ERCP for the treatment of biliary complications following cholecystectomy

ACHAIKI IATRIKI | 2021; 40(3):137–142

Original research article

Christos Konstantakis1, Georgios Theocharis1, Georgios Skroubis2, Ioannis Kehagias2, Christos Triantos1, Konstantinos Thomopoulos1

1Division of Gastroenterology, Department of Internal Medicine, University Hospital of Patras, Patras, Greece
2Department of Surgery, University Hospital of Patras, Patras, Greece

Received: 10 Jan 2021; Accepted: 02 Mar 2021

Corresponding author: Christos Konstantakis, MD, University Hospital of Patras, Rio 26504, Patras, Greece, Tel.: +30 6974 563157, Fax: +30 261 0999518, E-mail:

Key words: ERCP, biliary complications, biliary leak, biliary stenosis, management, cholecystectomy, guidelines



Background: The purpose of this study is to evaluate the feasibility and efficacy of endoscopic treatment of biliary complications in patients undergoing cholecystectomy, as represented by a decade of data.

Methods: During the 01/2010 – 12/2019 period, 4465 ERCPs were performed in our department. We selected and studied retrospectively cases with complications after cholecystectomy. We evaluated mainly patients with postoperative biliary leak and biliary stenosis. Patients with choledocholithiasis found after cholecystectomy were not included in the study unless coexisting with the above conditions. All data were retrieved from patients´ files and electronic records.

Results: A total of 86 ERCPs (1.9%) were performed in 65 unique patients (31 male) for biliary complications following cholecystectomy. Patients range in age from 27 to 90. Forty-eight patients (73.8%) presented with Amsterdam type A injuries (leakage from cystic duct, duct of Luschka or peripheral biliary radicals), 8 (12.3%) type B (major bile duct leakage), 5 (7.6%) type C (an isolated ductal stricture) and 4 (6.1%) type D (complete transection of the bile duct). In one patient (0.65%) selective canulation of the bile duct was not possible. In 60 out of the 65 patients’ permanent resolution of the biliary injury was achieved giving an overall success rate of 92.3 % following one to five procedures.

Conclusion: Endoscopic treatment of postoperative complications of cholecystectomy is both feasible and highly effective, accompanied by a very high success rate. However, multiple sessions / hospitalizations are often required, and in cases such as complete duct transection the solution remains surgical.


Cholecystectomy (CH) is considered the gold standard for the treatment of symptomatic cholelithiasis among other indications (acalculous cholecystitis, gallbladder polyps, porcelain gallbladder). It has an excellent safety profile. In population-based studies, post-operative mortality after CH for gallstone disease ranged between 0.1% – 0.7%. Complications following CH are rare but not insignificant. They can be identified intra-operatively or present in the immediate or late post-operative period. Surgical site infection, haemorrhage, bowel perforation, post cholecystectomy syndrome, cystic duct remnant, residual and recurrent bile duct stones and iatrogenic bile duct injuries (bile leak and biliary strictures) are among the most common.

Biliary injuries (BI) are perhaps the most feared complications following cholecystectomy. Less than 40% of these injuries are recognized during operation. The majority of BI will present during the 1st post – operative week (leaks) but can present years after index CH as biliary stenosis. Patients with BI are at an increased 1y mortality (4% vs. 1%) with a hazard ratio of 1.92 (95% CI = 1.24–2.97) when compared with uncomplicated CH. The introduction of laparoscopic cholecystectomy (LC) has not changed the frequency of complications, as it was comparable in a systematic review between LC and open cholecystectomy (OC). Despite increasing experience with laparoscopy, an unchanging incidence of bile duct injury between 0.42 and 1.1% has been reported in earlier studies. Although, true incidence may not be known due to underreporting bias [1-4].

Endoscopic retrograde cholangiopancreatography (ERCP) has been safely and effectively used for the diagnosis and management of postsurgical biliary leaks and stenoses [1]. Several endoscopic techniques have been implemented in the management of bile leaks. Biliary sphincterotomy (BS) alone, biliary stenting with or without sphincterotomy, and nasobiliary drainage with or without sphincterotomy have all been tried. The main principle of these methods is to reduce the pressure gradient between the biliary tree and the duodenum, thus facilitating bile flow into the duodenum instead of extravasation via the leak site. The principal treatment of bile duct stricture, with or without bile leak, is gradual dilatation of the strictured segment with pneumatic balloon dilators and stenting, with stents being exchanged every 3 months, till the site of narrowing disappeared. Stenting is essential for the preservation of the dilatation effect and usually multiple endoscopic sessions are required. Both multiple plastic stents and single fully covered metallic stent have been used to achieve this [3-5].

There are many classifications of post – CH BI. The Amsterdam classification is widely accepted because it links directly BI and treatment options, which is quite practical to use for endoscopic purposes (Table 1) [6].

The purpose of this study was to evaluate the feasibility and efficacy of endoscopic treatment of biliary complications in patients undergoing CH, as illustrated by a decade of data.

Materials and methods

All patients who underwent ERCP in our department for BI complications following both LC and OC between 01/2011 and 12/2020 were identified and included in our study. The ethics board of our hospital approved the performance of this study.

All data (table 1) regarding the baseline patients’ demographics, the characteristics of the biliary injury, the type of endoscopic management, the presence of common bile duct stones, the diagnostic and therapeutic endoscopic interventions and follow-up (resolution of BI / adverse events / surgical treatment) were retrospectively retrieved from patients´ files and electronic records.

Follow-up: In our hospital (tertiary centre) alone, there are more than 500 cholecystectomies performed annually. Furthermore, our hepatobiliary unit acts as regional referral centre. The likelihood of patients being referred to another unit would be truly improbable. On the contrary, BI patients are referred to our department from other hospitals too. That means that most outcomes (including any long – term outcomes on biliary strictures) could be recovered from our records (especially those requiring further endoscopic evaluation / intervention).

Patients with postoperative biliary leaks and/or biliary strictures were included. Patients with choledocholithiasis found after cholecystectomy were not included in the study unless coexisting with BI.

These patients were referred by the attending surgeon for persistent bilious-looking drainage from the surgical drain or following investigation of clinically relevant post-operative symptoms (abdominal pain, fever, nausea, vomiting, jaundice, gut distension).

Suspected bile duct injuries were investigated with the use of imaging modalities, like abdominal computed tomography (CT), abdominal ultrasound (US), and/or magnetic resonance cholangiopancreatography (MRCP) and were confirmed by fluoroscopy during ERCP. Patients with strong clinical suspicion underwent ERCP without prior investigation.

The type of the biliary leak / injury was classified by reviewing the cholangiography acquired during the endoscopic session, and it was graded according to the Amsterdam criteria [6].

Type A: Leakage from cystic duct or peripheral radicals / aberrant ducts.

Type B: Common bile duct injury, leakage (with or without stricture)

Type C: Bile duct stricture, without leakage

Type D: Complete transection or excision of the common bile duct with inability to visualize the proximal biliary radicles and failure of passage of guide wire proximally.

The endoscopic therapeutic interventions performed for the management of complication were recorded and classified into sphincterotomy, stenting or combination therapy. All procedures were performed by experienced pancreatobiliary endoscopists. The performed technique in each case was mainly based on endoscopist’s preference. Usually, a 10 Fr plastic stent was inserted proximal to the site of leakage or stricture to eliminate the transpapillary biliary-duodenal pressure gradient. The stents were removed after four weeks in patients with leaks. For patients with bile duct stricture, dilatation with 6-10 mm pneumatic biliary balloons and placement of a gradually increased number of 10Fr plastic stents was performed, with three-month intervals, until stricture resolution.

If residual bile duct stones were identified, they were also removed by balloon extraction, following sphincterotomy.

Success of endoscopic management was defined as:

In the immediate post – ERCP period: Resolution of symptoms, normalization of liver enzymes and significant reduction / cessation of drainage output in patients with a percutaneous drain.

On follow-up: No adverse outcomes recorded following a) removal of the drain in patients with a leak, b) resolution of the stricture in cholangiography when the stents were removed if a stricture was present.

ERCP failure was defined as the need for salvage biliary surgery. Note that need for repeat ERCP in a given patient was sometimes necessary and is not considered a failure if the objectives stated above were eventually achieved.


A total of 86 out of 4465 ERCPs (1.9%) were performed during the study period for biliary complications following cholecystectomy in 65 unique patients. Mean age was 61.7 (27 – 90), with relatively equal gender distribution (31 men, 34 women).

Out of the 65 patients, 48 patients (73.8%) presented with Amsterdam type A injuries (leakage from cystic duct, duct of Luschka or peripheral biliary radicals), 8 (12.3%) with type B (major bile duct leakage), 5 (7.6%) with type C (an isolated ductal stricture) and 4 (6.1%) with type D (complete transection of the bile duct) (Table 1).

In the 4 patients with type D injuries, ERCP contributed strictly diagnostic input, confirming the diagnosis of a complete CBD transection. All patients in this group were referred for surgical treatment. In one patient, presenting with jaundice attributed to a post-operative biliary stricture (type C BI), selective canulation of the bile duct wasn’t possible, even after two attempts. This patient was also treated surgically.

Seven (7) out of the 65 patients (10.7%) had underlying choledocholithiasis (undetected at baseline evaluation) and bile duct clearance of stones was performed following sphincterotomy.

Four (4) patients (6.1%) underwent only sphincterotomy (1 patient was eventually classified as type D) and 57 patients (87%) a combination of sphincterotomy and stent placement. Plastic stents were used in the majority, while in three patients total, a fully covered self-expandable metallic stent was placed in the first (one patient) or second ERCP session (in two patients – rescue therapy).

In five patients (7.6%) more than one ERCP was performed. One patient required repeat ERCP due to bile duct cannulation failure. The rest four patients due to failure to completely resolve the problem at first attempt. Additionally, patients with bile duct stricture required 3 to 5 sessions, performed at 3-month intervals, were treated with insertion of an increasing number of stents (max number of sessions was 5 and maximum number of simultaneous stents was 4).

In 60 out of the 65 patients, permanent resolution of the BI was achieved giving an overall success rate of 92.3%.

Safety profile: There were 4 cases of post-ERCP pancreatitis (mild and moderately severe), three cases of infectious complications / cholangitis but no case of bleeding, perforation, or mortality related to endoscopic treatment.


Endoscopic management of post cholecystectomy biliary injuries is highly successful. In our study 92.3 % of injuries were treated successfully following one or more endoscopic sessions. Only 4 out of the 65 patients eventually required surgical intervention. Surgery is required in patients with type D injuries according to the Amsterdam classification [6]. In cases with complex biliary injury restoration of the continuity of the biliary tree is nearly impossible with an endoscopic / radiologic approach (only few case reports) and these patients usually require operative management. Timing and type of surgery depends on patient condition and local expertise. The most common type of repair surgery is the (Roux en Y) hepaticojejunostomy.

Although the role of endoscopic management of post cholecystectomy bile duct injuries has been established as the preferred option, it still remains unclear which is the optimal endoscopic approach [7-9]. Endoscopic sphincterotomy, stenting, and combination therapy have all been used. In our study group the majority of patients were managed with both techniques.

The effectiveness of both endoscopic stenting and sphincterotomy alone in the treatment of biliary leaks after cholecystectomy has been a topic of research for many authors. It is an established fact that ES carries an additional risk of complications (pancreatitis, bleeding, perforation). Besides these drawbacks it is a very attractive option because it:

  1. obviates the need for repeat endoscopy (as opposed to stent insertion) and thus
  2. reduces the cost of the procedure. However, this must be weighed against potential severe and costly ES complications.
  3. finally, cannulation of the papilla is easier following sphincterotomy, thus facilitating easier stent insertion especially in cases of guide wire loss.

However, several studies report that ES is inferior to only stenting in all kind of leaks [9-12] with the exception maybe of Amsterdam A leaks where ES was reported to be non-inferior to combination treatment (ES plus stent) [3]. In other studies, endoscopic sphincterotomy alone has been found effective in treating bile leak patients [13,14]. Sandha et al., [14] showed in their study of 207 bile leak patients that ES alone is an effective treatment for the low-grade (LG) leaks. For high-grade leaks, they still recommended biliary stenting. Aksoz et al., [12] showed in their 31-patient study that ES is an effective treatment (87% success rate) when treating LG biliary leak. They recommend stenting as the primary treatment for high-grade leaks and for LG leaks only in the event that ES fails. Mavrogiannis et al., in a prospective study found that small-diameter biliary stent alone can be as effective and safe as endoscopic sphincterotomy followed by insertion of a large-diameter stent in bile leak patients [15]. Different results between studies may be influenced by the extent of sphincterotomy. If the ES is the only treatment, it has to be done properly. In our unit, ES is performed to cut the sphincter completely by extending the incision to the superior margin of the intramural bile duct. Smaller ES may not be effective enough when treating leaks, and this may have been the reason of poorer outcomes in previous studies. Moreover, extensive ES in these patients was carried out in our unit by expert endoscopists. The above represent the experience from our centre and in no way constitute a general recommendation. ES (especially extended) carries a non-negligible risk for serious complications even in the hands of the most experienced endoscopist.

Endoscopic (European) society guidelines (ESGE) advocate the use of stent insertion. Stent diameter (7-Fr vs 10-Fr) does not seem to be an issue [16].

Despite frequent sphincterotomies the risk of pancreatitis, bleeding, and perforation has been low in our study and similar results have been reported in previous studies [10-14].

Even following endoscopic sphincterotomy and stenting, some leaks cannot be ameliorated and these patients need additional intervention. Placement of a fully covered self-expandable metallic stent may reduce extravasation [17,18]. Nasobiliary drainage may be helpful in some patients [19].

Second look ERCP and cholangiography is not required for removal of the stent, provided that there is no sign of clinical, biochemical or radiological abnormality persisting. Stents can be removed with normal upper endoscopy using either an end- or side-viewing endoscope without performing fluoroscopy – ERCP, and this could help reduce costs and adverse events [20].

Endoscopic management has the advantage that the interval between BI and index ERCP for treatment does not seem to affect the outcome [16]. With that in mind, we believe that ERCP must be performed as soon as possible in order to prevent peritonitis and/or sepsis. Drain placement (of bilomas) either during the primary operation in case of injury suspicion or radiologically post-surgery is beneficial and it is recommended if there are difficulties with the anatomy of the biliary tree during the operation.

Our study has several limitations which mainly arise from its retrospective design. Several methodological issues regarding patient follow-up and data collection exist. Furthermore, the overwhelming majority of these patients are diagnosed and referred from other departments or even hospitals. All this further hinders acquisition of useful input.

In conclusion, ERCP is an effective procedure for the treatment of post cholecystectomy biliary complications.

Conflict of interest disclosure

None to declare

Declaration of funding sources

None to declare

Author Contributions

C. Konstantakis, K. Thomopoulos conception and design; G. Theocharis, G. Skroubis, I. Kehagias analysis and interpretation of the data; C. Konstantakis, K. Thomopoulos drafting of the article; C. Triantos, C. Konstantakis critical revision of the article for important intellectual content; K. Thomopoulos final approval of the article.


1 .Adler DG, Papachristou GI, Taylor LJ, McVay T, Birch M, Francis G, et al. Clinical outcomes in patients with bile leaks treated via ERCP with regard to the timing of ERCP: A large multicenter study. Gastrointest Endoscop. 2017;85(4):766-72.
2 .Vlaemynck K, Lahousse L, Vanlander A, Piessevaux H, Hindryckx P. Endoscopic management of biliary leaks: A systematic review with meta-analysis. Endoscopy. 2019;51(11):1074-81.
3 .Rainio M, Lindström O, Udd M, Haapamäki C, Nordin A, Kylänpää L. Endoscopic Therapy of Biliary Injury After Cholecystectomy. Dig Dis Sci. 2018;63(2):474-80.
4 .Way LW, Stewart L, Gantert W, Liu K, Lee CM, Whang K, et al. Causes and prevention of laparoscopic bile duct injuries: analysis of 252 cases from a human factors and cognitive psychology perspective. Ann Surg. 2003;237(4):460-9.
5 .Abbas A, Sethi S, Brady P, Taunk P. Endoscopic management of postcholecystectomy biliary leak: When and how? A nationwide study. Gastrointest Endoscop. 2019;90(2):233-41.
6 .Keulemans YC, Bergman JJ, de Wit L T, Rauws E A, Huibregtse K, Tytgat G N, et al. Improvement in the management of bile duct injuries? Am Coll Surg. 1998;187(3):246-54.
7 .Agarwal N, Sharma B, Garg S, Kumar R, Sarin S. Endoscopic management of postoperative bile leaks. Hepatobiliary Pancreat Dis Int. 2006;5(2):273-7.
8 .Do IN, Kim JC, Park SH, Lee JY, Jung SW, Cha JM, et al. The outcome of endoscopic treatment in bile duct injury after cholecystectomy. Korean J Gastroenterol. 2005;46(6):463-70.
9 .Kaffes J, Hourigan L, De Luca N, Byth K, Williams SJ, Bourke M. Impact of endoscopic intervention in 100 patients with suspected postcholecystectomy bile leak. Gastrointest Endosc. 2005;61(2):269-75.
10 .Marks JM, Ponsky JL, Shillingstad RB, Singh J. Biliary stenting is more effective than sphincterotomy in the resolution of biliary leaks. Surg Endosc. 1998;12(4):327-30.
11 .Youngelman DF, Marks JM, Ponsky T, Ponsky JL. Comparison of bile duct pressures following sphincterotomy and endobiliary stenting in a canine model. Surg Endosc. 1997;11(2):126-8.
12 .Dolay K, Soylu A, Aygun E. The role of ERCP in the management of bile leakage: endoscopic sphincterotomy versus biliary stenting. J Laparoendosc Adv Surg Tech A. 2010;20(5):455-9.
13 .Sandha GS, Bourke MJ, Haber GB, Kortan PP. Endoscopic therapy for bile leak based on a new classification: results in 207 patients. Gastrointest Endosc. 2004;60(4):567-74.
14 .Aksoz K, Unsal B, Yoruk G, Buyrac Ζ, Haciyanli Μ, Akpinaret Ζ, et al. Endoscopic sphincterotomy alone in the management of low-grade biliary leaks due to cholecystectomy. Dig Endosc. 2009;21(3):158-61.
15 .Mavrogiannis C, Liatsos C, Papanikolaou IS, Karagiannis S, Galanis P, Romanos A. Biliary stenting alone versus biliary stenting plus sphincterotomy for the treatment of post-laparoscopic cholecystectomy biliary leaks: a prospective randomized study. Eur J Gastroenterol Hepatol. 2006;18(4):405-9.
16 .Dumonceau JM, Tringali A, Papanikolaou IS, Blero D, Mangiavillano B, Schmidt A, et al. Endoscopic biliary stenting: indications, choice of stents, and results: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline – Updated October 2017. Endoscopy. 2018;50(9):910-30.
17 .Pausawasadi N, Soontornmanokul T, Rerknimitr R. Role of fully covered self-expandable metal stent for treatment of benign biliary strictures and bile leaks. Korean J Radiol. 2012;13 Suppl 1(Suppl 1): S67-73.
18 .Tringali A, Reddy N, Ponchon T, Neuhaus H, Ferrán González-Huix Lladó, Navarrete C, et al. Treatment of post-cholecystectomy biliary strictures with fully-covered self-expanding metal stents – results after 5 years of follow-up. BMC Gastroenterol. 2019;19(1):214.
19 .Pinkas H, Brady P. Biliary leaks after laparoscopic cholecystectomy: time to stent or time to drain. Hepatobiliary Pancreat Dis Int. 2008;7(6):628-32.
20 .Jain V, Yeasted N, Pooran N. Necessity of a repeat cholangiogram during biliary stent removal after postcholecystectomy bile leak. Can J Gastroenterol. 2012;26(10):701-4.