1
SURGERY FOR CHOLEDOCHAL CYST. A STUDY OF THE CLINICAL PROFILE AND CLINICAL OUTCOMES WHEN THE TWO COMMONLY USED HEPATICO-ENTERIC ANASTOMOSIS ARE
COMPARED
BEACH TRIAL – B illiary E nteric A nastomosis in CH oledochal cyst
A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE MCh BRANCH (PAEDIATRIC SURGERY) EXAMINATION
OF THE DR. M.G.R MEDICAL UNIVERSITY, TAMILNADU,
CHENNAI TO BE HELD IN AUGUST 2015
2
BONAFIDE CERTIFICATE
This is to certify that the work presented in this dissertation titled
“SURGERY OF CHOLEDOCHAL CYST. A STUDY OF THE CLINICAL PROFILE AND CLINICAL OUTCOMES WHEN THE TWO COMMONLY USED HEPATICO-ENTERIC ANASTOMOSIS ARE COMPARED” done towards fulfilment of the requirements of the Tamil Nadu DR M.G.R Medical University, Chennai for the MCh (Paediatric Surgery) exams to be conducted in August 2015, is a bonafide work of Dr Tarun John Kochukaleekal Jacob, Senior Post Graduate student in the department of Paediatric Surgery, Christian Medical College, Vellore under my Guidance and supervision. This thesis has not been submitted, fully or in part to any other board or University.
Dr Immanuel Sampath Karl Dr Alfred Job Daniel
Professor. Principal
Guide and Head of Department Christian Medical College Department of Paediatric Surgery Vellore 632004
Christian Medical College, Vellore -632004
Submission author:
Assignment title:
Submission title:
File name:
File size:
Page count:
Word count:
Character count:
Submission date:
Submission ID:
Digital Receipt
This receipt acknowledges that Turnitin received your paper. Below you will f ind the receipt inf ormation regarding your submission.
The f irst page of your submissions is displayed below.
181215064.. Mch - Paediatric Surge…
TNMGRMU EXAMINATIONS
Tarun Jacob Mch Paed Surgery - C…
Final_draf t_Tarun_thesis_11_Marc…
1.19M 93 18,293 94,995
17-Mar-2015 08:57AM 517227347
Co pyright 20 15 Turnitin. All rights reserved.
Turnitin
file:///C|/Users/DR7FE7~1.TAR/AppData/Local/Temp/Turnitin_Originality_Report_517227347-1.html[17-03-2015 13:07:30]
Turnitin Originality Report
Processed on: 2015年03月17日 08:59 IST ID: 517227347
Word Count: 18293 Submitted: 1
Tarun Jacob Mch Paed Surgery - Choledochal Cyst 2015 By 181215064.. Mch - Paediatric Surgery Dr Tarun John K. Jacob
Similarity Index
3% Internet Sources: 2%
Publications: 3%
Student Papers: 1%
Similarity by Source
Document Viewer
1% match (Internet from 15-Oct-2013) http://www.cmch-
vellore.edu/static/research/Files/Format%20of%20IRB%20Application%20for%20tests%20of%20Diagnostic%20Accuracy_060813.doc
< 1% match (publications)
Liem, Nguyen Thanh, Hien Duy Pham, Le Anh Dung, Tran Ngoc Son, and Hoan Manh Vu. "Early and Intermediate Outcomes of Laparoscopic Surgery for Choledochal Cysts with 400 Patients", Journal of Laparoendoscopic &
Advanced Surgical Techniques, 2012.
< 1% match (publications)
"Functional Bowel Disorders", The American Journal of Gastroenterology, 9/2006
< 1% match (publications)
H. Shimura. "Laparoscopic treatment of congenital choledochal cyst", Surgical Endoscopy, 10/01/1998
< 1% match (publications)
Ohashi, Taku, Toshifumi Wakai, Masayuki Kubota, Yasunobu Matsuda, Yuhki Arai, Toshiyuki Ohyama, Kengo Nakaya, Naoki Okuyama, Jun Sakata, Yoshio Shirai, and Yoichi Ajioka. "Risk of subsequent biliary malignancy in patients undergoing cyst excision for congenital choledochal cysts : Biliary malignancy after cyst excision", Journal of Gastroenterology and Hepatology, 2012.
< 1% match (Internet from 20-Aug-2014)
http://coursesindia.com/bsc-nursing-courses-at-tndmgrmu-chennai/
< 1% match (Internet from 25-Jan-2015)
http://www.cmch-vellore.edu/static/research/Files/research%20digest-jan_june_2014.pdf
< 1% match (Internet from 03-Dec-2008) http://www.emedicine.com/MED/topic349.htm
< 1% match (Internet from 26-Oct-2009) http://www.ijri.org/article.asp?issn=0971-
3026;year=2002;volume=12;issue=4;spage=491;epage=492;aulast=Ahluwalia
< 1% match (publications)
Alastair J.W. Millar. "Surgical Disorders of the Liver and Bile Ducts and Portal Hypertension", Diseases of the Liver and Biliary System in Children, 09/19/2008
3
ACKNOWLEDGEMENTS
To God who has granted us wisdom to learn and discern. All glory to Him.
To my wife and children who continually support me through my
training, who allow me to be absent from many responsibilities at home to complete the requirements of this course
To my Guide, teachers, colleagues and office staff who were patient and offered more of themselves than I often deserved.
I am humbled, grateful and indebted for all I have received.
4 ABSTRACT
Title: Surgery of choledochal cyst. A study of the clinical profile and clinical outcomes when the two commonly used hepatico-enteric anastomosis are compared.
Billiary Enteric Anastomosis in Choledochal cyst (BEACH TRIAL)
Aims/Objectives: To study the demography, presentations and outcomes in children who present for surgery of choledochal cyst. Statistical comparison of the two types of biliary enteric anastomosis – hepatico-duodenostomy (HD) and hepatico- jejunostomy (HJ) and discuss the early to medium term complications and outcomes.
The trial aims to determine if there is a difference in outcomes between these two anastomosis.
Materials and methods: 70 patients were studied – 35 in each arm. Their baseline characteristics were compared and their outcomes and complications noted.
Demographic data and data from investigations ordered were also analysed. Data was obtained from a patient’s history and clinical examination. Imaging, blood analysis and nuclear imaging data was also collected.
Results and conclusions: Both groups HD and HJ had similar baseline characteristics. The study was designed to allow for at least 6 months of follow up – however a mean follow up of 22 months in the HD and 25 months in the HJ group was obtained. Loss to follow up was less than 20%. It was interesting to note that the average bilirubin at presentation in both groups was over 2.1 mg%. The HJ group had more unusual presentations that required surgery. The outcomes were mostly similar;
both groups had similar rates of individual complication rates – such as intestinal obstruction, pancreatitis, cholangitis, pancreatic leak, jaundice, pelvic abscess and
5 wound infections. There were no statistical differences between the two groups.
However if all complications were combined, there is a 3.6 times greater chance to develop any complication in a HJ when compared to a HD. There was only stricture and that was in the HD group. It can be concluded that there is a role for both the anastomosis – early to intermediate follow up shows no significant differences between the two anastomosis. A randomized controlled trial that has at least a 10 year follow up can be justified on the basis of this trial.
6
Index. Pg No.
1. Introduction 7
2. Aims 9
3. Objectives 10
4. Materials and Methods 11
5. Review of Literature 18
6. Results and Analysis 47
7. Discussion 73
8. Conclusions 79
9. Bibliography 81
10. Appendix
Appendix 1 – Performa
Appendix 2- Master Sheet with data Appendix 3 –Consent Form
7
Introduction
Choledochal cysts or biliary cysts are a rare disease of childhood. The western world sees choledochal cysts in about 1:150,000 births. The disease is more common in the east(1,2). The treatment is a complete excision of the cyst followed by re-establishment of the biliary drainage - done by two commonly established procedures.
i. A direct anastomosis of the bile duct to the duodenum – hepatico- duodenostomy (HD)
ii. A Roux-en-Y anastomosis of the bile duct to jejunum in a Y fashion – Hepatico-jejunostomy (HJ)
The choice for which method is preferable is unclear. Various institutions follow their own choice, with their own reasons. There are a few series of Indian children in literature that describe demography, but none that compare the two common methods of re-establishing bile to intestinal continuity. We have performed both operations in our institution in the past.
This study will be looking at a cohort of children with choledochal cyst, to describe the demography and clinical features in our institution. This study will also attempt to compare the children’s outcomes after either of the two hepaticoenteric anastomosis and see if there are any significant differences between the early, late and combined complication rates of the two.
8
The patient interview during follow up and inpatient and outpatient charts will be screened for details of postoperative complications.
A descriptive analysis of the demography and presentation will be made. A description of the radiological features and complicated presentations will also be noted. Analysis that compare the outcomes of the two group of patients will be done – early, late and a combined complication rate.
At the end of the study we hope to describe how children present to us with choledochal cyst and how they do after they have undergone either of the two types of biliary enteric anastomosis. We hope to suggest if any one procedure is better and propose the initiation of a larger more structured trial to assess this question of surgical equipoise.
9
Aims of the study
• To understand the profile of the children who present for surgery of choledochal cysts in a tertiary care, paediatric surgical centre in south India.
• To determine if there is any benefit between the two standard anastomosis used for re-establishment of the biliary-enteric communication (Hepaticoduodenostomy Vs Roux En-Y Hepatico- jejunostomy) after excision of a choledochal cyst.
• To attempt to structure a future prospective study on this question of surgical equipoise based on our findings and structure a policy on hepatico-enteric anastomosis in our institution.
10
Objectives of the study
• To describe the age and sex distribution of children who present for choledochal cyst surgery in our institution.
• To objectively quantify the proportion of children with the various presentations in a tertiary paediatric surgery centre in South India (i.e.
– jaundice, pain, cholangitis, palpable mass etc.)
• To discuss the early, late and combined complication rates in children who undergo the two standard operations to re-establish biliary enteric anastomosis.
• To compare statistically if a hepatico-duodenal anastomosis is any different from a Roux-En- Y hepaticojejunostomy anastomosis in the re-establishment of bile continuity in choledochal cyst surgery.
11
MATERIALS AND METHODS
The following is the planned research methodology for the BEACH –Biliary Enteric Anastomosis after Choledochal cyst surgery trial.
Detailed research plan:
a. Setting: All children who underwent surgery for choledochal cyst in the department of paediatric surgery in a tertiary hospital, Vellore South India.
The last recruitment will be no later than Jan 2014 – this will allow for a minimum follow up period of 6 months.
b. Participants:
Eligibility criteria/ Inclusion criteria:
1. Children who underwent surgery of choledochal cyst in department of paediatric surgery.
2. Patients with only the open technique will be included. Children who undergo laparoscopic choledochal cyst surgery will be excluded from the trial – as outcomes and complication rates will be different and data difficult to interpret.
Exclusion criteria:
1. Patients who have undergone surgery elsewhere and present here for complications of the same.
2. Children with type 2, type 3 and type 5 (Todani’s Classification) Choledochal cysts as the surgical treatment is different and these are very rare presentations.
12 3. Patients who have undergone a hepaticoenteric anastomosis other than a Roux-en –Y hepaticojejunostomy or a Hepatico-duodenostomy. Variations of establishing hepaticobiliary continuity such as a jejunal transposition graft or an appendicular interposition will not be added in the two treatment arms.
c. Variables:
The demographic & clinical profile variables studied will be 1. Age at initial clinical presentation
2. Sex distribution in our patient cohort 3. Choice of hepaticoenteric anastomosis.
4. Presentation of disease (these may overlap) a. Number with pain
b. Number with Fever/ Cholangitis c. Number with a palpable mass d. Number with stones
e. Number with pancreatitis Immediate postoperative outcomes studied will be
5. Length of post-operative inpatient stay.
6. Bile leak if noted. And for how long.
7. Cholangitis if recorded.
8. Wound infections / dehiscence 9. Intestinal obstruction if present
13 The following data will be assessed from the laboratory data
10.Bile cultures if recorded 11.Biopsy reports of our series 12.Total bilirubin levels
Long term outcomes noted at follow up.
(Follow up will ideally be done at 6 months and after )
13.Number of episodes of Cholangitis after initial hospitalization. (Cholangitis defined here as fever with jaundice)
14.Number of episodes of intestinal obstruction – (Defined as Abdominal distension, bilious vomiting with abdominal pain, or the diagnosis of intestinal obstruction by a medical person at the child’s point of care)
15.Presence of gastritis –(Gastritis defined by epigastric pain relieved by antacids or PPI’s or food)
16.Follow up data analysed
a. Normal LFT – Y / N
b. Biliary dilatation on USG – Y/N (pneumobilia is acceptable)
c. HIDA scan –Any holdup of tracer – Y/N
17.Need for a secondary surgery related to the primary operation for choledochal cyst
14 Definitions for terms applied to outcomes –
The following are the definitions followed in this trial to avoid confusion and provide clarity when various terms are being discussed.
Choledochal cyst – defined as an abnormal dilatation of the common bile duct larger than the age appropriate size identified on radiology images – magnetic resonance cholangio-pancreatography (MRCP), ultrasonography or Computerised tomography (CT).
Children – All study participants are to be between infancy to 15 years at the time of first presentation.
Cholangitis – Defined with fever with jaundice. With or without associated upper abdominal pain.
Intestinal obstruction – In this study all children with persistent vomiting, abdominal distension, radiographic evidence of multiple air fluid levels, or a documented clinical impression of intestinal obstruction will be classified to have intestinal obstruction.
Follow up data of intestinal obstruction provided by the local treating physician will also be considered as intestinal obstruction.
Bile leak – as documented in the chart, progress notes or discharge summary.
Post-operative stay- the time in days between the index operation and the date of discharge from the ward.
15
Sample size
It is important to calculate an adequate sample size for an adequate power of the study. Outcomes in choledochal cysts are most importantly the complication rates of the operation. The operation with the better complication rate is the better preferred choice. Other parameters to be considered for calculating sample size could be the operating time and the post hospital stay duration. However these are of less significance to either the patient or the treating surgeon that the short and long term outcomes.
Hence in this trial the combined complication rate was considered as the basis for the calculation of the sample size.
The sample size was calculated with the 2 sample proportion testing method to show and significant difference in combined complication rates in the two groups.
The rates of combined complication rates were 42% in the Hepatico- duodenostomy based on the data from Shimohatakahara 2005 and was 7.1%
combined complication rate of the hepaticojejunostomy group, based on the data from Santore 2010(3,4).
Using this formula a minimum sample size at least 28 patients in each group was calculated. Hence a decision was made to choose 35 patients in each group to allow for missing data or loss to follow up. (Assuming a possible 20% loss to follow up – since many of our centre’s patients travel over 1500 km for treatment)
16 Both groups will be assessed for baseline characteristics and loss to follow up data for other groups can be compared if they are different.
In case there are major differences in baseline characteristics these need to be addressed – either by a test of heterogeneity or by stating the reasons during the analysis and discussion. For example if the baseline age for one group is neonates and the other is in the teenage – there needs to be extreme caution in interpreting the results of the study.
All baseline characteristics will be compared and a test of significance (p value) will be calculated for all dichotomous variables. In case there was a loss to follow up of more than 20%, more patients will be recruited in the arm where there was poor follow up.
17 Materials and Methods – algorithm of the BEACH trial.
35 cases in each arm are taken (Total sample size - 70 patients)
Children with HD (35) Children with Roux en Y –HJ(35)
Demographic data Radiological data
Clinical features
Post-operative follow up
Clinical, radiological, biochemical outcomes. Describe complications in detail
Analysis
Compare the two baseline characteristics – (p value) Compare the complication rates between 2 groups
18
LITTERATURE REVIEW
Choledochal cyst.
Choledochal cyst is described as an abnormal cystic dilatation of the biliary tree. The dilatation of the biliary tree can occur from the intra-parenchymal bile ducts to where it joins the duodenum. It has also been called as biliary cysts. The prevalence is much higher in Asia than the west. The incidence is described as from as common as 1 in 1000 in Asia to 1 in 2 million in the western world(5). The primary treatment for the disease is the surgical removal of the dilated portion to prevent complications of liver failure, reduce pain, reduce recurrent cholangitis and prevent the progression to malignancy. A choledochal cyst is most commonly detected in early infancy and only about 20% are undetected till their adulthood. Better antenatal medicine now allows for earlier diagnosis. Hence in the developed world a good number of diagnoses of choledochal cysts are made in the antenatal period itself.
CLINICAL PRESENTATION
The clinical presentations of choledochal cysts in the west and in Japan have been well documented. There are several case series that describe the clinical presentations and type of cyst, as well as the outcomes for various operations performed.(5,4,6) However the clinical presentation and profile of choledochal cysts in the Indian population has not been well documented. Small case series of various authors provide the Indian data on the disease(7,8). The presentation in literature is varied – and is a spectrum that ranges from the antenatal detection or early neonatal jaundice to the classical triad of fever, right upper quadrant pain and jaundice initially described by Alonso-Lej(9). In the Indian population, cholangitis may not be easily identified by primary care physicians or
19 investigated aggressively. Antenatal scans are rarely done in rural populations and if done – congenital anomalies are hardly ever looked for. This often means children with choledochal cysts are being missed and present late.
Apart from the common presentation of the classical triad, patients can present with mild epigastric pain. Acute fulminant cholangitis can be a presentation that is commonly investigated for and found to be due to a dilated and obstructed bile duct. Stones in the bile duct may also be seen on routine sonography.
Rupture of the choledochal cyst can be an unusual presentation. The rupture of a choledochal cyst is usually described at the cystic duct confluence as this is the weakest point of the biliary system(10).
Pancreatitis is a common presentation of choledochal cysts. The main pancreatic duct is often blocked with stones or thick protein like material. It is not uncommon for children to be treated for a chronic pancreatitis for a while only to be later diagnosed with a choledochal cyst. The pancreas often shows features of chronic pancreatitis – such as atrophy, main pancreatic duct dilatation and stones in the pancreatic duct. Pseudocysts and pancreatic leaks can also be seen.
Richa Lal et al. from SGPGI, Chandigarh published a series of choledochal cysts with complicated presentations in their institution(11). They saw a range of presentations such as malignancy in the biliary tract, gastric outlet obstruction, portal vein thrombosis and hepatolithiasis. These rare presentations were in addition to the more common ones we have discussed earlier.
Sex Ratio:
The ratio girls to boys who develop choledochal cysts have been well studied. It has classically been described as more commonly seen in women with the sex ratio of 2.4 to
20 4:1 female to male patients. Todani and colleagues in their landmark series of 37 cases, where they discussed the classification and operative procedures also noted that the disease was more common in girls(6). This ratio is not the same in the Indian series by U Podder et. al. at PGI Chandigarh where there was a higher male to female ration of 1.3:
1(8). The financial burden of the operation and treatment may possibly mean the ones who present for surgery may vary, especially in a culture where there is less financial commitment for the girl child.
Antenatally detected cysts:
Antenatally detected choledochal cysts have a higher propensity for hepatic fibrosis than those detected later in life. Mei Diao et al. demonstrated better outcomes for children who were randomized to early surgery (earlier than a month) than those who were delayed and taken up for definitive surgery for their antenatally detected choledochal cyst (after a month). They found that grade 3 and 4 hepatic fibrosis was more common in those operated after a month of age (late operation group) and in those who were antenatally diagnosed in the early antenatal period.(12) Antenatally detected biliary cysts can also be variations of biliary atresia and needs to be kept in mind when the diagnosis is made.
Common criteria to suspect choledochal cysts are bile ducts seen on antenatal screening or in the early neonatal period of more than 2 mm in size. Cysts in the abdomen diagnosed on antenatal screening are possibly choledochal cysts, ovarian, mesenteric or duplications. A close post natal watch is important – especially if a cystic biliary atresia is suspected. In a series in Taiwan Lin et al presented that when followed up all children with Common bile ducts more than 2 mm in the neonatal period, 87% of these children’s CBD dilatations resolved spontaneously.(13) These children were diagnosed as transient
21 CBD dilatation (TCBDD). It was found that 80% of TCBDD resolved by 6 months, If the cut off was increased to 3 mm the resolution of CBD size took longer. Also if the maternal age was over 35 years – the bile duct size resolution took longer.
It is therefore important to know that demography of the patients who present for choledochal cyst surgery in a tertiary care centre in south India and see if there are variations with the few case series that exist on this rare disease.
22 Colour plate No 1:
An intraoperative photograph of a choledochal cyst before its excision.
The cyst in this picture is a type 1 cyst and performed in a child with a palpable right upper quadrant mass preoperatively.
23
Etiology
There are various theories about the cause of choledochal cysts. These theories involve the presence of anatomical biliary obstruction and an anomalous pancreaticobiliary malformation (APBJ). These anatomical postulates cause the reflux of the pancreatic secretions into the biliary tree and a weakness of the biliary wall. As noted by Martin, all choledochal cyst theories are caused due to the alteration of the pressure compliance dynamic within the pancreatico-biliary system(14). Babbitt DP described a theory based on a series of preoperative and intraoperative cholangiograms of a common bile duct that enters the main pancreatic duct well before the sphincter of Oddi(15–17). He explained that the reflux of pancreatic secretions is the cause for chronic inflammation and dilatation of the common bile duct. The anomalous pancreatico-biliary junction is the most plausible and demonstrable defect in case series of choledochal cyst – however, the presence of such an anatomical anomaly is not absolute in every choledochal cyst. An abnormal pancreatico-biliary maljunction is seen only in 60- 90% of all choledochal cysts. Also there is no information on patients who have the pancreatico-biliary anomaly but do not have a choledochal cyst. Another counter to this theory is that neonatal pancreatic acini do no produce adequate pancreatic enzymes to cause an antenatal detected choledochal cyst. A long common channel can vary between 10 mm to 45 mm in literature. Komi et al. described 3 types of long common channel – the first is a right angle union without an accessory pancreatic duct. The second is an angled union without an accessory pancreatic duct. The third is a right angled or an angled anastomosis to an accessory pancreatic duct(18).
24 There is little to suggest a genetic predisposition to the disease, as choledochal cysts do not run in families. An abnormal epithelial proliferation of the biliary tree has been proposed by Yotsuyangi. An unequal cellular proliferation of the biliary tract can occur during early foetal life before the bile duct completes it canalization. (5)
Other theories for the etiology of choledochal cysts have been described. Tylor et al.
proposed a viral damage to the ganglion cells to the biliary tract may cause a motility disorder of the biliary system and dilatation – these findings are based on reoviral RNA levels in the biliary system(19).
Protein plugs, embryological webs at the sphincter of Oddi level have all been theories of choledochal cyst formation.
CLASSIFICATION
Alonso-Lej F et al. in 1959 were instrumental in describing the first series of choledochal cysts, where they analysed 94 cases and devised a classification system based on their series(9). Todani in 1977 added a 5th type to the existing 4 described by Alonso-Lej to create the most common classification system in use today. This classification has been demonstrated pictorially below. They classified type 1 cysts as those with a cystic dilatation of the extra hepatic biliary tree. Type 1 cysts do not involve the intrahepatic biliary tree. These were the most common in theirs as well as subsequent series’ of biliary cysts. The type 1 choledochal cyst can be either spherical or fusiform in appearance.
There has been a sub-classification as type 1a, type 1b and type 1c. The type 1 cysts include from 50% to 85% of all choledochal cysts and are the most common type of cyst encountered.
25 Type 1a cysts are usually associated with an anomalous pancreatico biliary junction (APBJ). In these cysts the common bile duct and hepatic ducts may be involved in a cystic dilatation. The cystic duct and gall bladder can be seen to arise from the dilated bile duct. Type 1 b cysts are usually a focal segmental dilation of the common bile duct.
This is usually distal to the insertion of the cystic duct. These are not associated with an anomalous pancreatic biliary junction (APBJ). Type 1c cysts are smooth and fusiform and can extend from the extra hepatic bile ducts up to the pancreaticobiliary junction.
Type 1c cysts are associated with APBJ malformations.
The type 2 cysts were classified as a cystic out-pouching from the bile duct in the extra- hepatic bile duct. Many believe these cysts are very rare and hardly are ever seen. Type 2 cysts are seen in only 2% of all series of choledochal cysts. Some authors believe that segmental type 1 cysts are often mislabelled as type 2 cysts. The type 3 cysts are rare and seldom seen in surgical series as they are unusual and can be treated endoscopically, without a major laparotomy. These are a cystic dilatation of the distal bile duct and also called a choledchocoele. Type 3 cysts or choledochocoele’s are seen in 5% of all biliary cysts. In 1977, Todani et. al. published their series of 37 cases from 1960 to 1976 and added to the classification if Alonso-Lej, based on new anatomical presentations in their series. They did not have any type 2 or type 3 cysts in their series, but introduced the following modifications – type IVa with an intrahepatic bile duct cysts as well as a choledochal cyst. Multiple cysts in the extrahepatic bile duct were classified as type IVb.
Type 4 cysts are seen in up to 20% of choledochal cyst series’. The operative strategy is similar in a type 4b as in a type 1 cyst. Purely intrahepatic disease was classified as type V.(6) or Caroli’s disease. Less than 2% of all biliary cysts have Caroli’s disease and the risk of malignant transformation remains high in this group.
26 Apart from the above well-established classification Shah et.al. has described a 6th type of cyst. A type 6 choledochal cyst is where the dilatation is limited to the cystic duct. This type can be easily treated with a cholecystectomy alone, no hepaticoenteric anastomosis is required.(20,21)
Ronald F Martin of the Marshfield clinic has suggested a simplified classification of the biliary cysts. It has been published in the surgical clinics of North America, 2014 (14).
The classification divides the choledochal cysts to 3 groups.
i) Intrahepatic cysts – If the child is a good surgical candidate – a segmental resection. Else a liver transplant for Caroli’s disease.
ii) Extrahepatic cysts – Resection of the extrahepatic cyst and restoration of the biliary enteric continuity
iii) Intraduodenal cysts - In small cysts de-roofing of the cyst by endoscopy or trans-duodenal approach.
27 Figure 1 – Classification of choledochal cyst. See text for details of each type.
28 Choledochal cysts warrant surgical treatment in view of their malignant potential and the fact that they tend to cause symptoms. Voyles et.al have described the malignant potential to be related to the age at the onset of symptoms. They felt that children presenting before the age of 10 years have a ‘minimal’ risk (0.7%), in contrast to those that present in the 11-20 age group where this risk increases to 6.8%, and in older subjects the risk is increased to 14.3%.(22)
INVESTIGATION ALGORITHM FOR WORKUP OF CHOLEDOCHAL CYST
Children who have choledochal cysts are usually diagnosed by a clinical suspicion based on the history or after an ultrasound- antenatal or an incidental ultrasound for abdominal pain or jaundice.
Imaging:
The mainstay of diagnosis rests with the diagnosis of a dilated common bile duct.
Ultrasound is the most commonly used modality of investigation when screening or looking for a choledochal cyst.
However ultrasound has its limitations – often a cyst in the abdomen is large, it cannot be made out from which organ it arises. Also a type 3 choledococoele cannot often be seen by a conventional ultrasound machine. Ultrasonography is limited by obese habitus, gaseous abdomens and uncooperative children. Tenderness in the abdomen may ensure that a child does not allow the sonologist to complete a satisfactory screening.
A common bile duct of more than 7 mm is considered as a high likelihood of a choledochal cyst, the accuracy reaching 71% and if 10 mm is used as the cut off the
29 accuracy is 97%.(23). Neonates have been closely followed up if their CBD is more than 2 mm as the normal CBD size in the first few months is about 1.3 -1 .7mm.(13).
The child with a suspected choledochal cyst is investigated with a thorough ultrasound and if required a MRCT – to determine anatomy of the biliary tree. Abnormalities of the pancreatico biliary junction can be made out in a well performed MRCT. Since the MRI delineates fluid there are certain important factors to keep in mind. The child needs to be fasted. Fluid within the stomach may present with an overlap of the 3 dimensional reconstruction. Also MRI scans often require a small infant to be anaesthetised for the duration of the study to prevent them from moving.
Rarely an Endoscopic retrograde cholangio-pancreatography (ERCP) is performed – usually after an episode of pancreatitis or when a type 3 choledochal cyst is suspected or identified. The choledochocoele or type 3 choledochal cyst is both diagnosed and can be treated by endoscopy by deroofing the wall into the duodenum. ERCP can rarely worsen an episode or trigger an episode of pancreatitis and needs to be done cautiously. ERCP can also be used to therapy stones in the bile duct by stent placements and sphincterotomy.
A rare test that can be mentioned for completion is a percutaneous transhepatic cholangiography. This is largely obviated by the intraoperative cholangiogram, though may have some use in the nonoperative diagnostic algorithm in very sick children. The same sitting can be used to perform a liver biopsy – in case biliary atresia was suspected.
30 Nuclear scintigraphy:
Another investigation that has a limited role and can be performed for the diagnosis of a choledochal cyst is the HIDA scan. The technetium 99 Hepatobiliary Imino-diacetic acid nuclear medicine study is a useful study as the dye is taken up by the functioning liver cells (hepatocytes) and is excreted unconjugated into the bile. There are certain criteria to determine or diagnose a choledochal cyst by HIDA. A cystic lesion in the right sub hepatic region when diagnosed by ultrasonography can be shown to be communicating with the biliary system by HIDA. More than 8 out of ten choledochal cysts will show an ovoid or spherical photon deficient area that gradually accumulates tracer over the next 2 hours. This region will then show a delay in the clearing of the activity for more than 24 hours. A non-filling structure needs to alert the surgeon that the diagnosis of a choledochal cyst is unlikely. The HIDA scan requires the technical expertise, equipment and exposes the child to radiation. It is reserved in cases where the diagnosis of a cystic structure in the right upper quadrant need further clarification.
Blood investigations:
Blood investigations for a child with choledochal cyst includes a complete liver function test – as long term obstruction leads to liver failure. Also children with choledochal cysts are usually jaundiced at presentation – either clinically or biochemically. We perform a prothrombin time with INR and routinely administer vitamin K1 to all children who are planned for choledochal cyst surgery as the Vitamin K dependant factors are usually deficient due to the obstructive jaundice that occurs with biliary cysts. The serum amylase and lipase are also checked as often children with choledochal cysts present with a pancreatitis.
31 In case a pancreatitis is diagnosed either by the symptoms or by biochemistry. The child is also worked up to assess the severity of pancreatitis and whether the pancreatitis will benefit from a surgical drainage. There have been children whose choledochal cyst surgery has been combined with a lateral pancreaticojejunostomy (LPJ) to decompress a chronic pancreatitis.
32 SURGERY FOR CHOLEDOCHAL CYST
The surgical options have evolved from performing early cyst drainage and a delayed definitive operation to the current practice of early complete excision with reconstruction of an adequate biliary enteric anastomosis. There is still a role for drainage of the cyst in sick and moribund patients, though data from several retrospective studies have shown that as far as possible a complete excision and hepaticoenterostomy is associated with far less morbidity and mortality(24). In difficult operations, where peri-cystitis causes concerns for safety to preserving vessels in the portal triad, the operative modification described by Lilly et al.(25) can be adopted. Here the choledochal cyst mucosa is excised without damage to the posterior wall of the cyst and hence keeping safely away from the portal vein and hepatic artery. Following successful excision of the mucosa, a hepatico- enteric anastomosis is performed. This technique eliminates the malignant potential of the mucosa and ensures anatomical safety for the creation of a patent hepatico-enteric anastomosis.
A good hepatico-enteric anastomosis requires no tension and must provide patent drainage of the proximally unaffected biliary tree. This may be by way of either a hepatico-duodenal or a hepatico-jejunal anastomosis. The hepatico-jejunal anastomosis is done by way of a roux loop of jejunum or the jejunal interposition that drains the biliary tree into the duodenum. There is no consensus regarding the ideal hepatico-enteric anastomosis. The ideal hepaticoenteric anastomosis presents itself as a question of surgical equipoise. There are various advantages to each procedure and the cited drawbacks are listed below. However, most series of comparisons are small. No such comparison has taken place in the Indian population. The complications of each procedure also are difficult to compare, since they are different. Hence the combined
33 complication rate is taken when comparisons are drawn between the types of hepatico- enteric anastomoses. The two commonly accepted methods to establish hepaticobiliary continuum are the Hepatico-duodenostomy and the Roux-En-Y hepatico-jejunostomy.
Hepatico-duodenostomy:
The operative steps of a Hepatico-duodenostomy are as follows. The abdomen is opened by a right subcostal incision. It is often not required to extend the incision across the midline. Some centres perform a midline incision for older children. The initial steps of the operation involve the identification and excision of the choledochal cyst. They are common irrespective to the type of anastomosis employed. The peritoneum over the gall bladder and the cyst is grasped and incised to release the gall bladder, common bile duct and the left and right hepatic duct. The gall bladder is released initially so as to perform an intraoperative cholangiogram from a small puncture on the gall bladder. A culture of the bile can be taken at the time the gall bladder is punctured. This helps to guide the future postoperative antibiotic plan if the child develops cholangitis and postoperative fever.
An intraoperative cholangiogram is done with a snugly fitting No 5 infant feeding tube or a scalp vein needle into the biliary system. It is important to flush out the air in these catheters to avoid artefacts due to air in the biliary tree that may look like stones. The following information is important to look for while performing an intraoperative cholangiogram.
i) The type of choledochal cyst can be confirmed. This will help determine the level of the normal bile duct and the abnormal biliary dilatation.
34 ii) The involvement of the hepatic ducts – to determine the various type 4
choledochal cysts
iii) The presence of accessory ducts and their relational anatomy before the bile duct is dissected and divided.
iv) The presence of an anomalous pancreatico-biliary junction or a long common channel. This is useful for a study of why the choledochal cyst occurred and the persistent long common channel is a reason that many children continue to have pancreatitis in the future.
v) The intra operative cholangiogram will help to identify radiolucent areas suggestive of stones in the system – these need removal by careful flushing to avoid problems later.
The cyst is next dissected, taking care so as to not damage the hepatic artery and the portal vein that lie in close relation to the cyst (behind it). In a grossly dilated cyst, the dissection can be made easier by opening the cyst. Once the cyst is open – the superior portion if followed superiorly to the highest part of the cyst and the lower portion is followed to where the cyst enters the pancreatic tissue. The cyst needs to be followed for a safe distance into the pancreatic tissue. The cyst is tied off as low as possible into the pancreatic parenchyma – but not too low so as not to damage the pancreatic duct.
Superiorly the cyst is dissected up to the normal common hepatic duct or at times may need to divide the right and left hepatic ducts individually. Care is taken not to leave any choledochal cyst mucosa behind. Also the dissection in the portal triad is just over the hepatic artery and portal vein. Utmost care is taken to avoid any damage to these structures.
35 Both the superior and inferior aspects are flushed with copious amounts of saline to flush out and dislodge any stones in the biliary system. Any retained stones are often a cause of grief at a later date.
Till this point the dissection has only involved the process of removal of the choledochal cyst. These steps are common to both types of hepatico-enteric anastomosis. After the choledochal cyst has been excised, what remains is to restore the continuity of the biliary drainage.
If the hepatico-duodenostomy is planned, the duodenum is released and it is seen if the duodenum can be easily brought up to the common hepatic duct. Release of the duodenum is performed by a cattle’s manoeuvre – where the right hepatic flexure of the colon is released. The duodenum is then dissected free and can be Kocherized by dividing its retroperitoneal adhesions and bands to the lateral peritoneal wall.
The site for the hepatico-duodenal anastomosis is usually chosen from the antero-lateral portion of the 1st and 2nd portion of the duodenum, taking care not to stretch the bile duct or injure the pancreatic duct. A small incision is made on the first part of the duodenum along the length of its course and this is held between silk or catgut stay sutures. This incision line is left to lie without any tension close to the severed bile duct. An interrupted biliary duodenal anastomosis is taken first in the center and then at the corners. These sutures are usually interrupted and taken with 4-0 Vicryl. The posterior wall is then completed with as many sutures that are required for an integral anastomosis by the halving suture technique. Knots of these absorbable sutures are left within the lumen of the duct.
36 Next the anterior wall is then closed with sutures that keep the knot outside the biliary system. If the draining lumen is sufficient and no leak detected after a few minutes of observation, a small tongue of omentum can be wrapped around the anastomosis. A peritoneal drainage is placed in the right sub hepatic space either with a corrugated rubber drain or a closed abdominal drain kit. Such a drain may be removed after the child starts to feed.
The hepaticoduodenal anastomosis is simple – there is only a single anastomosis between the duodenum and the biliary tree. The bile drains into the duodenum as it would have into the second part – except the sphincter of Oddi is now lost and the site of drainage is more proximal than the intended natural position.
The Hepatico jejunal anastomosis is created by the creation of a standard Roux –en- Y loop. The jejunum is divided about 15 to 20 cm’s from the duodenojeunal (DJ) flexure.
The loop distal to the division is taken through the mesocolon, and the luminal end of the jejunum could either be anastomosed directly as an biliary end to bowel end anastomosis (end to end). Alternatively the end of the jejunum could be closed and a biliary end to jejunum side (end to side) anastomosis can be performed. The Roux-en-Y loop needs completion by anastomosing the proximal jejunum at an appropriate spot on the jejunum after the loop. A 25-30 cm limb from the biliary anastomosis to the jejuno-jejunal anastomosis allows for a potential length to prevent reflux of the gastroduodenal content into the biliary anastomosis. Some centres also create an ‘access loop’. The access loop is an operative strategy that fixes a circumference of the Roux-en-Y loop to the peritoneal wall with a radio opaque ring like metal wire, so that later in the future the patency of the hepatico-jejunal anastomosis can be easily tested by radiological access. The access loop
37 can be accessed in a digital subtraction radiology suite (DSA) by a percutaneous puncture and a series of catheter dilatations to access both the proximal and the distal anastomosis.
Hepatico-duodenostomy is a technically easier operation as there is only a single anastomosis after the excision of the choledochal cyst. The divided biliary tree is anastomosed to a new opening on the antero-lateral duodenum. This has the added advantage of easy access for intervention if there is a stricture of the biliary enteric anastomosis that requires dilatation or documentation by endoscopy. There is a theoretical increase in the possibility of bile reflux gastritis in patients with a hepaticoduodenostomy, this has been documented with endoscopy and biopsy(3). Barium swallow studies show reflux into the biliary tree, raising the possible concern of higher cholangitis rates in this procedure; however literature does not support this worry. The rates of post-operative cholangitis are similar in both hepaticojejunostomy and hepatico- duodenostomy series. (4)
The advantages of the Roux en Y hepaticojejunostomy are the reduction of duodenal content refluxing into the biliary system. This translated to a theoretical reduction in cholangitis and possible long term malignancies. Shimotakahara et. al 2005(3) have stated their preference for the Roux en Y procedure as they were able to demonstrate endoscopically proven gastritis in 4/12 (33%) of their patients who were followed up after a Hepatico-duodenostomy. There are certain disadvantages as well with the hepatico-jejunostomy as well that need to be stated. The first and most obvious is the fact that the procedure is more time consuming and requires an additional suture line on the bowel as compared to a simple hepaticoduodeal anastomosis. The second is that there is a recorded larger incidence of intestinal obstruction since the bowel is handled and operated upon. The same series reported a 7.1% incidence of both cholangitis and
38 intestinal obstruction in those who underwent the excision of the choledochal cyst and underwent the hepatico-jejunal anastomosis. Laparoscopic hepaticojejunostomy has been performed with the excision of choledochal cyst, and the largest Indian series by Palanivelu et al, have promising results with the procedure.(26). A recent study more along the lines of comparison of the two biliary enteric anastomosis was done in the University of Hong Kong by Yeung et al(27). The outcomes of Hepatico-duodenostomy were compared with Hepaticojejunostomy in operations performed laparoscopically. It was found that the outcomes of hepatico-duodenostomy were not inferior to the hepaticojejunostomy anastomosis. The numbers were smaller than what we have recruited in the BEACH trial – only a total of 31 patients were chosen. 20 who had undergone hepaticojejunostomy and 11 who underwent hepatico-duodenostomy. It was found that the operation times were shorter, ICU stay and the post-operative hospital stay was shorter in the hepatico-duodenostomy group. The hepaticojejunostomy group also required 5 additional operations for complications and a stricture of the anastomosis.
Other operations attempted have been substitution of the common bile duct with a jejunal loop and the use of the appendix as a conduit. Jejunal substitution was first described by Lopez Gibert in 1963 and in this technique various lengths of jejunum for substitution have been described – the early series used shorter lengths of jejunum 20- 25 cm for the interposition – however Pappalarado and colleagues modified the loop to a length of 40- 50 cm to decrease the reflux of the duodenal content into the biliary system. The rationale for a jejunal interposition is the development of peptic ulcers in 1.7 to 22% (28,29) of Roux-en-Y hepaticojejunostomy patients and up to 52% in animal studies(30). It has been postulated that this occurs since the gastric acid is not neutralized by the duodenal content after the construction of the Roux-en Y loop.
39 Figure 2- A line diagram that pictorially depicts the two types of anastomosis in choledochal cysts. The hepaticojejunostomy is shown as more versatile for example when
the right and left hepatic ducts need to be anastomosed to bowel.
40
Systematic review on the two methods of hepaticoenteric anastomosis
In the planning of the BEACH trial, the COCHRANE collaboration was contacted to see if it was possible to conduct a systematic review to go with the results of the BEACH trial. We were refused as it was pointed out at the time that since there was no randomised control trial between hepaticojejunostomy and hepaticoduodenostomy and the COCHRANE collaboration was not interested in allowing the relase of a protocol and initiating a meta analysis without a Randomised control trial (RCT).
SK Narayan and colleagues, Westmead Australia with Yong Chen and colleagues at the KK Womens and Childrens hospital, Singapore have put together a systematic review of the availiable cohort studies that compared the two anastomosis – Hepatico duodenostomy and hepaticojejunostomy(31). Their inclusion criteria were fairly strict – the trials were to consist of only these two biliary enteric anastomosis (HD and HJ).
Trials that were unclear or included a jejunal interposition graft were not accepted.
Institutions that were publishing multiple papers with overlappng data were only considered for their most recent data. All papers needed to describe the individual complication data in each group – such as strictures, adhesive obstuctions, cholangitis and reflux/gastritis.
These were the findings of the systematic review. Bile leaks – there were no difference in the leak rates between the two groups. HD had a 2.1% rate and HJ had a 2.9% rate with a p value of 0.46). Cholangitis – When the data was pooled the HD and HJ groups showed no difference in the rates of cholangitis (HD- 2.47% and HJ 2.42%). However, when individual studies were looked at there was a statistical difference between the HD and
41 HJ groups between the individual trials. Two trial favoured HJ and three trials favoured HD. This data helps demonstrate the advantage of pooled data in the meta analysis method.
Anastomotic stricture- Three studies had data about anastomotic strictures. The rate of stricture was 1.21% (4/330) in HD and 1.47% (3/204) in the group of hepaticojejunostomy. It was not possible to demontrate that anastomotic strictures were more common in either type of anastomosis. Reflux / gastritis – four studies looked at gastritis – it was assessed either clinically or by endoscopy. What stood out was the hepaticojejunostomy group had no children with gastritis or reflux, while 18 out of 306 children had reflux. This was found to be the only significant clinical outcome between the two biliary enteric anastomosis. (The p value in this particular instance was 0.002).
The intestinal obstruction rates though described as high in many trials that describe hepaticojejunostomy were 5.12% in comparison to none in the hepaticoduodenostomy.
Though these numbers are large, there was no significant difference in either those who developed an obstruction or those that required a reopeation in both the groups.( p values 0.39 and 0.20 respectively)
Other parameters looked at though not strictly the outcome rates were length of hospital stay and operative time. There was a significant difference in the average post operative stay where the average time of stay and resumption of normal diet was 4.8 days in the HD group in comparison to 6.1 days in the HJ group. The time nasogastric decompression were however similar in both groups.
The systematic review finally concluded with caution that the hepaticoduodenostomy is a viable alternative to the hepaticojejunostomy – particularly in the laparoscopic era. The
42 results are described to be interpreted with ‘caution’ as the number of the studies in the meta- analysis are few and the quality of the studies are insufficient. The quality of a systematic review that relies only on cohort trials are bound to show bias – though limited by the fact that there are no randomised controlled trials.
If the results of the BEACH trial as described later are looked at,we have found in essence a similar constellation of findings as the systematic review. The systematic review did not however mention a combined complication rate as we have.
43 Table 1. A table of compiled studies that list early and late complication rates and the various surgical hepaticoenterostomy offered is given below.
Study (Year)
combined complicat ion %
Leak rate Bile gastritis
Intestinal obstruction
Stricture Cholangitis Wound dehisence
Min Ju Li (2002)
HJ 9.3%
10/172 (5.78%)
1/172 (0.5%)
1/172 (0.5%)
3/172 (1.7%)
Shimotakaha ra 2005
HD 42%
4/12 33%
Shimotakaha ra 2005
RY HJ 7.1%
2/28 (7.1%)
Saing1997 RY HJ 7% - 5%
Yamataka 1997
RY HJ 12.5%
3%
Elalaby 2005 RY HJ 18%
7% 11%
Elalaby 2005 HD 11%
11%
Santore 2010 HD 7.6%
5% 2.6% 0%
Santore 2010 RY HJ 20%
10% 5% 5%
Yeung 2014
HJ HD
5%
Nil
1/20 (5%) Nil
5
reoperations Nil
44
Risk of malignancy in choledochal cysts
The reason choledochal cysts require complete excision is that the biliary tract is at a high risk of malignancy(32–34). Older procedures like a drainage or incomplete excision of the cyst wall predispose to this.
A review of literature regarding the formation of cholangiocarcinoma in the biliary system revealed a pooled data from 32 patients. This review was done in 2013 by Ohashi et al(35). He looked at data from previous publications – usually single case reports and data from his own institution. He compiled the case reports and found the following distribution of bile tract malignancies. He then compared the literature to what he found in his single institution practice. The results have been tabulated below.
Literature review Ohashi series 2013
Average Age at cyst excision 28 (0.4 – 68) 22.75
Average age of malignancy detected
41 (18 – 70) 43.5
SITE OF MALIGNANCY
Hilar 17 1
Intrabiliary 9 2
Pancreatic 6 1
45 These changes and results seem dismal. The outcomes for these patients too were far from satisfactory. There were no 4 year survivors from any of these patients.
It is obvious that the entire disease with the reflux of the biliary and pancreatic secretions causes a field change and the intrahepatic, hilar and pancreatic ducts are not spared. Long term follow up with surveillance and a high index of suspicion is indicated for all children who undergo choledochal cyst surgery, even more so for those with abnormalities of the pancreatico – biliary junction (APBJ).
Todani et al reported that choledochal cysts that are treated with internal drainage operations(36) are usually related to dysplasia and metaplasia of the epithelium. The risk of malignancy is higher in type 1 and type 4 cysts and not common in type 3 and 2 cysts.
Ohashi et al applied the Kaplan-Meier method to calculate the possibility of developing a biliary tract malignancy in patients after choledochal cyst excision. They were able to suggest that there is a 1.6% risk at 15 years, 3.9 % at 20 years and a 11.3% risk at 25 years.
46 Why do this study?
Surgery for choledochal cyst is associated with an over 90% success rate. The short term complications are anywhere between 2.5% to 27%, and the long term complications are up to 25%. The method of hepatico-enteric anastomosis largely depends on the institution where it is practiced and there remains a controversy of surgical equipoise.
The choice of restoration of biliary continuity is commonly debated in various forums, conferences and meetings. One of the largest series in India by Dr Richa Lal et al at the SGPGI, Chandigarh that has published a series of 144 patients (11), when asked about the choice of choledochal cyst hepaticoenterostomy in their institution – replied that they chose hepaticojejunostomy by convention.
In our institution, we have done a fair number of both procedures (Hepatico- duodenostomy and Hepatico-jejunostomy) over the years. This is a rare disease and numbers of the patients who come for surgery of choledochal cyst are few. However if well conducted, a study of existing data in our population, when compared statistically – may throw some light on our results and help frame future recommendations for our surgical practice.
Also in the era where minimally invasive surgery is taking off – clarity about the outcomes of hepatico-duodenostomy will help simplify and reduce time taken in laparoscopic choledochal cyst surgery. Many see the resorting to a hepatico- duodenostomy as an easy way out – which may be far from the truth.
This study can also form the basis of a well conducted randomized controlled trial for choledochal cyst surgery that compares the two types of biliary enteric anastomosis.
47
Results & Analysis
Recruitment of cases for the BEACH trial.
The BEACH (Biliary Enteric Anastomosis in Choledochal cyst) was passed after scrutiny by the Institutional review board (IRB Min No: 8670, dated 19.2.14). The required sample size was collected by a careful study of the operation records and discharge summary records of the department of Paediatric surgery. Patients were collected both prospectively and retrospectively. However patients were not recruited beyond January 2014 to allow time for an adequate follow up.
The decision for which biliary-enteric anastomosis (Hepatico-duodenostomy Vs.
Hepatico- jejunostomy) was made was not influenced or directed by the study – as the aim was to describe and analyse outcomes in each type of study without randomizing the patients into either arm. The type of anastomosis was left to the senior surgeon’s personal preference in each case.
35 cases of each type of biliary-enteric anastomosis – Hepatico-duodenostomy and Hepatico-jejunostomy were chosen. Cases where other modalities were used – such as the Lilly’s procedure, drainage of cysts and a jejunal conduit were excluded from our sample collection. Laparoscopic surgery though performing the required anastomosis for our comparison was excluded from the study as there was a potential for bias in their selection and outcomes.
48
Baseline characteristics:
The two groups of patients chosen for the two surgical comparisons were comparable in their baseline characteristics. (See table 1).
The sex ratio is a carefully studied baseline characteristic in choledochal cysts. Various countries and series have different sex ratios described. The Hepatico-duodenostomy group had 12 boys and 23 girls, while the Hepaticojejunostomy group had 8 boys and 27 girls.
Table 2. GROUP vs. SEX – CROSS TABULATION
Sex
Total Female Male
GROUP HD n/35 count 23 12 35
% within group 65.7% 34.3% 100.0%
HJ n/ 35 count 27 8 35
% within group 77.1% 22.9% 100.0%
TOTAL Count 50 20 70
% within group 71.4% 28.6% 100.0%
The sex ratio is an important detail when characterising the distribution of the disease with the male and female ratio for different populations. The p value was 0.290, which was not significant for a difference. The ratio of male to female patient in our population was 1:2.5.
49 The age of presentation of symptoms in the hepatico-duodenostomy group was an average of 53 months with a range of 3 months to 192 months at the first presentation.
The hepaticojejunostomy group had an average age of presentation of 69 months with a range of 4 months to 156 months. (P value 0.163)
The clinical presentation of the patients in both the groups are described below. It stands to reason a Hepatico-duodenostomy being a technically easier operation is performed in more stable and straightforward anatomy, whereas the hepatico-jejunostomy may be sometimes reserved for a more hostile anatomy and sometimes the sicker patient. In one instance a child’s parents specifically requested the hepaticojejunostomy based on their personal reading and choice. In most other cases the operation was chosen based on various factors – such as inability to bring the duodenum for a satisfactory anastomosis.
The need to anastomose the right and left ducts separately or in some cases an accessory duct that required a separate anastomosis often required a Hepatico-jejunal roux en Y loop to be created.
50
Baseline characteristics table:
Table 3
Baseline char. Hep Duo Hepatico-jejunostomy P value
Age in months 53 69 0.163
Sex (F/M) 23/12 27/8 0.29
Mass RUQ 3 2 1.00
Presenting complaint
Hep Duo Hepatico-jejunostomy P value
Pain 26/35 24/35 0.584
Jaundice 13/35 17/35 0.337
Antenatally detected
2/35 0/35 0.47
Cholangitis 4/35 4/35 1.00
Vomiting 6/36 3/35 0.172
Pancreatitis 2/35 3/35 1.00
Other presenting features
-Spontaneous rupture CBD in past
-Perforation and peritonitis -Post Lap chole pain
-Forme Fruste with ERCP and stenting for stones
0.05
51 Clinical presentations
Palpable upper quadrant mass.
The typical presentation of a right upper quadrant mass is described in many a text book.
In our series of children – we found that that this so called ‘classical presentation’ as a clinical findings is unusual only 3 of our 35 in the hepatico-duodenostomy group and 2 out of the 35 children in the hepaticojejunostomy group had a palpable mass in the right upper quadrant.
This means that a palpable mass is seen in only 7% of our patients who present for surgical correction of a choledochal cyst.
Pancreatitis:
A radiological, biochemical or clinical suspicion of pancreatitis was made in 2 out of 35 of the hepatico-duodenostomy group and 3 out of 35 children in the hepaticojejunostomy group. The average serum amylase for children without proven pancreatitis in the hepatico-duodenostomy group was 157 IU and in the hepaticojejunostomy group was 235 IU. The drawback in the analysis of the serum amylase levels are that they are not routinely ordered for all patients preoperatively. Hence it would be of poor power to draw conclusions on the amylase levels of patients in the two groups.
Other rare presentations of choledochal cyst:
It is interesting to note that all the unusual presentations of choledochal cyst – not mentioned above were seen in the hepaticojejunostomy group of patients. It is this reason that this is the only significant baseline characteristic that is different between the 2 groups of patients. The hepatico-duodenostomy group, other than the common and
52 standard clinical presentations – had no rare or unusual presentations. This may explain later why the hepaticojejunostomy group had a higher complication rate.
The unusual presentations are as follows:
One child gave a history of a laparotomy for a spontaneous rupture CBD in past. This was operated in our institution. A laparotomy was performed that showed a rupture of the CBD in infancy – this was repaired – and later the child was diagnosed with a choledochal cyst. Another child had a perforation of the choledochal cyst and presented with peritonitis. It was at this sitting that a hepaticojejunostomy was performed after removing the choledochal cyst. A third child had undergone laparoscopic cholecystectomy elsewhere. There was a persistent pain, when evaluated revealed a choledochal cyst that was missed on the initial surgery and workup. The last child who had an unusual presentation was labelled forme – fruste, as the child presented with repeated abdominal pain. The pain was suspected to be due to stones in the biliary tree (choledocholithiasis) – an ERCP and stenting was done. At ERCP and MRI a 6 mm bile duct was diagnosed to possibly be a choledochal cyst – and the parents, being doctors specifically requested for a hepatico jejunostomy.
Size of the Cyst:
The size of the choledochal cyst is an important indicator of the type of disease. The forme fruste variety choledochal cysts may behave differently from massive cysts that have a lot of peri-cystic adhesions. Hence it is important to know if both groups dealt with a similar size of cyst – so that the results are comparable.
