Lilly's Technique for Delayed Hemorrhage After Choledochal Cyst Radical Surgery

Abstract

Background:

Choledochal cysts (CCs) are characterized by dilations of the extra- and/or intrahepatic bile ducts. Surgery (cyst excision and Roux-en-Y hepaticojejunostomy) remains the gold standard for treatment. However, delayed hemorrhage can occur postoperatively, and although rare, it can be life-threatening. This study aimed to determine the risk factors and corresponding prevention of delayed hemorrhage after radical CC surgery, and to apply a technique to lower its incidence.

Materials and Methods:

This retrospective study enrolled 267 patients who received CC surgery between June 2016 and December 2020 at Shenzhen Children's Hospital. Univariate and multivariate logistic regression analyses were performed to identify risk factors for delayed hemorrhage.

Results:

Eleven (4.1%) patients had delayed hemorrhage after laparoscopic radical surgery. The most common hemorrhage site was the dissected surface between the cyst and adjacent structures with chronic severe adhesions, postoperatively. The occurrence of recurrent CC-associated complication and excessive total blood loss during surgery were risk factors for delayed hemorrhage after CC radical surgery. Length of disease course, operation when cholangitis/pancreatitis still existed, cyst diameter, and application of trypsin inhibitor after the surgery were not significantly different between the two groups.

Conclusion:

For patients without adhesions, complete cyst resection is the gold standard. However, for those with intensive adhesions, in cases of delayed hemorrhage on the dissection surface and malignancy transformation risk, the Lilly's technique with Roux-en-Y hepaticojejunostomy could be an alternative.

Introduction

Choledochal cysts (CCs) are cystic dilatations of the intrahepatic and/or extrahepatic biliary ducts. It has a higher incidence in Asian people than in Western populations (1 in 13,000 versus 1 in 100,000 to 1 in 150,000), with a characteristic female predominance (∼4:1).^1–3 CCs are diagnosed incidentally due to complications, such as abdominal pain or pancreatitis, and they mostly occur in children. Conventionally, for patients with CCs, the standard approach is complete cyst excision and Roux-en-Y hepaticojejunostomy to ameliorate both short- and long-term sequelae.^4,5 Delayed hemorrhage is a relatively rare but severe postoperative complication that generally requires emergency surgery. However, to date, studies focused on this life-threatening complication are limited. Consequently, whether delayed hemorrhage occurs intraoperatively or immediately postoperatively and its etiology remain unclear.

This study aimed to determine the risk factors for delayed hemorrhage after CC radical surgery and corresponding preventions. We also aimed to identify the most common hemorrhage site and attempted to raise another consideration for the avoidance of delayed hemorrhage in patients with CCs with the identified risk factors.

Materials and Methods

Patient selection

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Shenzhen Children's Hospital. Informed consent was obtained from all the guardians of all the patients. Data were collected from the case system of Shenzhen Children's Hospital. All cases were preoperatively diagnosed using ultrasound, computed tomography, or magnetic resonance cholangiopancreatography according to the Todani classification.

Altogether, 267 patients with CC who received laparoscopic radical surgery performed by the same experienced surgical team were included in the study. Inclusion criteria: A dilated extra- and/or intrahepatic bile duct seen intraoperatively during cholangiography, and pathological confirmation of CC. Exclusion criteria: Patients with other accompanying diseases, patients diagnosed with pathology other than CC, and incomplete history or clinical data of the patients. Of the patients, 11 had delayed hemorrhage, whereas 256 had none.

Data on age, gender, length of disease course, occurrence of recurrent cholangitis and pancreatitis (or other complications associated with CC) preoperatively, the presence of cholangitis and pancreatitis (or other complications associated with CC) during the surgery, cyst type, cyst diameter, total blood loss intraoperatively, timing of hemorrhage-related symptoms, clinical manifestations caused by hemorrhage, use of trypsin inhibitor postoperatively, postoperative serum amylase level, timing of second (or subsequent) surgery, concrete procedure of the second (or subsequent) surgery, and location of the hemorrhage were recorded and analyzed.

Surgical procedure

The patients were in supine and placed under general anesthesia. A vertical incision through the umbilicus was made to insert a 10 mm 30° laparoscope, and a carbon dioxide pneumoperitoneum was created (8–12 mm Hg). Trocars were then placed at the intersection of the right anterior axillary line and costal margin, midpoint of the right rectus abdominis outer rim, and left upper abdominal rectus outer rim. The ligamentum teres hepatis was sutured and lifted externally to expose the porta hepatis. The ligamentum hepatoduodenale was cut using an electrotome, the gallbladder was dissected from its bed, and the cystic artery was ligated while temporarily maintaining the gallbladder. The duodenum was pushed down and separated from the cyst antetheca.

If the cyst was sufficiently small, it was completely drained by fine-needle aspiration to reduce its size. The distal cyst wall was raised and closely separated at the confluence of the bile duct and pancreatic duct, followed by ligation of the distal bile duct; thus, the distal cyst wall was resected from the pancreas. The paries posterior of the cyst was carefully dissociated from the porta hepatis, followed by removing the dilated cyst along with cystic duct, gallbladder, and the distal part of the common hepatic duct. After the excision, a trumpet-shaped terminal of the common hepatic duct was created for better anastomosis with the jejunum. If the left and right hepatic ducts were excessively narrow, these structures were also enlarged.

After the ligament of Treitz was identified, the jejunum was held with an intestinal grasper 15 cm distal to the ligament of Treitz. The 10-mm umbilical trocar was then extracted, and the umbilical incision was expanded to 1.5–2.5 cm. The jejunum was retracted from the abdomen, and excised into two (proximal and distal) parts. While the distal jejunum opening was closed, the proximal jejunum was anastomosed end-to-side with the distal jejunum, ∼25–30 cm distal to the previous closure of the distal jejunum end (end-to-side Roux-en-Y anastomosis).

The mesenteric incision created for Roux limb construction was closed to avoid internal hernia, before these structures were placed back into the abdominal cavity. The umbilical trocar was repositioned, and pneumoperitoneum was re-established. The transverse colon mesentery was incised to course the Roux limb through the hepatic hilum behind the colon. A small incision was made on the antimesenteric side distal to the closure end of jejunum for end-to-side hepaticojejunostomy. Posterior and anterior walls of the hepaticojejunostomy were created using continuous suture using 5-0 absorbable sutures. A drainage tube was placed near the anastomotic site, and the abdominal cavity was flushed and closed.

Statistical analysis

The SPSS software (version 26.0) was used for statistical processing. Categorical variables are presented as frequencies and percentages. Continuous variables are presented as mean ± standard deviation according to their distribution. Differences between groups were compared using the t-test, Mann–Whitney U test, and Wilcoxon signed-rank test, as appropriate. To determine statistical differences between ordinal values, the chi-square or Fisher's exact test was used as appropriate. Categorical variables and continuous variables between the two groups were assessed by univariate analysis, and variables with P value <.1 on univariate analysis were included in the multivariate logistic regression analysis to explore risk factors associated with delayed hemorrhage after CC surgery. A forward stepwise method was adopted for the logistic regression analysis to reduce the number of independent variables entering the analysis model. P < .05 was considered significant.

Results

Risk factors of delayed hemorrhage after CC radical surgery

Between June 2016 and December 2020, 267 patients with CC were enrolled in the study. Of these, 11 (4.1%) involved delayed hemorrhage after laparoscopic radical surgery. Demographic characteristics are summarized in Table 1. The mean age was 6.09 ± 4.16 years for the delayed hemorrhage group and 3.01 ± 2.94 years for the control group. Five (45.5%) male patients were enrolled in the delayed hemorrhage group, whereas 58 (22.7%) were enrolled in the control group. No significant differences in age or gender were observed between the two groups (P > .05).

Table 1.

Demographics of Choledochal Cyst Patients

	Delayed hemorrhage (n = 11)	Non-hemorrhage (n = 256)	P value
Age (years)	6.09 ± 4.16	3.01 ± 2.94	>.05
Gender (male)	5 (45.5)	58 (22.7)	>.05

Nine (81.8%) patients in the hemorrhage group and 107 (41.8%) in the control group had recurrent cholangitis and pancreatitis (or other complications associated with CC) preoperatively (P = .049, OR 8.45, 95% CI 1.000–71.384). Intraoperative total blood loss in the delayed hemorrhage group was 95 ± 140 mL, whereas that in the control group was 9 ± 6 mL (P = .008, OR 1.078; 95% CI 1.019–1.140). The results indicated that the occurrence of recurrent CC-associated complications and total blood loss during surgery were risk factors for delayed hemorrhage after CC radical surgery (Table 2).

Table 2.

Univariate and Multivariate Logistic Regression Analysis for Odds to Delayed Hemorrhage

Variables	Delayed hemorrhage (n = 11)	Non-hemorrhage (n = 256)	Univariate analysis P value	Multivariate analysis P value	OR	95% CI
Length of disease course (m)	2.50 ± 3.46	5.74 ± 11.35	.348	N/A	N/A	N/A
Recurrent preoperative cholangitis/pancreatitis (n)	9 (81.8)	107 (41.8)	.021	.049	8.45	1.000–71.384
Operation when cholangitis/pancreatitis still existed (n)	6 (54.5)	47 (18.4)	.010	.050	N/A	N/A
Total blood loss during the first surgery (mL)	95 ± 140	9 ± 6	.019	.008	1.078	1.019–1.140
Cyst diameter (cm)	4.7 ± 1.7	3.4 ± 2.0	.011	.253	N/A	N/A
Trypsin inhibitor intervention after the surgery (n)	1 (9.1)	71 (27.7)	.309	N/A	N/A	N/A

CI, confidence interval; OR, odds ratio.

Length of disease course, existence of cholangitis and pancreatitis (or other complications associated with CC) intraoperatively, cyst diameter, and use of trypsin inhibitor postoperatively were not significantly different between the two groups (P > .05) (Table 2).

Most common hemorrhage site of delayed hemorrhage

Tables 3 and 4 provide more information on each patient in the delayed hemorrhage group. In particular, clinical manifestations and treatment have been emphasized. The timing of the appearance of hemorrhage-related symptoms varied from 30 minutes to 12 days. The most common clinical manifestations of hemorrhage include abdominal pain, pale skin, and bloody fluid from the drainage tube. Amylase levels were measured postoperatively in only 5 (45.5%) cases, whereas only 1 (9.1%) case was treated with trypsin inhibitor postoperatively. Nine (81.8%) patients received a second surgery, whereas 8 (72.7%) received laparotomy.

Table 3.

The Details of Patients Who Suffered Delayed Hemorrhage After Choledochal Cyst Radical Surgery

	Cyst type	Length of onset of disease (m)	Preoperative recurrent cholangitis/pancreatitis (n)	Cholangitis/pancreatitis existed during operation (n)	Cyst diameter (cm)	Total blood loss during the first surgery (mL)	Occurrence time of hemorrhage-related symptoms	Postoperative application of trypsin inhibitor	Postoperative amylase level (IU/L)	Interval for subsequent surgery	Subsequent surgery after hemorrhage	Hemorrhage site
Pt 1, F, 4 years	1a	4.00	Yes	No	3.0	50	7\11\12 days	No	Normal^a 98,886^b	8\11\14 days	Laparoscopy\laparotomy\DSA	Dissection surface of the gallbladder bed, dissection surface of the cyst, anastomosis, gastroduodenal artery
Pt 2, M, 11 years	1c	0.25	No	Yes	4.2	200	11 hours\8 days	No	N/A	14 hours\8 days	Laparotomy, DSA	Dissection surface of the pancreas, arteria pancreaticoduodenalis, right gastroepiploic artery
Pt 3, M, 7 years	1c	1.00	Yes	No	1.4	10	23 hours	No	N/A	2 days	Laparoscopy transferred to laparotomy	Dissection surface of the cyst
Pt 4, M, 2 years	1a	12.00	Yes	No	4.0	10	2 days	No	Normal^a	2 days	Laparoscopy transferred to laparotomy	Dissection surface of the pancreas
Pt 5, M, 13 years	1c	4.00	Yes	No	2.6	10	30 minutes	No	Normal^a	1 hour	Laparoscopy	Dissection surface of the pancreas
Pt 6, F, 1 year	1a	1.00	Yes	Yes	3.8	400	5 hours	No	Normal^a	5 hours	Laparotomy	Thrombosis in the portal vein
Pt 7, F, 5 years	1a	3.00	Yes	Yes	4.4	5	6 hours	No	N/A	1 day	Laparotomy	Dissection surface of the gallbladder bed, dissection surface of the cyst
Pt 8, F, 11 years	1c	0.50	Yes	Yes	4.4	50	1 day	No	N/A	1 day	DSA+laparotomy	Gastroduodenal artery, right hepatic artery, dissection surface of the cyst
Pt 9, F, 5 years	1c	0.10	No	Yes	2.1	5	8 hours	No	Normal^a	1 day	Laparotomy	Dissection surface of the pancreas, arteria pancreaticoduodenalis
Pt 10, M, 7 years	1a	0.67	Yes	Yes	6.5	300	2 days	Yes	N/A	N/A	N/A	N/A
Pt 11, F, 1 year	1a	1.00	Yes	No	3.4	5	1 day	No	N/A	N/A	N/A	N/A

Peripheral blood.

Ascites.

DSA, digital subtraction angiography; F, female; M, male; N/A, not applicable; Pt, patient.

Table 4.

Clinical Manifestations Caused by Delayed Hemorrhage

	Abdominal pain (n = 5)	Pale (n = 6)	Bloody fluid from drainage tube (mL)
Pt 1, F, 4 years	Yes	Yes	No
Pt 2, M, 11 years	Yes	Yes	100
Pt 3, M, 7 years	No	Yes	60
Pt 4, M, 2 years	No	Yes	No
Pt 5, M, 13 years	Yes	No	No
Pt 6, F, 1 year	No	No	No
Pt 7, F, 5 years	No	Yes	10
Pt 8, F, 11 years	Yes	Yes	110
Pt 9, F, 5 years	Yes	No	300
Pt 10, M, 7 years	No	No	25
Pt 11, F, 1 year	No	No	No

F, female; M, male; Pt, patient.

During the second (or subsequent) surgery, the dissected surface, the area used for adherence between the dissected bile duct and adjacent tissues, was the most common site of delayed hemorrhage. However, bleeding did not occur intraoperatively; instead, it occurred postoperatively because of slower errhysis on the dissected surface. Others involved branches of the gastroduodenal artery, pancreaticoduodenal artery, right gastroepiploic artery, right hepatic artery, and portal vein. Not every patient with delayed hemorrhage developed symptoms: only 5 out of 9 patients had abdominal pain, 6 out of 9 had pale skin and bloody fluid from the drainage tube, and 2 patients demonstrated no relevant symptoms.

Discussion

In this study, we identified two risk factors for delayed hemorrhage after radical CC surgery: recurrent CC-associated complications preoperatively and increased intraoperative total blood loss. These two factors are both associated with adhesion.

Patients with recurrent cholangitis and pancreatitis, the most common complications associated with CC, tend to have severe adhesions, leading to difficulty in dissecting the cyst from surrounding tissues during surgery. It is commonly acknowledged that blunt dissection is better than sharp dissection because the former technique helps surgeons clearly identify the boundaries between tissues and thus avoid unnecessary injury.

However, if, preoperatively, patients suffered from recurrent inflammation, severe adhesion between the cyst and surrounding structures could be observed intraoperatively, which made the boundaries obscure; in this situation, sharp dissection was preferably applied. The adjacent vessel wall, capsule of the pancreas, and even the pancreas itself were at a higher risk of injury during the dissociation process. Therefore, more total blood loss occurred intraoperatively under these conditions. From our data analysis, the delayed hemorrhage site at the dissection surface of the pancreas further confirmed this. Therefore, if we could avoid injury to these sites in CC with severe adhesion, delayed hemorrhage would be well controlled.

This theory was confirmed by the findings of the second surgery. Nine patients received second, third, and fourth surgeries. Most of the subsequent surgeries were laparotomy, or some patients received laparoscopy first but were then made to receive laparotomy (Table 3). Hence, laparotomy might be a better option for patients with delayed hemorrhage. Digital subtraction angiography (DSA) could also be considered when laparotomy is less effective. The hemorrhage and dissection surfaces were always identical, particularly the surface where the posterior wall of the cyst was dissected.

However, whether any bleeding on the dissection surface completely prevents delayed hemorrhage in patients with CC with severe adhesions was unclear, and our findings suggest otherwise. Therefore, understanding why delayed hemorrhage still occurs postoperatively with well-managed hemorrhage intraoperatively helps us to establish strategies to avoid it.

Nowadays, sharp dissection, as stated earlier, is more widely accepted than usual. The electrotome, one of the most extensively used equipment for hemostasis, has become an ideal choice when there is extensive oozing of blood on the dissection surface by forming an eschar at the bleeding site. However, an excessively thick eschar was noted to fall off postoperatively, resulting in delayed hemorrhage. This most likely occurred in relatively major vessels, such as the branches of the gastroduodenal artery, pancreaticoduodenal artery, right gastroepiploic artery, right hepatic artery, and portal vein. High blood pressure requires a thick eschar; otherwise, blood underneath the eschar could easily break through it postoperatively.

Therefore, although hemorrhage is well managed intraoperatively, delayed hemorrhage can still occur postoperatively. Previous researchers have also reported that any attempt to completely separate the cyst may result in injury to the portal vein.⁶ Once injured, the repair process can result in further complications that can be easily ignored by surgeons. Patient 6 sustained portal vein injury during radical surgery; thus, to suture the breach, the portal vein was temporally clamped, which led to thrombosis, requiring emergency surgery 5 hours later.

Moreover, pancreatic fistulas caused by pancreatic injury may also be associated with delayed hemorrhage. This may be due to the leakage of trypsin resulting in digesting of the vascular wall. During CC radical surgery, the pancreas was assumed to be intact, and severe adhesion between the cyst and pancreas made it difficult to dissociate the cyst wall from the pancreas.

Since pancreatic fistula is typically considered a complication of pancreatic surgeries,⁷ most of the patients included in this study did not have trypsin inhibitor or amylase level measurements postoperatively. Among the 9 patients who received a second surgery, blood amylase levels (all within the normal range) were measured in 5 patients, whereas the amylase level in the ascites was measured in only 1 patient, of which the result (98,886 IU/L) indicated that the pancreatic fistula was complicated. During the second surgery, 4 of 9 patients had active and oozing bleeding at the surface of the pancreas. Moreover, trypsin leakage could not only digest the minor vessels on the pancreatic surface but also the surrounding tissues or eschar.

Therefore, all of the aforementioned reasons can be attributed to the surgeons always pursuing complete dissection of the cyst while ignoring severe adhesions.

In addition, the early identification of delayed hemorrhage is of key importance. Ten patients who experienced delayed hemorrhage developed symptoms within 48 hours after the first surgery (Table 3). Although abdominal pain, pale skin, and bloody fluid from the drainage tube are common symptoms of delayed hemorrhage, not every patient has the aforementioned manifestations, especially in the early stage of hemorrhage. Two patients did not demonstrate any symptoms until decreased hemoglobin levels were detected in their blood tests (Table 4). Therefore, regular postoperative blood tests should be closely monitored.

Carcinogenesis is one of the most severe long-term complications of CC.⁸ Thus, complete excision of the cyst is recommended.^9,10 However, for patients with severe adhesion between the cyst and the surrounding tissues, whether pursuing total excision or keeping the posterior wall of the cyst to avoid the risk (Lilly's technique) is controversial, considering the potential risk of tissue damage and subsequent delayed hemorrhage.

The Lilly's technique, which refers to mucosal excision through curettage or cautery, while the serosa remains attached to the adhering structures instead of removing the entire, has been proposed for >40 years.¹¹ However, studies on this procedure remain few. Previously, the Lilly's technique was successfully performed in 11 patients with intensive adhesions.¹²

Although a residual cyst could result in carcinogenesis, it generally suggests preserving the distal CC or bile duct.^13,14 An Indian team performed the Lilly's technique in 4 patients, and their follow-up revealed that none of them developed malignancy.¹⁵ Since carcinogenesis starts from the mucosal lining,⁹ if the intraoperative pathological section indicates no sign of mucosal cancer, while the adhesion is severe, the Lilly's technique should be considered a safe option. By completely removing all the mucosa of the cyst wall that severely adheres to adjacent structures, the malignancy risk is largely decreased. However, for patients with CC without severe adhesions, conventional complete cyst resection remains the gold standard treatment.

Conclusions

Patients with recurrent cholangitis, pancreatitis, and excessive blood loss during radical CC surgery tend to have delayed hemorrhage. For patients without adhesions, complete cyst resection is the gold standard. However, for those with the aforementioned risk factors, delayed hemorrhage mostly occurred at the dissection surface, where the posterior cyst wall was tightly stuck to adjacent structures postoperatively due to inconspicuous chronic bleeding. In cases of malignancy and delayed hemorrhage, the pathological cyst should be completely excised, but at the severe adhesion site, complete removal of the mucosal lining of the cyst by cauterization with the serosa remaining attached to the adhering structures (Lilly's technique) together with Roux-en-Y hepaticojejunostomy might be a better surgical strategy to deal with the severe adhesion surface for patients with CC with risk factors.

Footnotes

Authors' Contributions

Review and article drafting by W.W. Provision of study materials or patients by J.Z. Collection and assembly of data by X.Z., Q.M., and J.G. Data analysis and interpretation by W.W., J.Z., L.H., and Y.Y. Review and editing by Y.Y. and B.W. Conception and design by B.W. All authors contributed to the article and approved the submitted version of the article.

Disclosure Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Information

This study was supported by the Sanming Project of Medicine in Shenzhen (grant No. SZSM201812055).

References

Sela-Herman

, Scharschmidt

. Choledochal cyst, a disease for all ages. Lancet (London, England), 1996; 347(9004):779; doi:10.1016/s0140-6736(96)90864-8

Soares

, Arnaoutakis

, Kamel

, et al. Choledochal cysts: presentation, clinical differentiation, and management. J Am Coll Surg, 2014; 219,(6):1167–1180; doi:10.1016/j.jamcollsurg.2014.04.023

Diao

, Li

, Cheng

. Timing of choledochal cyst perforation. Hepatology (Baltimore, MD), 2020; 71,(2):753–756; doi:10.1002/hep.30902

She

, Chung

, Lan

, et al. Management of choledochal cyst: 30 years of experience and results in a single center. J Pediatr Surg, 2009; 44,(12):2307–2311; doi:10.1016/j.jpedsurg.2009.07.071

Ishibashi

, Shimada

, Kamisawa

, et al. Japanese clinical practice guidelines for congenital biliary dilatation. J Hepatobiliary Pancreat Sci, 2017; 24,(1):1–16; doi:10.1002/jhbp.415

Martinez-Ferro

, Esteves

, Laje

. Laparoscopic treatment of biliary atresia and choledochal cyst. Semin Pediatr Surg, 2005; 14,(4):206–215; doi:10.1053/j.sempedsurg.2005.06.003

Kawaida

, Kono

, Hosomura

, et al. Surgical techniques and postoperative management to prevent postoperative pancreatic fistula after pancreatic surgery. World J Gastroenterol, 2019; 25,(28):3722–3737; doi:10.3748/wjg.v25.i28.3722

Tashiro

, Imaizumi

, Ohkawa

, et al. Pancreaticobiliary maljunction: retrospective and nationwide survey in Japan. J Hepatobiliary Pancreat Surg, 2003; 10,(5):345–351; doi:10.1007/s00534-002-0741-7

Watanabe

, Toki

, Todani

. Bile duct cancer developed after cyst excision for choledochal cyst. J Hepatobiliary Pancreat Surg, 1999; 6,(3):207–212; doi:10.1007/s005340050108

10.

Lü

, Shi

, Wang

, et al. Technical points of total laparoscopic choledochal cyst excision. Chin Med J (Engl), 2013; 126,(5):884–887; doi:10.3760/cma.j.issn.0366-6999.20111678

11.

Lilly

. Total excision of choledochal cyst. Surg Gynecol Obstet, 1978; 146,(2):254–256.

12.

Tang

, Yang

, Wang

, et al. Laparoscopic choledochal cyst excision, hepaticojejunostomy, and extracorporeal Roux-en-Y anastomosis: a technical skill and intermediate-term report in 62 cases. Surg Endosc, 2011; 25,(2):416–422; doi:10.1007/s00464-010-1183-y

13.

Stain

, Guthrie

, Yellin

, et al. Choledochal cyst in the adult. Ann Surg, 1995; 222,(2):128–133; doi:10.1097/00000658-199508000-00004

14.

Lee

, Yoon

, Han

, et al. Laparoscopic excision of type II choledochal cyst arising from the intrapancreatic common bile duct in an adult. J Hepatobiliary Pancreat Sci, 2020; 27,(10):789–790; doi:10.1002/jhbp.808

15.

Senthilnathan

, Patel

, Nair

, et al. Laparoscopic management of choledochal cyst-technical modifications and outcome analysis. World J Surg, 2015; 39,(10):2550–2556; doi:10.1007/s00268-015-3111-8