The Role of Prophylactic Endoscopic Sphincterotomy for Prevention of Postoperative Bile Leak in Hydatid Liver Disease: A Randomized Controlled Study

Abstract

Background:

Bile leak is the main cause of morbidity and mortality after surgery for hydatid liver cysts. Aim was to assess the role of prophylactic endoscopic sphincterotomy (ES) in reducing postoperative bile leak in patients undergoing partial cystectomy.

Methods:

Fifty-four patients with hepatic hydatid cyst met inclusion criteria, 27 were excluded or declined to participate. Twenty-six women and 28 men (mean age 44.6 ± 10.1, range: 22–61 years) were randomly assigned to either group I with ES (n = 27) or group II without ES (n = 27).

Results:

Demographics and clinical, laboratory, and radiological characteristics of cysts were not statistically different between two groups. Group I had a significant decrease in bile leak rate compared with group II (11.1% versus 40.7%, P = .013), with significantly shorter duration of hospital stay (P < .0001). Biliary fistula in group I had significantly lower daily output (100 mL/day versus 350 mL/day) with gradual reduction till stoppage of leak in 3–4 days without intervention. Biliary fistula in group II had a significantly higher need for biliary intervention through postoperative endoscopic retrograde cholangiopancreatography with ES compared with biliary fistula in group I (^FEP = .002), with significantly longer mean time of fistula closure (P = .011) and longer time to drain removal (P < .0001). Nonbiliary complications were comparable between two groups.

Conclusion:

Prophylactic ES provides significant reduction in postoperative bile leak rate with shorter hospital stay after partial cystectomy of hydatid cyst. Biliary fistula in patients with ES has significantly lower daily output with shorter time of drain removal and shorter time to closure than patients without ES.

Introduction

Hydatid disease or cystic echinococcosis is caused by adult or larval stages of tapeworms that belong to the genus Echinococcus. It represents an increasing public health and socioeconomic concern in many areas of the world^1–3 and is currently considered an endemic zoonose in the Mediterranean region. Infection with Echinococcus granulosus primarily affected the liver through the development of one or more unilocular hydatid cysts. In rare occasions, other organs of the body are affected.

One of the common and serious complications is rupture of hydatid cyst into the intrahepatic bile ducts, which occurs in 2%–42% of cases.^4–7 The resulting cystobiliary communication can be either occult or frank.⁸ The frank type occurs in about 5%–17% of the cases and usually presents itself clearly in the preoperative period through patients' complaint, clinical status, and laboratory and radiological finding.^9,10 The occult type occurs in 13%–37% of cases,¹¹ and is unlikely to be identified preoperatively as it can be asymptomatic and usually presents itself as bile leak, external biliary fistula, biloma, biliary peritonitis, cavity infection, or abscess in postoperative period.^11–13

Postoperative bile leak represents the main cause of morbidity and mortality after conservative surgical procedures (partial cystectomy) performed for the treatment of hydatid liver disease.^14,15 Radical surgery, including liver resection or pericystectomy, is associated with a much lower rate of bile leak due to eradication of chances of cystobiliary communications through applications of the principles of liver resection.^16–18 However, the latter requires operative expertise and not always feasible in patients with hydatid liver disease due to size, location, or multiplicity.¹⁹

Although there are several surgical techniques described for intraoperative detection of these minor cystobiliary communications ranging from exploring the cyst cavity under magnification with placement of white surgical gauze to cholecystectomy, intraoperative cholangiography, and common bile duct explorations, however, 10%–32% of cases eventually develop bile leak.^20,21

The therapeutic role of endoscopic retrograde cholangiopancreatography (ERCP) is established in the preoperative management of patients with frank intrahepatic biliary rupture, as well as in the postoperative management of bile leak after partial cystectomy due to unidentified minor cystobiliary communications. However, the role of prophylactic endoscopic sphincterotomy (ES) for prevention of postoperative bile leak in patients with hydatid liver disease treated by partial cystectomy remains controversial.

Therefore, the aim of our prospective study is to assess the role of preoperative prophylactic endoscopic sphincterotomy in reducing bile leak rates in patients undergoing partial cystectomy for hydatid liver disease versus those without endoscopic sphincterotomy.

Patients and Methods

This study was conducted at the Surgery Department, Alexandria Main University Hospital, Egypt. The scientific and ethics committee at our institute approved the study and treatment protocol. An informed consent was taken from all patients who agreed to participate in this study. All patients presenting to our department from January 2015 till September 2017 with diagnosis of hepatic hydatid disease were discussed in weekly multidisciplinary meeting and assessed for eligibility to participate in the study. Inclusion criteria included patients presenting with uncomplicated hydatid liver cysts, with a diameter of 10 cm or more, with cyst type, including type 2, type 3, symptomatic type 4, or with viable content on ultrasound basis. Exclusion criteria were patients with cholangitis, elevated serum bilirubin, history of radiological or surgical intervention for hydatid disease, patients presenting to the emergency department with complicated hydatid disease, and patients who will undergo liver resection or pericystectomy.

A total of 81 patients were admitted to our surgery department because of clinical or radiological diagnosis of hydatid liver disease. All patients underwent routine laboratory tests, including serum bilirubin (total and direct), liver function tests, alkaline phosphatase, gamma glutamyl transferase, and serological tests for hydatid antibodies. Abdominal ultrasound was performed to delineate site, number, and type of hydatid cyst followed by dynamic abdominal CT scan. Magnetic resonance cholangiopancreatography was performed if patient's laboratory tests showed elevated serum bilirubin, serum alkaline phosphatase, or if the patient had a history of jaundice or dilated biliary radicles on ultrasound or CT scan. Eighteen patients were excluded as they did not meet our inclusion or had an exclusion criterion, whereas an additional 9 patients declined to participate in the study. The remaining 54 patients met our selection criteria. Patients were randomly assigned after consent to either group I (ES), or group II (no ES). Randomization was performed using a pseudorandom number generator with individual assignments concealed in sequentially numbered sealed envelopes that were opened in order when assignments were made. An independent observer managed patient allocation in either group.

Sample size estimation

Sample size calculation was conducted using Epi-save software to conduct a comparative study to detect difference in occurrence of postoperative bile leak in patients with hydatid cyst on using preoperative ERCP and ES. Sample size was estimated to be 25 patients in each group, totally 50 subjects included in the study to detect reduction in incidence of bile leak among the studied patients from 46% to 10% by using preoperative ERCP (36% reduction of incidence).¹² The estimated sample size is made at assumption of 95% confidence level and 80% power of study.

The primary endpoint was the occurrence of postoperative bile leak. The diagnosis of the bile leak was made through the drainage results from the drain placed at surgery. Postoperative bile leak was defined as ongoing bilious drainage of any amount through the drainage tube lasting for more than 2 days. The secondary endpoints were duration of postoperative hospital stay, mortality, and morbidity classified according to the Clavien–Dindo classification.

All patients underwent exploration under general anesthesia through inverted L incision. The site, number, and size of hepatic hydatid cysts were confirmed using intraoperative ultrasound. The liver was mobilized partially or completely as needed to allow safe and secure access to the cyst. Any adhesions encountered surrounding the cyst wall were completely and meticulously divided to avoid perforation of the cyst wall or spillage of its contents. Green surgical gauzes soaked with hypertonic saline (10%) or betadine were placed to cover the entire surrounding surgical field and isolate the cyst from surrounding viscera. Using a wide bore syringe, aspiration of the cyst was performed followed by injection of the hypertonic saline 10%. After 10 minutes, a small incision was made in the cyst wall with aspiration of contents using wide bore suction tubes to evacuate all cyst contents. The cyst cavity was again refilled with hypertonic saline to remove any left debris inside the cyst. Harmonic scalpel was used to excise the wall of cyst following the plane between the cyst wall and the compressed surrounding liver parenchyma with the edges oversewn using continuous locked polypropylene 2/0. A white piece of gauze was gently used to scrub the cyst cavity in a trial to detect points of bile leak. Any visible area of bile leak was thoroughly irrigated and scraped to identify the exact orifice of the cystobiliary communication and was sutured using 4/0 or 5/0 polypropylene depending on the diameter of orifice. In a trial to increase intrabiliary pressure, the gall bladder was squeezed in combination with partial occlusion of the common bile duct to further enhance the possibility of identifying additional areas of bile leak through minor cystobiliary communications. Two 24F tube drains were routinely placed within the cyst cavity after fixing omental flap to cut edges of the cyst to reduce the dead space inside the residual cavity.

Postoperative management and follow-up

Patients were transferred to the surgical ward except when the clinical situation indicated the need for admission to the intensive care unit. Postoperative management protocol was the same for all patients in both groups and included intravenous fluid administration, perioperative antibiotic prophylaxis, and proton pump inhibitors. All patients were encouraged to mobilize early and resume feeding as soon as possible. Albendazole 400 mg twice daily was started as soon as the patient tolerated oral feeding and continued for 3 months postoperatively. Albendazole therapy was continued for an additional 6 months in patients with soiling or cyst content spillage at the time of surgery. Discharge criteria were the ability to tolerate a soft or regular hospital diet, pain control with oral analgesics, and no surgical complication. Drains were removed before discharge if their contents are nonbilious and the daily output is less than 100cc. Patients with nonbilious drain output of more than 100cc daily were discharged with the drain and followed at the outpatient clinic twice weekly. Patients with bilious drain output resume their hospital stay with daily monitoring of the drain output until the bile output stopped either spontaneously or through intervention. Postoperative morbidity and mortality were defined as events occurring during the same hospital stay or within 3 months of resection and graded following the Clavien–Dindo classification.²² Postoperative mortality was defined as any death within 30 days after the procedure was performed.

Statistical analysis

The raw data were coded and entered into SPSS system files (SPSS package version 18). Analysis and interpretation of data were conducted. The following statistical measures were used: descriptive statistics, including frequency, distribution, mean, median, standard deviation, and interquartile range, were used to describe different characteristics. The Kolmogorov–Smirnov test was used to examine the normality of data distribution. Univariate analyses, including Student's t-test and Mann–Whitney test, were used to test the significance of results of quantitative variables. Chi-square test, Monte Carlo test, and Fisher's exact test were used to test the significance of results of qualitative variables. The significance of the results was at the 5% level of significance.

Results

A total of 54 patients (26 women and 28 men) ranging from 22 to 61 years in age (mean 44.6 ± 10.1 years) were randomly assigned to group I with prophylactic ES (n = 27) or group II to surgery without ES (n = 27). Demographic features and clinical, preoperative laboratory, and radiological characteristics of the patients are shown in Table 1. There was no statistically significant difference in age, gender, and BMI between the two groups. Preoperative clinical characteristics, including history of jaundice, cholangitis, existence of comorbid conditions, were comparable between the groups. There was no significant difference in preoperative laboratory results, including serum bilirubin (P = .754), aspartate transaminase (AST) (P = .927), alanine transaminase (ALT) (P = .606), alkaline phosphatase (P = .651), and gamma-glutamyl transferase (GGT) (P = .333). Preoperative radiological characteristics of hydatid cysts were comparable in the number, size, and type between the two groups. The cysts were located in the right lobe of the liver in 29 (53.7%) patients, left lobe in 18 (33.3%), and in both lobes in 7 (13%). A total of 77 hepatic cysts were treated in the 54 patients. All characteristics of the cysts were not statistically different between two groups, as shown in Table 1.

Table 1.

Demographics and Clinical, Preoperative Laboratory, and Radiological Characteristics of Patients

	Group (I) with ES (n = 27)		Group (II) no ES (n = 27)
Personal and clinical characteristics	n	%	n	%	Significance
Age (years)
Minimum–maximum	22.0–60.0		25.0–61.0		t = 0.144
Mean ± SD	43.6 ± 10.2		44.9 ± 10.4		P = .886
Gender
Male	14	51.9	14	51.9	χ² = 0.001
Female	13	48.1	13	48.1	P = 1.0
History of jaundice
No	22	81.5	23	85.2	^FEP = 1.0
Yes	5	18.5	4	14.8
Diameter of cyst
Minimum–maximum	10.2–19		10.5–20.3		t = 0.716
Mean ± SD	13.7 ± 2.9		14.2 ± 2.4		P = .477
Number of cysts
One	18	66.7	18	66.7	χ² = 0.277
Two	7	25.9	6	22.2	^MCP = 1.0
Three	2	7.4	3	11.1
Type of cysts
Type 2	5	18.5	7	25.9	χ² = 0.833
Type 3	17	63.0	17	63.0	^MCP = .678
Type 4	5	18.5	3	11.1
AST
Minimum–maximum	16–39		18–35		t = 0.093
Mean ± SD	24.2 ± 5.8		26.4 ± 5.9		P = .927
ALT
Minimum–maximum	17–38		18–36		t = 0.520
Mean ± SD	28.4 ± 5.7		29.2 ± 5.3		P = .606
GGT
Minimum–maximum	22–52		23–60		t = 0.978
Mean ± SD	38.3 ± 11.7		41.4 ± 11.6		P = .333
ALP
Minimum–maximum	234–382		241–380		t = 0.454
Mean ± SD	269.6 ± 30.8		279.8 ± 36.6		P = .651
Bilirubin
Minimum–maximum	0.5–1.3		0.7–1.2		t = 0.315
Mean ± SD	0.8 ± 0.2		0.9 ± 0.4		P = .754

χ²: chi-square test; ^MCP: Monte Carlo corrected P-value; ^FEP: Fisher's exact test; t: Student's t-test.

ALT, alanine transaminase; ALP, alkaline phosphatase; AST, aspartate transaminase; ES, endoscopic sphincterotomy; GGT, gamma-glutamyl transferase; SD, standard deviation.

Prophylactic ERCP with ES was technically successful in all patients in group I. Cholangiogram revealed no case of major or minor cystobiliary communication. Post-ERCP complication occurred in 2 patients; mild pancreatitis settled with conservative treatment in one patient and ascending cholangitis required antibiotic therapy for 10 days in the other patient.

The color of cyst fluid aspirated during operation was comparable between the two groups (P = .784), as shown in Table 2. In group I, the cyst fluid was serous in 19 cysts, brown in 10 cysts, and watery in 9 cysts, compared with group II, where the fluid was serous in 22, brown in 10, and watery in 7 cysts. There was no case in which the aspirated fluid yielded a bile tinged aspirate. Intraoperative identification of minor leaking bile orifice was successful in 10 patients in group II compared with 5 patients in group I (P = .129), they were managed by suture ligation. All swab tests at closure were negative in both groups at the end of the operation before drain insertion and wound closure.

Table 2.

Intraoperative Findings and Nonbiliary Complications Among the Studied Groups of Patients Subjected to Partial Cystectomy

	Group (I) with ES (n = 27)		Group (II) no ES (n = 27)
Intraoperative findings	n	%	n	%	Significance ^*
Identified bile orifice
No	22	81.5	17	63.0	χ² = 2.308
Yes	5	18.5	10	37.0	P = .129
Suture ligation
No	22	81.5	17	63.0	χ² = 2.308
Yes	5	18.5	10	37.0	P = .129
Color of cyst fluid
Serous	19	44.4	22	51.9	χ² = 0.487
Brown	10	29.7	10	29.6	P = .784
Watery	9	25.9	7	18.5
Nonbiliary complications:
Yes	5	18.5	4	14.8	χ² = 1.810
No	22	81.5	23	85.2	P = .178
Paralytic ileus	0	0	1	3.7	^FEP = 1.0
Wound	4	14.8	2	94.4	P = .701
Chest	1	3.7	0	0	^FEP = 1.0
Pleural effusion	0	0	1	3.7	^FEP = 1.0
Clavien–Dindo classification
No complications	22	81.5	23	85.2
Grade I	1	3.7	1	3.7	χ² = 1.612
Grade II	4	14.8	3	11.1	^MCP = .819
Grade III a	0	0	0	0
Grade III b	0	0	0	0

All swab tests at closure were negative. χ²: chi-square test; ^FEP: Fisher's exact test.

Significant at P ≤ .05.

ES, endoscopic sphincterotomy.

Group I with ES showed a significant decrease in bile leak rate compared with group II (P = .013), as illustrated in Table 3. Bile leak occurred postoperatively in only 3 patients (11.1%) in group I compared with 11 patients (40.7%) in group II. Group I patients had a significantly less median fistula daily output (100 mL/day, range: 80–120 mL/day) than group II patients (350 mL/day, range: 280–540 mL/day). All biliary fistulas in group I were low output that settled with conservative management with daily reduction in bile drain output until stoppage of leak in 3–4 days without any intervention. Biliary fistula in group II had a high output in 9/11 patients (81.8%) and low output in 2/11 patients (18.2%). Biliary fistula in group II had a significant increase needed for biliary intervention through postoperative ERCP with ES compared with biliary fistula in group I (^FEP = .002). Nine patients in group II required ES after mean duration of 5.6 ± 1.0 days (range: 4–7 days) with the remaining 2 patients settled with conservative treatment. Biliary leakage in all cases remained as controlled fistula, and no patient developed postoperative intra-abdominal biloma, biliary peritonitis, or cyst cavity biliary abscess. The mean time of fistula closure in group II was significantly longer (11 days, range: 6–15 days) compared with group I (3 days, range: 3–4 days), with P = .011. This resulted in a significant increase (P < .0001) in the time needed for drain removal (3 days, range: 2–5 days) in group I versus (5 days, range: 3–16 days) group II. Postoperative duration of hospital stay was significantly shorter in group I (3 days, range: 2–5 days) compared with (6 days, range: 3–17 days) group II with P < .0001.

Table 3.

Postoperative Bile Leak Data and Management Among the Studied Groups of Patients After Partial Cystectomy

	Group (I) with ES (n = 27)		Group (II) no ES (n = 27)
Postoperative characteristics	n	%	n	%	Significance
Bile leak in drain
No	24	88.9	16	59.3	χ² = 6.171
Yes	3	11.1	11	40.7	P = .013^*
Amount of leak	n = 3		n = 11
Minimum–maximum	80–120		280–540		Z = 2.185
Median (IQR)	100		350		P = .029^*
Duration of leak	n = 3		n = 11
Minimum–maximum	3–4		7–15		Z = 2.586
Median (IQR)	3 (3–4)		12 (9–14)		P = .010^*
Time for fistula closure	n = 3		n = 11
Minimum–maximum	3–4		6–15		Z = 2.553
Median (IQR)	3 (3–4)		11 (9.5–13.3)		P = .011^*
Time of drain removal
Minimum–maximum	2–5		3–16		Z = 5.554
Median (IQR)	3 (2–3)		5 (4–12)		P < .0001^*
Duration of hospital stay
Minimum–maximum	2–5		3–17		Z = 5.134
Median (IQR)	3 (2–3)		6 (3–13)		P < .0001^*
Need for postoperative ERCP
No	27	100.0	18	66.7	^FEP = .002^*
Yes	0	0.0	9	33.3
Complications of ERCP
No	25	92.6	7	77.8	^FEP = .255
Yes	2	7.4	2	22.2
Communication in ERCP			n = 9
No	27	100.0	3	33.3	^FEP < .0001^*
Yes	0	0.0	6	66.7

χ²: chi-square test; Z: Mann–Whitney test; ^FEP: Fisher's exact test.

Significant at P ≤ .05.

ERCP, endoscopic retrograde cholangiopancreatography; ES, endoscopic sphincterotomy; IQR, interquartile range.

The postoperative ERCP performed for bile leak management in group II was successful in identification of the site of leak in 6/9 patients (66.7%). No patient required stent insertion. None of these leaks involved main right or left hepatic ducts. After ERCP, mild biochemical pancreatitis occurred in 2 patients and settled with conservative treatment.

There was no mortality in both groups. The nonbiliary complication rate was comparable between two groups (P = .178). Five patients (18.5%) in group I developed nonbiliary complications; 4 patients with wound infection requiring antibiotic therapy and repeated dressing and one patient with chest infection. In group II, 4 patients (14.8%) had nonbiliary complications; 2 patients with superficial wound infection required dressing, one with paralytic ileus and one with pleural effusion managed conservatively.

Discussion

The cystobiliary communication is commonly due to an occult intrabiliary rupture in 10%–37% resulting in minor communication between the cyst and biliary system.^11,23,24 If the patient is asymptomatic, it is difficult to identify it preoperatively. It usually presents itself during the postoperative period in patients treated by partial cystectomy with bile leak and fistula causing an increase in morbidity. Large hepatic hydatid cyst with minor communication, if left untreated by cystectomy or if only managed by medical treatment and ES, would lead to further progression in the size and width of cytobiliary communication resulting in frank rupture of cyst content in biliary tree. The latter will pave the way for the cyst content, driven by the pressure gradient, to be emptied into the biliary tree causing intermittent or total obstruction. Patients will present with biliary colic, jaundice, and cholangitis, which can worsen, resulting in liver abscess, septicemia, or anaphylaxis. Less frequently, complications such as acute cholecystitis and pancreatitis can also occur.

In the present study, prophylactic ES provided a significant reduction (P = .013) in postoperative bile leak rate after conservative treatment of hydatid cyst in group I patients (11.1% bile leak) compared with group II patients who had no preoperative sphincterotomy (40.7% bile leak). Biliary fistulas, when developed in patients with previous ES (3 patients in group I), were low-output fistulas (100 mL/day, range: 80–120 mL/day) that closed spontaneously with conservative treatment highlighting the effect of biliary decompression. The fistulas healed shortly after a median time of 3 days (range: 3–4 days). However, patients in group II without preoperative biliary decompression had significant higher fistula output (350 mL/day, range: 280–540 mL/day) with prolonged duration (12 days, range: 7–15 days) that needed endoscopic intervention to heal in 81.8% of cases (9/11 patients) with only 2 patients closed spontaneously with conservative treatment. Biliary fistulas caused by cystobiliary communication in hydatid disease usually persist in 4%–27.5% of cases and rarely close spontaneously without decompression of the biliary tract²⁵ compared with most external biliary fistulas that usually close spontaneously.²⁶ Endoscopic treatment is indicated for persistent high-output fistulas with daily output exceeding 300 mL for more than 1 week and in low-output fistulas of duration more than 3 weeks.^12,27

Size of cyst exceeding 10 cm,²⁸ elevated levels of alkaline phosphatase (ALP),^23,27–29 GGT^23,28,29 bilirubin,^23,27,29 and history of cholangitis have been reported as risk factors for an occult cystobiliary communication in literature. In this study, the mean size of the cysts exceeded 10 cm and was comparable in both groups (13.7 ± 2.9 and 14.2 ± 2.4 in group I and group II, respectively, P = .477). The size of the cyst correlates with the intracystic pressure and can lead to a spontaneous intrabiliary rupture.²⁸ The large hydatid cysts grow by expansion leading to compression atrophy of the surrounding hepatic parenchyma.²³ These weak points of contact when encountering a bile duct can lead to spontaneous rupture forming a cystobiliary communication.²³ They are usually visualized during operation or detected through biliary leakage after hydatid liver surgery.²⁴ Passage of cyst fluid carrying possible hydatid sand and debris to the biliary system does occur with high intracystic pressure exceeding the biliary pressure. According to the size of the communication and the nature of debris, obstructive jaundice may or may not be apparent. None of our patients had elevated serum bilirubin, and all preoperative laboratory values, including GGT and ALP, showed no significant difference.

In view of our results, it was not uncommon for the surgeon not to be able to find bile in the cavity despite the presence of occult cystobiliary communication, denoting that nonbilious-stained cyst fluid does not mean absence of cystobiliary communication. This could be explained by pressure difference across the communication. The biliary system pressure is about 15–20 cm H₂O compared with intracystic pressure ranging from 30 to 80 cm H₂O according to the size of the cyst.³⁰ The pressure inside the cyst correlates directly with cyst diameter. The high intracystic pressure prevents the bile from entering the cyst despite the presence of minor cystobiliary communications, with the direction of bile flow usually toward the biliary system due to the pressure gradient in these cases.^29,30 If cystobiliary communication is not seen and repaired through suture closure, the pressure gradient is reversed and bile flows into the residual cavity rather than through the biliary tree, with bile leak appearing postoperatively through drain output.

It is not always successful to detect the sites of cystobiliary communication intraoperatively despite the many techniques described in literature. This may be explained by the following. First, bile secretion may not be leaking through these orifices at the time of operation either due to temporary blockage by sediments that accumulated and maintained over time by the high intracystic pressure or changes in pressure gradient in favor of increase in the biliary pressure across the communication may cause a cyst-biliary communication to leak later as the sediments dislodge and bile leaks. Second, the inner wall of the cyst is always corrugated and has inner folds that could temporary hide these minor communications especially in deep-seated cavities. Third, there is always a fear from the surgeon to brush inner folds of cyst cavity for fear from bleeding or injury of underlying major hepatic vasculature especially hepatic veins.

Suture ligation was performed in 5 patients in group I and 10 patients in group II in an attempt to control bile leak. No published literature has reported the success rate of perioperative suture ligation to prevent biliary leaks especially when the fistula orifices are deeply located in the cyst cavity. Our results in group II indicate that there is a significant failure rate for an attempt to control a detected biliary fistula intraoperatively if that was used as the only method to control the cystobiliary without biliary decompression. Minor cystobiliary communication even if discovered intraoperatively represents a challenge to the surgeon in suturing. The leaking orifice inside the cyst does not represent the type and place of the hole in the bile duct, and these biliary holes do not necessarily have to be the terminal end of the bile duct ending within the cyst but they can represent a lateral defect on a major bile duct. Controlling the sites of cystobiliary communication by sutures may not be successful in many cases whether due to a superficial suture that does not close the communication or too deep suture that may lead to partial or complete occlusion of a major duct as reported in one study.³¹ If the biliary pressure is high, the leak will continue from the suture line.

Postoperative ERCP performed for the management of bile leak in group II succeeded in identifying the sites of cystobiliary communication in 6/9 patients (66.7%), in contrast to preoperative ERCP that failed to identify any of these communications. In group I patients, minor cystobiliary communications were detected in 5 patients intraoperatively and were suture ligated, and preoperative ERCP failed to detect these communications. The integrity of the biliary system was falsely confirmed despite the presence of cystobiliary communication. This could be explained by the small size of these communications, the pressure effect of cyst on biliary system, and the change in pressure gradient across the communication. When the cyst is present with its high pressure, it prevents the identification of these communications in preoperative ERCP in group I, while after partial cystectomy with reversal of pressure gradient in favor of bile leaking inside the cyst cavity, postoperative ERCP easily identified the sites of these communications.

Some authors have reported techniques involving cholecystectomy with dissection of the common bile duct and performing a leakage test to delineate cystobiliary communication during partial cystectomy for hydatid liver disease.^21,32 In their experience there was no added morbidity or long-term biliary problems, however, the reported bile leak rate was significantly lower (2.6%) than what commonly reported in literature (8.2%–26%).^14,21,28 They have claimed that the added maneuver was less risky compared to morbidities caused by bile leak. However, manipulation, dissection, and temporary occlusion of the common bile ducts always carry a theoretical risk for developing biliary stricture. The required equipment and surgeon skills for the added biliary intervention through bile duct manipulation and intraoperative cholangiogram may not be available in hospitals that practice the conservative management for hydatid liver disease. This might lead to different results with added complications. Preoperative ERCP eliminates the need for any added biliary intervention that facilitated the operative steps for the average general surgeon who usually performs this type of operation.

In recent years, with growing experience and skills in the world of endoscopy, endoscopic biliary intervention became a commonly practiced procedure with accepted morbidity and mortality. Preoperative ERCP has some advantages. First, it decompresses the biliary system and lowers the biliary pressure, the main cause of leak after relief of the high intracystic pressure after partial cystectomy. Second, it clears the biliary tree from debris that may have passed through the cystobiliary communications. Third, it helps to delineate the biliary anatomy and guide the surgeon to possible branches of involvement in the biliary system, which might not be possible intraoperatively due to the distorted anatomy caused by the cyst pressure. Fourth, it eliminates the need to perform cholecystectomy or common bile duct (CBD) maneuver to perform an intraoperative cholangiogram or leakage test, which may be not possible for average surgeons and that potentially confers further morbidity.

The present study has shown that the endoscopic method is an effective and safe modality causing a significant reduction in postoperative biliary fistula after partial cystectomy in patients with hydatid liver disease. Although preoperative ERCP may fail in delineating the sites of minor cystobiliary communication, the ES decompresses the biliary system preventing bile leaks from these minor communications after partial cystectomy with drop in intracystic pressure. Postoperative ERCP performed for management of bile leak after partial cystectomy is more informative about the sites of minor cystobiliary communication and presents a highly effective method in reducing the pressure gradient between the bile duct and the duodenum that is maintained by an intact sphincter of Oddi, and diverts bile away from the site of the leak, resulting in healing of the fistula.

In conclusion, prophylactic ES before partial cystectomy of hydatid liver cyst provides a significant reduction in postoperative biliary fistula, daily fistula output, shorter time to fistula closure without intervention, and a significant reduction in duration of hospital stay than patients without ES.

Footnotes

Disclosure Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. No competing financial interests exist.

References

Eckert

, Conraths

, Tackmann

. Echinococcosis: An emerging or re-emerging zoonosis?. Int J Parasitol, 2000; 30:1283–1294.

Garippa

, Varcasia

, Scala

. Cystic echinococcosis in Italy from the 1950s to present. Parassitologia, 2004; 46:387–391.

Grosso

, Gruttadauria

, Biondi

, Marventano

, Mistretta

. Worldwide epidemiology of liver hydatidosis including the Mediterranean area. World J Gastroenterol, 2012; 18:1425–1437.

Akcan

, Sozuer

, Akyildiz

, Ozturk

, Atalay

, Yilmaz

. Predisposing factors and surgical outcome of complicated liver hydatid cysts. World J Gastroenterol, 2010; 16:3040–3048.

Paksoy

, Karahasanoglu

, Carkman

, Giray

, Senturk

, Ozcelik

, et al. Rupture of the hydatid disease of the liver into the biliary tracts. Dig Surg, 1998; 15:25–29.

Ramia

, Figueras

, De la Plaza

, Garcia-Parreno

. Cysto-biliary communication in liver hydatidosis. Langenbecks Arch Surg, 2012; 397:881–887.

Unalp

, Baydar

, Kamer

, Yilmaz

, Issever

, Tarcan

. Asymptomatic occult cysto-biliary communication without bile into cavity of the liver hydatid cyst: A pitfall in conservative surgery. Int J Surg, 2009; 7:387–391.

Hankins

. Management of complicated hepatic hydatid cysts. Ann Surg, 1963; 158:1020–1034.

Kornaros

, Aboul-Nour

. Frank intrabiliary rupture of hydatid hepatic cyst: Diagnosis and treatment. J Am Coll Surg, 1996; 183:466–470.

10.

Akkiz

, Akinoglu

, Colakoglu

, Demiryurek

, Yagmur

. Endoscopic management of biliary hydatid disease. Can J Surg, 1996; 39:287–292.

11.

Becker

, Frieling

, Saleh

, Haussinger

. Resolution of hydatid liver cyst by spontaneous rupture into the biliary tract. J Hepatol, 1997; 26:1408–1412.

12.

Skroubis

, Vagianos

, Polydorou

, Tzoracoleftherakis

, Androulakis

. Significance of bile leaks complicating conservative surgery for liver hydatidosis. World J Surg, 2002; 26:704–708.

13.

Magistrelli

, Masetti

, Coppola

, Messia

, Nuzzo

, Picciocchi

. Surgical treatment of hydatid disease of the liver. A 20-year experience. Arch Surg, 1991; 126:518–522; discussion 23.

14.

Balik

, Basoglu

, Celebi

, Oren

, Polat

, Atamanalp

, et al. Surgical treatment of hydatid disease of the liver: Review of 304 cases. Arch Surg, 1999; 134:166–169.

15.

El Malki

, El Mejdoubi

, Souadka

, Mohsine

, Ifrine

, Abouqal

, et al. Predictive factors of deep abdominal complications after operation for hydatid cyst of the liver: 15 years of experience with 672 patients. J Am Coll Surg, 2008; 206:629–637.

16.

Yagci

, Ustunsoz

, Kaymakcioglu

, Bozlar

, Gorgulu

, Simsek

, et al. Results of surgical, laparoscopic, and percutaneous treatment for hydatid disease of the liver: 10 years experience with 355 patients. World J Surg, 2005; 29:1670–1679.

17.

Agaoglu

, Turkyilmaz

, Arslan

. Surgical treatment of hydatid cysts of the liver. Br J Surg, 2003; 90:1536–1541.

18.

El Malki

, Souadka

, Benkabbou

, Mohsine

, Ifrine

, Abouqal

, et al. Radical versus conservative surgical treatment of liver hydatid cysts. Br J Surg, 2014; 101:669–675.

19.

Zeybek

, Dede

, Balci

, Coskun

, Ozerhan

, Peker

, et al. Biliary fistula after treatment for hydatid disease of the liver: When to intervene. World J Gastroenterol, 2013; 19:355–361.

20.

Galati

, Sterpetti

, Caputo

, Adduci

, Lucandri

, Brozzetti

, et al. Endoscopic retrograde cholangiography for intrabiliary rupture of hydatid cyst. Am J Surg, 2006; 191:206–210.

21.

Kayaalp

, Aydin

, Olmez

, Isik

, Yilmaz

. Leakage tests reduce the frequency of biliary fistulas following hydatid liver cyst surgery. Clinics, 2011; 66:421–424.

22.

Dindo

, Demartines

, Clavien

. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg, 2004; 240:205–213.

23.

Atli

, Kama

, Yuksek

, Doganay

, Gozalan

, Kologlu

, et al. Intrabiliary rupture of a hepatic hydatid cyst: Associated clinical factors and proper management. Arch Surg, 2001; 136:1249–1255.

24.

Kayaalp

, Bostanci

, Yol

, Akoglu

. Distribution of hydatid cysts into the liver with reference to cystobiliary communications and cavity-related complications. Am J Surg, 2003; 185:175–179.

25.

Rodriguez

, Sanchez del Rio

, Alguacil

, De Dios Vega

, Fugarolas

. Effectiveness of endoscopic sphincterotomy in complicated hepatic hydatid disease. Gastrointest Endosc, 1998; 48:593–597.

26.

Dolay

, Akcakaya

, Soybir

, Cabioglu

, Muslumanoglu

, Igci

, et al. Endoscopic sphincterotomy in the management of postoperative biliary fistula A complication of hepatic hydatid disease. Surg Endosc, 2002; 16:985–988.

27.

Demircan

, Baymus

, Seydaoglu

, Akinoglu

, Sakman

. Occult cystobiliary communication presenting as postoperative biliary leakage after hydatid liver surgery: Are there significant preoperative clinical predictors?. Can J Surg, 2006; 49:177–184.

28.

Kayaalp

, Bzeizi

, Demirbag

, Akoglu

. Biliary complications after hydatid liver surgery: Incidence and risk factors. J Gastrointest Surg, 2002; 6:706–712.

29.

Ozaslan

, Bayraktar

. Endoscopic therapy in the management of hepatobiliary hydatid disease. J Clin Gastroenterol, 2002; 35:160–174.

30.

Yalin

, Aktan

, Yegen

, Dosluoglu

. Significance of intracystic pressure in abdominal hydatid disease. Br J Surg, 1992; 79:1182–1183.

31.

Peker

, Gumusoglu

, Seyit

, Kabuli

, Salik

, Gonenc

, et al. Prevention of postoperative bile leak in partial cystectomy for hydatid liver disease: Tricks of the trade. J Gastrointest Surg, 2015; 19:2228–2234.

32.

Silva

, Mirza

, Bramhall

, Mayer

, McMaster

, Buckels

. Treatment of hydatid disease of the liver. Evaluation of a UK experience. Dig Surg, 2004; 21:227–233; discussion 33–34.