Indian J Surg DOI 10.1007/s12262-012-0722-2
ORIGINAL ARTICLE
Management of Bile Duct Injury at Various Stages of Presentation: Experience from a Tertiary Care Centre Md. Ibrarullah & S. Sankar & K. Sreenivasan & S. R. K. Gavini
Received: 7 April 2012 / Accepted: 3 September 2012 # Association of Surgeons of India 2012
Abstract The clinical presentation, management and outcome of all patients with bile duct injury who presented to our tertiary care centre at various stages after cholecystectomy were analyzed. The patients were categorized into three groups: group A–patients in whom the injury was detected during cholecystectomy, group B–patients who presented within 2 weeks of cholecystectomy and group C–patients who presented after 2 weeks of cholecystectomy. Our team acted as rescue surgeons and performed ‘on-table’ repair for injuries occurring in another unit or in another hospital. Strasberg classification of bile duct injury was followed. In group A, partial and complete transections were managed by repair over T-tube and high hepaticojejunostomy, respectively. Patients in group B underwent endoscopic retrograde cholangiogram and/ or magnetic resonance cholangiogram to evaluate the biliary tree. Those with intact common bile duct underwent endoscopic papillotomy and stenting in addition to drainage of intraabdominal collection when present. For those with complete transection, early repair was considered if there was no sepsis. In presence of intra-abdominal sepsis an attempt was made to create controlled external biliary fistula. This was followed by hepatico jejunostomy at least after 3 months. Group C patients underwent hepaticojejunostomy at least 6 weeks after the injury. The outcome was graded into three categories: grade A–no clinical symptoms, normal LFT; grade B–no clinical symptoms, mild derangement of LFT or occasional episodes of pain or fever; grade C–pain, cholangitis and abnormal LFT; grade D–surgical revision or dilatation required. Fifty nine patients were included in the study and the distribution was group A–six patients, group B–33 patients and group C–20 patients. In group A, one patient with complete transection of M. Ibrarullah (*) : S. Sankar : K. Sreenivasan : S. R. K. Gavini Department of Surgical Gastroenterology, Sri Ramachandra Medical College and Research Institute, Porur Chennai 600116, India e-mail: [email protected]
the right hepatic duct (type C) and partial injury to left hepatic duct (LHD) underwent right hepaticojejunostomy and repair of the LHD over stent. Two patients with type D and three patients with type E 2 injury underwent repair over T-tube and hepaticojejunostomy, respectively. In group B, all except one of the 18 patients with type A injury underwent endoscopic papillotomy and stenting. The bile leak subsided at a mean interval of 8 days in all, except one patient who died of fulminant sepsis. Of the 15 patients with type E injury, five underwent hepaticojejunostomy after a minimum gap of 3 months. Early repair was considered in 10 patients. Twenty patients in group C underwent hepaticojejunostomy. In a mean follow-up of 40 months, the outcome was grade A in 54 patients, grade B in three patients (one from each of the three groups) and grade D in one patient (group C). The latter patient with a type E3 injury developed recurrent stricture and cholangitis necessitating percutaneous transhepatic dilatation. The high success rate of bile duct repair in the present study can be attributed to the appropriate timing, meticulous technique and the tertiary care experience. Keywords Cholecystectomy . Benign biliary stricture . Bile duct injury Cholecystectomy is the commonest abdominal surgery performed across the world. The reported incidence of bile duct injury (BDI) following open cholecystectomy is 0.1 %– 0.2 %. In the present era of laparoscopic cholecystectomy, the incidence has gone up to 0.4 %–0.7 % [1]. Though some of the injuries can be identified during the primary surgery, majority however, present in the postoperative period [2–4]. Proper identification of the injury, appropriate timing, technique of repair and experience of the surgeon are paramount in the management of these patients. Some of the recent studies have laid much emphasis on repair by specialist hepatobiliary surgeons and considered this to be one of the
Indian J Surg
most important factors in the successful outcome of BDI [5–7]. The present study is a retrospective analysis of all patients with BDI who presented to us during or at a variable period after the index cholecystectomy. Their clinical presentation, management and outcome were studied.
Patients and Methods Ours is a tertiary care referral centre with special interest in hepatobiliary surgery. Hospital records of all patients with postcholecystectomy BDI, managed between April 2004 and April 2012, were entered in a prospectively maintained database. The study included patients who sustained injuries while being operated in our hospital and also those who were operated elsewhere and referred to our centre after the injury was detected or suspected. Depending upon the time of presentation, the patients were categorized into following groups: group A–patients whose injury was detected ‘ontable’ during cholecystectomy, group B–patients who presented in the early postoperative period, i.e. within 2 weeks of cholecystectomy and group C–patients who presented 2 weeks after cholecystectomy. Our team acted as rescue surgeons for injuries detected in another unit in the same hospital or another hospital, when promptly informed. The latter group of patients was followed up in active consultation with the primary surgeon. We followed Strasberg classification of BDI [8]. For group A patients, laparoscopy was converted to open laparotomy by a wide subcostal or midline incision. All the ligatures or clips barring those applied on vessels were removed. Intraoperative cholangiogram was performed to delineate the entire biliary tree. After removal of the ligature/clip if the common bile duct (CBD) was found to be intact, a choledochotomy was performed and an appropriate size T-tube was placed. Partial transection was repaired over appropriate size stent/s or T-tube using 40 polygalactine (Vicryl) suture. In case of complete transection, an attempt was made to delineate the proximal cut end of the bile duct. The stoma was extended proximally so as to perform a high hepaticojejunostomy Roux-en-Y (HJ). When required, the hilar plate was lowered and a stoma of approximately 2.5 cm was created by extending the ductal incision to left hepatic duct (LHD). HJ was performed using single layer, interrupted 3-0 or 4-0 Vicryl sutures. Transjejunal stent was used for very thin-walled and undilated bile duct. A wide bore tube drain was placed in the subhepatic space. The stent or T-tube was kept for 3 months and removed after obtaining satisfactory cholangiogram. For group B patients, injury was suspected on clinical grounds, i.e. tachycardia, fever, jaundice, abdominal distension or bilious drainage. Ultrasound (US) was performed to locate any collection, check the integrity of CBD and for any evidence of distal obstruction. Patients with external
biliary fistula (EBF) without any signs of intra-abdominal sepsis underwent endoscopic retrograde cholangiogram (ERC) or magnetic resonance cholangiogram (MRC). MRC was the investigation of choice if the patient could afford or else ERC was performed since it offered one time therapeutic advantage as well. Patients showing bile leakage but intact CBD were subjected to endoscopic papillotomy (EPT) and insertion of 10f plastic stent. For patients whose ERC showed complete ‘cut-off’, the proximal biliary tree was evaluated on MRC. The MRC was analyzed for the proximal stump, the sectoral branches and any evidence of intraabdominal collection. Distance between the proximal and distal CBD stump was also carefully studied to ascertain if there was any segmental loss or it was only a case of ligated/ clipped CBD (Fig. 1). Early laparotomy and repair was considered if the patient was stable, did not show any signs of intra-abdominal collection or sepsis. At laparotomy, the approach to a ligated or transected CBD was similar to what has been described for group A patients. In patients with suspected bile collection and/or peritonitis, an attempt was made to drain all the collection and create a controlled EBF. These patients underwent US or contrast enhanced CT scan examination of the abdomen. A localized collection was drained by US guided percutaneous catheter. In case of diffuse or excessive collection, patients were subjected to relaparoscopy or laparotomy. Preceding this, if the patient was hemodynamically stable and could afford, MRC was performed to delineate the intra- and extrahepatic biliary tree. If the latter investigation was not available, ERC was performed in the same sitting prior to laparoscopy/ laparotomy. EPT and bile duct stenting was carried out for patients showing bile leak but intact CBD. All the bile collections were thoroughly drained and tube drains inserted in the subhepatic space and pelvis. The biliary stent was removed 4 weeks after the stoppage of bile leak. For patients with complete CBD ‘cut-off’ on MRC/ERC and gross peritoneal contamination, exploratory laparotomy
Fig. 1 MRC showing segmental loss of the common duct (marked by the arrows) just below the hilar confluence
Indian J Surg
(not laparoscopy) was performed to drain the peritoneal cavity and create a controlled EBF. At laparotomy, the proximal leaking stump was identified. Gentle effort was made to look for the distal divided end of the CBD by removing any ligature or clip in the hepatoduodenal ligament distally. If both ends could be identified, biliary continuity was established by inserting a T-tube of appropriate size into both the stumps (Fig. 2). Otherwise, the proximal stump was drained by an infant feeding tube of adequate size (tube hepaticostomy). A large bore tube drain was placed in the subhepatic space to drain out the leaking bile. Very sick patients were managed in the intensive care unit. They were initially administered broad spectrum and later culture specific antibiotics. In patients with tube hepaticostomy or T-tube, the subhepatic drain was removed once it stopped draining. The T-tube was clamped if cholangiogram showed free flow of dye into duodenum. The patients were discharged home either with the tube hepaticostomy or the clamped T-tube. They were kept on regular follow-up and a definitive repair, i.e. HJ was performed at least after 3 months. In group C patients, BDI and the resultant stricture was suspected on clinical grounds. US examination of abdomen was performed as routine to evaluate the biliary tree as well as to rule out any abdominal collection. The proximal biliary tree was evaluated by MRC or fistulogram (internal/external). The internal fistulogram was performed in the radiology suite by standard end-viewing upper GI endoscope. The duodenum was intubated and the strictured hepaticoduodenostomy stoma was visualized. A standard ERCP cannula
was negotiated through the strictured opening, 10 ml contrast (gastrograffin) was injected into the biliary tree and cholangiogram was obtained. Preoperative preparation included correction of coagulation parameters and broad spectrum antibiotics. The patients underwent HJ following the above-mentioned technique at least 6 weeks after the index cholecystectomy. Transjejunal stents were used selectively if the anastomosis appeared tenuous due to thin-walled, undilated ducts and local inflammation. Stents were removed 3 months later. The follow-up protocol for these patients included clinical assessment and liver function test (LFT) every 6 months. The outcome was graded into three categories: grade A–no clinical symptoms, normal LFT; grade B–no clinical symptoms, mild derangement of LFT or occasional episodes of pain or fever; grade C–pain, cholangitis and abnormal LFT; grade D–surgical revision or dilatation required [9].
Results The study group comprised 59 patients: 26 males and 33 females (mean age: 52 years (range 22–63 years)). The indication for cholecystectomy was acute cholecystitis in 24 patients and chronic cholecystitis in 35 patients. The injury occurred following open cholecystectomy and laparoscopic cholecystectomy in 22 and 37 patients, respectively. Group A comprised 6 patients, group B 33 patients and group C 20 patients. The distribution of patients as per Strasberg classification in various groups and with respect to laparoscopic or open cholecystectomy has been presented in Table 1. Group A: Of the six patients, two were from our own unit and two from another unit in the same hospital. The remaining two were from other hospitals in the same town located within 10 km radius. The injury in the latter two patients was recognized by the respective operating surgeons and communicated to us over telephone. Our team reached the concerned hospital and carried out the biliary repair. All the patients had sustained injuries during laparoscopic cholecystectomy. One patient had complete transection of the RHD (type C) and partial injury to LHD. He underwent right hepaticojejunostomy and repair of the LHD over stent. Two patients had partial transection of the common duct (type D) that was repaired over T-tube. Three patients had complete transection of the common duct (type E2) who underwent HJ over transjejunal stent. The postoperative recovery in all the patients was uneventful, except in one who had bile leak that lasted for 2 weeks.
Fig. 2 T-tube cholangiogram delineating the proximal and distal bile duct stump. Note the intervening missing segment
Group B: Eighteen patients presented with type A injury. The clinical presentation was (Table 2) controlled EBF in 13,
Indian J Surg Table 1 Distribution of patients as per Strasberg classification in various groups and with respect to open (OC) or laparoscopic cholecystectomy (LC)
Strasberg
Groups
Type
Nature of injury
A
B
C
OC
LC
A C D E1 E2 E3 E4 E5 Total
Cystic duct leak or leak from ducts in liver bed Transection of right hepatic duct Lateral injury to major bile duct Cut-off 2 cm below hilar confluence Cut-off within 2 cm below hilar confluence Cut-off just below confluence Separation of left and right hepatic duct Stricture involving common duct and Rt sectoral duct
– 01 02 – 03 – – – 06
18 – – 03 06 05 01 – 33
– – – 02 07 08 01 02 20
01 – – 03 07 10 01 – 22
17 01 02 03 09 03 01 01 37
biliary peritonitis in three, subhepatic bilioma in one and bile ascites, i.e. abdominal distension without significant systemic signs in one patient. Fifteen patients had type E injury, which was further classified into E1–three patients, E2–six patients, E3–five patients and E4–one patient. The clinical presentation in these patients was EBF in seven, biliary peritonitis in three, progressive jaundice in three and multiple intra-abdominal bile collections in two patients. Type A injury was diagnosed on ERC in all but one patient. Only one patient in this subgroup was found to have a stone in the CBD as a possible cause of pressure build-up and cystic duct blow-out. ERC was followed by EPT and insertion of 10f plastic stent into the CBD. Failure of ERC in one patient was due to pre-existing pyloric stenosis. He was managed expectantly after confirming the diagnosis on MRC. Relaparoscopy and drainage of the peritoneal cavity were performed concomitantly in patients with biliary peritonitis and bile ascites. Subhepatic bilioma was drained by US guided percutaneously placed pig-tail catheter. One patient developed mild post-ERC pancreatitis. One patient with biliary peritonitis died in the postoperative period due Table 2 Clinical presentation in groups B and C Clinical presentation
External biliary fistula Biliary peritonitis Subhepatic bilioma Bile ascites Multiple intra-abdominal collections Progressive jaundice Cholangitis Abdominal pain Night blindness
Group B
Group C
N033
N020
20 6 1 1 2 3 – 11 –
2 – – – – 10 12 10 1
to fulminant sepsis and multiorgan dysfunction. Bile leak in the remaining patients subsided in a mean period of 8 days (2 days–4 weeks). Type E injury was diagnosed on ERC in six patients and on MRC in nine patients. Patients with biliary peritonitis (n03) and intra-abdominal bile collection (n02) underwent exploratory laparotomy. At laparotomy, the Ttube was inserted into the proximal and distal stumps of the CBD in three patients (Fig. 3a). In two patients, only the proximal stump could be identified and tube hepaticostomy was performed. These two patients underwent hepaticojejunostomy 3 months later. The patients with the T-tube underwent hepaticojejunostomy after an interval of 3 months, 6 months and 2 years, respectively. During this interval, the T-tube remained clamped in two patients and replaced with endoscopically placed plastic stent after 6 months in the third patient. This stent was changed twice at 6 monthly intervals till she could be motivated to undergo HJ for the persistent stricture (Fig. 3b). Definitive repair was attempted in 10 patients at a median period of 8 days (range 4 days–2 weeks). Two of these patients had ligated but intact CBD. In one, the CBD had got trapped in the cystic duct ligature taken distal to the cystic lymph node. CBD in the second patient had got ligated by a hemostatic suture in the hepatoduodenal ligament. Both the injuries were suspected on MRC that showed the proximal and the distal CBD stump in close proximity. Removal of the offending ligature resulted in bile flow through the cystic duct. Intraoperative cholangiogram through the cystic duct stump confirmed intact CBD. A 12-f T-tube was inserted and retained for 3 months. In the remaining eight patients, one presented with ligated bile duct and other seven presented with controlled EBF through the drain. HJ was performed over transjejunal stent. Three patients had bile leak in the postoperative period, which stopped on conservative management.
Indian J Surg Fig. 3 a T-tube cholangiogram in a patient who had transected CBD. The T-tube was placed into the transected ends during laparotomy for biliary peritonitis. b MRC performed in the same patient 2 years later, showing stricture at the site of injury
Group C: Twenty patients were included in group C. The clinical presentation was (Table 2) cholangitis in 12, jaundice in 10, abdominal pain in four, biliary fistula in two and night blindness in one patient. Eleven patients had prior intervention in the form of end-to-side hepaticoduodenostomy in four, T-tube repair in two, T-tube repair and angioembolization of right hepatic artery pseudoaneurysm in one and suture closure of bile leak in two patients. Median serum bilirubin was 6 mg/dl (range 1.8–18), serum aminotransferase was 110 IU/dl (range 30–270), serum alanine transferase was 120 IU/dl (35–380) and serum alkaline phosphatise was 440 IU/dl (range 66–1025). Cholangiographic evaluation of the proximal biliary tree was done by MRC in 17 patients, internal fistulogram in two patients and external fistulogram in one patient. The median interval between index cholecystectomy and definitive repair (HJ) was 8 months (range 6 weeks–10 years). One patient had postoperative bile leak and two had wound infection. In a mean follow-up of 40 months (range 6–90 months), the outcome was grade A in 54 patients, grade B in three patients (one from each of the three groups) and grade D in one patient (group C). The latter patient with a type E3 injury developed recurrent stricture and cholangitis necessitating percutaneous transhepatic dilatation of the stricture by the radiologist.
Discussion In the present era of laparoscopic cholecystectomy, it is interesting to find that a sizable proportion of our patients (22/59) had undergone open cholecystectomy. This underscores the fact that in a developing country like India, open cholecystectomy is still performed in large numbers in peripheral hospitals. BDI possibly reflects inadequacy of infrastructure and experience of the operating surgeon in these hospitals. In our
observation, type A injury was more common with laparoscopic cholecystectomy whereas type E3 was more common with open cholecystectomy. This contradicts the earlier observation that laparoscopic surgery is associated with severe and more complex BDI [10, 11]. A majority of BDI go unrecognized during the index cholecystectomy [2–4]. When recognized, the basic principles of management are: convert to open in case of laparoscopic cholecystectomy, assess the injury, perform on-table cholangiogram, seek help from more experienced surgeon and carry out an optimal repair [1]. Partial injury should be repaired over the T-tube [11]. Complete transection of the bile duct is best repaired by hepaticojejunostomy for a better long-term outcome. The latter procedure can pose technical difficulty because the bile duct, at this stage, is thin walled and undilated. It is, therefore, desirable that the procedure should always be performed by tertiary care specialist hepatobiliary surgeon. In one study, only 17 % of the repair was successful in the hands of primary surgeons compared to 94 % when performed by tertiary care surgeons [5]. Similar results have also been reported by other authors [6]. We managed six patients (group A) in whom the injury was detected during cholecystectomy. Our approach in these patients was similar to what has been recommended by other authors [1]. Four of the six patients were from other units or hospitals. Our team members reached out to offer expert help and carried out the required repair. The feasibility and positive outcome of such ‘specialist outreach service’ have earlier been reported [7]. In the latter report, the ‘traveling surgeon’, when called upon to help, carried along with him some specialized surgical instruments for performing the repair. Such an approach is feasible in India, notwithstanding the poor infrastructural facilities present in some of these hospitals where the BDI has occurred. In the present series, type A (cystic duct leak) was the commonest form of injury seen in 18 patients belonging to
Indian J Surg
group B. This may represent referral bias for ERCP that we routinely perform. In only one patient, CBD stone could be identified as the factor for cystic duct leak. In the remaining patients, clip failure was the possible cause. This contradicts the earlier perception that unsuspected CBD stone is the commonest predisposing factor for cystic duct ‘blow out’ [12]. Retained CBD stone has been reported in about 20 % of patients with cystic duct leak [13–16]. Management of type A injury includes ERC, EPT alone for low-grade leaks or with bile duct stenting. Additionally, drainage of intraabdominal collection, when present is achieved by US or CT guided percutaneously placed catheter or relaparoscopy [13–16]. Percutaneous placement of catheter precedes ERC whereas relaparoscopy follows ERC during the same anesthesia. By this approach, four of the five patients in group B could be successfully salvaged. Conventional approach to type E injuries includes control of sepsis, creation of controlled EBF and definitive repair after a waiting period ranging from 6 weeks to several months. The latter is to allow the inflammation to subside fibrosis at the cut end to stabilize and the bile duct to dilate for an optimal hepaticojejunostomy [17–20]. An attempt at early repair was associated with high failure rates [4, 21–23]. Five patients in group B were managed in this manner. All these patients underwent laparotomy to drain out the peritoneal cavity as well as to create controlled EBF. At laparotomy if feasible, we prefer to locate both the divided ends of the bile duct by gentle dissection and place the T-Tube. This was possible in three of the five patients. The remaining patients had tube hepaticostomy only. In our opinion the Ttube has definite advantages over tube hepaticostomy. In the initial stages, it maintains external as well as internal drainage of bile. After cessation of peritubal/subhepatic bile leak, the T-tube could be clamped, thereby diverting the entire bile volume into the duodenum. The fluid and electrolyte loss could be minimized in addition to the obvious advantages of restoring the normal physiology of bile. A clamped T-tube had a better acceptance with the patients in delaying the definitive repair to an optimum time. This was in contrast to the externally draining tube (hepaticostomy) that acted as a constant source of irritation to the patients who kept on nagging for an early repair. The delay also helped in better preparing the patients physically, mentally and financially for the definitive surgery. Notwithstanding the conventional practice, some recent studies have highlighted the role of early repair of BDI in a select group of patients [6, 24, 25]. The result of early repair was comparable to the delayed approach in a stable patient in whom intra-abdominal sepsis, vascular and cautery injury could be excluded and more importantly when the surgery was performed by ‘specialist hepatobiliary surgeon’. The exact timing of repair with respect to ‘early’ was not clearly defined but the repair was carried out within 2 weeks of
cholecystectomy. In the present study, early repair was carried out in 10 of the group B patients. All these patients had none or minimally inflamed hepatic hilum either because of the completely ligated bile duct (three patients) or a controlled EBF (five patients). Two patients in group B having intact but ligated CBD were treated successfully by removal of the offending ligature and long-term T-tube placement. We could not find mention of this type of injury and, therefore, support this kind of approach in any of the available literature. End-to-end repair of transected CBD over the T-tube is associated with a high failure rate, hence not recommended by some authors [5, 26]. Proponents, on the other hand, proposed T-tube repair if there is no segmental loss of the bile duct [27, 28]. In addition to simplicity of the technique, strictures following T-tube repair is salvageable either by radiological intervention or revision surgery, in the form of HJ without much difficulty [28]. Simple ligature around the CBD, in our opinion, devitalizes lesser segment of the bile duct compared to transected one. This segment should get revascularized over a period of time if patency of its lumen is maintained by an indwelling stent. However, a longer follow-up and more number of cases will be required to prove this point. No attempt was made by us to exclude vascular injury in any of our patients. The implications of concomitant vascular injury with respect to repair of the bile duct have been highlighted by various authors [29–32]. These authors recommend that reconstruction of the arterial injury should be attempted whenever feasible. However, we strongly believe that a high anastomosis extending onto LHD in a well collateralized hilum obviates the need for any such vascular repair that can otherwise be tedious and time consuming. This view has also been shared by other authors [33]. Nine of the 20 patients in group C had prior operative intervention by the operating surgeon to correct the injury. Four of them had hepaticoduodenostomy. Published reports suggest that repair by the primary or the ‘injuring surgeon’ carries a poor outcome [5, 34]. Moreover, hepaticoduodenostomy almost always fails in the long run [35]. The only advantage the latter anastomosis was that the opening in the duodenum provided an access for obtaining cholangiogram (internal fistulogram). Such modality of cholangiogram has not been reported earlier. The repair was performed at least 6 weeks after the index cholecystectomy. The reasons for this waiting were similar to what has been stated for patients in group B who underwent delayed repair. The high success rate of bile duct repair in the present study can be attributed to the appropriate timing, meticulous technique and the tertiary care experience, a fact that has been highlighted by several authors [1, 18, 20, 35]. The only limitation in the present study, however, is the short follow-up period, i.e. 40 months. A follow-up of at least 5 years or more is desirable to make a final pronouncement of the results [36].
Indian J Surg
References 1. Lau WE, Lai ECH, Lau SHW (2010) Management of bile duct injury after laparoscopic cholecystectomy: a review. ANZ J Surg 80:75–81 2. Slater K, Strong RW, Wall DR, Lynch SV (2002) Iatrogenic bile duct injury: the scourge of laparoscopic cholecystectomy. ANZ J Surg 72:83–88 3. Sicklick JK, Camp MS, Lillemore KD, Melton GB, Yeo CJ, Campbell KA et al (2005) Surgical management of bile duct injuries sustained during laparoscopic cholecystectomy. Ann Surg 241:786–795 4. De Reuver PR, Rauws EA, Bruno MJ (2007) Survival in bile duct injury patients after laparoscopic cholecystectomy: a multidisciplinary approach of gastroenterologists, radiologists and surgeons. Surgery 142:1–9 5. Stewart L, Way LW (1995) Bile duct injuries during laparoscopic cholecystectomy. Factors that influence the results of treatment. Arch Surg 130:1123–1128 6. Perera MTPR, Silva MA, Hegab B, Muralidharan V, Bramhall SR, Mayer AD et al (2011) Specialist early and immediate repair of post-laparoscopic cholecystectomy bile duct injuries is associated with an improved long-term outcome. Ann Surg 253:553–560 7. Silva MA, Coldham C, Mayer AD, Bramhali SR, Buckels J, Mirza DF (2008) Specialist outreach service for on-table repair of iatrogenic bile duct injuries—a new kind of ‘travelling surgeon’. Ann R Coll Surg Engl 90:243–246 8. Strasberg SM, Hertl M, Soper NJ (1995) An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 180:101–125 9. McDonald MI, Farnell MB, Nagorney DM, IIstrup DM, Kutch JM (1995) Benign biliary stricture: repair and outcome with a contemporary approach. Surgery 118:582–591 10. Chaudhary A, Manisegran M, Chandra A (2001) How do bile duct injuries sustained during laparoscopic cholecystectomy differ from those during open cholecystectomy. J Laparoendosc Adv Surg Tech 11:187–191 11. Sikora SS, Kumar A, Dash NR, Sarkari A, Saxena R, Kapoor VK (2001) Laparoscopic bile duct injuries-spectrum at a tertiary care centre. J Laparoendosc Adv Surg Tech 11:63–68 12. Corvera CU, Jarnagin WR, Blumgart LH (2007) Biliary fistula. In: Blumgart LH (ed) Surgery of the liver, biliary tract and pancreas. Saunders Elsevier, Philadelphia, pp 655–684 13. Sandha GS, Bourke MJ, Haber GB, Kortan PP (2004) Endoscopic therapy for bile leak based on a new classification; results in 207 patients. Gastrointest Endosc 60:567–574 14. Eisenstein S, Greenstein AJ, Kim U, Divino CM (2008) Cystic duct stump leaks; after the learning curve. Arch Surg 143:1178–1183 15. Pinkas H, Brady PG (2008) Bile leaks after laparoscopic cholecystectomy: time to stent or time to drain. Hepatobiliary Pancreat Dis Int 7:628–632 16. Aksoz K, Unsal B, Yoruk G, Buyrac Z, Haciyanli M, Akpinar Z, Alper E (2009) Endoscopic sphincterotomy alone in the management of low-grade biliary leaks due to cholecystectomy. Dig Endosc 21:158–161 17. Lillemoe KD, Pitt HA, Cameron JL (1990) Postoperative biliary stricture. Surg Clin N Am 70:1355–1388
18. Sikora SS, Pottakkat B, Srikanth G, Kumar A, Saxena R, Kapoor VK (2006) Postcholecystectomy benign biliary strictures-long-term results. Dig Surg 23:304–312 19. Walsh RM, Henderson JM, Vogt DP (2007) Long-term oucome of biliary reconstruction for bile duct injuries from laparoscopic cholecystectomies. Surgery 142:450–456 20. Kapoor VK (2007) Bile duct injury repair: when? what? who? J Hepatobiliary Pancreat Surg 14:476–479 21. Schol FP, Go PM, Gouma DJ (1995) Outcome of 49 repairs of bile duct strictures after laparoscopic cholecystectomy. World J Surg 19:168–172 22. Schmidt SC, Lagrehr JM, Hintze RE, Neuhaus P (2005) Long-term results and risk factors influencing outcome of major bile duct injuries following cholecystectomy. Br J Surg 92:76–82 23. deReuver PR, Grossmann I, Busch OR, Obertop H, van Gulik TM, Gouma DJ (2007) Referral pattern and timing of repair are risk factors for complications after reconstructive surgery for bile duct injury. Ann Surg 245:763–770 24. Mercado MA (2007) Early versus late repair of bile duct injuries. Surg Endosc 20:1644–1647 25. Thomson BNJ, Parks RW, Madhavan KK, Wigmore SJ, Garden OJ (2006) Early specialist repair of biliary injury. Br J Surg 93:216–220 26. Wudel LJ Jr, Wright JK, Pinson CW, Herline A, Debelak J, Seidel S, Revis K, Chapman WC (2001) Bile duct injury following laparoscopic cholecystectomy: a cause of continuing concern. Am Surg 67:557–563 27. Bismuth H, Majno PE (2001) Biliary strictures: classification based on the principles of surgical treatment. World J Surg 25:1241–1244 28. de Reuver PR, Busch ORC, Rauws EA, Lameris JS, van Gulik TM, Gouma DJ (2007) Long-term results of a primary end-to-end anastomosis in preoperative detected bile duct injury. J Gastrointest Surg 11:296–302 29. Schmidt SC, Settmacher U, Langrehr JM, Neuhaus P (2004) Management and outcome of patients with combined bile duct and hepatic arterial injuries after laparoscopic cholecystectomy. Surgery 135:613–618 30. Gupta N, Solomon H, Fairchild R, Kaminski DL (1998) Management and outcome of patients with combined bile duct and hepatic artery injuries. Arch Surg 133:176–181 31. Koffron A, Ferrario M, Parsons W, Nemcek A, Saker M, Abecassis M (2001) Failed primary management of iatrogenic biliary injury: incidence and significance of concomitant hepatic arterial disruption. Surgery 130:722–731 32. Bachellier P, Nakano H, Weber J-C, Lemarque P, Oussoultzoglou E, Candau C et al (2001) Surgical repair after bile duct and vascular injuries during laparoscopic cholecystectomy: when and how? World J Surg 25:1335–1345 33. Tzovaras G, Dervenis C (2006) Vascular injuries in laparoscopic cholecystectomy: an underestimated problem. Dig Surg 23:370– 374 34. Flum DR, Cheadle A, Prela C, Dellinger EP, Chan L (2003) Bile duct injury during cholecystectomy and survival in medicare beneficiaries. JAMA 290:2168–2173 35. Connor S, Garden OJ (2006) Bile duct injury in the era of laparoscopic cholecystectomy. Br J Surg 93:158–167 36. Sikora S (2012) Management of post-cholecystectomy benign biliary strictures: review. Ind J Surg 74:22–28
ORIGINAL ARTICLE
Management of Bile Duct Injury at Various Stages of Presentation: Experience from a Tertiary Care Centre Md. Ibrarullah & S. Sankar & K. Sreenivasan & S. R. K. Gavini
Received: 7 April 2012 / Accepted: 3 September 2012 # Association of Surgeons of India 2012
Abstract The clinical presentation, management and outcome of all patients with bile duct injury who presented to our tertiary care centre at various stages after cholecystectomy were analyzed. The patients were categorized into three groups: group A–patients in whom the injury was detected during cholecystectomy, group B–patients who presented within 2 weeks of cholecystectomy and group C–patients who presented after 2 weeks of cholecystectomy. Our team acted as rescue surgeons and performed ‘on-table’ repair for injuries occurring in another unit or in another hospital. Strasberg classification of bile duct injury was followed. In group A, partial and complete transections were managed by repair over T-tube and high hepaticojejunostomy, respectively. Patients in group B underwent endoscopic retrograde cholangiogram and/ or magnetic resonance cholangiogram to evaluate the biliary tree. Those with intact common bile duct underwent endoscopic papillotomy and stenting in addition to drainage of intraabdominal collection when present. For those with complete transection, early repair was considered if there was no sepsis. In presence of intra-abdominal sepsis an attempt was made to create controlled external biliary fistula. This was followed by hepatico jejunostomy at least after 3 months. Group C patients underwent hepaticojejunostomy at least 6 weeks after the injury. The outcome was graded into three categories: grade A–no clinical symptoms, normal LFT; grade B–no clinical symptoms, mild derangement of LFT or occasional episodes of pain or fever; grade C–pain, cholangitis and abnormal LFT; grade D–surgical revision or dilatation required. Fifty nine patients were included in the study and the distribution was group A–six patients, group B–33 patients and group C–20 patients. In group A, one patient with complete transection of M. Ibrarullah (*) : S. Sankar : K. Sreenivasan : S. R. K. Gavini Department of Surgical Gastroenterology, Sri Ramachandra Medical College and Research Institute, Porur Chennai 600116, India e-mail: [email protected]
the right hepatic duct (type C) and partial injury to left hepatic duct (LHD) underwent right hepaticojejunostomy and repair of the LHD over stent. Two patients with type D and three patients with type E 2 injury underwent repair over T-tube and hepaticojejunostomy, respectively. In group B, all except one of the 18 patients with type A injury underwent endoscopic papillotomy and stenting. The bile leak subsided at a mean interval of 8 days in all, except one patient who died of fulminant sepsis. Of the 15 patients with type E injury, five underwent hepaticojejunostomy after a minimum gap of 3 months. Early repair was considered in 10 patients. Twenty patients in group C underwent hepaticojejunostomy. In a mean follow-up of 40 months, the outcome was grade A in 54 patients, grade B in three patients (one from each of the three groups) and grade D in one patient (group C). The latter patient with a type E3 injury developed recurrent stricture and cholangitis necessitating percutaneous transhepatic dilatation. The high success rate of bile duct repair in the present study can be attributed to the appropriate timing, meticulous technique and the tertiary care experience. Keywords Cholecystectomy . Benign biliary stricture . Bile duct injury Cholecystectomy is the commonest abdominal surgery performed across the world. The reported incidence of bile duct injury (BDI) following open cholecystectomy is 0.1 %– 0.2 %. In the present era of laparoscopic cholecystectomy, the incidence has gone up to 0.4 %–0.7 % [1]. Though some of the injuries can be identified during the primary surgery, majority however, present in the postoperative period [2–4]. Proper identification of the injury, appropriate timing, technique of repair and experience of the surgeon are paramount in the management of these patients. Some of the recent studies have laid much emphasis on repair by specialist hepatobiliary surgeons and considered this to be one of the
Indian J Surg
most important factors in the successful outcome of BDI [5–7]. The present study is a retrospective analysis of all patients with BDI who presented to us during or at a variable period after the index cholecystectomy. Their clinical presentation, management and outcome were studied.
Patients and Methods Ours is a tertiary care referral centre with special interest in hepatobiliary surgery. Hospital records of all patients with postcholecystectomy BDI, managed between April 2004 and April 2012, were entered in a prospectively maintained database. The study included patients who sustained injuries while being operated in our hospital and also those who were operated elsewhere and referred to our centre after the injury was detected or suspected. Depending upon the time of presentation, the patients were categorized into following groups: group A–patients whose injury was detected ‘ontable’ during cholecystectomy, group B–patients who presented in the early postoperative period, i.e. within 2 weeks of cholecystectomy and group C–patients who presented 2 weeks after cholecystectomy. Our team acted as rescue surgeons for injuries detected in another unit in the same hospital or another hospital, when promptly informed. The latter group of patients was followed up in active consultation with the primary surgeon. We followed Strasberg classification of BDI [8]. For group A patients, laparoscopy was converted to open laparotomy by a wide subcostal or midline incision. All the ligatures or clips barring those applied on vessels were removed. Intraoperative cholangiogram was performed to delineate the entire biliary tree. After removal of the ligature/clip if the common bile duct (CBD) was found to be intact, a choledochotomy was performed and an appropriate size T-tube was placed. Partial transection was repaired over appropriate size stent/s or T-tube using 40 polygalactine (Vicryl) suture. In case of complete transection, an attempt was made to delineate the proximal cut end of the bile duct. The stoma was extended proximally so as to perform a high hepaticojejunostomy Roux-en-Y (HJ). When required, the hilar plate was lowered and a stoma of approximately 2.5 cm was created by extending the ductal incision to left hepatic duct (LHD). HJ was performed using single layer, interrupted 3-0 or 4-0 Vicryl sutures. Transjejunal stent was used for very thin-walled and undilated bile duct. A wide bore tube drain was placed in the subhepatic space. The stent or T-tube was kept for 3 months and removed after obtaining satisfactory cholangiogram. For group B patients, injury was suspected on clinical grounds, i.e. tachycardia, fever, jaundice, abdominal distension or bilious drainage. Ultrasound (US) was performed to locate any collection, check the integrity of CBD and for any evidence of distal obstruction. Patients with external
biliary fistula (EBF) without any signs of intra-abdominal sepsis underwent endoscopic retrograde cholangiogram (ERC) or magnetic resonance cholangiogram (MRC). MRC was the investigation of choice if the patient could afford or else ERC was performed since it offered one time therapeutic advantage as well. Patients showing bile leakage but intact CBD were subjected to endoscopic papillotomy (EPT) and insertion of 10f plastic stent. For patients whose ERC showed complete ‘cut-off’, the proximal biliary tree was evaluated on MRC. The MRC was analyzed for the proximal stump, the sectoral branches and any evidence of intraabdominal collection. Distance between the proximal and distal CBD stump was also carefully studied to ascertain if there was any segmental loss or it was only a case of ligated/ clipped CBD (Fig. 1). Early laparotomy and repair was considered if the patient was stable, did not show any signs of intra-abdominal collection or sepsis. At laparotomy, the approach to a ligated or transected CBD was similar to what has been described for group A patients. In patients with suspected bile collection and/or peritonitis, an attempt was made to drain all the collection and create a controlled EBF. These patients underwent US or contrast enhanced CT scan examination of the abdomen. A localized collection was drained by US guided percutaneous catheter. In case of diffuse or excessive collection, patients were subjected to relaparoscopy or laparotomy. Preceding this, if the patient was hemodynamically stable and could afford, MRC was performed to delineate the intra- and extrahepatic biliary tree. If the latter investigation was not available, ERC was performed in the same sitting prior to laparoscopy/ laparotomy. EPT and bile duct stenting was carried out for patients showing bile leak but intact CBD. All the bile collections were thoroughly drained and tube drains inserted in the subhepatic space and pelvis. The biliary stent was removed 4 weeks after the stoppage of bile leak. For patients with complete CBD ‘cut-off’ on MRC/ERC and gross peritoneal contamination, exploratory laparotomy
Fig. 1 MRC showing segmental loss of the common duct (marked by the arrows) just below the hilar confluence
Indian J Surg
(not laparoscopy) was performed to drain the peritoneal cavity and create a controlled EBF. At laparotomy, the proximal leaking stump was identified. Gentle effort was made to look for the distal divided end of the CBD by removing any ligature or clip in the hepatoduodenal ligament distally. If both ends could be identified, biliary continuity was established by inserting a T-tube of appropriate size into both the stumps (Fig. 2). Otherwise, the proximal stump was drained by an infant feeding tube of adequate size (tube hepaticostomy). A large bore tube drain was placed in the subhepatic space to drain out the leaking bile. Very sick patients were managed in the intensive care unit. They were initially administered broad spectrum and later culture specific antibiotics. In patients with tube hepaticostomy or T-tube, the subhepatic drain was removed once it stopped draining. The T-tube was clamped if cholangiogram showed free flow of dye into duodenum. The patients were discharged home either with the tube hepaticostomy or the clamped T-tube. They were kept on regular follow-up and a definitive repair, i.e. HJ was performed at least after 3 months. In group C patients, BDI and the resultant stricture was suspected on clinical grounds. US examination of abdomen was performed as routine to evaluate the biliary tree as well as to rule out any abdominal collection. The proximal biliary tree was evaluated by MRC or fistulogram (internal/external). The internal fistulogram was performed in the radiology suite by standard end-viewing upper GI endoscope. The duodenum was intubated and the strictured hepaticoduodenostomy stoma was visualized. A standard ERCP cannula
was negotiated through the strictured opening, 10 ml contrast (gastrograffin) was injected into the biliary tree and cholangiogram was obtained. Preoperative preparation included correction of coagulation parameters and broad spectrum antibiotics. The patients underwent HJ following the above-mentioned technique at least 6 weeks after the index cholecystectomy. Transjejunal stents were used selectively if the anastomosis appeared tenuous due to thin-walled, undilated ducts and local inflammation. Stents were removed 3 months later. The follow-up protocol for these patients included clinical assessment and liver function test (LFT) every 6 months. The outcome was graded into three categories: grade A–no clinical symptoms, normal LFT; grade B–no clinical symptoms, mild derangement of LFT or occasional episodes of pain or fever; grade C–pain, cholangitis and abnormal LFT; grade D–surgical revision or dilatation required [9].
Results The study group comprised 59 patients: 26 males and 33 females (mean age: 52 years (range 22–63 years)). The indication for cholecystectomy was acute cholecystitis in 24 patients and chronic cholecystitis in 35 patients. The injury occurred following open cholecystectomy and laparoscopic cholecystectomy in 22 and 37 patients, respectively. Group A comprised 6 patients, group B 33 patients and group C 20 patients. The distribution of patients as per Strasberg classification in various groups and with respect to laparoscopic or open cholecystectomy has been presented in Table 1. Group A: Of the six patients, two were from our own unit and two from another unit in the same hospital. The remaining two were from other hospitals in the same town located within 10 km radius. The injury in the latter two patients was recognized by the respective operating surgeons and communicated to us over telephone. Our team reached the concerned hospital and carried out the biliary repair. All the patients had sustained injuries during laparoscopic cholecystectomy. One patient had complete transection of the RHD (type C) and partial injury to LHD. He underwent right hepaticojejunostomy and repair of the LHD over stent. Two patients had partial transection of the common duct (type D) that was repaired over T-tube. Three patients had complete transection of the common duct (type E2) who underwent HJ over transjejunal stent. The postoperative recovery in all the patients was uneventful, except in one who had bile leak that lasted for 2 weeks.
Fig. 2 T-tube cholangiogram delineating the proximal and distal bile duct stump. Note the intervening missing segment
Group B: Eighteen patients presented with type A injury. The clinical presentation was (Table 2) controlled EBF in 13,
Indian J Surg Table 1 Distribution of patients as per Strasberg classification in various groups and with respect to open (OC) or laparoscopic cholecystectomy (LC)
Strasberg
Groups
Type
Nature of injury
A
B
C
OC
LC
A C D E1 E2 E3 E4 E5 Total
Cystic duct leak or leak from ducts in liver bed Transection of right hepatic duct Lateral injury to major bile duct Cut-off 2 cm below hilar confluence Cut-off within 2 cm below hilar confluence Cut-off just below confluence Separation of left and right hepatic duct Stricture involving common duct and Rt sectoral duct
– 01 02 – 03 – – – 06
18 – – 03 06 05 01 – 33
– – – 02 07 08 01 02 20
01 – – 03 07 10 01 – 22
17 01 02 03 09 03 01 01 37
biliary peritonitis in three, subhepatic bilioma in one and bile ascites, i.e. abdominal distension without significant systemic signs in one patient. Fifteen patients had type E injury, which was further classified into E1–three patients, E2–six patients, E3–five patients and E4–one patient. The clinical presentation in these patients was EBF in seven, biliary peritonitis in three, progressive jaundice in three and multiple intra-abdominal bile collections in two patients. Type A injury was diagnosed on ERC in all but one patient. Only one patient in this subgroup was found to have a stone in the CBD as a possible cause of pressure build-up and cystic duct blow-out. ERC was followed by EPT and insertion of 10f plastic stent into the CBD. Failure of ERC in one patient was due to pre-existing pyloric stenosis. He was managed expectantly after confirming the diagnosis on MRC. Relaparoscopy and drainage of the peritoneal cavity were performed concomitantly in patients with biliary peritonitis and bile ascites. Subhepatic bilioma was drained by US guided percutaneously placed pig-tail catheter. One patient developed mild post-ERC pancreatitis. One patient with biliary peritonitis died in the postoperative period due Table 2 Clinical presentation in groups B and C Clinical presentation
External biliary fistula Biliary peritonitis Subhepatic bilioma Bile ascites Multiple intra-abdominal collections Progressive jaundice Cholangitis Abdominal pain Night blindness
Group B
Group C
N033
N020
20 6 1 1 2 3 – 11 –
2 – – – – 10 12 10 1
to fulminant sepsis and multiorgan dysfunction. Bile leak in the remaining patients subsided in a mean period of 8 days (2 days–4 weeks). Type E injury was diagnosed on ERC in six patients and on MRC in nine patients. Patients with biliary peritonitis (n03) and intra-abdominal bile collection (n02) underwent exploratory laparotomy. At laparotomy, the Ttube was inserted into the proximal and distal stumps of the CBD in three patients (Fig. 3a). In two patients, only the proximal stump could be identified and tube hepaticostomy was performed. These two patients underwent hepaticojejunostomy 3 months later. The patients with the T-tube underwent hepaticojejunostomy after an interval of 3 months, 6 months and 2 years, respectively. During this interval, the T-tube remained clamped in two patients and replaced with endoscopically placed plastic stent after 6 months in the third patient. This stent was changed twice at 6 monthly intervals till she could be motivated to undergo HJ for the persistent stricture (Fig. 3b). Definitive repair was attempted in 10 patients at a median period of 8 days (range 4 days–2 weeks). Two of these patients had ligated but intact CBD. In one, the CBD had got trapped in the cystic duct ligature taken distal to the cystic lymph node. CBD in the second patient had got ligated by a hemostatic suture in the hepatoduodenal ligament. Both the injuries were suspected on MRC that showed the proximal and the distal CBD stump in close proximity. Removal of the offending ligature resulted in bile flow through the cystic duct. Intraoperative cholangiogram through the cystic duct stump confirmed intact CBD. A 12-f T-tube was inserted and retained for 3 months. In the remaining eight patients, one presented with ligated bile duct and other seven presented with controlled EBF through the drain. HJ was performed over transjejunal stent. Three patients had bile leak in the postoperative period, which stopped on conservative management.
Indian J Surg Fig. 3 a T-tube cholangiogram in a patient who had transected CBD. The T-tube was placed into the transected ends during laparotomy for biliary peritonitis. b MRC performed in the same patient 2 years later, showing stricture at the site of injury
Group C: Twenty patients were included in group C. The clinical presentation was (Table 2) cholangitis in 12, jaundice in 10, abdominal pain in four, biliary fistula in two and night blindness in one patient. Eleven patients had prior intervention in the form of end-to-side hepaticoduodenostomy in four, T-tube repair in two, T-tube repair and angioembolization of right hepatic artery pseudoaneurysm in one and suture closure of bile leak in two patients. Median serum bilirubin was 6 mg/dl (range 1.8–18), serum aminotransferase was 110 IU/dl (range 30–270), serum alanine transferase was 120 IU/dl (35–380) and serum alkaline phosphatise was 440 IU/dl (range 66–1025). Cholangiographic evaluation of the proximal biliary tree was done by MRC in 17 patients, internal fistulogram in two patients and external fistulogram in one patient. The median interval between index cholecystectomy and definitive repair (HJ) was 8 months (range 6 weeks–10 years). One patient had postoperative bile leak and two had wound infection. In a mean follow-up of 40 months (range 6–90 months), the outcome was grade A in 54 patients, grade B in three patients (one from each of the three groups) and grade D in one patient (group C). The latter patient with a type E3 injury developed recurrent stricture and cholangitis necessitating percutaneous transhepatic dilatation of the stricture by the radiologist.
Discussion In the present era of laparoscopic cholecystectomy, it is interesting to find that a sizable proportion of our patients (22/59) had undergone open cholecystectomy. This underscores the fact that in a developing country like India, open cholecystectomy is still performed in large numbers in peripheral hospitals. BDI possibly reflects inadequacy of infrastructure and experience of the operating surgeon in these hospitals. In our
observation, type A injury was more common with laparoscopic cholecystectomy whereas type E3 was more common with open cholecystectomy. This contradicts the earlier observation that laparoscopic surgery is associated with severe and more complex BDI [10, 11]. A majority of BDI go unrecognized during the index cholecystectomy [2–4]. When recognized, the basic principles of management are: convert to open in case of laparoscopic cholecystectomy, assess the injury, perform on-table cholangiogram, seek help from more experienced surgeon and carry out an optimal repair [1]. Partial injury should be repaired over the T-tube [11]. Complete transection of the bile duct is best repaired by hepaticojejunostomy for a better long-term outcome. The latter procedure can pose technical difficulty because the bile duct, at this stage, is thin walled and undilated. It is, therefore, desirable that the procedure should always be performed by tertiary care specialist hepatobiliary surgeon. In one study, only 17 % of the repair was successful in the hands of primary surgeons compared to 94 % when performed by tertiary care surgeons [5]. Similar results have also been reported by other authors [6]. We managed six patients (group A) in whom the injury was detected during cholecystectomy. Our approach in these patients was similar to what has been recommended by other authors [1]. Four of the six patients were from other units or hospitals. Our team members reached out to offer expert help and carried out the required repair. The feasibility and positive outcome of such ‘specialist outreach service’ have earlier been reported [7]. In the latter report, the ‘traveling surgeon’, when called upon to help, carried along with him some specialized surgical instruments for performing the repair. Such an approach is feasible in India, notwithstanding the poor infrastructural facilities present in some of these hospitals where the BDI has occurred. In the present series, type A (cystic duct leak) was the commonest form of injury seen in 18 patients belonging to
Indian J Surg
group B. This may represent referral bias for ERCP that we routinely perform. In only one patient, CBD stone could be identified as the factor for cystic duct leak. In the remaining patients, clip failure was the possible cause. This contradicts the earlier perception that unsuspected CBD stone is the commonest predisposing factor for cystic duct ‘blow out’ [12]. Retained CBD stone has been reported in about 20 % of patients with cystic duct leak [13–16]. Management of type A injury includes ERC, EPT alone for low-grade leaks or with bile duct stenting. Additionally, drainage of intraabdominal collection, when present is achieved by US or CT guided percutaneously placed catheter or relaparoscopy [13–16]. Percutaneous placement of catheter precedes ERC whereas relaparoscopy follows ERC during the same anesthesia. By this approach, four of the five patients in group B could be successfully salvaged. Conventional approach to type E injuries includes control of sepsis, creation of controlled EBF and definitive repair after a waiting period ranging from 6 weeks to several months. The latter is to allow the inflammation to subside fibrosis at the cut end to stabilize and the bile duct to dilate for an optimal hepaticojejunostomy [17–20]. An attempt at early repair was associated with high failure rates [4, 21–23]. Five patients in group B were managed in this manner. All these patients underwent laparotomy to drain out the peritoneal cavity as well as to create controlled EBF. At laparotomy if feasible, we prefer to locate both the divided ends of the bile duct by gentle dissection and place the T-Tube. This was possible in three of the five patients. The remaining patients had tube hepaticostomy only. In our opinion the Ttube has definite advantages over tube hepaticostomy. In the initial stages, it maintains external as well as internal drainage of bile. After cessation of peritubal/subhepatic bile leak, the T-tube could be clamped, thereby diverting the entire bile volume into the duodenum. The fluid and electrolyte loss could be minimized in addition to the obvious advantages of restoring the normal physiology of bile. A clamped T-tube had a better acceptance with the patients in delaying the definitive repair to an optimum time. This was in contrast to the externally draining tube (hepaticostomy) that acted as a constant source of irritation to the patients who kept on nagging for an early repair. The delay also helped in better preparing the patients physically, mentally and financially for the definitive surgery. Notwithstanding the conventional practice, some recent studies have highlighted the role of early repair of BDI in a select group of patients [6, 24, 25]. The result of early repair was comparable to the delayed approach in a stable patient in whom intra-abdominal sepsis, vascular and cautery injury could be excluded and more importantly when the surgery was performed by ‘specialist hepatobiliary surgeon’. The exact timing of repair with respect to ‘early’ was not clearly defined but the repair was carried out within 2 weeks of
cholecystectomy. In the present study, early repair was carried out in 10 of the group B patients. All these patients had none or minimally inflamed hepatic hilum either because of the completely ligated bile duct (three patients) or a controlled EBF (five patients). Two patients in group B having intact but ligated CBD were treated successfully by removal of the offending ligature and long-term T-tube placement. We could not find mention of this type of injury and, therefore, support this kind of approach in any of the available literature. End-to-end repair of transected CBD over the T-tube is associated with a high failure rate, hence not recommended by some authors [5, 26]. Proponents, on the other hand, proposed T-tube repair if there is no segmental loss of the bile duct [27, 28]. In addition to simplicity of the technique, strictures following T-tube repair is salvageable either by radiological intervention or revision surgery, in the form of HJ without much difficulty [28]. Simple ligature around the CBD, in our opinion, devitalizes lesser segment of the bile duct compared to transected one. This segment should get revascularized over a period of time if patency of its lumen is maintained by an indwelling stent. However, a longer follow-up and more number of cases will be required to prove this point. No attempt was made by us to exclude vascular injury in any of our patients. The implications of concomitant vascular injury with respect to repair of the bile duct have been highlighted by various authors [29–32]. These authors recommend that reconstruction of the arterial injury should be attempted whenever feasible. However, we strongly believe that a high anastomosis extending onto LHD in a well collateralized hilum obviates the need for any such vascular repair that can otherwise be tedious and time consuming. This view has also been shared by other authors [33]. Nine of the 20 patients in group C had prior operative intervention by the operating surgeon to correct the injury. Four of them had hepaticoduodenostomy. Published reports suggest that repair by the primary or the ‘injuring surgeon’ carries a poor outcome [5, 34]. Moreover, hepaticoduodenostomy almost always fails in the long run [35]. The only advantage the latter anastomosis was that the opening in the duodenum provided an access for obtaining cholangiogram (internal fistulogram). Such modality of cholangiogram has not been reported earlier. The repair was performed at least 6 weeks after the index cholecystectomy. The reasons for this waiting were similar to what has been stated for patients in group B who underwent delayed repair. The high success rate of bile duct repair in the present study can be attributed to the appropriate timing, meticulous technique and the tertiary care experience, a fact that has been highlighted by several authors [1, 18, 20, 35]. The only limitation in the present study, however, is the short follow-up period, i.e. 40 months. A follow-up of at least 5 years or more is desirable to make a final pronouncement of the results [36].
Indian J Surg
References 1. Lau WE, Lai ECH, Lau SHW (2010) Management of bile duct injury after laparoscopic cholecystectomy: a review. ANZ J Surg 80:75–81 2. Slater K, Strong RW, Wall DR, Lynch SV (2002) Iatrogenic bile duct injury: the scourge of laparoscopic cholecystectomy. ANZ J Surg 72:83–88 3. Sicklick JK, Camp MS, Lillemore KD, Melton GB, Yeo CJ, Campbell KA et al (2005) Surgical management of bile duct injuries sustained during laparoscopic cholecystectomy. Ann Surg 241:786–795 4. De Reuver PR, Rauws EA, Bruno MJ (2007) Survival in bile duct injury patients after laparoscopic cholecystectomy: a multidisciplinary approach of gastroenterologists, radiologists and surgeons. Surgery 142:1–9 5. Stewart L, Way LW (1995) Bile duct injuries during laparoscopic cholecystectomy. Factors that influence the results of treatment. Arch Surg 130:1123–1128 6. Perera MTPR, Silva MA, Hegab B, Muralidharan V, Bramhall SR, Mayer AD et al (2011) Specialist early and immediate repair of post-laparoscopic cholecystectomy bile duct injuries is associated with an improved long-term outcome. Ann Surg 253:553–560 7. Silva MA, Coldham C, Mayer AD, Bramhali SR, Buckels J, Mirza DF (2008) Specialist outreach service for on-table repair of iatrogenic bile duct injuries—a new kind of ‘travelling surgeon’. Ann R Coll Surg Engl 90:243–246 8. Strasberg SM, Hertl M, Soper NJ (1995) An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 180:101–125 9. McDonald MI, Farnell MB, Nagorney DM, IIstrup DM, Kutch JM (1995) Benign biliary stricture: repair and outcome with a contemporary approach. Surgery 118:582–591 10. Chaudhary A, Manisegran M, Chandra A (2001) How do bile duct injuries sustained during laparoscopic cholecystectomy differ from those during open cholecystectomy. J Laparoendosc Adv Surg Tech 11:187–191 11. Sikora SS, Kumar A, Dash NR, Sarkari A, Saxena R, Kapoor VK (2001) Laparoscopic bile duct injuries-spectrum at a tertiary care centre. J Laparoendosc Adv Surg Tech 11:63–68 12. Corvera CU, Jarnagin WR, Blumgart LH (2007) Biliary fistula. In: Blumgart LH (ed) Surgery of the liver, biliary tract and pancreas. Saunders Elsevier, Philadelphia, pp 655–684 13. Sandha GS, Bourke MJ, Haber GB, Kortan PP (2004) Endoscopic therapy for bile leak based on a new classification; results in 207 patients. Gastrointest Endosc 60:567–574 14. Eisenstein S, Greenstein AJ, Kim U, Divino CM (2008) Cystic duct stump leaks; after the learning curve. Arch Surg 143:1178–1183 15. Pinkas H, Brady PG (2008) Bile leaks after laparoscopic cholecystectomy: time to stent or time to drain. Hepatobiliary Pancreat Dis Int 7:628–632 16. Aksoz K, Unsal B, Yoruk G, Buyrac Z, Haciyanli M, Akpinar Z, Alper E (2009) Endoscopic sphincterotomy alone in the management of low-grade biliary leaks due to cholecystectomy. Dig Endosc 21:158–161 17. Lillemoe KD, Pitt HA, Cameron JL (1990) Postoperative biliary stricture. Surg Clin N Am 70:1355–1388
18. Sikora SS, Pottakkat B, Srikanth G, Kumar A, Saxena R, Kapoor VK (2006) Postcholecystectomy benign biliary strictures-long-term results. Dig Surg 23:304–312 19. Walsh RM, Henderson JM, Vogt DP (2007) Long-term oucome of biliary reconstruction for bile duct injuries from laparoscopic cholecystectomies. Surgery 142:450–456 20. Kapoor VK (2007) Bile duct injury repair: when? what? who? J Hepatobiliary Pancreat Surg 14:476–479 21. Schol FP, Go PM, Gouma DJ (1995) Outcome of 49 repairs of bile duct strictures after laparoscopic cholecystectomy. World J Surg 19:168–172 22. Schmidt SC, Lagrehr JM, Hintze RE, Neuhaus P (2005) Long-term results and risk factors influencing outcome of major bile duct injuries following cholecystectomy. Br J Surg 92:76–82 23. deReuver PR, Grossmann I, Busch OR, Obertop H, van Gulik TM, Gouma DJ (2007) Referral pattern and timing of repair are risk factors for complications after reconstructive surgery for bile duct injury. Ann Surg 245:763–770 24. Mercado MA (2007) Early versus late repair of bile duct injuries. Surg Endosc 20:1644–1647 25. Thomson BNJ, Parks RW, Madhavan KK, Wigmore SJ, Garden OJ (2006) Early specialist repair of biliary injury. Br J Surg 93:216–220 26. Wudel LJ Jr, Wright JK, Pinson CW, Herline A, Debelak J, Seidel S, Revis K, Chapman WC (2001) Bile duct injury following laparoscopic cholecystectomy: a cause of continuing concern. Am Surg 67:557–563 27. Bismuth H, Majno PE (2001) Biliary strictures: classification based on the principles of surgical treatment. World J Surg 25:1241–1244 28. de Reuver PR, Busch ORC, Rauws EA, Lameris JS, van Gulik TM, Gouma DJ (2007) Long-term results of a primary end-to-end anastomosis in preoperative detected bile duct injury. J Gastrointest Surg 11:296–302 29. Schmidt SC, Settmacher U, Langrehr JM, Neuhaus P (2004) Management and outcome of patients with combined bile duct and hepatic arterial injuries after laparoscopic cholecystectomy. Surgery 135:613–618 30. Gupta N, Solomon H, Fairchild R, Kaminski DL (1998) Management and outcome of patients with combined bile duct and hepatic artery injuries. Arch Surg 133:176–181 31. Koffron A, Ferrario M, Parsons W, Nemcek A, Saker M, Abecassis M (2001) Failed primary management of iatrogenic biliary injury: incidence and significance of concomitant hepatic arterial disruption. Surgery 130:722–731 32. Bachellier P, Nakano H, Weber J-C, Lemarque P, Oussoultzoglou E, Candau C et al (2001) Surgical repair after bile duct and vascular injuries during laparoscopic cholecystectomy: when and how? World J Surg 25:1335–1345 33. Tzovaras G, Dervenis C (2006) Vascular injuries in laparoscopic cholecystectomy: an underestimated problem. Dig Surg 23:370– 374 34. Flum DR, Cheadle A, Prela C, Dellinger EP, Chan L (2003) Bile duct injury during cholecystectomy and survival in medicare beneficiaries. JAMA 290:2168–2173 35. Connor S, Garden OJ (2006) Bile duct injury in the era of laparoscopic cholecystectomy. Br J Surg 93:158–167 36. Sikora S (2012) Management of post-cholecystectomy benign biliary strictures: review. Ind J Surg 74:22–28
