Introduction

Biliary fistulae can result from various conditions like post biliary and liver surgery, abdominal trauma, and rupture of liver abscesses and cysts in the biliary system. Clinically significant bile leak after cholecystectomy is infrequent but constitutes a serious complication and poses difficulties in management. The rate of bile duct injury during open cholecystectomy has been estimated to be 0.1%-0.2% and 0.3%-0.6% in patients after laparoscopic cholecystectomy. [1-4] Surgical management of biliary fistulae is associated with high morbidity and mortality. Biliary endoscopic procedures have become the treatment of choice for the management of biliary fistulae. Depending on the situation identified cholangiographically, various endoscopic methods including biliary sphincterotomy, nasobiliary drainage (NBD) and biliary stent placement are used. We studied prospectively the endoscopic management of biliary fistulae after surgery.

Methods

During a 48-month period (July 2001-June 2005), we treated 90 patients with bile leaks and biliary fistulae after cholecystectomy.  All patients were clinically evaluated and subjected to routine investigations such as hemography, liver function test, determination of coagulation parameters, abdominal ultrasonography, and endoscopic retrograde cholangiopancreaticography (ERCP). Bile leaks and fistulae were diagnosed by the presence of clinical features like abdominal pain, fever, abdominal distension, jaundice and radiological findings (ultrasound and CT scan of the abdomen). The pre-sence of bile leaks and fistulae were confirmed by ERCP and the appearance of bile in percutaneous drainage of abdominal collections. Endoscopic procedures were performed with a side-viewing duodenoscope (TJF 20, Olympus Optical Co., Tokyo, Japan). Prophylactic antibiotics were prescribed 24 hours before the procedure and continued for next 24-48 hours. Of the 90 patients with bile leaks after surgery, 18 patients had complete transection of the common bile duct. They were subjected to bilioenteric anastomosis and excluded from analysis.  In the remaining patients after cholangiography and localization of the site of bile leaks, therapeutic procedures like sphincterotomy, biliary stenting and nasobiliary drainage (NBD) were performed. If residual stones were seen in the common bile duct, sphincterotomy was followed by stone extraction using dormia basket. Nasobiliary drain or stents of 7F size were placed according to the standard techniques. The NBD and stents were positioned so that their proximal end lies above the site of the leaks. NBD was removed when bile leakage stopped and closure of the fistula confirmed cholangiographically. The stents were removed after an interval of 6-8 weeks.

The study protocol was approved by the hospital ethical committee. Also, written consent was obtained from all the patients before their enrollment.

Results

The study patients included 56 women and 34 men with mean age of 36 years (range 18-62 years). All the   patients had bile leaks and fistulae after cholecystectomy (open cholecystectomy in 45 patients, cholecystectomy with common bile duct exploration in 20, and laparoscopic cholecystectomy in 25).

Of the 90 patients, 26 had bile leak during surgery and were subjected to peroperative subhepatic drain. The amount of daily drain output was less than 150 ml in 12 patients and more than 150 ml in 14 patients. The mean interval between index surgery and detection of bile leak was 4.5 days (range 0-26 days). Presenting clinical features included biliocutaneous fistula (36 patients), biloma formation (22), bile ascites (4) and jaundice with cholangitis (24). In the 36 patients with biliocutaneous fistula, bile flowed to the skin surface through the T-tube tract (20 patients) or through and around existing drains (16). All the 22 patients with biloma were subjected to percutaneous catheter drainage. The mean duration of percutaneous catheter drainage before ERCP was 12.8?0 days.

Of the 90 patients with postoperative bile leak, 18 patients underwent complete transection of the common bile duct on ERCP and were subjected to bilioenteric anastomosis. The site of leak was demonstrated in 72 patients with bile leak after surgery. Bile leak from the cystic duct was found in 38 patients (Fig.1), from the common bile duct in 30, and from the right hepatic duct in 4. Of the 72 patients, 24 had associated retained common bile duct stones (single in 16 patients and multiple in 8 patients) and 1 had ascaris in the common bile duct. All the 72 patients were subjected to therapeutic procedures including sphincterotomy with stone extraction followed by biliary stenting (24 patients), removal of ascaris and biliary stenting (1), sphincterotomy with biliary stenting (18), sphincterotomy with NBD (12), biliary stenting alone (12), and NBD alone (5) (Figs. 2 and 3). Stents and NBD of 7F size were used to bypass the site of leak. Bile leak stopped in 3 days in 42 patients, in 4-7 days in 18 patients and in 7-16 days in 12 patients. Bile drainage through subhepatic drains placed peroperatively and percutaneous catheters placed postoperatively stopped at a mean interval of five days (range 3-16 days) after endoscopic procedures. Subhepatic drains and percutaneous catheters could be removed at a mean interval of 6 days (range 5-20 days) after endoscopic procedures. During follow-up, complete cessation of bile leak was documented cholangiographically in all the patients. NBD was removed at an interval of 6-22 days after placement while stents were removed at a 6-8 weeks interval. There was no morbidity or mortality associated with initial endoscopic procedures or follow up procedures documenting healing and removal of stents.

There was no significant difference in efficacy and in time for the treatment of bile leak by sphincterotomy with biliary endoprosthesis (4.6?.2, range 3-16 days) or endoprosthesis alone (4.2?.8, range 4-11 days) in patients with bile leak after surgery.

Discussion

We have shown that endoscopic procedures are simple, safe and highly effective in the management of biliary fistulae after surgery.  In the present series, bile leak stopped and biliary fistulae healed in all patients after endoscopic therapy.

Operative bile duct injury, bile leak, and biliary fistula may occur in the setting of acute or chronic inflammation, in the presence of anatomic variants, and at the time of injury to the accessory bile duct, the liver during dissection of the gallbladder or small biliary radicles entering the gallbladder fossa, ligation and/or transection of the bile duct, and dislodgement or malposition of surgical clips. [5-14] Bile duct injury has been classified into 3 types: clean incision of the bile duct (type 1), misplacement of a surgical clip (type 2), and transection (type 3).  About 49% of bile duct injuries are recognized intraoperatively and 51% at a later time. [15]

Biliary leak post cholecystectomy has also been classified according to the severity of leak.  Sandha et al [16] classified the leak into low grade (demonstration of leak only after opacification of intrahepatic biliary radicles) and high grade (demonstration of leak even without opacification of the intrahepatic system).

Treatment options available for biliary fistula include surgical repair, percutaneous biliary drainage, and endoscopic biliary drainage.  Earlier, biliary fistula has been treated by surgical repair, but surgery is associated with high morbidity (22%-37%) and mortality (3%-18%). [17-20] Fistula may recur in one-third of patients and strictures may occur in 37% to 50% of patients after reoperation. [21, 22] Percutaneous transhepatic biliary drainage also carries a high morbidity rate owing to hemorrhage and bile leak related to liver puncture. Apart from these complications, there are also technical problems in puncturing a non-dilated biliary system. [23]

Endoscopic techniques reduce the bile duct-duodenal pressure gradient maintained by the intact sphincter of Oddi and divert bile away from the site of leak, resulting in healing of fistula. NBD or stent bridges the defect at the site of leak, physically occluding it while providing a conduit for bile flow.  NBD and stent may also prevent stricture formation during the healing. [5-9] There is no evidence that stent or NBD placement proximal to the site of leak is required or beneficial, and this may not be possible in the majority of patients with peripheral parenchymal bile leak. However, in the absence of randomized trials, we opted to secure the upper end of endoprosthesis above the site of leak. Nasobiliary drains have advantages of providing visual confirmation of biliary decompression and facilitating repeat cholangiography. Nasobiliary drains also allow gravity-assisted drainage to siphon bile from the duct. They have the disadvantages of being more uncomfortable, a risk of displacement and they deprive the patient of a significant proportion of their intestinal bile.

Bile leak can be managed safely and effectively by sphincterotomy with or without biliary endoprosthesis (NBD/stent) or biliary endoprosthesis alone, as shown in the present series and others. In a series of 43 patients with bile leak, endoscopic treatment was successful in all but one patient who had sustained an injury of the main right hepatic duct. [24] Foutch et al [6] treated 23 patients with bile leak by sphincterotomy alone, sphincterotomy with stenting, stent alone, and sphincterotomy with NBD.  All patients improved and stents were removed at an average of 8 weeks.

The role of endoscopic management of biliary fistula appears rational but the choice of the procedure remains uncertain.  All specific endoscopic treatment modalities have been found to achieve equally good results. Endoscopic sphincterotomy alone, stent alone or sphincterotomy with stent are equally effective.

Sandha et al [16] targeted their endoscopic therapy for bile leak accordingly to severity: endoscopic sphincterotomy (EST) only for low grade leak and biliary stenting (with or without EST) for high grade leak. We did not use NBD because of concerns of patient抯 discomfort. In contrast, we placed NBD in 17 patients (12 after EST and 5 without EST). Although, NBD placement is associated with patient抯 discomfort, we feel that it has distinct advantage of allowing cholangiography without doing endoscopy and no need for repeat endoscopy while removal of the prosthesis.

In a recent study by Kaffes et al [25] endoscopic therapy was carried out in 97 of 100 patients wtih biliary leak. The endoscopic procedures included stent insertion alone (40 patients), sphincterotomy alone (18), and combination of stent and sphincterotomy (31). There were significantly more treatment failures in the EST group than in the stent group or combination group. However, treatment in this study was categorized according to the severity of leak, so it is not exactly known that patients who failed EST treatment had major or high grade leak. Further, randomized trails of sufficient sample size would be needed to compare the results of different endoscopic modalities for the treatment of biliary leak. [26]

In patients with postoperative bile leak in the present study, we compared sphincterotomy with prosthesis versus prosthesis alone and found that there was no significant difference in stoppage of bile leak. Both of the modalities stopped bile leak at the same mean interval and were equally effective in stopping bile leak. It has been proposed that stent placement may only be suitable for bile leak associated with bile duct stricture. In case of distal obstruction owing to stone, sphincterotomy and stone extraction are treatment of choice. In present study distal obstruction caused by stones and ascaris was seen in 25 of the 72 patients (35%) in comparison with those (40%-70%) reported in other studies. Stone extraction was needed in 24 of the 72 patients. 

In conclusion, endoscopic therapy is a safe and effective mode of management for bile leak and fistula after surgery. Sphincterotomy with endoprosthesis or endoprosthesis alone is equally effective in management of postoperative bile leak.

Funding: None.

Ethical approval: Not needed.

Contributors: AN made study plans, collected and analyzed data, reviewed literature and wrote the paper. SBC contributed to the study design, performance of endoscopic procedures. GS and KR collected data. SSK contributed to the study design and performance of endoscopic procedures.

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

References

1 Schlumpf R, Klotz HP, Wehrli H, Herzog U. A nation抯 experience in laparoscopic cholecystectomy. Prospective multicentric analysis of 3722 cases. Surgical Endosc 1994;8:35-41.

2 Collet D, Edye M, Perissat J. Conversion and complications of laparoscopic cholecystectomy.  Results of a survey conducted by French Society of endoscopic surgery and interventional radiology. Surg Endosc 1993;7:334-338.

3 Cuschieri A, Dubois F, Mouiel J, Mouret P, Becker H, Buess G, et al. The European experience with laparoscopic cholecystectomy. Am J Surg 1991;161:385-387.

4 Meyers WC, Jones RS. Preoperative and postoperative biliary problems.  Textbook of liver and biliary surgery.  Philadelphia: J. B. Lippincott;1990:373-390.

5 Brooks DC, Becker MJ, Connors PJ, Carr-lock DL.  Management of bile leaks following laparoscopic cholecystectomy. Surg Endosc 1993;7:292-295.

6 Foutch GP, Harlen JR, Hoefer M. Endoscopic therapy for patients with a postoperative biliary leak. Gastrointest Endosc 1993;39:416-421.

7 Davids PH, Rauws EA, Tytgat GN, Huibregste K. Postoperative bile leakage: endoscopic management. Gut 1992;33:1118-1122.

8 Binmoeller KF, Katon RM, Shneidman R. Endoscopic management for patients with a post operative biliary leak: Review of 77 cases and report of 2 cases with biloma formation. Am J Gastroenterol 1991;86:227-231.

9 Ponchon T, Gallez JF, Vallete PJ, Chavaillon A, Bory R. Endoscopic treatment of biliary tract fistulas. Gastrointest Endosc 1989;35:490-498.

10 Peter JH, Gibbon GD, Innes JT, Nichols KE, Front ME, Roby SR, et al. Complications of laparoscopic cholecystectomy. Surgery 1991;110:769-778.

11 Kozarek R, Gannan R, Baerg R, Wagonfeld J, Ball T. Bile leak after laparoscopic cholecystectomy: Diagnostic and therapeutic application of ERCP. Arch Intern Med 1992;152:1040-1043.

12 Soper NJ, Flye MW, Brunt LM. Dark lining of silver cloud: Biliary complications of laparoscopic cholecystectomy. Gastroenterology 1992;38:A572.

13 Ponsky JL. Complications of laparoscopic cholecystectomy. Am J Surg 1991;161:393-395.

14 McQuillan T, Monalas SC, Haymen JA. Surgical significance of the bile duct of Luschka. Br J Surg 1989;76:696-698.

15 Deziel DJ, Millikan KW, Brunt LM. Complications of laparoscopic cholecystectomy: Results of a national survey of 4292 hospitals and analysis of 77604 cases. Gastroenterology 1992;38:A569.

16 Sandha GS, Bourke MJ, Haber GB, Kortan PP. Endoscopic therapy for bile leak based on a new classification: result in 207 patients. Gastrointest Endosc 2004;4:567-574.

17 Browder IW, Dowling JB, Koontz KK, Litwin MS. Early management of postoperative injury of the extra hepatic biliary tract. Ann Surg 1987;205:649-658 .

18 Hills TM, Westbrook KC, Caldwell FT, Read R. Surgical injury of common bile duct. Am J Surg 1977;134:712-716.

19 Castirini G, Pappalardo G. Iatrogenic stricture of the bile duct: our experience with 66 cases.  World J Surg 1981;5:763-768.

20 Kune G. Bile duct injury during cholecystectomy; causes, prevention and surgical repair in 1979. Aust NZ J Surg 1979;49:35-40.

21 Czerniak A, Thompson JN, Soreido O, Benzamin IS, Blumgart LH. The management of fistulas of biliary tract after injury to the bile duct during cholecystectomy. Surg Gynecol Obstet 1988;167:33-38.

22 Sandberg AA, Johanson B, Bengmark S. Accidental lesions of common bile duct at cholecystectomy: Results of treatment. Ann Surg 1985;209:452-455.

23 Von Sonnenberg E, Giovana C, Wittich GR, Christensen R, Varney RR, Neff CC, et al. The role of interventional radiology for one complications of cholecystectomy. Surgery 1990;107:632-638.

24 Schmitt C, Affronti JP, Branch MS. Endoscopic stenting is the preferred first line management for bile duct leaks. Gastrointest Endosc 1992;38:139A.

25 Kaffes AJ, Hourigan L, Luca ND, Byth K, Williams SJ, Bourke MJ. Impact of endoscopic intervention in 100 patients with suspected postcholecystectomy bile leak. Gastrointest Endosc 2005;61:269-275.

26 Huang CS, Lichtenstein DR. Post cholecystectomy bile leak: what is the optimal treatment? Gastrointest Endosc 2005;61:276-278.

Received November 21, 2005

Accepted after revision March 7, 2006