ClinicalRadiology (1992) 46, 333-336
Post-Laparoscopic Cholecystectomy Bilomas- Preliminary Experience A. P. B R A D Y ,
F. P. M c G R A T H ,
D . J. M O O T E *
a n d D . E. M A L O N E
Department of Radiology, MeMaster University Medical Centre, Hamilton, and *Department of Radiology, Victoria Hospital, London, Ontario, Canada The incidence of biloma formation after laparoscopic cholccystectomy (LC) is, in the early experience of many centres, significantly higher than after open cholecystectomy. We describe four cases representative of the possible spectrum of post-LC bile leaks and review the literature regarding the radiologic investigation and management of this complication. We suggest that post-LC bilomas should initially be drained percutaneously. Endoscopic cholangiography (ERC) should then be used to identify the source of the leak, but sphincterotomy and/or stent placement may be best reserved for those whose leaks do not resolve after 10 days of free drainage. Surgery is recommended only for major biliary injury and for leaks that remain unresolved after sphincterotomy and/or stenting. An algorithmic summary of this approach is presented. B r a d y , A , P . , M c G r a t h , F . P . , M o o t e , D . J . & M a l o n e , D . E . (1992). Clinical Radiology 46, 333 336. P o s t - L a p a r o s c o p i c Cholecystectomy Bilomas Preliminary Experience
Accepted for Publication 11 June 1992
L a p a r o s c o p i c c h o l e c y s t e c t o m y ( L C ) w a s first p e r f o r m e d in F r a n c e in 1987 [1,2], a n d h a s r a p i d l y b e c o m e a c c e p t e d surgical practice. The technique has many advantages over open surgery, but carries a significantly higher risk of p o s t - o p e r a t i v e m o r b i d i t y [3], w i t h s o m e c o m p l i c a t i o n s s u c h as b i l o m a f o r m a t i o n w h i c h a r e a m e n a b l e t o i n t e r ventional radiologic management. We report our prelim i n a r y e x p e r i e n c e in t h e i n v e s t i g a t i o n a n d m a n a g e m e n t of bile leaks following LC, and summarize this in an algorithmic approach. C A S E REPORTS Case 1. Six days after LC, a 48-year-old woman developed abdominal pain and swelling. Ultrasound (US) showed a large fluid collection in the right subphrenic and subhepatic spaces, extending to the right lower quadrant (RLQ). A drainage catheter (10.3F All-purpose drainage (APD); Meditech, Watertown, MA, USA) was placed percutaneously under US control in the dependent portion of the collection (RLQ) and 31 of bilious fluid drained over the next 48 h. Drainage then stopped and the surgeon removed the tube without further imaging. The collection rapidly reaccumulated, and repeat percutaneous RLQ drainage (12F Ring-McLean catheter, Cook Inc., Stouffville, Ontario, Canada) was performed, producing 1700 ml of bilious fluid. A HIDA scan showed abnormal isotope accumulation over the lower part of the right lobe of the liver, but did not identify the source of the leak. Endoscopic retrograde cholangiography (ERC) showed a cystic duct leak (Fig. 1) and sphincterotomy was performed. Four days later, catheter drainage was minimal but the patient complained of dyspnoea and upper abdominal pain. US and computed tomography (CT) showed large collections in the right upper quadrant (RUQ) and the left subphrenic area, and also demonstrated an intraperitoneal gallstone which had been too large to remove via the trocar puncture at surgery (Fig. 2). Despite drainage of the RUQ collection with a second 12 F Ring-McLean catheter, and a second ERC with failure to place a stent across the cystic duct orifice (because of the high junction of the cystic duct with the common hepatic duct), drainage continued, and a laparotomy was performed 26 days after LC. The cystic duct stump "&as ligated, loculations in the RUQ collection were broken down and drained and the residual gallstone was removed from the peritoneal cavity. Correspondence to: Dr. D. E. Malone, Radiology Department, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario L8N 3Z5, Canada.
Post-operatively, the patient remained febrile, and needed further percutaneous drainage of collections in the left upper quadrant and the subhepatic space from which Candidaalbicanswas cultured. Drainage continued for 8 weeks. The patient remains well 5 months after catheter removal, 8 months after her initial LC. Case 2. A 29-year-old woman developed RUQ pain 5 days after LC. US demonstrated a collection of mixed echogenicity in the gall-bladder (GB) fossa, and a right subhepatic and subphrenic fluid collection. A drainage catheter (8.3 F APD) was inserted into the GB fossa under CT control, as close as possible to the surgical clips. Five-hundred ml of bile, which grew Bacillus species on culture, drained over the next 24 h. Broad-spectrum antibiotics were given intravenously for 5 days. ERC, performed the day after drainage, showed leakage of contrast from a biliary radicle in the GB bed (Fig. 3). The surgeon commented that, in retrospect, at the time of surgery he had noted a small amount of bile leakage fom the hepatic surface during gall-bladder dissection but had not thought it significant. Sphincterotomy was not performed. The catheter tip was repositioned under fluoroscopy to lie within 1 cm of the leaking bile duct. Drainage volume gradually diminished over the next 3 days. Follow-up US showed complete resolution of the collection. The drain was removed 2 weeks after its initial insertion, and the patient remains well 6 months later. Case 3. A 54-year-old man presented 3 days after LC with increasing abdominal pain and pyrexia. Prior to his surgery, he had had an endoscopic sphincterotomy with removal of common bile duct stones. US demonstrated a large right subphrenic collection and several small calculi in the GB fossa (Fig. 4), spilled from the GB during removal at LC. A percutaneous drainage catheter (12F Ring-McLean) was inserted under US and fluoroscopic control via a low intercostal approach. A litre of bile was aspirated, and drainage gradually diminished over the following 5 days. The catheter was removed 11 days after insertion, and the patient remains well 7 months later. Case 4. A 54-year-old woman had LC performed at a district hospital, during which the surgeon commented that he had 'torn' the cystic duct. Post-operatively, large volumes of bile flowed from a Jackson-Pratt surgical drain (Bard, Mississauga, Ontario, Canada) placed intraoperatively. The patient was transferred to a tertiary-care centre the day after LC. Drainage of bile continued, but no localized intra-abdominal biloma developed. ERC demonstrated that a large portion of the GB neck, containing a calculus, and the cystic duct had not been removed at surgery. The leak was from the GB remnant (Fig. 5). Papillotomy was performed and a naso-biliary catheter inserted. Large volumes of bile continued to discharge through the Jackson-Pratt drain. The GB remnant was removed at a second session of laparoscopic surgery 5 days after ERC. No further bile drainage occurred, and the patient remains well 6 months later.
334
CLINICAL RADIOLOGY
Fig. 3 - Case 2. ERC: there is leakage of contrast from a biliary radicle in the GB bed (arrow), beside the surgical clips (arrowheads). The percutaneous drainage catheter lies just below the source o f the leak.
Fig. 1 - C a s e 1. E R C P shows contrast leaking from the cystic duct stump (arrow) at the site of a surgical clip (arrowhead).
Fig. 4 - Case 3. US o f the subhepatic space identifying small retained gallstones (arrows).
Fig. 2 - Case 1. C T demonstrates the retained intraperitoneal gallstone (arrowhead) and the extensive bilomas (arrows).
DISCUSSION LC is largely replacing open cholecystectomy and the non-operative means of GB stone managment in many centres [1,4]. Compared to open cholecystectomy, LC has advantages of short hospitalization and minimal invasiveness [2,5,6]. The surgical literature to date suggests that these advantages have, in many centres, been achieved at the expense of increased operative and post-operative morbidity. The incidence of bile duct injury following open cholecystectomy is 0.1-0.3%, while the incidence
following common bile duct (CBD) exploration rises to 1-2% [7,8]. The incidence of complications following LC has been reported as being up to three to five times that of open cholecystectomy, particularly during the early experience of many surgeons [3], with a reported incidence of bile duct injury of 0.5-1.4% [3,9-11]. Difficulty in identifying biliary anatomy is the predominant cause of this higher biliary injury rate [12]. The GB fossa should not contain any significant fluid collections after uncomplicated LC [13]. Any such fluid collection seen in the early post-operative period most likely represents either blood or bile. If the patient's haemoglobin level has not fallen post-operatively, the collection probably represents a biloma. The source of the leak may be the cystic duct stump (due to slippage of the lateral clip or ligature placed at surgery [14] (Case 1)), the
335
POST-LAPAROSCOPIC CHOLECYSTECTOMYBILOMAS
Post-LCcollection Percutaneousdrainage (cathetertip nearsurgicalclips) Bile obtained ERC
Majorbiliary injury
Cysticduct
G8 bed
t e a ~ i
Drainageperiists> 10days t / l / Surgery
Noleak
~
~
Sphincterotomy+/- stent Drainagep e ~ ~
~-'~-~
Drainageiesolves SinographyandUSshow res°luti°n °i c°llecti°n Removecatheter
Fig. 6 An algorithmic approach to management of post-LC bile leaks. Fig. 5 Case 4. ERC demonstrates the residual cystic duct, neck of GB containing a calculus (arrowhead), and leaking contrast (short arrows). The surgical clips are seen lateral to the gallbladder remnant (long arrow).
GB bed (following dissection of the GB from the liver surface [3,15] (Case 2)), or from operative biliary tract injury [3,10,11] as in Case 4, where the GB was incompletely removed. If bile does not continue to leak actively, the origin of the biloma may not be identifiable (Case 3). Leaking bile conforms to adjacent natural boundaries and, by virtue of the detergent and tissue-destroying properties of certain bile acids, elicits a low-grade inflammatory response. This results in a thin capsule or in adhesions involving the mesentery and omentum, isolating the collection. The inflammatory effect of bile acids may subsequently lead to biliary stricture formation (16]. Bile is an excellent culture medium; an undrained biloma is very liable to become purulent [17]. Early drainage is important in optimal management of bilomas. Bilomas are generally confined to the upper two thirds of the abdomen [16,18], but occasionally present in the RLQ (Case 1). Because of their tendency to 'wall-off', a drainage catheter with its tip in the dependent portion (RLQ) of a large biloma may not control a leak originating in the RUQ, as in Case 1. Multiple catheters may be required. It is suggested that the first catheter be placed with its tip near the presumed site of leakage, as otherwise continued leakage of bile worsens the situation. Positioning of the drain is less important in those with sealed-off bile collections [17,18] (Case 3). The most effective way to treat post-LC bilomas may be to assume that an active leak is present in all cases, and to position the drainage catheter tip as near as possible to the surgical clips marking the cystic duct ligation site. This will usually result in rapid resolution of drainage [17,18]. Had the first percutaneous catheter been placed near that leakage site in Case 1, much of the morbidity from continued leakage
of bile and its subsequent infection might have been a avoided. An algorithmic management strategy is outlined in Fig. 6. Once drainage has been established, the site and activity of the bile leak is investigated without delay. Cholescintigraphy has been recommended to identify the source of bile leaks after surgery [3,8,17,18], but the spatial resolution of the images may not permit exact localization of the source of the leak (Case 1); ERC is the method of choice for exact localization (Cases 1, 2, 4). Some post-operative biliary leaks have been successfully treated endoscopically [7,15]; these include some leaks occurring after LC [3,19]. ERCP alone has a complication rate of 2 7%, with a mortality rate of between 0.001% and 0.8% [20]. Acute pancreatitis is the commonest serious complication of ERCP, and is mainly related to overfilling of the pancreatic ductal system with contrast, and trauma to and oedema of the ampulla [21]. In ERC alone, as performed in our patients, without a pancreatogram, the incidence of pancreatitis should be lower. Sphincterotomy has a complication rate of 5-10%, the most serious complications being haemorrhage, pancreatitis, cholangitis and perforation of the duodenal wall. This complication rate is greater than that of LC [22]. The mortality rate of sphincterotomy is 0.5-1% [2325]. Therefore, it is suggested that a diagnostic ERC initially, followed by a second ERC and sphincterotomy when necessary (for continuous large-volume percutaneous drainage), is safer than proceeding to sphincterotomy in all patients at the initial ERC. Most leaks from the GB bed are likely to resolve with percutaneous drainage alone [7,16]. Leaks from sources other than the GB bed are more likely to require eventual sphincterotomy, with or without biliary stenting, to produce resolution, but a trial of percutaneous drainage alone is probably worthwhile before sphincterotomy and/or stent placement. If the diagnostic ERC demonstrates CBD stones, sphincterotomy should be performed at this time.
336
CLINICAL RADIOLOGY
After diagnostic E R C , catheter position should be adjusted, if necessary, to place the tip as n e a r as possible to the site of the leak. I f the catheter r e t u r n s do n o t d i m i n i s h after 10 days, s p h i n c t e r o t o m y a n d / o r stent p l a c e m e n t should be p e r f o r m e d to equalize the pressures in the d u o d e n u m a n d the biliary tree [7,16] (Cases 1, 4). I n s e r t i o n of a n endoprosthesis provides a c o n d u i t past the leak a n d may physically occlude the ductal defect [7,16]. This usually requires s p h i n c t e r o t o m y b u t stent insertion is increasingly being p e r f o r m e d w i t h o u t s p h i n c t e r o t o m y [7,26]. Early resolution of the leak in Case 3 m a y have been c o n t r i b u t e d to by remote prior sphincterotomy. P o s t - L C bile leaks due to m a j o r surgical t r a u m a (Case 4) are best treated surgically unless the p a t i e n t is unfit for operation. Surgical repair should otherwise be reserved for those leaks n o t resolving after endoscopic sphincterotomy, as in Case 1 [3,11,12,27]. T w o of our four cases occurred d u r i n g the first 6 m o n t h s of the relevant surgeons' experience with LC. As surgical experience with LC increases, it is h o p e d that the incidence of complications will a p p r o a c h that of open cholecystectomy. REFERENCES
1 Brandon JC, Velez MA, Teplick SK, Mueller PR, Rattner DW, Broadwater JR Jr et al. Laparoscopic cholecystectomy: Evolution, early results, and impact on nonsurgical gallstone therapies. American Journal of Roentgenology 1991;157:235-239. 2 Cushieri A, Dubois F, Mouiel J, Mouret P, Becker H, BuessG et al. The European experiencewith laparoscopic cholecystectomy.American Journal of Surgery 1991;161:385 387. 3 Peters JH, Gibbons GD, Innes JT, Nichols KE, Front ME, Roby SR et al. Complications of laparoscopic cholecystectomy. Surgery 1991;110:769 778. 4 Boyce HW Jr, Lightdale CJ. The rush to laparoscopic cholecystectomy: a word of caution. Gastrointestinal Endoscopy 1991;37:92 94. 5 Grace PA, Quereshi A, Coleman J, Keane R, McEntee G, Broe Pet al. Reduced postoperative hospitalization after laparoscopic cholecystectomy. British Journal of Surgery 1991;78:160-162. 6 Grace P, Quereshi A, Darzi A, McEntee G, Leahy A, Osbourne H et al. Laparoscopic cholecystectomy: a hundred consecutive cases. Irish Medical Journal 1991;84:12-14. 7 Binmoeller KF, Katon RM, Shneidman R. Endoscopic management of postoperative biliary leaks: review of 77 cases and report of two cases with biloma formation. American Journal of Gastroenterology 1991;86:227-231. 8 Czerniak A, Thompson JN, Soreide O, Benjamin IS, Blumgart LH. The management of fistulas of the biliary tract after injury to the bile duct during cholecystectomy. Surgery, Gynecology and Obstetrics
1988;167:33 38. 9 Gadacz-TR, Talamini MA. Traditional versus laparoscopic cholecystectomy. American Journal of Surgery 1991;16l:336 338, 10 Salky BA, Bauer JJ, Kreel I, Gelernt IM, Gorfine SR. Laparoscopic cholecystectomy: an initial report. Gastrointestinal Endoscopy 1991;37:1-4. 11 Southern Surgeons Club. A prospective analysis of 1518 laparoscopic cholecystectomies. New England Journal of Medicine 1991;324:1073-1078. 12 Nottle PD, Wale RJ, Johnson WR. Percutaneous cholecystectomy: the first fifty. Australian and New Zealand Journal o f Surgery 1991;61:254-260. 13 Quinn SF, Sangster W, Standage B, Schuman E, Gross G. Biliary complications related to laparoscopic cholecystectomies: clinical presentation and radiological diagnosis and management. Radiology 1991;181(P):291 (abstr.). 14 Nathanson LK, Shimi S, Cushieri A. Laparoscopic cholecystectomy: the Dundee technique. British Journal o f Surgery 1991;78:155 159. 15 Cotton PB, Baillie J, Pappas TN, Meyers WS. Laparoscopic cholecysteetomy and the biliary endoscopist. Gastrointestinal Endoscopy 1991;37:94 96, l 6 Ligoury C, Vitale GC, LefebreJF, BonnelD, Cornud F. Endoscopic treatment of postoperative biliary fistulae. Surgery 1991;110:779784. 17 Mueller PR, Ferrucci JT Jr, Simeone JF, Cronan JJ, Wittenberg J, Neff CC e t al. Detection and drainage of bilomas: special considerations. American Journal of Roentgenology 1983;140:715 720. 18 Vazquez JL, Thorsen MK, Dodds WJ, Quiroz FA, Martinez ML, Lawson TL et al. Evaluation and treatment of intraabdominal bilomas. American Journal of Roentgenology 1985;144:933 938. 19 Walker AT, Shapiro AW, Brooks DC, Braver JM, Tumeh SS. Bile duct disruption and biloma after laparoscopic cholecystectomy: imaging evaluation. American Journal of Roentgenology 1992;158:785 789. 20 Ferguson DR, Sivak MV. Indications, contraindications and complications of ERCP. In: Sivak MV, ed. Gastroenterologic endoscopy. Philadelphia: WB Saunders, 1987:581-599. 21 Huibregtse K, Tytgat GNJ. Endoscopic retrograde cholangiopancreatography. In': Lygiadakis NJ & Tytgat GNJ, eds. Hepatobiliary and pancreatic malignancies. Stuttgart: Thieme, 1989:100 114. 22 McLoughlin RF, Gibney RG, Mealy K, Hyland J. Radiological investigationin laparoscopic compared with conventional cholecystectomy - an early assessment. Clinical Radiology 1992;45:267-270. 23 Classen M. Endoscopic papillotomy. In: Sivak MV, ed. Gastroenterologic endoscopy. Philadelphia: WB Saunders, 1987:631-652. 24 Safrany L, Stenzel M. Endoscopic papillotomy and stone removal. In: Bennett JR & Hunt RH, eds. Therapeutic endoscopy and radiology of the gut. London: Chapman & Hall, 1990:222-234. 25 SchillerKFR, Cockel R, Hunt RH. A colour atlas of gastrointestinal endoscopy. London: Chapman & Hall, 1986:170. 26 Kozarek RA, Traverso LW. Endoscopic stent placement for cystic duct leak after laparoscopic cholecystectomy. Gastrointestinal Endoscopy 1991;37:71 73. 27 Colovic R, Perisec-Savic M. Retroperitoneal biloma secondary to operative common bile duct injury. HPB Surgery 1991;3:193 197.
Post-Laparoscopic Cholecystectomy Bilomas- Preliminary Experience A. P. B R A D Y ,
F. P. M c G R A T H ,
D . J. M O O T E *
a n d D . E. M A L O N E
Department of Radiology, MeMaster University Medical Centre, Hamilton, and *Department of Radiology, Victoria Hospital, London, Ontario, Canada The incidence of biloma formation after laparoscopic cholccystectomy (LC) is, in the early experience of many centres, significantly higher than after open cholecystectomy. We describe four cases representative of the possible spectrum of post-LC bile leaks and review the literature regarding the radiologic investigation and management of this complication. We suggest that post-LC bilomas should initially be drained percutaneously. Endoscopic cholangiography (ERC) should then be used to identify the source of the leak, but sphincterotomy and/or stent placement may be best reserved for those whose leaks do not resolve after 10 days of free drainage. Surgery is recommended only for major biliary injury and for leaks that remain unresolved after sphincterotomy and/or stenting. An algorithmic summary of this approach is presented. B r a d y , A , P . , M c G r a t h , F . P . , M o o t e , D . J . & M a l o n e , D . E . (1992). Clinical Radiology 46, 333 336. P o s t - L a p a r o s c o p i c Cholecystectomy Bilomas Preliminary Experience
Accepted for Publication 11 June 1992
L a p a r o s c o p i c c h o l e c y s t e c t o m y ( L C ) w a s first p e r f o r m e d in F r a n c e in 1987 [1,2], a n d h a s r a p i d l y b e c o m e a c c e p t e d surgical practice. The technique has many advantages over open surgery, but carries a significantly higher risk of p o s t - o p e r a t i v e m o r b i d i t y [3], w i t h s o m e c o m p l i c a t i o n s s u c h as b i l o m a f o r m a t i o n w h i c h a r e a m e n a b l e t o i n t e r ventional radiologic management. We report our prelim i n a r y e x p e r i e n c e in t h e i n v e s t i g a t i o n a n d m a n a g e m e n t of bile leaks following LC, and summarize this in an algorithmic approach. C A S E REPORTS Case 1. Six days after LC, a 48-year-old woman developed abdominal pain and swelling. Ultrasound (US) showed a large fluid collection in the right subphrenic and subhepatic spaces, extending to the right lower quadrant (RLQ). A drainage catheter (10.3F All-purpose drainage (APD); Meditech, Watertown, MA, USA) was placed percutaneously under US control in the dependent portion of the collection (RLQ) and 31 of bilious fluid drained over the next 48 h. Drainage then stopped and the surgeon removed the tube without further imaging. The collection rapidly reaccumulated, and repeat percutaneous RLQ drainage (12F Ring-McLean catheter, Cook Inc., Stouffville, Ontario, Canada) was performed, producing 1700 ml of bilious fluid. A HIDA scan showed abnormal isotope accumulation over the lower part of the right lobe of the liver, but did not identify the source of the leak. Endoscopic retrograde cholangiography (ERC) showed a cystic duct leak (Fig. 1) and sphincterotomy was performed. Four days later, catheter drainage was minimal but the patient complained of dyspnoea and upper abdominal pain. US and computed tomography (CT) showed large collections in the right upper quadrant (RUQ) and the left subphrenic area, and also demonstrated an intraperitoneal gallstone which had been too large to remove via the trocar puncture at surgery (Fig. 2). Despite drainage of the RUQ collection with a second 12 F Ring-McLean catheter, and a second ERC with failure to place a stent across the cystic duct orifice (because of the high junction of the cystic duct with the common hepatic duct), drainage continued, and a laparotomy was performed 26 days after LC. The cystic duct stump "&as ligated, loculations in the RUQ collection were broken down and drained and the residual gallstone was removed from the peritoneal cavity. Correspondence to: Dr. D. E. Malone, Radiology Department, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario L8N 3Z5, Canada.
Post-operatively, the patient remained febrile, and needed further percutaneous drainage of collections in the left upper quadrant and the subhepatic space from which Candidaalbicanswas cultured. Drainage continued for 8 weeks. The patient remains well 5 months after catheter removal, 8 months after her initial LC. Case 2. A 29-year-old woman developed RUQ pain 5 days after LC. US demonstrated a collection of mixed echogenicity in the gall-bladder (GB) fossa, and a right subhepatic and subphrenic fluid collection. A drainage catheter (8.3 F APD) was inserted into the GB fossa under CT control, as close as possible to the surgical clips. Five-hundred ml of bile, which grew Bacillus species on culture, drained over the next 24 h. Broad-spectrum antibiotics were given intravenously for 5 days. ERC, performed the day after drainage, showed leakage of contrast from a biliary radicle in the GB bed (Fig. 3). The surgeon commented that, in retrospect, at the time of surgery he had noted a small amount of bile leakage fom the hepatic surface during gall-bladder dissection but had not thought it significant. Sphincterotomy was not performed. The catheter tip was repositioned under fluoroscopy to lie within 1 cm of the leaking bile duct. Drainage volume gradually diminished over the next 3 days. Follow-up US showed complete resolution of the collection. The drain was removed 2 weeks after its initial insertion, and the patient remains well 6 months later. Case 3. A 54-year-old man presented 3 days after LC with increasing abdominal pain and pyrexia. Prior to his surgery, he had had an endoscopic sphincterotomy with removal of common bile duct stones. US demonstrated a large right subphrenic collection and several small calculi in the GB fossa (Fig. 4), spilled from the GB during removal at LC. A percutaneous drainage catheter (12F Ring-McLean) was inserted under US and fluoroscopic control via a low intercostal approach. A litre of bile was aspirated, and drainage gradually diminished over the following 5 days. The catheter was removed 11 days after insertion, and the patient remains well 7 months later. Case 4. A 54-year-old woman had LC performed at a district hospital, during which the surgeon commented that he had 'torn' the cystic duct. Post-operatively, large volumes of bile flowed from a Jackson-Pratt surgical drain (Bard, Mississauga, Ontario, Canada) placed intraoperatively. The patient was transferred to a tertiary-care centre the day after LC. Drainage of bile continued, but no localized intra-abdominal biloma developed. ERC demonstrated that a large portion of the GB neck, containing a calculus, and the cystic duct had not been removed at surgery. The leak was from the GB remnant (Fig. 5). Papillotomy was performed and a naso-biliary catheter inserted. Large volumes of bile continued to discharge through the Jackson-Pratt drain. The GB remnant was removed at a second session of laparoscopic surgery 5 days after ERC. No further bile drainage occurred, and the patient remains well 6 months later.
334
CLINICAL RADIOLOGY
Fig. 3 - Case 2. ERC: there is leakage of contrast from a biliary radicle in the GB bed (arrow), beside the surgical clips (arrowheads). The percutaneous drainage catheter lies just below the source o f the leak.
Fig. 1 - C a s e 1. E R C P shows contrast leaking from the cystic duct stump (arrow) at the site of a surgical clip (arrowhead).
Fig. 4 - Case 3. US o f the subhepatic space identifying small retained gallstones (arrows).
Fig. 2 - Case 1. C T demonstrates the retained intraperitoneal gallstone (arrowhead) and the extensive bilomas (arrows).
DISCUSSION LC is largely replacing open cholecystectomy and the non-operative means of GB stone managment in many centres [1,4]. Compared to open cholecystectomy, LC has advantages of short hospitalization and minimal invasiveness [2,5,6]. The surgical literature to date suggests that these advantages have, in many centres, been achieved at the expense of increased operative and post-operative morbidity. The incidence of bile duct injury following open cholecystectomy is 0.1-0.3%, while the incidence
following common bile duct (CBD) exploration rises to 1-2% [7,8]. The incidence of complications following LC has been reported as being up to three to five times that of open cholecystectomy, particularly during the early experience of many surgeons [3], with a reported incidence of bile duct injury of 0.5-1.4% [3,9-11]. Difficulty in identifying biliary anatomy is the predominant cause of this higher biliary injury rate [12]. The GB fossa should not contain any significant fluid collections after uncomplicated LC [13]. Any such fluid collection seen in the early post-operative period most likely represents either blood or bile. If the patient's haemoglobin level has not fallen post-operatively, the collection probably represents a biloma. The source of the leak may be the cystic duct stump (due to slippage of the lateral clip or ligature placed at surgery [14] (Case 1)), the
335
POST-LAPAROSCOPIC CHOLECYSTECTOMYBILOMAS
Post-LCcollection Percutaneousdrainage (cathetertip nearsurgicalclips) Bile obtained ERC
Majorbiliary injury
Cysticduct
G8 bed
t e a ~ i
Drainageperiists> 10days t / l / Surgery
Noleak
~
~
Sphincterotomy+/- stent Drainagep e ~ ~
~-'~-~
Drainageiesolves SinographyandUSshow res°luti°n °i c°llecti°n Removecatheter
Fig. 6 An algorithmic approach to management of post-LC bile leaks. Fig. 5 Case 4. ERC demonstrates the residual cystic duct, neck of GB containing a calculus (arrowhead), and leaking contrast (short arrows). The surgical clips are seen lateral to the gallbladder remnant (long arrow).
GB bed (following dissection of the GB from the liver surface [3,15] (Case 2)), or from operative biliary tract injury [3,10,11] as in Case 4, where the GB was incompletely removed. If bile does not continue to leak actively, the origin of the biloma may not be identifiable (Case 3). Leaking bile conforms to adjacent natural boundaries and, by virtue of the detergent and tissue-destroying properties of certain bile acids, elicits a low-grade inflammatory response. This results in a thin capsule or in adhesions involving the mesentery and omentum, isolating the collection. The inflammatory effect of bile acids may subsequently lead to biliary stricture formation (16]. Bile is an excellent culture medium; an undrained biloma is very liable to become purulent [17]. Early drainage is important in optimal management of bilomas. Bilomas are generally confined to the upper two thirds of the abdomen [16,18], but occasionally present in the RLQ (Case 1). Because of their tendency to 'wall-off', a drainage catheter with its tip in the dependent portion (RLQ) of a large biloma may not control a leak originating in the RUQ, as in Case 1. Multiple catheters may be required. It is suggested that the first catheter be placed with its tip near the presumed site of leakage, as otherwise continued leakage of bile worsens the situation. Positioning of the drain is less important in those with sealed-off bile collections [17,18] (Case 3). The most effective way to treat post-LC bilomas may be to assume that an active leak is present in all cases, and to position the drainage catheter tip as near as possible to the surgical clips marking the cystic duct ligation site. This will usually result in rapid resolution of drainage [17,18]. Had the first percutaneous catheter been placed near that leakage site in Case 1, much of the morbidity from continued leakage
of bile and its subsequent infection might have been a avoided. An algorithmic management strategy is outlined in Fig. 6. Once drainage has been established, the site and activity of the bile leak is investigated without delay. Cholescintigraphy has been recommended to identify the source of bile leaks after surgery [3,8,17,18], but the spatial resolution of the images may not permit exact localization of the source of the leak (Case 1); ERC is the method of choice for exact localization (Cases 1, 2, 4). Some post-operative biliary leaks have been successfully treated endoscopically [7,15]; these include some leaks occurring after LC [3,19]. ERCP alone has a complication rate of 2 7%, with a mortality rate of between 0.001% and 0.8% [20]. Acute pancreatitis is the commonest serious complication of ERCP, and is mainly related to overfilling of the pancreatic ductal system with contrast, and trauma to and oedema of the ampulla [21]. In ERC alone, as performed in our patients, without a pancreatogram, the incidence of pancreatitis should be lower. Sphincterotomy has a complication rate of 5-10%, the most serious complications being haemorrhage, pancreatitis, cholangitis and perforation of the duodenal wall. This complication rate is greater than that of LC [22]. The mortality rate of sphincterotomy is 0.5-1% [2325]. Therefore, it is suggested that a diagnostic ERC initially, followed by a second ERC and sphincterotomy when necessary (for continuous large-volume percutaneous drainage), is safer than proceeding to sphincterotomy in all patients at the initial ERC. Most leaks from the GB bed are likely to resolve with percutaneous drainage alone [7,16]. Leaks from sources other than the GB bed are more likely to require eventual sphincterotomy, with or without biliary stenting, to produce resolution, but a trial of percutaneous drainage alone is probably worthwhile before sphincterotomy and/or stent placement. If the diagnostic ERC demonstrates CBD stones, sphincterotomy should be performed at this time.
336
CLINICAL RADIOLOGY
After diagnostic E R C , catheter position should be adjusted, if necessary, to place the tip as n e a r as possible to the site of the leak. I f the catheter r e t u r n s do n o t d i m i n i s h after 10 days, s p h i n c t e r o t o m y a n d / o r stent p l a c e m e n t should be p e r f o r m e d to equalize the pressures in the d u o d e n u m a n d the biliary tree [7,16] (Cases 1, 4). I n s e r t i o n of a n endoprosthesis provides a c o n d u i t past the leak a n d may physically occlude the ductal defect [7,16]. This usually requires s p h i n c t e r o t o m y b u t stent insertion is increasingly being p e r f o r m e d w i t h o u t s p h i n c t e r o t o m y [7,26]. Early resolution of the leak in Case 3 m a y have been c o n t r i b u t e d to by remote prior sphincterotomy. P o s t - L C bile leaks due to m a j o r surgical t r a u m a (Case 4) are best treated surgically unless the p a t i e n t is unfit for operation. Surgical repair should otherwise be reserved for those leaks n o t resolving after endoscopic sphincterotomy, as in Case 1 [3,11,12,27]. T w o of our four cases occurred d u r i n g the first 6 m o n t h s of the relevant surgeons' experience with LC. As surgical experience with LC increases, it is h o p e d that the incidence of complications will a p p r o a c h that of open cholecystectomy. REFERENCES
1 Brandon JC, Velez MA, Teplick SK, Mueller PR, Rattner DW, Broadwater JR Jr et al. Laparoscopic cholecystectomy: Evolution, early results, and impact on nonsurgical gallstone therapies. American Journal of Roentgenology 1991;157:235-239. 2 Cushieri A, Dubois F, Mouiel J, Mouret P, Becker H, BuessG et al. The European experiencewith laparoscopic cholecystectomy.American Journal of Surgery 1991;161:385 387. 3 Peters JH, Gibbons GD, Innes JT, Nichols KE, Front ME, Roby SR et al. Complications of laparoscopic cholecystectomy. Surgery 1991;110:769 778. 4 Boyce HW Jr, Lightdale CJ. The rush to laparoscopic cholecystectomy: a word of caution. Gastrointestinal Endoscopy 1991;37:92 94. 5 Grace PA, Quereshi A, Coleman J, Keane R, McEntee G, Broe Pet al. Reduced postoperative hospitalization after laparoscopic cholecystectomy. British Journal of Surgery 1991;78:160-162. 6 Grace P, Quereshi A, Darzi A, McEntee G, Leahy A, Osbourne H et al. Laparoscopic cholecystectomy: a hundred consecutive cases. Irish Medical Journal 1991;84:12-14. 7 Binmoeller KF, Katon RM, Shneidman R. Endoscopic management of postoperative biliary leaks: review of 77 cases and report of two cases with biloma formation. American Journal of Gastroenterology 1991;86:227-231. 8 Czerniak A, Thompson JN, Soreide O, Benjamin IS, Blumgart LH. The management of fistulas of the biliary tract after injury to the bile duct during cholecystectomy. Surgery, Gynecology and Obstetrics
1988;167:33 38. 9 Gadacz-TR, Talamini MA. Traditional versus laparoscopic cholecystectomy. American Journal of Surgery 1991;16l:336 338, 10 Salky BA, Bauer JJ, Kreel I, Gelernt IM, Gorfine SR. Laparoscopic cholecystectomy: an initial report. Gastrointestinal Endoscopy 1991;37:1-4. 11 Southern Surgeons Club. A prospective analysis of 1518 laparoscopic cholecystectomies. New England Journal of Medicine 1991;324:1073-1078. 12 Nottle PD, Wale RJ, Johnson WR. Percutaneous cholecystectomy: the first fifty. Australian and New Zealand Journal o f Surgery 1991;61:254-260. 13 Quinn SF, Sangster W, Standage B, Schuman E, Gross G. Biliary complications related to laparoscopic cholecystectomies: clinical presentation and radiological diagnosis and management. Radiology 1991;181(P):291 (abstr.). 14 Nathanson LK, Shimi S, Cushieri A. Laparoscopic cholecystectomy: the Dundee technique. British Journal o f Surgery 1991;78:155 159. 15 Cotton PB, Baillie J, Pappas TN, Meyers WS. Laparoscopic cholecysteetomy and the biliary endoscopist. Gastrointestinal Endoscopy 1991;37:94 96, l 6 Ligoury C, Vitale GC, LefebreJF, BonnelD, Cornud F. Endoscopic treatment of postoperative biliary fistulae. Surgery 1991;110:779784. 17 Mueller PR, Ferrucci JT Jr, Simeone JF, Cronan JJ, Wittenberg J, Neff CC e t al. Detection and drainage of bilomas: special considerations. American Journal of Roentgenology 1983;140:715 720. 18 Vazquez JL, Thorsen MK, Dodds WJ, Quiroz FA, Martinez ML, Lawson TL et al. Evaluation and treatment of intraabdominal bilomas. American Journal of Roentgenology 1985;144:933 938. 19 Walker AT, Shapiro AW, Brooks DC, Braver JM, Tumeh SS. Bile duct disruption and biloma after laparoscopic cholecystectomy: imaging evaluation. American Journal of Roentgenology 1992;158:785 789. 20 Ferguson DR, Sivak MV. Indications, contraindications and complications of ERCP. In: Sivak MV, ed. Gastroenterologic endoscopy. Philadelphia: WB Saunders, 1987:581-599. 21 Huibregtse K, Tytgat GNJ. Endoscopic retrograde cholangiopancreatography. In': Lygiadakis NJ & Tytgat GNJ, eds. Hepatobiliary and pancreatic malignancies. Stuttgart: Thieme, 1989:100 114. 22 McLoughlin RF, Gibney RG, Mealy K, Hyland J. Radiological investigationin laparoscopic compared with conventional cholecystectomy - an early assessment. Clinical Radiology 1992;45:267-270. 23 Classen M. Endoscopic papillotomy. In: Sivak MV, ed. Gastroenterologic endoscopy. Philadelphia: WB Saunders, 1987:631-652. 24 Safrany L, Stenzel M. Endoscopic papillotomy and stone removal. In: Bennett JR & Hunt RH, eds. Therapeutic endoscopy and radiology of the gut. London: Chapman & Hall, 1990:222-234. 25 SchillerKFR, Cockel R, Hunt RH. A colour atlas of gastrointestinal endoscopy. London: Chapman & Hall, 1986:170. 26 Kozarek RA, Traverso LW. Endoscopic stent placement for cystic duct leak after laparoscopic cholecystectomy. Gastrointestinal Endoscopy 1991;37:71 73. 27 Colovic R, Perisec-Savic M. Retroperitoneal biloma secondary to operative common bile duct injury. HPB Surgery 1991;3:193 197.
