Pediatr Surg Int (2011) 27:829832 DOI 10.

1007/s00383-011-2877-5

ORIGINAL ARTICLE

Is it necessary to ligate distal common bile duct stumps after excising choledochal cysts?
Mei Diao Long Li Wei Cheng

Accepted: 23 February 2011 / Published online: 24 March 2011 Springer-Verlag 2011

Abstract Purpose After excision of choledochal cysts (CDC), the distal common bile duct (CBD) stumps are conventionally ligated. Yet, the distal common bile duct stump mobilization and ligation carries certain risk of pancreatic duct injury. The current study investigates the feasibility of selectively leaving distal stump unligated in CDC children with stenotic distal CBD. Methods Between 2001 and 2010, 270 CDC children successfully underwent laparoscopic cyst excision and Roux-en-Y hepatojejunostomy and were recruited into this study. The distal stump management was based on its radiological morphology. The patients were divided into two groups: (1) non-ligation group (n = 207), where the distal stump was stenotic and was left unligated; (2) ligation group (n = 63), where the distal stump was not stenotic and was ligated. The pancreatic juice leakage rates were compared. Results The median follow-up periods in the non-ligation and ligation group were 36 and 33 months, respectively. There was no pancreatic juice leakage in either group.

Conclusion Not ligating distal stump is a feasible approach for managing CDC with stenotic distal CBD. It simplies the operative procedure and may minimize pancreatic duct injury. Keywords Choledochal cyst Distal common bile duct Pancreaticobiliary malunion Pancreatic duct injury Pancreatic juice leakage

Introduction Conventionally, the distal common bile duct (CBD) stumps are ligated or sutured after excision of choledochal cysts (CDC) [17]. However, in CDC children with stenotic CBD, distal stump mobilization carries certain risk. Excessive dissection close to pancreas and ligation of friable distal stump may result in pancreatic duct injury and consequent pancreatic juice leakage. Previous reports in China showed that 26% of CDC operations were complicated with pancreatic duct damage [810]. In centers where laparoscopic excision of CDC is adopted, dissection of the distal stump is technically demanding. The aim of the current study is to investigate the feasibility of selectively leaving distal stump unligated in those CDC children with stenotic distal CBD. This practice started accidently during a difcult identication and dissection of distal stump in some patients with very stenotic distal CBD, even though careful explorations were performed by experienced surgeons. Surprisingly, no pancreatic juice leakage occurred. We thought it was safer to leave the stump alone rather than proceeding with dissection into the pancreatic tissue and decided to start a new practice. We herewith reviewed the results of this practice.

M. Diao L. Li (&) Department of Pediatric Surgery, Capital Institute of Pediatrics, Beijing, Peoples Republic of China e-mail: lilong22@hotmail.com W. Cheng (&) Department of Paediatric Surgery, Monash Childrens, Southern Health, Department of Paediatrics and Department of Surgery, Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, Victoria, Australia e-mail: wei.cheng@monash.edu

123

830

Pediatr Surg Int (2011) 27:829832

Methods We carried out a retrospective study of 270 CDC patients (mean age 3.72 years) who successfully underwent laparoscopic choledochal cyst excision and Roux-en-Y hepatojejunostomy between October 2001 and October 2010. Ethics approval from the Ethics Committee of Capital Institute of Pediatrics was obtained. Written informed consents were obtained from the parents of CDC patients prior to the study. Pre-operative ultrasonography, computed tomography, magnetic resonance cholangiopancreatography and intraoperative cholangiography were performed. A 5 mm 30o laparoscope was inserted via an umbilical port after induction of general anesthesia. The second and third ports were placed below the costal margin bilaterally under videoscopic guidance. An additional port was introduced in the right lower quadrant. Intraoperative cholangiograms were carried out to establish the anatomy of pancreaticobiliary conjunction and the sites of the possible calculi. Cholecystectomy and cyst mobilization were performed as previously described [11, 12]. Distal dissection was continued to a level proximal to the pancreaticobiliary conjunction to minimize injury of the pancreatic duct [1]. In patients with non-stenotic distal CBD, the common channels were examined and irrigated using an 8F neonatal urethroscope or 8F urethral catheter. Protein plugs clearance was conrmed by either direct vision (via a urethroscope) or additional cholangiogram [13]. The distal stump ligation was performed selectively accordingly to the radiological morphology of the distal CBD: (1) non-ligation group: in patients with stenotic distal CBD (n = 207), which was dened as the diameter of distal CBD on cholangiogram being less than 2 mm or no distal contrast ow, the CBD stumps were left unligated; (2) ligation group: in patient with non-stenotic distal CBD (n = 63), after irrigation of the common channels, the distal stumps were ligated with a 4/0 absorbable suture to prevent post-operative pancreatic juice leakage. Extra-corporal jejunal Roux loop [14] and laparoscopic end-to-side hepaticojejunostomy were then carried out as previously described [11, 12]. A subhepatic drain was placed at the end of the procedure. All patients were followed-up 1, 3, 6, 12 months postoperatively and 6 monthly thereafter. Physical examination, abdominal ultrasonographic studies, and laboratory tests were carried out at each visit. The operative time and blood loss, post-operative hospital stay, drainage duration, post-operative complications, and peri-operative ultrasonographic ndings and laboratory results were assessed.

Statistical analysis Data were collected and analyzed using SPSS (Chicago, IL) 13.0 software. Data were presented as mean SD. Students t tests were used to compare the mean maximal CDC diameters and mean minimal diameters of distal CBD between the non-ligation and ligation group. v2 tests were applied to compare the morbidities of ductal inammation, adhesions between cysts and surrounding tissues, associated pancreatic duct anomalies, and post-operative complications between the two groups. Paired t tests were used to compare the pre-operative and post-operative laboratory values. P \ 0.05 was considered statistically signicant.

Results The mean maximal CDC diameter of the non-ligation group was 5.71 3.12 cm (2.521 cm), which was signicantly greater than 1.51 0.53 cm (0.82.5 cm) of the ligation group (P \ 0.001). The mean minimal diameter of distal CBD of the non-ligation group was 0.18 0.12 cm, which was signicantly narrower than 0.35 0.17 cm of the ligation group (P \ 0.001). Inammations, evidenced by duct wall hyperemia and edema as well as biopsy nding of copious chronic inammatory cells inltration in the cyst walls, were found in 94.7% (196/207) patients from the non-ligation group, which was signicantly higher than 84.9% (53/63) from the ligation group (P \ 0.01). Dense adhesions between cysts and surrounding tissues were detected more frequently in the non-ligation group (163/207 (78.7%) vs. 31/63 (49.2%), P \ 0.001). Pancreatic duct anomalies were found in 78.7% (163/207) patients of the non-ligation group, including pancreaticobiliary malunion (156/163 patients, 95.7%), complete pancreas divisum (2/163 patients, 1.2%), and incomplete pancreas divisum [15] (6/163 patients, 3.6%). In contrast, all patients in the ligation group had associated pancreatic duct anomalies, including pancreaticobiliary malunion (58/63 patients, 92.1%), complete pancreas divisum (1/63 patient, 1.6%), and incomplete pancreas divisum [15] (3/63 patients, 4.7%). There was no statistically signicant difference between the two groups in the frequencies of pancreaticobiliary malunion and complete/incomplete pancreas divisum (P = 0.18, 0.83, and 0.71, respectively). All patients recovered uneventfully. The median followup periods in the non-ligation and ligation group were 36 and 33 months, respectively. Neither mortality nor complications of pancreatic duct injury, pancreatic juice leakage, pancreatitis, or stone formation was observed in either group. Liver function tests and serum amylase levels normalized post-operatively (P \ 0.001).

123

Pediatr Surg Int (2011) 27:829832

831

Discussion The distal CBD mobilization and ligation are routinely performed in surgical corrections of CDC. However, open CDC excisions are associated with 26% rate of pancreatic juice leakage in China [810]. The pancreatic juice leakage rate in laparoscopic excision of CDC is unknown, but it is reasonable to assume a similar if not higher rate with the laparoscopic approach. In addition to faulty CBD stump ligation, excessive dissection of distal stump and injury of pancreaticobiliary junction may also contribute to pancreatic juice leakage. A previous study reviewing 112 CDC children (mean age 5.2 years) who underwent surgery between 1980 and 1999 reported that 4% of the patients (n = 5) experienced pancreatic duct injury during dissection of intra-pancreatic portion of CBD and ligation of distal stumps [8]. The common features to these patients included: (1) all of CDC were cystic type with stenotic distal CBD; (2) the distal stenotic segments were extremely short (0.10.5 cm in length); (3) most of the patients had severe CBD dilatations (cyst size in 4 out of 5 patients: 1016 cm in diameter and 610 cm in length; cyst size of the another patient: 3 cm in diameter and 8 cm in length); (4) all patients had considerable ductal inammation and edema; (5) all patients had dense adhesions between cysts and surrounding tissues and (6) all patients had pancreatic duct anomalies, including pancreaticobiliary malunion (n = 3), complete pancreas divisum (n = 1), and incomplete pancreas divisum (n = 1). The accessory pancreatic duct, which is often situated close to the distal stenotic segment, was believed to be transected during dissection. This experience provides the evidence of our current practice of leaving the stenotic distal CBD unligated. In the current study, biliary dilatation in the non-ligation group was more pronounced than that in the ligation group, whereas the distal CBD in the former group was signicantly narrower than that in the latter group. This suggests that the non-ligation group of patients mainly suffered from cystic type of CDC. We speculate that severe biliary dilatation weakens duct wall. Furthermore, inammation compounded the ductal wall damage with mural hyperemia and edema. In this situation, distal stump in patients with stenotic distal CBD is often short, friable and more likely to bleed. In addition, severe adhesions and associated pancreatic duct anomalies make the distal stump mobilization much more demanding and it may increase the risk of pancreatic duct injury. Hence, we question whether it is necessary to ligate distal stumps for all CDC. Our results have demonstrated the feasibility and outcome of a new surgical practice whereby distal CBD stump is left unligated for CDC children with stenotic distal CBD. The evidence from the current large series suggests that (1) no clinically detectable post-operative pancreatic juice

leakage occurred even if the distal stump was left unligated in those patients with stenotic distal CBD; (2) iatrogenic injury of pancreatic ducts may be minimized and (3) the practice simplies the surgical procedure. This study is weakened by its retrospective nature. Moreover, despite being one of largest series, it fails to demonstrate a difference in pancreatic duct injury rate, as we stopped the practice of dissection of stenotic distal CBD and thus did not have an appropriate control group. A prospective randomized study of comparison between ligation and non-ligation group is warranted in future. In conclusion, in the patients who have cystic CDC with stenotic distal CBD, the ligation of the distal CBD stump is optional. On the other hand, in the patients who have fusiform CDC with non-stenotic distal CBD stone debris in the common channel, distal CBD stump must be ligated to prevent pancreatic juice leakage.
Conict of interest Dr. Mei Diao, Prof. Long Li, and Prof. Wei Cheng have no conicts of interest or nancial ties to disclose.

References
1. Miyano T, Li L, Yamada K (2009) Choledochal cyst. In: Gupta D, Sharma S, Azizkhan R (eds) Pediatric Surgerydiagnosis and management. Jayee Brother Medical Publishers (P) LTD, New Delhi, pp 10131025 2. Lee H, Hirose S, Bratton B, Farmer D (2004) Initial experience with complex laparoscopic biliary surgery in children: biliary atresia and choledochal cyst. J Pediatr Surg 39:804807, discussion 804807 3. Le DM, Woo RK, Sylvester K, Krummel TM, Albanese CT (2006) Laparoscopic resection of type 1 choledochal cysts in pediatric patients. Surg Endosc 20:249251 4. Ure BM, Nustede R, Becker H (2005) Laparoscopic resection of congenital choledochal cyst, hepaticojejunostomy, and externally made Roux-en-Y anastomosis. J Pediatr Surg 40:728730 5. Nguyen Thanh L, Hien PD, Dung le A, Son TN (2010) Laparoscopic repair for choledochal cyst: lessons learned from 190 cases. J Pediatr Surg 45:540544 6. Palanivelu C, Rangarajan M, Parthasarathi R, Amar V, Senthilnathan P (2008) Laparoscopic management of choledochal cysts: technique and outcomesa retrospective study of 35 patients from a tertiary center. J Am Coll Surg 207:839846 7. Chowbey PK, Katrak MP, Sharma A, Khullar R, Soni V, Baijal M, Vashistha A, Dhir A, Dewan A (2002) Complete laparoscopic management of choledochal cyst: report of two cases. J Laparoendosc Adv Surg Tech 12:217221 8. Li SL, Zhang DR, Li YC, Li ZD, Niu AG (2000) Prevention and treatment for pancreatic duct injury during the excision of choledochal cyst. Chin J Pediatr Surg 21:211213 9. He XB, Feng D (2008) Surgical treatment for 48 adults with choledochal cysts. J Hepatopancreatobiliary Surg 20:357358 10. Li ZF (2004) Post-operative complications in children with choledochal cysts. J Anhui Health Coll 3:3435 11. Li L, Feng W, Jing-Bo F, Qi-Zhi Y, Gang L, Liu-Ming H, Yu L, Jun J, Ping W (2004) Laparoscopic-assisted total cyst excision of choledochal cyst and Roux-en-Y hepatoenterostomy. J Pediatr Surg 39:16631666

123

832 12. Diao M, Li L, Cheng W (2010) Laparoscopic versus Open Rouxen-Y hepatojejunostomy for children with choledochal cysts: intermediate-term follow-up results. Surg Endosc [Epub ahead of print] 13. Diao M, Li L, Zhang JS, Cheng W (2010) Laparoscopic-assisted clearance of protein plugs in the common channel in children with choledochal cysts. J Pediatr Surg 45:20992102 14. Diao M, Li L, Zhang JZ, Cheng W (2010) A shorter loop in Roux-Y hepatojejunostomy reconstruction for choledochal cysts

Pediatr Surg Int (2011) 27:829832 is equally effective: preliminary results of a prospective randomized study. J Pediatr Surg 45:845847 15. Li L, Yamataka A, Wang YX, Wang DY, Wang K, Li ZX, Shimizu T, Yamashiro Y, Zhang JZ, Lane GJ, Miyano T (2003) Anomalous pancreatic duct anatomy, ectopic distal location of the papilla of Vater and congenital biliary dilatation: a new developmental triad? Pediatr Surg Int 19:180185

123