Introduction

The living donor liver transplantation (LDLT) was introduced to solve the problem of donor shortage especially in Western countries [1]. It is the only source of liver grafts where deceased donors is not allowed based on religious, legal, or social basis, such as in many Asian countries [2, 3]. The first reported LDLT was by Raia in Brazil in 1988 [4], and in 1989, the first LDLT was performed in Japan [5]. The first LDLT series from Japan was reported on right lobe LDLT in 1994 [6, 7]. Worldwide deceased donor liver shortages made LDLT to be adopted in Europe and the USA [8, 9]. Although with the advances in the surgical techniques and the proper understanding of the blood supply of the bile ducts, biliary complications remain the most common problem after LDLT [10,11,12]. Variations of the bile ducts and the number of bile ducts with different blood supply increases the challenges for biliary complications for any LDLT program [13]. Biliary complications affect the outcome of LDLT regarding both patient and graft survival and also the recipient morbidity rate [14, 15]. The incidence of biliary complications ranges from 5.3% to 40.6%. Leaks occur in 0% to 21.9% while strictures occur in 3.7% to 25.3% [16,17,18,19]. Anastomotic strictures represent the majority of strictures after LDLT. The incidence in most of the reports was >90% [20,21,22]. The incidence of both biliary leaks and biliary stricture are still higher in recipients with duct-to-duct anastomosis [23,24,25].

Methods

Study patients

During the period from 1999 to 2004, 505 patients underwent 518 LDLT in the Department of Liver Transplantation and Immunology, Kyoto University Hospital, Japan. The data was collected and analyzed retrospectively.

Preoperative workup of donors

The selection of donors for LDLT is one of the most important aspects of this procedure. The aim is to ensure that the donor operation may be performed safely and the donor is able to yield a suitable graft for the recipient. A meticulous and extensive workup for the donors was done. This was reported previously [26]. High resolution duplex to detect the vascular anatomy, the spiral computed tomography (CT) with intravenous contrast to determine the graft volume and hepatic venous and portal vein anatomy. The anatomy of the biliary tree was provided in all donors by operative cholangiogram which provided sufficient information for safe surgery in most cases. Magnetic resonance (MR) cholangiography was indicated preoperatively to diagnose bile duct anomalies in patients proved to have portal venous variants by the 3D CT portography.

Surgical techniques and the perioperative management of the recipient have been reported elsewhere [27].

Biliary reconstruction

The bile duct of the graft is either anastomosed to the jejunum (hepaticojejunostomy) or duct-to-duct anastomosis.

A. Hepaticojejunostomy [28]

This type of anastomosis was done in all pediatric patients and in adults with primary sclerosing cholangitis associated with extrahepatic lesions, adult biliary atresia, and in some cases, needing multiple duct anastomoses.

Technique

The jejunum is divided about 20 cm from the Treitz’s ligament and the distal limb is elevated to the hilum of the graft using the retrocolic route. The proximal limb is anastomosed end-to-side to the elevated distal limb at about 40 cm. The previously created Roux-en-Y limb can be used for the reconstruction in cases of post-Kasai operation, if it has not been greatly damaged during hepatectomy.

A small hole is made in the Roux-en-Y limb close to the proximal end. A 4-French polyvinyl alcohol tube is inserted through the hole into the intestinal lumen and then through the intestinal wall again. After the anastomosis of the posterior wall, the stent tube is inserted into the graft bile duct. The stent tube is secured to the anterior wall of the bile duct. When the orifice of the graft bile duct is small, an incision is made along the axis on the anterior wall of the bile duct to make it wider. After closure of the anterior wall, the stent tube is secured on the intestinal wall and is fixed to the anterior abdominal wall as an external stent. This tube can be removed around 8 weeks after surgery.

B. Duct-to-duct anastomosis [28,29,30]

This type of anastomosis was done to the majority of right lobe cases, with the exception of such situations as primary sclerosing cholangitis with extrahepatic lesions or adult biliary atresia.

Technique

The recipient bile duct was dissected out to obtain enough length for the duct-to-duct anastomosis. Care was taken to preserve the blood supply to the duct by avoiding removal of the connective tissue in the 3 o’clock and 9 o’clock locations [31]. The good mucosal color and arterial bleeding from the cutting stump guarantee favorable blood supply in the duct. The technique of the anastomosis was performed as previously described in hepaticojejunostomy. An external stent was used in small-size ducts.

In cases of grafts with multiple hepatic duct orifices, biliary anastomosis was achieved using the following techniques:

  1. 1-

    When the two orifices are close together, a ductoplasty with single anastomosis to a single orifice of the proper hepatic duct or Roux-en-Y limb.

  2. 2-

    If the two orifices are further apart in distance, they were reconstructed individually. In right lobe grafts, they were anastomosed to the recipient right and left hepatic ducts, or the proper hepatic duct and the cystic duct.

  3. 3-

    If the recipient biliary duct has no part suitable for two anastomoses, the duct-to-duct procedure was used for one orifice and hepaticojejunostomy was used for the other orifice or both orifices.

  4. 4-

    In left-side grafts, two orifices were anastomosed to the Roux-en-Y limb individually [28, 29].

Enterostomy tube was put as a routine in all pediatric cases and most adult cases. It was put in a jejunal loop about 30 cm from Trietz’s ligament. Food and nutrients were given (in addition to oral intake) for the patient postoperatively (about 1000 cc hypertonic solutions), for the general support. In recipients who had no complications, it was removed after 1 month postoperatively.

The graft vasculature and biliary tree were assessed by Doppler ultrasonography daily for 4 days and then weekly intervals until discharge, or whenever abnormalities in liver function occurred.

Assessment of the relation of the graft size for recipients by the ratio of graft weight/recipient body weight (graft body weight ratio) [27]. Arterial reconstruction was described elsewhere [27].

Diagnosis of biliary complications

Biliary complications were suspected clinically by presence of bile in the drain, elevated temperature and/or rigors indicating cholangitis, peritonitis or sepsis, and biochemically by elevated serum bilirubin or alkaline phosphatase. Diagnosis was confirmed radiologically by ultrasonography followed by cholangiography, CT scan, or biliary scintigraphy as indicated.

Management of biliary complications

Minor leaks stopped spontaneously without further management. Substantial persistent leak required drainage by either endoscopic nasobiliary drainage (ENBD) or percutaneous transhepatic biliary drainage (PTBD). Large or non-loculated leaks required operative intervention (in cases of biliary leak from Roux-en-Y biliary anastomosis, enterostomy was done to isolate the anastomosis from the intestinal juice, then the enterostomy was taken down after the stoppage of the leak and in cases of biliary leak from duct-to-duct anastomosis, biliary reconstruction may be redone by Roux-en-Y choledochojejunostomy in one or two settings according to the condition of the recipient). Bilomas were managed by percutaneous ultrasound-guided drainage. Biliary strictures were initially managed by either percutaneous transhepatic cholangiography (PTC) or endoscopic retrograde cholangiography (ERC), followed by balloon dilatation and stent insertion. If balloon dilatation was impossible or recurrence of the stricture was several times within 2 to 3 weeks in spite of treatment, operative intervention was resorted to, to anastomose the dilated bile duct to Roux-en-Y limb or resection of the stricture and duct-to-duct reanastomosis.

Statistical analysis

Statistical calculations for mean values and standard deviations were performed using the SPSS 17.0 software package (SPSS Inc., Chicago, IL, USA). Results were expressed as the mean and standard deviation (SD) after verification of normal distribution or median (interquartile range) for quantitative variables. ANOVA procedure and Pearson correlation coefficient were used to compare between different values. A value of p<0.05 was considered statistically significant.

Results

During the period from May 1999, to May 2004, 505 patients underwent 518 living donor liver transplantation (LDLT) in the Department of Liver Transplantation and Immunology, Kyoto University Hospital, Japan.

Recipients’ characteristics (Table 1)

The recipient gender for the 518 cases of LDLT was 261 males (50.4%) and 257 females (49.6%). Pediatric cases (<18 years old) were 230 (44.4%) and adult cases (≥18 years old) were 288 (55.6%). The median age of the recipients was 25 years, range (0.7–69). The median body weight was 48.3 kg, range (2.78–108).

Table 1 Demographic characteristics in 518 cases of living donor liver transplantation
Full size table

The preoperative performance status of the recipients was at home in 241 patients (46.5%), hospitalized in 220 patients (42.5%), and in the intensive care unit (ICU) in 57 patients (11%). Child-Pugh classification was Child A in 57 cases (11%), Child B in 222 patients (42.9%), and Child C in 239 patients (46.1%). United Network for Organ Share (UNOS) status 1 was present in 50 patients (9.7%), UNOS status 2A in 131 patients (25.3%), UNOS status 2B in 259 patients (50%), and UNOS status 3 in 78 patients (15.1%).

ABO blood type compatibility was identical in 348 patients (67.2%), compatible in 104 patients (20.1%), and incompatible in 66 patients (12.7%).

The indications for LDLT were cholestatic liver diseases in 220/518 patients (42.5%), chronic hepatocellular liver diseases (CHD) in 69/518 patients (13.3%), hepatocellular carcinoma (HCC) with or without liver cirrhosis in 96/518 patients (18.5%), other tumors in 14/518 patients (2.7%), fulminant hepatic failure (FHF) in 59/518 patients (11.4%), metabolic liver diseases and genetic liver diseases in 21/518 patients (4.1%), vascular causes in 2/518 patients (0.4%), and graft failure and retransplantation in 37/518 patients (7.1%).

Biliary atresia constituted the most common indication for LDLT. It was the indication in 149/518 patients (28.8%). Almost all patients underwent at least a single previous unsuccessful portoenterostomy (Kasai’s operation). Other cholestatic liver diseases were the indication in 71 patients; they included primary biliary cirrhosis (PBC) in 31/518 patients (6.0%), primary sclerosing cholangitis (PSC) in 19/518 patients (3.5%), Alagille’s syndrome in 10/518 patients (1.9%), Caroli’s syndrome in 4/518 patients (0.8%), Byler’s disease in 6/518 patients (1.2%), and TPN-induced cirrhosis in one patient (0.2%).

Donor characteristics (Table 2)

The number of donors was 519 (one recipient received a dual graft from his twin daughters). The donor gender was 258 males (49.7%) and 261 females (50.3%). The median age of the donors was 37 years, range (18–66 years). The median body weight was 61 kg, range (39–107.2 kg). The median body mass index (BMI) was 22.51 kg/m2, range (16–37.65 kg/m2). The median liver/spleen (L/S) ratio was 1.18, range (0.76–1.78). Donors were relatives of the recipients up to the third degree civil relationship or legal spouse. Mother was the donor in 154 cases (29.7%), father in 120 cases (23.1%), son in 64 cases (12.3%), daughter in 23 cases (4.4%), brother in 31 cases (6.0%), sister in 29 cases (5.6%), grandmother in 3 cases (0.6%), uncle in 4 cases (0.8%), aunt in 2 cases (0.4%), nephew in 2 cases (0.4%), cousin in 2 cases (0.4%), husband in 35 cases (6.7%), and wife in 50 cases (9.6%).

Table 2 Demographic characteristics in 519 donors
Full size table

The BMI was >25 in 103 donors (19.9%), while the liver/spleen (L/S) ratio was ≤1.1 in 97 donors (18.7%). These donors were considered to have mild to moderate steatosis.

Graft characteristics (Table 3)

The type of the graft transplanted was reduced monosegment in 25 cases (4.8%), left lateral segment in 150 cases (30.0%), left lateral segment and part of segment 4 (S4) in 5 patients (1.0%), left lobe without the caudate lobe in 28 patients (5.4%), whole left lobe in 2 patients (0.4%), the right lobe in 259 patients (50.0%), extended right lobe in 39 patients (5.8%), posterior segment of the right lobe in 4 patients (0.8%), auxiliary partial orthotopic liver transplantation (APOLT) in 5 patients (1.0%), and dual graft (from twin daughters to their father consisting of the left lateral segment from one daughter and the right lobe from the other) in one patients (0.2%).

Table 3 The type of the transplanted grafts
Full size table

Graft recipient body weight ratio (Table 4)

According to the GRWR, grafts were divided into four groups. Small-for-size (<0.8%) in 25 patients (4.8%), accepted size (0.8% to <1.0%) in 83 patients (16.0%), optimum size (1.0% to 4.0%) in 386 patients (74.5%), and large-for-size (>4.0%) in 24 patients (4.6%).

Table 4 Groups of graft recipient body weight ratio (GRWR)
Full size table

Operative characteristics (Table 5)

The operation was done electively in 461 patients (89.0%) and as emergency in 57 patients (11.0%).

Table 5 Operative characteristics
Full size table

The median operative time was 10.51 h, range (5.29–23.25). The median blood loss was 2060 cc, range (81–60,000).

Portal vein graft was needed in 42 patients (8.1%): donor left ovarian vein graft in 23 patients (54.8%), donor inferior mesenteric vein (IMV) graft in 13 patients (31.0%), donor internal jugular vein graft in 2 patients (4.8%), recipient external iliac vein (EIV) graft in 3 patients (7.1%), and recipient EIV graft and donor IMV graft in one patient (2.4%).

The number of bile ducts in the graft and the number of anastomoses performed were as follows: one duct/one anastomosis in 415 patients (80.1%), two ducts/one anastomosis in 64 patients (12.4%), two ducts/two anastomoses in 36 patients (6.9%), three ducts/two anastomoses in 2 patients (0.4%), and three ducts/one anastomosis in one patient (0.2%). The median hospital stay of survivors was 52 days, range (18–371). The median period of follow up was 30 months, range (12–72).

Biliary complications (Table 6)

  • During hospital stay: occurred in 99 patients (19.1%); biliary leakage in 74 patients (14.3%), biloma in 7 patients (1.4%), biliary leak followed by biliary stricture in 6 patients (1.1%), and biliary stricture in 12 patients (2.3%).

  • Within 3 months from discharge: occurred in 36 patients (6.9%); biliary leakage occurred in 3 patients (0.6%), biloma in 2 patients (0.4%), biliary leak followed by stricture in 2 patients (0.4%), and biliary stricture in 29 patients (5.6%).

  • After 3 months from discharge: occurred in 67 patients (13.0%); biliary leakage in 2 patients (0.4%), biloma in 4 patients (0.8%), biliary leakage followed by biliary stricture in one patient (0.2%), and biliary stricture in 60 patients (11.6%).

Table 6 Biliary complications
Full size table

Management of biliary complications:

  • Biliary leakage: the total number of patients who developed biliary leakage was 79/518 patients (15.4%). They were managed as follows:

  1. 1.

    Simple follow up was adopted in 50/79 patients (63.2%), where there was gradual decrease in the amount of leak until eventually disappeared.

  2. 2.

    ENBD and tube insertion was resorted to in 16/79 patients (20.3%), but failed in one patient which needed Roux-en-Y operation.

  3. 3.

    Roux-en-Y hepaticojejunostomy was performed in 4/79 patients (5.1%).

  4. 4.

    External diversion of already made choledochjejunostomy (external stoma) in 9/79 patients (11.4%).

Failure of the treatment of biliary leakage occurred in 7/79 patients (8.9%) who developed graft failure. Six patients died from septic cholangitis, septicemia, and septic shock and one patient underwent retransplantation. The success rate of management of biliary leakage was 72/79 patients (91.1%).

  • Biliary stricture: the total number of patients who developed biliary stricture was 110/518 patients (21.1%), 9 of them were preceded by biliary leak. They were managed as follows:

  1. 1.

    Simple follow up was adopted in 2/110 patients (1.8%) which showed no progressive intrahepatic biliary dilatation as detected by follow up US.

  2. 2.

    Percutaneous transhepatic cholangiography with balloon dilatation and drainage (PTCD) and insertion of a tube (external or internal) was adopted in 62/110 patients (56.4%). This management failed in 3 patients and were managed by Roux-en-Y hepaticojejunostomy.

  3. 3.

    Endoscopic retrograde cholangiography with bile duct dilatation (ERBD) and stent insertion was resorted to in 45/110 patients (40.9%). This management failed in one patient and was managed by Roux-en-Y hepaticojejunostomy.

  4. 4.

    Roux-en-Y hepaticojejunostomy was resorted to in one patient (0.9%).

  • Biloma: the total number of patients who developed biloma were 13/518 patients (2.5%). They were managed by US-guided drainage (percutaneous drainage), which was successful in all patients.

In the current study, many factors were studied as potential risk factors for the occurrence of biliary complications including patient age, gender, preoperative functional status, number of bile ducts, presence or absence of stent insertion, ABO blood group incompatibility, the way of anastomosis, graft size, graft type, operative time, and hepatic artery thrombosis. The only factors which found to be significant for biliary complications were hepatic artery thrombosis (p value 0.001) and the way of bile duct anastomosis (p value 0.03). Early leakage was associated with late stricture but insignificantly (p value 0.3).

As regards the number of bile ducts in the graft, there was one bile duct in 80.1% of patients, two bile ducts in 19.3% of patients, and three bile ducts in 0.6% of patients. The number of anastomoses performed was one duct/one anastomosis in 80.1% of patients, two ducts/one anastomosis in 12.4% of patients, and two ducts/two anastomoses in 6.9% of patients. Biliary complications occurred in patients with one bile duct anastomosis in 28.4% and in patients with two bile ducts in 34.8%. Although this result is insignificant (p value 0.57), yet, it shows less biliary complications in cases of single duct anastomosis. As regards the way of anastomosis of the bile ducts, Roux-en-Y hepaticojejunostomy was performed in 45% of patients; duct-to-duct anastomosis was performed in 48% of patients, one duct by Roux-en-Y hepaticojejunostomy, and one duct by duct-to-duct anastomosis in 7% of patients. Biliary complications in patients with Roux-en-Y anastomosis occurred in 17.6% and in patients with duct-to-duct anastomosis occurred in 39.6%. This result shows that biliary complications were significantly more in patients with duct-to-duct anastomosis with p value 0.03. As regards the use of external stents during bile duct anastomosis, they were used in 26.4%. In patients where no stent was put, biliary complications occurred in 29.9% while in patients where stent was put, biliary complications occurred in 30.4%.

Discussion

At present, liver transplantation proved to be a real breakthrough in surgery as the only effective intervention to deal with otherwise fatal liver diseases. In Western countries, most of the organs used for transplantation are obtained from brain stem-dead and heart-beating cadaveric donors. However, the number of organs required to satisfy the needs of transplantation far exceeds the number of cadaveric organs available. This has prompted a relaxation in deceased donor selection criteria and the use of organs from so called “marginal donors.” This expansion could not solve the donor shortage and may increase recipient morbidity and mortality. The need to resort to living-donor liver donation arose as a natural response to a growing demand for liver transplantation and a constant undersupply of grafts from brain-dead donors. This need is even greater in countries where deceased donors are not allowed based on religious, legal, or social basis, such as in many Asian countries where living-donor liver donation is the only source of liver grafts. Therefore, it is not surprising that the largest series in LDLT all over the world come from Asian centers especially the center in University of Kyoto where most of the innovations and techniques of LDLT are perfected [32]. As LDLT is relatively recent, information about the long-term risks of the procedure are lacking. The present work addresses both the early and late risk factors for biliary complications and their management as experienced in the biggest center in LDLT in the world.

Biliary complications occurring after LDLT had decreased by the advancement in the surgical techniques but they are still a major problem as they are associated with high rate of morbidity and mortality [33]. Many biliary complications can occur but the most serious are the biliary leaks which occur mainly in the immediate postoperative and biliary stricture which mainly occur late during the follow up period [34, 35]. Early diagnosis and proper management is the key to prevent graft failure [36].

The presence of good blood supply especially in duct-to-duct anastomosis can decrease the incidence of biliary complications and improve the posttransplant results [37, 38]. Ischemic biliary complication can lead to graft failure without proper and timely treatment. [39] Proper understanding of the blood supply of the bile ducts can guarantee lower biliary complications [31, 40, 41]. Dissection along the donor bile ducts should not be too close to the wall in order to keep the fine arterial plexus that supply the graft in the recipient side proximal to the anastomosis. So shorter donor ducts can improve the blood supply around the anastomosis and this is the procedure adopted in our center [29]. Also, the dissection around the bile duct in the recipient should preserve the ascending arterial plexus as much as possible [31]. Vellar described that the arterial plexus in the recipient came from the gastrodudenal artery, then the superior pancreaticoduodenal artery, and then the ascending marginal arteries [42].

Biliary leakage mainly occurs at early posttransplantation and mainly at the anastomotic site. It occurs mostly due to ischemia [43]. Double or triple hepaticojejunostomies are risk factors for higher incidence of leakage in patients with Roux-en-Y anastomosis [25, 44, 45]. In deceased donor liver transplantation, the incidence of biliary leakage with T-tube insertion was 33%, while in the patients without T-tube insertion it was 15.5% [46]. In right lobe LDLT with duct-to-duct anastomosis was 4.7% to 7.3% weather with or without draining tube [23,24,25].

Percutaneous drainage is the first step management of biliary leakage, while surgery was resorted to in patients with large biliary leakage or complete anastomotic disruption [43]. In patients with duct to duct anastomosis, ERC can be performed as a diagnostic tool. It can detect the site of leakage and ENBD or endoscopic biliary drainage (EBD) can be inserted proximally to reduce the leakage till healing [25, 47, 48]. EBD was successful in the management of biliary leakage. One study showed 100% success rate [48]. Another study showed 50% was treated successfully with ENBD and 40% required surgical intervention and 10% died from sepsis [25]. Surgery was resorted to in failure of ENBD, EBD or ERCP, and sphincterectomy [43].

Biliary stricture usually occurs late [23,24,25]. Biliary stricture mainly occurs in the anastomotic area. The predisposing factors are ischemia and secondly, to biliary leakage [25]. Other causes include hepatic artery complications and cytomegalovirus infection while blood type incompatibility is not a significant factor in biliary stricture [25]. Biliary stricture in case of LDLT may develop in multibranched fashion because of multiple anastomoses compared with deceased donor liver transplantation which is single [24, 25, 49, 50]. This may lead to some problems in repairing post-LDLT strictures especially in cases of multiple anastomoses.

PTC with balloon dilatation and drainage (PTCD) with or without stenting has proved its effectiveness at the site of anastomosis especially in pediatric patients in whom Roux-en-Y hepaticojejunostomy was performed [51,52,53,54,55]. In the present study, PTCD was adopted in 62/110 patients (56.4%). This procedure was repeated with the recurrence of biliary stricture. This management failed in 3 patients where recurrence occurred after three series of interventions. These patients were managed by Roux-en-Y hepaticojejunostomy. This result may be comparable or even better than some other reports [51,52,53,54,55]. Greif et al. reported more strictures in cases of Roux-en-Y hepaticojejunostomy [56]. Other studies showed lower incidence being less than 15% [33, 57]. In posttransplantation pediatric patients with Roux-en-Y hepaticojejunostomy percutaneous transhepatic biliary interventions were preferred for easier access [58, 59]. In patients with duct to duct anastomosis, endoscopic biliary stenting appears to be of benefit in treating multibranched strictures or single strictures by inserting single or multiple stents. The success rate reached 74.5% to 75% which can be compared with deceased donor liver transplantation 68% to 90% [25, 48, 60,61,62].

In the current study, most of biliary leaks (74/79) occurred early and 50/79 were managed conservatively and there was gradual decrease in the amount of leak until eventually disappeared. Biliary strictures mostly occurred after discharge and during the follow up period in 89/101. Mostly biliary strictures were treated radiologically in 97%. These results were similar to other report in cadaveric liver transplantation [63]. Greif et al. reported in his study that strictures occurred early were due to technical issues while late strictures were due to fibrosis resulted from the healing process [56].

The procedure of biliary anastomosis we adopt during this study with insertion of biliary stent was found to be the least to cause biliary complications. This technique was found to be easy, and save both time and money. Scatton et al. concluded that biliary complications decreased by T-tube insertion, but the incidence of leakage is high after removal of the tube. They also detected high incidence of biliary stricture in the non T-tube insertion recipients [46]. These results are in accordance with other group [64]. The removal of the stent in the current study was associated with leakage in only 3% of patients. The stent serves as a mean of decompression of the biliary tree and to decrease biliary complications as well as performing postoperative cholangiography, if needed, although it is safe but may be stressful for the patients.

In conclusion, biliary complications although decreased, yet, still occur. They can be treated effectively with interventional radiological procedures successfully before subjecting the patient to invasive surgery. Preventive measures were suggested to decrease the rate of these complications. As it is known that ischemic changes around the anastomosis is a major cause of anastomotic stenosis, greater precaution should be taken to preserve the peribiliary plexus around the resected bile duct in the donor. Although duct-to-duct reconstruction still has certain rate of complications, it is still applied to the majority of right lobe cases because it is more physiological (no reflux, and maintained bowel continuity allowing early oral intake) than hepaticojejunostomy. It also has the added advantage of allowing the use of ERCP postoperatively, if required. Although the use of external stents during bile duct anastomosis does not decrease the incidence of biliary complications, yet, its presence is important for performing postoperative cholangiography and timely diagnosis of biliary complications especially in cases of Roux-en-Y anastomosis and in critically ill patients.