscopic stents into the duodenum, without leakage of contrast.

Part I | Chapter 3

60

could be removed leaving a patent bile duct. When patients had to be referred for surgery during the period of stenting, this was considered secondary failure of the treatment. If no (recurrent) stenosis occurred after a minimum of 12 months of follow-up, the RV treatment was considered as long-term successful. Follow-up data were obtained through regular outpatient visits and mail and telephone surveys with general practitioners and referring doctors.Overall mortality within 90 days, RV related mortality and BDI related mortality were assessed. Procedure-related adverse events that occurred within 90 days after RV with Clavien-Dindo grade 2 or more were evaluated. These included false passage, major bleeding requiring blood transfusion or intervention, pneumothorax, stent clogging, stent dislocation, cholangitis, post-ERCP pancreatitis, sepsis requiring ICU observation and (intra-abdominal or intrahepatic) abscess formation. As the RV procedure was followed by a period of progressive stenting which could also lead to adverse events and some patients suffered from more than one adverse event, we assessed the overall morbidity rate of the RV treatment using the Comprehensive Complication Index (CCI) [23]. The CCI summarizes all complications after an intervention, weighted for their severity. Through a formula, the combination of complications for each patient is calculated and ranked on a scale ranging from 0 to 100. In calculating the CCI we also encountered ‘minor’ adverse events such as stent-related pain (Clavien-Dindo I), to fully assess the impact of the RV treatment in patients. Adverse events that had occurred before RV or after HJ were not included in calculation of the CCI.

Statistical analysisData derived from patient characteristics, management and outcomes are represented in numbers and percentages. Means with standard deviation (SD) or median values with minimum and maximum values or interquartile range (IQR) are presented, when appropriate. Comparison between groups was performed by a Chi-square test, Fisher’s exact test or Mann-Whitney U test when appropriate. Data analyses were performed using SPSS® software (SPSS, Chicago, Illinois, USA). A two-sided p value of < 0.05 was considered statistically significant.

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

61

Results

Patient characteristics From our cohort of 812 consecutive BDI patients, a total of 55 RV procedures were performed in 47 patients (5.8%). Patient characteristics, indications for cholecystectomy, type of cholecystectomy and type of injury are summarized in table 1. Thirty-one patients (66%) had a complete transection of a major bile duct (Amsterdam type D/Strasberg type E). In all patients with a lateral defect of a major bile duct (Amsterdam type B/Strasberg type D), the defect was reported to be ‘relatively large’. In patients with a complete transection of a bile duct, the BDI was significantly more often recognized during the initial operation (10/31 patients, 32%) than in patients with a partial bile duct injury (1/16 patients, 6%; p=0.046). Consequently, a statistically significant difference was observed in the median time between initial operation and diagnosis of type D and B injuries (4 days, IQR 0-15 days versus 16 days, IQR 6-58 days; p=0.025).

Therapeutic interventions The time interval between diagnosis of BDI and RV procedure, as well as therapeutic interventions performed prior to RV are listed in table 2. The median period between diagnosis of BDI and RV was 38 days (IQR 24-93 days). Two patients had a time interval of 7 and 10 months respectively between diagnosis of BDI and RV procedure. This delay can be explained by multiple ERCP attempts in the referring hospital, without satisfactory effects. Another patient had a time interval of 7 days between LC and diagnosis of BDI but the RV was performed at 39 months. This is explained by the late development of a CBD stenosis after LC with a Amsterdam type B injury initially treated by percutaneous drainage of the biloma only. Significantly more patients with a complete transection underwent surgical repair during initial operation than patients with a partial BDI. In all of these patients primary repair of the injury was attempted by constructing an end-to-end anastomosis or by direct suturing of a partial defect, however ongoing bile leakage indicated further treatment. Nine patients (19%) underwent a relaparotomy, with no difference in reoperation rate between patients with a complete or partial injury. In all of these patients, only surgical drainage was performed during laparotomy.

Part I | Chapter 3

62

Table 1. Characteristics of patients who underwent an RV procedureTotal

N = 47 (%)

Complete

transection

N = 31

Partial

injury

N = 16 P value

Age at cholecystectomy, mean in years (SD) 51 (18) 52 (18) 50 (17) 0.63Female sex 32 (68) 23 (74) 9 (56) 0.21Preoperative ASA classification ≥ 2 20 (43) 11 (36) 9 (56) 0.17

Indication for cholecystectomy:Symptomatic cholecystolithiasis 32 (68) 23 (75) 10 (63)

0.55Acute cholecystitis 10 (21) 6 (19) 4 (25)Delayed cholecystectomy after acute cholecystitits 2 (4) 1 (3) 1 (6)Biliary pancreatitis 2 (4) 1 (3) 1 (6)

Initial procedure

Conversion to open procedure 17 (36) 11 (36) 6 (38) 0.89Intraoperative recognition of BDI 11 (23) 10 (32) 1 (6) 0.046

Time between LC and diagnosis (days), median (IQR) 7 (0-26) 4 (0-15) 16 (6-58) 0.025

Postoperative sepsis* 9 (19) 5 (16) 5 (31) 0.46Vascular injury 6 (13) 4 (13) 2 (13) 0.97Initial type of injury

Amsterdam ClassificationType A (cystic duct leakage) 1 (2)Type B (bile duct leakage) 14 (30)Type C (bile duct stricture) 1 (2)Type D (complete transection of major bile duct) 31 (66)

Strasberg ClassificationType A 1 (2)Type B 0

Type C 3 (6) Type D 12 (26) Type E 31 (66) Type E I 0 Type E II 9 (19) Type E III 8 (17) Type E IV 5 (11) Type E V 6 (13)*for which ICU observation was requiredRHD, right hepatic duct; RV, rendezvous procedure.

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

63

All patients underwent one or more attempts at either ERCP, PTBD or both prior to RV. Main reasons for failure of adequate treatment by PTBD and/or ERCP were failed canalization in 13 patients (28%), bile duct disruption without obstruction in 11 patients (23%), and bile duct disruption with obstruction in 22 patients (49%). The latter was significantly more often the reason for failure of PTBD and/or ERCP in patients with a complete transection (19/47, 61% versus 3/16, 19%, p=0.015). In one patient, ERCP repeatedly failed due to inability to cannulate the papilla, which was located in a duodenal diverticulum. This patient had a cystic duct leakage (Amsterdam and Strasberg type A BDI). No RV procedure or reconstructive surgery was performed before referral to our center.

Table 2. Therapeutic Interventions prior to RV (N = 47).

Total,

N = 47, (%)

Complete

transection

N = 31

Partial

injury

N = 16 P value

Surgical interventions

Repair during initial cholecystectomy 11 (23) 10 (32) 1 (6) 0.046 Relaparotomy: Abdominal lavage without repair 9 (19) 6 (19) 3 (19) 0.96Non-surgical interventions

Percutaneous drainage of fluid collection 19 (40) 12 (39) 7 (44) 0.74ERCP 40 (85) 26 (84) 14 (88) 0.74

With papillotomy 23 (49) 16 (52) 7 (44) 0.61With stent 10 (21) 6 (19) 4 (25) 0.65

Number of ERCPs per patient (median, range) 1 (0-7) 1 (0-7) 1 (0-3) 0.71 PTBD 43 (92) 30 (97) 13 (81) 0.07

Number of PTBDs per patient (median, range) 1 (0-9) 1 (0-6) 1 (0-9) 0.89Time from diagnosis to RV (days), median (IQR) 38 (24-93) 38 (24-93) 44 (21-99)0.90Main reason for failure of PTBD or ERCP

Failure to cannulate the papilla 1 (2) 0 1 (6)0.015Bile duct disruption with obstruction* 22 (49) 19 (61) 3 (19)

Bile duct disruption without obstruction 11 (23) 4 (13) 7 (44)Inability to pass stenosis (failed canalization) 13 (28) 8 (26) 5 (31)

* Obstruction: due to a stenosis or occluding clip. BDI, bile duct injury; ERCP, endoscopic retrograde cholangiopancreatography; LC, laparoscopic cholecystectomy; PTBD, percutaneous transhepatic biliary drainage; RV, rendezvous procedure.

Part I | Chapter 3

64

Rendezvous procedureIndication for RV, time from diagnosis to RV procedure, details about the RV procedure itself and procedure-related adverse events are listed in table 4. A total of 55 RV procedures were performed in 47 patients. Two patients in whom the first RV attempt failed were successfully treated in a second RV procedure a few days later; six other patients underwent a second RV procedure to exchange an internally draining PTBD for endoscopically placed stents after the first RV had established bile duct continuity. Of all 55 procedures, 50 were successful (91%). Of 47 patients, RV was eventually successful in 44, leading to a technical success rate of 94%. There was no difference in success rate between antegrade or retrograde RV (13/14, 93% versus 36/39, 92%; p = 0.12). The site of RV was extraluminal in 27 procedures (49%) and intraluminal in 27 procedures (49%); in one patient, data on location of RV could not be retrieved. An occluding clip was present during 28 procedures (51%), significantly more in patients with a complete transection (23/38, 61% versus 5/17, 29%; p=0.03). In 5 procedures (9 %) the RV attempt failed. Reasons for failure were inability to pass the stricture in the bile duct (N=2), inability to make a connection between the two wires (N=2) and a broken guidewire (N=1). Eventually, RV was successful in 44/47 patients (94%). A median of 1 stent with a range of 1 to 5 stents were placed during the initial RV procedure. No 90-day mortality occurred. Overall procedure-related adverse events occurred in 10 patients (18%), with two patients suffering from more than one adverse event. Cholangitis, intra-abdominal abscess and liver abscess were the most common adverse events. There were no significant differences in the incidence of adverse events between patients with a complete transection and patients with a partial injury.

Long-term outcome after RV procedure Of 44 patients with a successful RV, a total of 39 patients subsequently underwent stent revisions, see table 5 and figure 2. In 5 patients reconstructive surgery was performed directly after a primary successful RV; reasons were ‘continuing bile leakage 2 months after RV’ (N=1) and ‘preference of the doctor’(N=4). All of these patients were treated before the year 2003; at that time the success rate of endoscopic treatment for extensive lesions was not yet established and repair with a HJ was considered the optimal option for definitive treatment.

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

65

Table 3. Details of RV procedure and short-term outcomeTotal,

N=55 (%)

Complete

transection

N = 38 (%)

Partial injury

N = 17 (%)

P value

Time from diagnosis BDI to RV (days), median (IQR)

38 (24-93) 38 (24-93) 44 (21-99) 0.90

Side of entrance of the hepatic system Right 28 (51) 20 (53) 8 (47)

0.67 Left 22 (40) 14 (37) 8 (47) Both 4 (7) 4 (11) 0

Location of RVExtraluminal 27 (49) 19 (50) 8 (47) 0.54Intraluminal 27 (49) 18 (47) 9 (53)Unknown 1 (2) 1 (3) 0

Direction of RV* Retrograde 39 (71) 24 (63) 15 (88) 0.08Antegrade 14 (26) 13 (34) 1 (6)Unknown 2 (4) 1 (3) 1 (6)

Presence of obstructing clip 28 (51) 23 (61) 5 (29) 0.03

Dilatation performed 26 (47) 21 (55) 5 (29) 0.08Number of stents inserted, median (range) 1 (0-5) 1 (0-5) 1 (0-3) 0.52Adverse Events

Overall morbidity 10 (18) 7 (18) 3 (18) 0.9590-day mortality 0 0 0 -Major bleeding 1 (2) 0 1 (6) 0.13False passage 0 0 0 -Pneumothorax 1 (2) 1 (3) 0 0.5Intra-abdominal Abscess 2 (4) 1 (3) 1 (6) 0.55Liver abscess 2 (4) 2 (5) 0 0.34Sepsis** 1 (2) 1 (3) 0 0.5Post-ERCP pancreatitis** 1 (2) 1 (3) 0 0.5Cholangitis 4 (7) 3 (8) 1 (6) 0.79

Technically successful procedure 50 (91) 34 (90) 16 (94) 0.58* Direction in which the snare or basket was inserted ** for which ICU observation was requiredBDI, bile duct injury; RV, rendezvous procedure.

Part I | Chapter 3

66

Page 1 of 2 Reference ID: Error! Unknown document property name.

Total N=47

Primary Success N=44

Primary Failure N=3

Progressive stenting N=39

Surgery N=5

Surgery N=3

Secondary failure N=8

Secondary success N=31

Long-term success N=26

Recurrent stenosis N=5

Surgery N=5

Surgery N=4

Permanent stent N=4

Figure 3. Flow diagram of long-term outcome of patients who underwent an RV.The median number of stent revisions per patient was 3 with a range of 1 to 6 revisions. The median period between RV and removal of the last stent was 7 months, ranging from 2 to 19 months. Median CCI (on a scale from 0 to 100) for the RV and subsequent stenting treatment was 21 with a range from 0 to 47, with no significant differences between the two groups. The most common adverse events during the period of progressive stenting contributing to the CCI were stent clogging (N=3), cholangitis (N=3), stent migration (N=3) and pain without signs of cholestasis (N=3). In 1 patient a false passage was made for which no additional interventions were required.

In 31 patients (66%), RV and subsequent stenting treatment were secondary successful. Four patients were referred for surgery during this period; in 2 patients this was due to prolonged stent dependency (> 1 year), the other 2 patients had a tight stenosis which could not be dilated adequately. In 4 patients (9%) with a persisting stenosis, surgery was indicated but not performed due to significant comorbidity or poor general condition of the patient; these patients were treated with a permanent, fully covered metal stent. Median follow-up after stent removal was 40 months (IQR 23-54 months). During follow-up, 5 patients (11%) developed a recurrent or new stenosis requiring surgical reconstruction. Altogether, in 17 patients (36%) RV was followed by surgical reconstruction. BDI related mortality occurred in 2 patients: the first patient, a 54 years old female, had undergone an unsuccessful RV attempt; after a period of external drainage of bile through a PTBD a HJ was performed 4 months after the RV attempt. She died 2 weeks

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

67

after surgery, post-mortem examination could not reveal the cause of death but sepsis from an unknown source was suggested. The second patient, a 69 years old male, died 2 years after RV from liver failure that resulted from severe cirrhosis due to longstanding stenosis of the CBD. Liver cirrhosis was already present at the time of RV; in this patient a permanent self-expandable metal stent was placed for a persisting stenosis. No significant differences in long-term success were found between patients with complete transection of the bile duct and partial BDI.

Table 4. Progressive stenting and long-term outcome of RV procedureTotal, N=47

(%)

Complete

transection,

N= 31 (%)

Partial

injury,

N= 16 (%) P-value

Primary success

Technically successful procedure 44 (94) 28 (90) 16 (100) 0.54 RV followed by progressive stenting 39 (83) 23 (74) 12 (75) 0.95

Adverse Events during period of stenting (N=39) CCI* (median, range) 21 (0-47) 20 (0-47) 13 (0-33) 0.57

Secondary success

Bile duct patency after stent removal 31 (66) 19 (61) 12 (75) 0.35 Number of stent revisions (median, range) 3 (1-6) 3 (1-6) 3 (1-6) 0.64 Period of progressive stenting, months (median,

range)7 (2-19) 7 (2-18) 11 (2-19) 0.01

Long-term success

Follow-up after stent removal, months (median, IQR)

40 (23-54) 41 (24-54) 38 (22-60) 0.77

Successful treatment 26 (55) 16 (52) 10 (63) 0.48Recurrent stenosis during follow-up 5 (11) 3 (10) 2 (13) 0.95 Time to recurrent stenosis, months (median, range)

6 (2-19) 11 (2-19) 6 (4-7) 0.99

Overall number of patients referred for surgery 17 (36) 13 (42) 4 (25) 0.25Overall number of patients treated with permanent stenting

4 (9) 4 (13) 0 0.28

RV related mortality 0 0 0 -BDI related mortality 2 (4) 2 (7) 0 0.3*All adverse events of RV and of subsequent stent revisions were included in calculating the CCIBDI, bile duct injury; CCI, comprehensive complication index; RV, rendezvous procedure.

Part I | Chapter 3

68

Discussion The present study shows that RV has a 94% primary success rate in patients with complex BDI following LC, with low morbidity and no observed procedure-related mortality. After successful consecutive stent revisions in 31 patients (66%) we found a long-term surgery-free and stent-free success rate of 55% (26 patients), after a median follow up of 40 months. The burden of this treatment seems acceptable with a relatively low median CCI of 21 on a scale of 0 to 100 [23]. In patients in whom a HJ was chosen in the course of treatment, either primarily or secondarily to treat refractory or recurrent stenoses , the RV acted as a bridge to a more elective surgical repair.An extraluminal RV can be challenging. In case of an occluding clip it can be accomplished using a catheter / guidewire combination to perforate the bile duct at a site close to the clip to enter the subhepatic space. This tract may then be dilated. Through injecting contrast the other end of the bile duct may be visualized and sometimes can be entered without needing RV. If a RV is required, a snare is introduced into the subhepatic space either percutaneously or endoscopically to pick up the wire introduced from the other end. After externalisation of the guidewire the defect must be bridged either with a percutaneously inserted internal/external, multiple side hole catheter into the duodenum, or with endoscopically placed plastic biliary stents. As the endoscopic route allows multiple biliary stents to be placed, thus supporting the bile duct wall from the inside and diminishing the risk of long-term stenosis [9,24], this method is preferred. However, this is not always possible in the first session, requiring a repeated RV a few weeks later to replace the PTBD for endoscopic stents. Modified methods of RV have been reported, such as extra-anatomical intraduodenal RV [25] and RV using magnetic compression anastomosis[26]. In our series, two patients underwent a modification of extraluminal RV by making a connection between the cystic duct and a right segmental branch. In both patients, this treatment resulted in a long-term success. Intraluminal RV for failed canalization can be required in case of severe iatrogenic stenosis [27]. Intraluminal RV for failed cannulation of the papilla is more commonly practiced. EUS-guided RV has been described and may be very useful in these patients[28-30].In literature there is limited evidence available concerning the use of the RV procedure in BDI after laparoscopic cholecystectomy. Although the RV was first described in 1987 [31], literature consists of case reports and small series [16-18,25,27,32-38]. In a series of 23 patients with miscellaneous diseases undergoing RV, Tomizawa et al reported a technical success rate of 88% and an overall morbidity of 22% [18]. Similar results were reported by de Calvo in 14 patients [19]. Little is known regarding long-term outcome

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

69

of the RV; Fiocca et al reported a remarkably high technical success rate of 100% and long-term success rate of 82%, with low morbidity, in a series of 22 patients with complete transection of the CBD after cholecystectomy [17]. The difference between long-term success rate in this study and our series might be due to a more aggressive treatment protocol with a longer duration of stent treatment.

Endoscopic treatment is usually the first step in the management of BDI; ERCP is used to identify the injury and classify the severity of the lesion. Simple leakage of the cystic duct or CBD can be treated with endoscopic stents, however in case of complete or nearly complete transection of the CBD, both ERCP and PTBD have a high risk of failure, due to loss of biliary continuity [9,11]. In these cases an RV procedure provides a useful treatment option. Endoscopic treatment of partial BDI is reported to be successful in 66-89% [9,11,20,24]. The difference in long-term success rate of the RV treatment in the present study compared to usual endoscopic treatment of BDI can be explained by the fact that our patient population merely consists of complex cases of BDI. With complete transection of the bile duct the vasculature of the duct itself is also severed, potentially impairing healing and increasing the risk of stricture formation in the long term. However in the present study we observed that even in case of complete transection bile duct patency could be restored in a majority of patients. Limitations of this study are the retrospective design and wide timespan of the study over 20 years. In this time, common practice regarding work-up and treatment of BDI changed. Furthermore, we describe a single center experience. Although every patient was discussed thoroughly in a multidisciplinary team, some decisions in the course of treatment were subjective to preferences of the treating physician or based on the judgement of a single physician, such as the presence or absence of a stenosis during ERCP and consequently, the decision to pursue stenting or remove stents. The small sample size of our study was another limitation. As the RV is a rescue procedure with limited indication, only a low percentage of our cohort of patients with BDI underwent an RV. We used both the Amsterdam and the Strasberg/Bismuth classifications to classify BDI [20-22]. No correlation between complete transection of the bile duct and long-term outcome could be demonstrated. Only a Bismuth type III location of BDI was significantly associated with long-term treatment failure [data not shown]. We observed a tendency towards association of vascular injury with long-term failure of treatment. As our BDI patients did not routinely undergo angiography to exclude vascular injury, these data are somewhat unreliable. However, this possible association is in line with literature reporting on the consequences of vascular injury occurring along with BDI [39]. The number of stents inserted at first intervention has been reported to correlate with

Part I | Chapter 3

70

long-term outcome in endoscopic treatment of BDI [9,24]. We however, could not demonstrate a similar association in RV patients. In our opinion, the RV can be considered as primary treatment in all patients with a disrupted bile duct, irrespective of the type of BDI, unless there is loss of bile duct tissue of more than 1 centimeter. Due to the retrospective nature of this study we cannot provide data regarding the loss of tissue in our series. However, it is reasonable to believe that BDI with a defect of >1cm, which will have to be overgrown by fibrous tissue, will result in focal stenosis. In these patients RV potentially serves as a bridge to surgery but also has much added value as it allows for internal drainage of bile, preventing dehydration and electrolyte loss. RV is particularly useful in patients who present with a disrupted major bile duct in whom the diagnosis is made one week or more after initial cholecystectomy. Patients in whom BDI is recognized intraoperatively or within a few days after cholecystectomy may benefit from an early surgical repair by means of a HJ [13]. In patients who undergo surgical repair between 1 and 6 weeks after cholecystectomy, complications and risk of long-term stenosis are reported to be higher than in patients undergoing delayed surgical repair [12-15]BDI is a burdensome complication of a commonly performed operation and comes with detrimental short- and long-term quality of life (QoL) [6,40]. This emphasizes the importance of optimal and patient-centered treatment. In a previous publication reporting on our cohort of BDI patients, duration and type of treatment did not affect QoL, suggesting that from a patient perspective, the longer duration of endoscopic and radiological treatments are acceptable compared to surgery [40]. In summary, the RV procedure has a high primary success rate with low morbidity and no observed 90-day mortality, when performed in an expert center. When ERCP or PTBD by itself fails to pass the injury, the RV procedure should be considered as the next step in treatment before moving on to surgery. RV can act as a bridge to surgery in the not acute setting, and be the final treatment with a long-term success rate of 55%, irrespective of the severity of the BDI.

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

71

List of abbreviations

ASA American Society of AnesthesiologistsBDI                 Bile duct injuryCBD                Common bile ductCCI                  Comprehensive complication indexERCP               Endoscopic retrograde cholangiopancreatographyHJ                   HepaticojejunostomyICU                 Intensive care unitIQR                 Interquartile rangeLC                    Laparoscopic cholecystectomyPTBD              Percutaneous transhepatic biliary drainageQoL Quality of lifeRHD                Right hepatic ductRV                   Rendezvous procedureSD                   Standard deviation

Part I | Chapter 3

72

References

1. Calvete J, Sabater L, Camps B et al. Bile duct injury during laparoscopic cholecystectomy: myth or reality of the learning curve? Surg Endosc 2000; 14: 608-611

2. Go PM, Schol F, Gouma DJ. Laparoscopic cholecystectomy in The Netherlands. Br J Surg 1993; 80: 1180-1183

3. Rystedt J, Lindell G, Montgomery A. Bile Duct Injuries Associated With 55,134 Cholecystectomies: Treatment and Outcome from a National Perspective. World J Surg 2016; 40: 73-80

4. Booij KA, de Reuver PR, Yap K et al. Morbidity and mortality after minor bile duct injury following laparoscopic cholecystectomy. Endoscopy 2015; 47: 40-46

5. Strasberg SM, Gouma DJ. ‘Extreme’ vasculobiliary injuries: association with fundus-down cholecystectomy in severely inflamed gallbladders. HPB (Oxford) 2012; 14: 1-8

6. Booij KAC, de Reuver PR, van Dieren S et al. Long-term Impact of Bile Duct Injury on Morbidity, Mortality, Quality of Life, and Work Related Limitations. Ann Surg 2017, DOI: 10.1097/SLA.0000000000002258:

7. de Reuver PR, Rauws EA, Bruno MJ et al. Survival in bile duct injury patients after laparoscopic cholecystectomy: a multidisciplinary approach of gastroenterologists, radiologists, and surgeons. Surgery 2007; 142: 1-9

8. Flum DR, Cheadle A, Prela C et al. Bile duct injury during cholecystectomy and survival in medicare beneficiaries. JAMA 2003; 290: 2168-2173

9. de Reuver PR, Rauws EA, Vermeulen M et al. Endoscopic treatment of post-surgical bile duct injuries: long term outcome and predictors of success. Gut 2007; 56: 1599-1605

10. Janssen JJ, van Delden OM, van Lienden KP et al. Percutaneous balloon dilatation and long-term drainage as treatment of anastomotic and nonanastomotic benign biliary strictures. Cardiovasc Intervent Radiol 2014; 37: 1559-1567

11. Rauws EA, Gouma DJ. Endoscopic and surgical management of bile duct injury after laparoscopic cholecystectomy. Best Pract Res Clin Gastroenterol 2004; 18: 829-846

12. de Reuver PR, Grossmann I, Busch OR et al. Referral pattern and timing of repair are risk factors for complications after reconstructive surgery for bile duct injury. Ann Surg 2007; 245: 763-770

13. Dominguez-Rosado I, Sanford DE, Liu J et al. Timing of Surgical Repair After Bile Duct Injury Impacts Postoperative Complications but Not Anastomotic Patency. Ann Surg 2016; 264: 544-553

14. Stilling NM, Fristrup C, Wettergren A et al. Long-term outcome after early repair of iatrogenic bile duct injury. A national Danish multicentre study. HPB (Oxford) 2015; 17: 394-400

15. Ismael HN, Cox S, Cooper A et al. The morbidity and mortality of hepaticojejunostomies for complex bile duct injuries: a multi-institutional analysis of risk factors and outcomes using NSQIP. HPB (Oxford) 2017; 19: 352-358

16. Donatelli G, Vergeau BM, Derhy S et al. Combined endoscopic and radiologic approach for complex bile duct injuries (with video). Gastrointest Endosc 2014; 79: 855-864

17. Fiocca F, Salvatori FM, Fanelli F et al. Complete transection of the main bile duct: minimally

Percutaneous-endoscopic rendezvous for bile duct injury

Chap

ter 3

73

invasive treatment with an endoscopic-radiologic rendezvous. Gastrointest Endosc 2011; 74: 1393-1398

18. Tomizawa Y, Di Giorgio J, Santos E et al. Combined interventional radiology followed by endoscopic therapy as a single procedure for patients with failed initial endoscopic biliary access. Dig Dis Sci 2014; 59: 451-458

19. Calvo MM, Bujanda L, Heras I et al. The rendezvous technique for the treatment of choledocholithiasis. Gastrointest Endosc 2001; 54: 511-513

20. Bergman JJ, van den Brink GR, Rauws EA et al. Treatment of bile duct lesions after laparoscopic cholecystectomy. Gut 1996; 38: 141-147

21. Bismuth H, Majno PE. Biliary strictures: classification based on the principles of surgical treatment. World J Surg 2001; 25: 1241-1244

22. Strasberg SM, Hertl M, Soper NJ. An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 1995; 180: 101-125

23. Slankamenac K, Graf R, Barkun J et al. The comprehensive complication index: a novel continuous scale to measure surgical morbidity. Ann Surg 2013; 258: 1-7

24. Parlak E, Disibeyaz S, Odemis B et al. Endoscopic treatment of patients with bile duct stricture after cholecystectomy: factors predicting recurrence in the long term. Dig Dis Sci 2015; 60: 1778-1786

25. Meduri B, Derhy S, Dhumane P et al. Extra-anatomical intraduodenal endoscopic-radiologic biliary rendezvous for treatment of iatrogenic complete stenosis of the common bile duct. Endoscopy 2015; 47 Suppl 1: E565-566

26. Perez-Miranda M, Aleman N, de la Serna Higuera C et al. Magnetic compression anastomosis through EUS-guided choledochoduodenostomy to repair a disconnected bile duct in orthotopic liver transplantation. Gastrointest Endosc 2014; 80: 520-521

27. Tomoda T, Kato H, Mizukawa S et al. Biliary Anastomotic Stricture After Adult Living Donor Liver Transplantation With Duct-to-Duct Reconstruction: Outcome After Endoscopic Treatment Including Rendezvous Procedure. Transplantation 2016; 100: 1500-1506

28. Iwashita T, Yasuda I, Mukai T et al. EUS-guided rendezvous for difficult biliary cannulation using a standardized algorithm: a multicenter prospective pilot study (with videos). Gastrointest Endosc 2016; 83: 394-400

29. Kawakubo K, Kuwatani M, Kato S et al. Direct puncture of the ampulla as a modified Endoscopic ultrasound-guided rendezvous technique. Endosc Ultrasound 2017, DOI: 10.4103/eus.eus_31_17:

30. Nakai Y, Isayama H, Matsubara S et al. A novel “hitch-and-ride” deep biliary cannulation method during rendezvous endoscopic ultrasound-guided ERCP technique. Endoscopy 2017, DOI: 10.1055/s-0043-113444:

31. Sommer A, Burlefinger R, Bayerdorffer E et al. [Internal biliary drainage in the “rendezvous” procedure. Combined transhepatic endoscopic retrograde methods]. Dtsch Med Wochenschr 1987; 112: 747-751

32. Benner KG, Ivancev K, Porayko MK et al. Re-establishment of biliary tract continuity by a combined ERCP and PTC approach after iatrogenic common bile duct ligation. Gastrointest

Part I | Chapter 3

74

Endosc 1992; 38: 506-50933. Nasr JY, Hashash JG, Orons P et al. Rendezvous procedure for the treatment of bile leaks and

injury following segmental hepatectomy. Dig Liver Dis 2013; 45: 433-43634. Curcio G, Miraglia R, Ligresti D et al. Intraperitoneal rendezvous: a mini-invasive biliary

reconstruction. Gastrointest Endosc 2016; 84: 184-18535. Gronroos JM. Unsuccessful endoscopic stenting in iatrogenic bile duct injury: remember rendezvous

procedure. Surg Laparosc Endosc Percutan Tech 2007; 17: 186-18936. Miller T, Singhal S, Neese P et al. Non-operative repair of a transected bile duct using an endoscopic-

radiological rendezvous procedure. J Dig Dis 2013; 14: 509-51137. Odemis B, Oztas E, Akpinar MY et al. An alternative treatment for biliary injuries characterized

by complete transection of the common bile duct: intraperitoneal rendezvous. Therap Adv Gastroenterol 2017; 10: 521-523

38. Rana S, Sharma V, Arun AC et al. A modification of rendezvous technique for endoscopically treating transected common bile duct following cholecystectomy. Journal of Digestive Endoscopy 2014; 5: 129

39. Stewart L, Robinson TN, Lee CM et al. Right hepatic artery injury associated with laparoscopic bile duct injury: incidence, mechanism, and consequences. J Gastrointest Surg 2004; 8: 523-530; discussion 530-521

40. de Reuver PR, Sprangers MA, Rauws EA et al. Impact of bile duct injury after laparoscopic cholecystectomy on quality of life: a longitudinal study after multidisciplinary treatment. Endoscopy 2008; 40: 637-643