Our experiences in 1000 case single-centre endoscopic retrograde cholangiopancreatography

Abstract

Background:

Endoscopic retrograde cholangiopancreatography (ERCP) has a very important role in both diagnosis and treatment of pancreaticobiliary pathologies. The aim of this study was to review the indications, technical success complications and results of ERCP procedures performed in our centre.

Materials and Methods:

In the study, the data of 1000 patients who were diagnosed with extrahepatic cholestasis, obstructive cholangitis/pancreatitis or bile leakage in clinical, medical and radiological data and who underwent ERCP between May 2019 and November 2021 were evaluated retrospectively.

Results:

The age distribution was between 14 and 109 years, and the average age was 57.97 years for women (14–109) and 57.48 for men (19–95). Gender distribution was as follows: 552 (55.2%) women and 448 (44.8%) men. ERCP indications in this study were as follows: choledocholithiasis, malignant bile duct obstruction, odysphincter dysfunction (ODS), post-operative bile leakage, hepatic hydatid cyst rupture into the biliary tract and bile duct stenosis. Eight hundred and seventy-one (87.1%) patients had common bile duct stones, 30 (3%) pancreatic head tumour, 22 (2.2%) common bile duct tumour and 20 (2%) cholestasis due to papilla tumour and/or obstructive icterus. Twenty-two (2.2%) patients were treated for sphincter dysfunction (ODS) of Oddi. ERCP was performed in 12 (1.2%) patients with the diagnosis of bile leakage after liver hydatid surgery, 10 (1%) after gall bladder surgery and 8 (0.8%) with the diagnosis of biliary tract stenosis after gall bladder surgery. ERCP was performed in 5 (0.5%) patients due to biliary tract obstruction as a result of rupture of hydatid liver cyst into the biliary tract. Successful cannulation was achieved in 1000 patients listed. Ninety-seven patients who were included in the ERCP procedure but could not be cannulated were not included in the study. Endoscopic sphincterotomy was performed on all patients in the ERCP procedure. In patients with choledochal stones, stone extraction from the common bile duct and/or plastic stent placement in the common bile duct was added. Plastic stent was placed in the common bile duct in patients with pancreatic head tumour, common bile duct tumour, bile duct leakage and common choledochal stenosis. Endoscopic biopsy was taken from all patients with suspected papillary tumour. The processing time varied between 15 and 90 min. The overall complication rate was 17.4%. After ERCP, 93 patients developed amylasaemia that did not require treatment, while 50 patients were diagnosed with clinical and laboratory acute pancreatitis. Intraoperative bleeding, which did not require blood transfusion and could be controlled with adrenaline injection and/or balloon pressure, developed in 10 patients. In one patient, basket and stone were stuck in the papilla during stone extraction. There was no operative mortality.

Conclusion:

ERCP is a complex procedure that uses special equipment and must be performed by experienced specialists. To increase the reliability of the ERCP procedure, it is necessary to determine the risk factors for ERCP complications very well. Unnecessary ERCP should be avoided. ERCP should not be performed, especially in patients with low probability of stone or stricture, patients with normal bilirubin and patients who do not show other signs of biliary disease. The use of non-invasive imaging methods as much as possible instead of diagnostic ERCP will reduce the complications associated with ERCP.

INTRODUCTION

Endoscopic retrograde cholangiopancreatography (ERCP) is a widely used method in the diagnosis and treatment of biliary and pancreatic diseases. ERCP has high sensitivity and specificity in detecting stones in the common bile duct [Figures 1–3].

Figure 1.

Choledochal injury after gall bladder surgery. Contrast material leakage in cholangiography in endoscopic retrograde cholangiopancreatography

Figure 3.

Our plastic stent placed in the common bile duct

Figure 2.

A plastic stent was placed into the common bile duct in endoscopic retrograde cholangiopancreatography (ERCP). Before ERCP, the accumulation of bile in the abdomen was evacuated with a percutaneously placed drain

Endoscopic sphincterotomy and stone extraction are successful in more than 90% of cases.[1] For the success of this invasive procedure, which is extremely beneficial, careful evaluation of the patient before the procedure, the correct indication of the intervention to be performed, the procedure performed in experienced hands and the follow-up of the patient after the intervention are required. Although ERCP was initially used as a diagnostic procedure, in the past 30 years, diagnostic ERCP has mostly been replaced by radiological imaging methods such as magnetic resonance (MR) imaging and is now performed almost entirely for therapeutic purposes.[2] Today, ERCP is mostly used for the removal of bile duct stones, stent placement in malignant or benign strictures and treatment of complications after gall bladder or bile duct surgeries. Since it is an invasive procedure, ERCP-related complications (up to 10%) have been reported in the literature.[3,4]

1%–2% of these complications are serious complications, and mortality rates <1% have been reported.[5] Complications specific to ERCP mainly include pancreatitis, bleeding, inadequate drainage, cholangitis and retroperitoneal perforation. Patient-related factors and the experience and expertise of the endoscopist are important in the occurrence of ERCP-related complications.

The aim of this study was to determine the indications of patients who underwent ERCP procedure in a single centre, our diagnostic tools, the consistency of diagnosis with imaging methods during the procedure and the frequency of complications after ERCP and to present the results of the ERCP cases included in the study. The secondary purpose is to determine the risk factors in patients who develop complications with the ERCP procedure, as well as to determine the appropriate diagnostic and therapeutic approach.

MATERIALS AND METHODS

Ethics committee approval of the study was obtained from the Clinical Research Ethics Committee of Harran University (29 November 2011/21/13.30). All patients who underwent ERCP between May 2019 and November 2021 in the General Surgery ERCP Unit of Şanlıurfa Mehmet Akif Inan Training and Research Hospital were evaluated in this study. In addition to clinical evaluation and physical examination, the most common indication for the application of the procedure was ultrasonography (USG) and/or computed tomography (CT), MR cholangiography (MR cholangiopancreatography [MRCP]) and biochemistry examinations, as a result of the pancreatobiliary system obstruction or suspicion of obstruction. In addition, patients with bile leakage in clinical and imaging methods were included in the ERCP procedure. Before the procedure, all patients underwent routine haemogram and biochemistry tests. Age, gender, ERCP indication, procedures performed, success status, complications after ERCP, risk factors for complications, comparison of pre-operative MRCP reports and diagnoses in ERCP procedure were evaluated retrospectively. The data obtained in this study were taken from our hospital’s central database and patient files. All patients were informed about the procedure and its risks, and a consent form was obtained. ERCP process was performed under general anaesthesia under the control of an anaesthesiologist. Fujifilm Processor VP-3500HD was used as the device. Selective method was used for ERCP cannulation. In patients with biliary obstruction, the common bile duct was operated or stented if necessary. Balloon occlusion cholangiogram was taken to visualise the common bile duct and cystic duct. As a standard, 0.035-inch wire was used for cannulation. Cefoperazone + sulbactam (Sulperazon) and metronidazole 1 g were administered to each patient. Each patient was followed up for at least 1 day after the procedure. Patients with complications were followed up in the ward until their clinical condition improved. Bleeding from the sphincterotomy site, post-ERCP pancreatitis (PEP), cholangitis, perforation and cardiopulmonary (hypoxia, hypotension and arrhythmia) complications related to the ERCP procedure were recorded. SPSS for Windows Version 20 computer programme was used for statistical analysis. Categorical variables were expressed as numbers and percentages.

RESULTS

ERCP was performed on 1000 patients between May 2019 and November 2021. All patients who underwent successful cannulation were included in the study. The age distribution was between 14 and 109 years, and the average age was 57.97 for women (14–109) and 57.48 for men (19–95). Gender distribution was as follows: 552 (55.2%) women and 448 (44.8%) men. The processing time was between 15 and 90 min. The most common indication for ERCP was the choledochal stone in 871 (87.1%) patients with the presence of obstructive jaundice and/or high serum bilirubin level. Then, respectively, the indications were 30 (3%) pancreatic head tumours, 22 (2.2%) common bile duct tumours and 20 (2%) cholestasis due to peri-ampullary tumour. Twenty-two (2.2%) patients were treated for sphincter of Oddi dysfunction (SOD). ERCP was performed in 12 (1.2%) patients after liver hydatid surgery and 10 (1%) after gall bladder surgery due to bile leakage. ERCP was performed in 8 (0.8%) patients with the diagnosis of biliary tract stenosis after gall bladder surgery. ERCP was performed in 5 (0.5%) patients due to biliary tract obstruction as a result of rupture of hydatid liver cyst into the biliary tract [Table 1].

Table 1.

Endoscopic retrograde cholangiopancreatography indications

Indications	n	%
Choledochal stone	87	87.1
Pancreatic tumour	30	3
Bile duct tumour	22	2.2
Oddi sphincter dysfunction	22	2.2
Peri-ampullary tumour	20	2
Bile leakage after hydatid cyst surgery	10	1
Post-cholecystectomy bile leakage	8	0.8
Biliary tract stricture	5	0.5

ERCP was performed in 871 patients due to common choledochal stones. Endoscopic sphincterotomy (ES) was performed in all patients who were treated for choledochal stones. Stone extraction was performed with basket and/or balloon catheter in 797 (91.5%) of 871 patients. Stone extraction could not be performed in 74 patients who did not have adequate ES due to papillary stricture or location inside or near the diverticulum and/or had stones larger than 1.5–2 cm. Aplastic stent was placed in the common bile duct.

Common choledochal stenting was performed in 30 patients who underwent ERCP with successful cannulation due to pancreatic head tumour.

ERCP was performed on 22 patients due to common choledochal tumour. Two of these patients had proximal common bile duct tumours, and a stent extending to the proximal hepatic duct was placed in these patients. The other 20 patients who underwent ERCP with the diagnosis of common bile duct tumour were distal common bile duct tumours. These patients also underwent choledochal stenting after ES.

ERCP was performed in 20 patients with bile duct obstruction with the diagnosis of papilla tumour. Six of these patients were diagnosed during the procedure. Papillary biopsy was performed in all of these patients during ERCP. The biopsy results of 11 patients were malignant. Biopsies of nine patients could not be evaluated due to insufficient material intake. ERCP was performed in 22 patients with the diagnosis of SOD. Although manometry could not be performed due to technical limitations, these patients had clinical symptoms as well as biochemical abnormalities before the procedure. MRCP images were reported as narrowing at the distal end of the common bile duct and enlargement of the common bile duct. It was because of SOD.

Twelve patients were taken to the ERCP procedure due to high flow rate and prolonged bile leakage from the drain after liver hydatid surgery. These patients underwent plastic stent placement in the common bile duct after ES. Bile leakage of four patients stopped after 2 months of follow-up. Bile leakage of eight patients stopped after the 7^th day.

After gall bladder surgery, ERCP was performed on 10 patients with high drainage from the drain and unceasing bile leakage despite medical treatment. These patients underwent choledochal stenting after ES. In one of these patients, bile leakage lasted for a long time and closed after 1 month with medical treatment support. Bile leakage stopped in nine patients within 5 days.

Eight patients with bile duct stenosis were included in the ERCP procedure. In these patients, segmental stenosis of the common bile duct was observed in the cholangiography taken during the ERCP procedure. After ES, a plastic stent was placed in the common bile duct, passing the narrow segment. In the anamnesis of these patients, one had a history of liver hydatid surgery, and the other seven had a history of cholecystectomy surgery.

ERCP was performed in five patients due to biliary tract obstruction as a result of rupture of hydatid liver cyst into the biliary tract. These patients presented with the diagnosis of abdominal pain, obstructive jaundice and cholangitis. Two of them had previously been diagnosed with hydatid liver cyst. These patients underwent choledochal stenting after ES.

One hundred and forty of the patients who underwent ERCP presented with acute biliary pancreatitis (ABP). Choledochal stone/mud was detected in 139 of these patients. One patient had hydatid liver cyst that opened into the bile ducts. One hundred and thirty-nine patients with distal common bile duct stones had obstructive jaundice and their total bilirubin values were above 5 mg/dl at the time of admission. Liver function values were high. Of the 139 patients, 54 also had signs of cholangitis. ERCP was performed within 72 h in these 54 patients diagnosed with ABP (ABP) + cholangitis, and ERCP was performed within 5–7 days in patients with ABP without cholangitis findings.

ERCP processing time ranged from 15 to 90 min.

Successful cannulation was achieved in 1000 patients listed. Ninety-seven patients who were included in the ERCP procedure but could not be cannulated were not included in the study. The successful cannulation rate was determined as 91.1%.

The overall complication rate was 17.4% [Table 2]. After ERCP, 93 (9.3%) patients developed amylasaemia that did not require treatment. In these patients who did not require treatment, amylase values were <200 IU/L (normal values: 28–100 IU/L).

Table 2.

Endoscopic retrograde cholangiopancreatography complications

Complications	n	%
Amylasaemia	93	9.3
Pancreatitis	50	5
Bleeding	10	1
Stone basket blockage	1	0.1
Cholangitis	20	2.0
Perforation	0	0

These patients were discharged 1 day after the procedure. Acute pancreatitis (PEP) developed in 50 (5%) patients after the procedure. In these patients, amylase and lipase values were more than three times normal, and there was abdominal pain that did not regress despite treatment. Oral intake of these patients was stopped, and intravenous (IV) fluid support, IV antibiotics and analgesics were started. All patients with post-ERCP acute pancreatitis underwent abdominal USG and CT. CT scans of 44 of 50 patients who developed acute pancreatitis were reported as pancreatic oedema, peri-pancreatic and peri-hepatic fluid accumulation. These 44 patients were discharged after an average of 4 days of medical treatment. On CT scans of six patients, diffuse fluid accumulation in the abdomen with diffuse oedema of the pancreas was detected. These six patients were discharged after 10–15 days of medical treatment. Localised pancreatic necrosis was observed on CT in four patients who developed acute pancreatitis. The hospital stay of these patients lasted longer than 1 month. In these patients, pseudocyst occurred in the pancreas within 3–6 months. Only one patient was treated with endoscopic surgery. In the other five patients, the cyst regressed and was followed up. Of the patients who developed PEP, 38 (76%) were female and 12 (24%) were male. ERCP was performed in 45 (90%) cases of PEP due to choledochal stones, 3 (6%) malignancy and 2 (4%) due to bile leakage after liver hydatid surgery.

Intraoperative bleeding, which did not require blood transfusion and could be controlled with adrenaline injection and/or balloon pressure, developed in 10 (1%) patients. No significant decrease was observed in haemoglobin values in the post-procedure follow-ups. In 1 (0.1%) patient, basket and stone became stuck in the papilla during stone extraction. Open surgery was performed in this patient in the same session and the basket was removed together with the compressed stone.

Post-procedure cholangitis developed in 19 (1.9%) patients. In 1 (0.1%) of these, cholangitis was severe. A 35-year-old female patient who was taken to the intensive care unit died on the 4^th day after the procedure. This patient had a history of tumour originating from the peripheral nervous system in the lumbosacral region.

There was no operative mortality in this case series. More appropriate term could be ‘intra-procedure mortality’ (mortality during the procedure).

DISCUSSION

ERCP is considered an effective and safe method in the treatment of pancreatobiliary diseases. Bile duct leaks and fistulas are best demonstrated by ERCP and treated endoscopically. Strictures in the extrahepatic bile ducts should be treated endoscopically first. In this study, strictures were more than 2 cm away from the haptic bifurcation in patients who were stented with ERCP.

The success of the ERCP application reaches 90%–95% in good centres. The successful cannulation rate in our ERCP application was 91.1%. The most common factors preventing success in ERCP are failure of selective cannulation, fibrosis of sphincter of Oddi, peri-ampullary diverticulum and malignancy.[6] In our series, these reasons were frequently encountered in failure. We did not include 91 patients with unsuccessful cannulation in our series. PEP was developed in 28 of the patients with unsuccessful cannulation. In these patients, the length of the procedure, oedema due to trauma during papillary cannulation and traumatisation of the pancreatic duct were considered the causes of PEP. ES and stent applications are very useful in biliary tract problems such as remnant choledochal stones, post-operative biliary tract injuries, bile leakage and bile duct stenosis after cholecystectomy. ERCP is still considered the most important method for the evaluation of the pancreatic and biliary system in recent years; however, MRCP is an imaging technique that brightens stationary fluids such as bile and pancreatic fluid while darkening other tissues and moving fluids.[7] A systematic review of the literature has shown that when compared with ERCP, MRCP has an aggregated sensitivity, specificity, positive predictive value and negative predictive value of 85%, 93%, 87% and 82%, respectively.[8] ERCP and MRCP have comparable sensitivity and specificity for the diagnosis of choledocholithiasis.[9] In a study, the sensitivity, specificity, positive predictive value and negative predictive value of MRCP in the evaluation of all bile duct pathologies were 85.0%, 50.0%, 89.4% and 40.0%, respectively.[10] Parashari et al.[11] found MRCP sensitivity and specificity of 91.66% and 90.46% for choledocholithiasis and 85% and 71% for malignant stenosis, respectively. Suthar et al. found MRCP sensitivity and specificity of 100% and 100% for choledocholithiasis and 85.7% and 96.3% for malignant strictures, respectively.[12] MRCP is a non-invasive imaging method whose accuracy has been proven by studies in the detection and evaluation of the biliary system and pancreatic duct pathologies.[13] Abdominal USG and MRCP, which are imaging methods, were used in all patients included in this study. In some cases, CT of the abdomen with contrast was performed to confirm the diagnosis. The most common indication for ERCP in the literature is choledocholithiasis. In a study of 1072 cases, it was reported that 46.6% of ERCP procedures were performed with the indication of choledocholithiasis.[14] In our study, the rate of choledocholithiasis was 87.1%. Choledocholithiasis may occur in 8%–18% of patients with symptomatic gallstones.[15] In the literature, the success rate in removing common choledochal stones with ERCP is between 80% and 97%.[16] In this study, 1000 patients who underwent successful cannulation were reported. Ninety-seven patients whose cannulation failed were excluded from the record. Our successful cannulation rate was 91.1%. Stone extraction was performed with basket and/or balloon catheter in 797 (91.5%) of 871 patients who underwent ERCP for choledochal stones. Stone extraction could not be performed in 74 patients who did not have adequate ES due to papillary structure or location (inside or near the diverticulum) and/or had stones larger than 1.5–2 cm. A plastic stent was placed in the common bile duct. Our success rate in removing the common choledochal stone is 91.5%. This rate is compatible with the literature. In MRCP, bile duct stones are seen as multiple filling defects with angular contours.[17] As the diameter of the common bile duct decreases in MRCP, the diagnostic value also decreases. In ERCP, false-negative results may occur due to the covering of small stones with opaque material.[18] Although no stones were reported in the MRCP imaging of 30 (3%) patients, stones were extracted from the common bile duct in ERCP. The diagnosis of these 30 patients was made by clinical findings and laboratory results, and their common bile duct diameters were between 7 and 9 mm. The excavated stones were in the form of mud and/or in millimetres. In this study, the accuracy rate of MRCP for choledocholithiasis was determined as 96.5%. Injury in the biliary ducts is an accidental cause of stenosis. After injury, an inflammatory reaction develops, resulting in collagen deposition, fibrosis and focal narrowing.[19] The most common cause of benign biliary strictures (80%–90%) is previous hepatobiliary surgery.[20] MR cholangiopancreatography is as sensitive as direct cholangiography, and short segments in the common bile duct and common hepatic duct can show well-circumscribed stenosis.[21] With contrast-enhanced MRI, the narrowed segment is usually seen as thin, non-contrast-enhancing, well-defined walls.[22] In this series, eight patients had complaints of prolonged abdominal pain and laboratory findings of cholestasis. All of these patients had a history of cholecystectomy in the past. These patients had moderately elevated cholestasis enzymes and had frequent recurrent abdominal pain. Segmental common bile duct stenosis and distal collapse were demonstrated by MRCP performed for diagnosis. Segmental stenosis in patients who underwent ERCP was at the level of the cystic duct and was short segmented. Direct surgical repair was performed in two cases with severe stricture in the common bile duct after cholecystectomy. Right–left hepaticojejunostomy was performed in one patient with a stricture in the bifurcation. SOD is an acalculous obstructive disease of the sphincter of Oddi and is characterised by biliary and/or pancreatic pain. The most valid test in the diagnosis of SOD is manometry. Sphincter of Oddi manometry is an invasive test and it is seen that 4%–10% of pancreatitis and 0.4% mortality occur after manometry.[23] In patients with clinically suspected SOD, MR cholangiopancreatography findings; focal stenosis of the Oddi sphincter, tapering narrowing of the distal common bile duct and dilation (usually >12 mm) in the extrahepatic bile ducts.[24] In our study, MRCP examination performed on patients with suspected SOD presenting with biliary type pain and intermittent or persistent mild elevations in serum levels of liver and/or pancreatic enzymes revealed that the common bile duct was terminated by thinning at its distal end and there was a 10–12 mm widening of the common bile duct. In patients who underwent ERCP with the suspicion of SOD, the common bile duct was 10–12 mm wide or more in cholangiography. Only ES was performed in the procedure. Manometry could not be performed in these patients due to technical conditions. Clinical and laboratory improvement was achieved in patients after ERCP. Papilla tumours, pancreatic tumours, chronic pancreatitis and irritable bowel disease are included in the differential diagnosis of SOD. In our SOD cases, there was thinning at the distal end of the common bile duct in MRCP and ERCP. In these patients, clinical findings, MRCP, abdominal USG and CT images of pancreatic tumour and papillary tumour were not reported. Biochemical tumour markers were studied in suspicious cases. Tumour markers were normal. There was no information in the patient’s anamnesis about chronic pancreatitis. There was cholestasis in the laboratory values of the patients who had ERCP with the pre-diagnosis of SOD. These patients had no signs of irritable bowel disease. In addition, laboratory and clinical improvement was observed in patients after ERCP. For these reasons, the diagnosis of SOD was considered in these 22 patients who underwent ERCP. ERCP and percutaneous drainage are options in mechanical junds. In general, ERCP is preferred because it provides more diagnostic information can invisualise in the papilla and pancreatic duct along with the bile duct and has a wider treatment spectrum. Only 20% of patients with malignant obstruction can be operated. Half of them undergo resection and the other half undergo palliative bypass.[25] Today, drainage with ERCP is accepted as the standard treatment for malignant bile duct obstruction. Treatment of hilar cholangio cancer (CC) can be curative or palliative. Curative treatment is resective surgery. However, the number of patients who have this chance due to the location of the tumour is extremely low. Palliative surgery is applied in patients who cannot undergo curative surgery (such as peritoneal implant, local lymph node involvement) (bilioenteric bypass). In cases with advanced malignant hilar stenosis, stenting is performed with ERCP to relieve jaundice. Endoscopic stenting is an important alternative to surgical bilioenteric anastomosis and percutaneous transhepatic drainage due to its low morbidity and mortality. In CC, methods such as MRCP, MR and CT are frequently used today to show the tumour. MRCP with contrast MR imaging is particularly useful in the evaluation of CC due to its high soft-tissue resolution and its capacity to show peripheral ductal involvement.[26] Two-thirds of patients with CC are between the ages of 50 and 70. There is a slight male dominance. Clinical presentation is highly dependent on tumour localisation. More than 90% of patients present with obstructive jaundice.[27] The ages of our patients who underwent ERCP with bile duct malignancy in this series were between 55 and 82 years, and 12 of 22 patients were male and 10 were female. All patients had obstructive jaundice. Local invasion and/or peritoneal implants were detected in pre-operative imaging methods in patients who underwent ERCP stenting due to CC. They were not suitable for surgical resection. Our data were in agreement with the literature. We think that our high age range is due to the delay in diagnosis and/or the late presentation of patients to their health centres. In studies, ERCP has an 80%–90% success rate in the endoscopic palliative treatment of inoperable patients.[28] Seventy per cent of pancreatic tumours are seen in the head, neck and uncinate process, and it usually presents with obstructive jaundice due to narrowing of the intrapancreatic part of the common bile duct.[29] MRCP is a non-invasive method and should be performed before ERCP if needed. ERCP also allows to provide drainage in patients with obstructive jaundice and to take biopsy when necessary.[30] In MRCP, abrupt termination at the level of the mass in the common bile duct and common pancreatic duct and ductal enlargement above the obstruction are observed.[31] In this series, significant narrowing of the common bile duct segment with the mass was observed in patients with a diagnosis of pancreatic head tumour presenting with obstructive jaundice due to obstruction of the bile duct or metastases in the liver (intrahepatic bile duct obstruction and/or intrahepatic cholestasis). Thirty patients with inoperable pancreatic head tumour underwent common bile duct stenting with ERCP for palliation. Since papillary tumours cause early symptoms, they are often small tumours at the time of diagnosis, and the tumour itself may not be visualised. In such cases, joint enlargement of the bile ducts and pancreatic duct (double duct sign) can be seen in MRCP, while in some patients, there is only enlargement of the common bile duct. Dilatation in the choledoch usually ends bluntly.[19] In this study, ERCP was performed on 20 patients with the diagnosis of papilla tumour. Six of these patients were diagnosed during the procedure. Papillary biopsy was performed in all of these patients during ERCP. The biopsy results of 11 patients were malignant. Biopsies of nine patients could not be evaluated due to insufficient material intake. In the MRCP performed for the diagnosis of these cases, asymmetric narrowing at the papilla level and enlargement in the upper part of the distal common bile duct were observed. There was dilatation in the common bile duct and bile ducts. Patients treated with papilla tumour presented with abdominal pain, obstructive jaundice and elevated cholestatic enzymes. A common bile duct stent was placed in these patients with ERCP. The patients whose clinical findings, laboratory values, imaging methods, ERCP findings and biopsy results were confirmed as papillary tumor were transferred to the surgery and medical oncology clinics for further treatment and follow-up. The factor that initiates the event in ABP due to gallstones; permanent biliopancreatic duct obstruction or periampullary oedema that develops even if the stone falls. The longer the duration of permanent biliopancreatic duct obstruction, the greater the possibility of worsening pancreatic damage.[32] Whether there is an accompanying acute cholangitis picture in ABP is important in making an ERCP decision. In the presence of acute cholangitis, mortality decreased from 100% to 9%–40% after antibiotic treatment and biliary decompression. ERCP and ES and/or stenting are widely used in acute cholangitis and biliary pancreatitis.[33] In this series, 140 of the patients who underwent ERCP presented with ABP. Choledochal stone/mud was detected in 139 of these patients. One had hydatid liver cyst that opened into the bile ducts. The amylase and lipase values of the patients presenting with ABP were at least three times higher than normal, and oedema around the pancreas and free fluid in the abdomen were reported in abdominal USG and/or abdominal CT imaging. ERCP was applied to these patients on an average of 5 days after medical treatment was applied. These patients underwent ES in ERCP, and stone extraction and/or stenting was applied to those with common choledochal stones. ES and common bile duct stenting were performed in a patient who presented with a ruptured liver hydatid biliary tract. Complication rates after surgical treatment are reported to be 8%–80% in studies, and intra-abdominal infection or abscess, bleeding, infection of the cyst cavity, biliary fistula and surgical site infection are reported as complications.[34] Biliary complications occur in 14%–16% of patients following liver hydatid cyst surgery. In the study of Küçük et al., the rate of biliary fistula was reported as 3.2%.[35] In a study, this rate was recorded as 9.3% and it was shown that this biliary fistula in patients closed spontaneously in 64%.[36] Endoscopic treatment of biliary fistulas after liver hydatid surgery; sphincterotomy, removal of debris from the bile duct, and endoscopic stenting or nasobiliary drainage. In this study, ERCP were performed in 12 patients due to bile leakage after liver hydatid surgery. Nine of the patients were female and three were male. In the follow-up of the patients after liver hydatid cyst surgery, an average of 700 cc or more continued bile leakage. ERCP was performed due to the continuation of bile drainage and high flow rate. ES was performed in all patients. Bile ducts were drained into the duodenum and a plastic stent was placed in the common bile duct. Bile drainage decreased gradually in eight patients, and bile leakage ceased in an average of 10 days. Bile leakage in four patients lasted longer than 1 month. Bile leakage ceased in 1.5–2 months in three of these patients. Cholestasis symptoms and abdominal pain developed in one patient in the 2^nd month. This patient underwent repeat ERCP. Asecond plastic stent was placed next to the plastic stent. The patient is still being monitored. Bile leakage decreased below 100 cc per day. Bile duct injuries are one of the serious complications that cause increased morbidity and mortality in patients after cholecystectomy. While ERCP helps to determine the type of injury, intracanal pressure is reduced and fistula closure is accelerated by sphincterotomy and/or stenting. In this study, ERCP was performed on 10 patients with bile leakage after laparoscopic cholecystectomy. There were leakage from the common bile duct in five patients and leakage from the cystic duct in four patients. Leakage site could not be observed in one patient (biliary fluid came from the intra-abdominal drain of the patient, whose leak site could not be observed). In these patients, abdominal pain, increased cholestasis enzymes and free fluid (bile) in the abdomen were detected in the early post-operative period after cholecystectomy. Drains were placed in the intra-abdominal fluid region with the interventional radiological method before ERCP in two patients. Bile came from the drain [Figures 1–3]. Patients with post-operative bile leakage were taken to ERCP within an average of 3 days. ES was performed on the patients and a 10 Fr (12 cm) plastic stent was placed. In three of the patients, common bile duct stone extraction was also performed. Bile leakage was stopped in nine patients in an average of 7 days. In one patient, bile leakage lasted more than 1 month. Three patients with circumferential choledochal injuries who could not be treated with ERCP were not included in this study. These patients underwent hepaticojejunostomy surgery. Major complications are more likely to occur in patients with obesity and serious systemic diseases, as well as in patients undergoing pancreatic manometry and complex methods.[37] Complication rates after ERCP range from 3.8% to 20.8%.[38] In our study, the overall complication rate was 17.4%, which was consistent with the literature. A total of 20.8% complications were developed in a study in which Imbeth et al. examined 96 cases; pancreatitis was most commonly seen with 11.5% and the mortality rate was 2.1%.[38] Our study included 1000 cases, complications (17.4%) and PEP and post-ERCP amylasemia total (14.3%) rates were consistent with this study and intraoperative death was never observed. PEP is the most common complication of ERCP. In the literature, the development of pancreatitis after ERCP has been reported between 3%-10%.[6] Most cases are mild pancreatitis with a mortality rate of 0.7%. In our study, 93 (9.3%) patients developed amylasemia that did not require treatment. These patients were discharged 1 day after the procedure. Post-ERCP acute pancreatitis (PEP) developed in 50 (5%) patients. This rate is compatible with the literature. The risk of PEP increases in younger age (under 55 years) and female gender. Consistent with the literature, in our study, although PEP was seen more frequently in women and at a younger age (61.3 years vs. 63.5 years), there was no significant difference in rate. When ERCP was compared in terms of the number of ERCP procedures and the development of PEP, 46 (92%) cases who developed PEP developed among those who underwent the first ERCP procedure, and 4 (8%) among those who underwent a second or more repeated ERCP procedure. The development of PEP is observed significantly more frequently in those who underwent the first ERCP. The most common infectious complication of ERCP is ascending cholangitis. In various publications, the incidence of cholangitis after ERCP, which is clinically important, is reported to be between 0.4% and 10%.[39] In this study, post-procedural cholangitis developed in 20 (2%) patients. In one of these, cholangitis was severe and a 35-year-old female patient died. This patient had a history of tumours originating from the peripheral nervous system in the lumbosacral region. Bleeding after diagnostic ERCP is rare, and the main cause of bleeding is sphincterotomy. In this way, the rate of bleeding has been reported as 0.3%–2% in the literature.[40] In the June 2020 review of Tringali et al., ERCP-related serious bleeding complications are reported between 1% and 2%.[41] In our series, bleeding occurred during sphincterotomy in 10 (1%) patients. Bleeding was stopped during the procedure with balloon pressure and adrenaline injection. No bleeding was observed in the late period. There was no need for blood transfusion after the procedure. If the platelet count of patients with risky comorbidities is <50,000/mm³, the number should be kept between 50 and 80,000 by giving platelet suspension with appropriate timing.[9] Although it is recommended in review studies that keeping the international normalised ratio (INR) value below 1.2 is safe, there are also studies claiming that routine INR measurements are not required and that it does not affect bleeding after ERCP.[42]

Pre-procedural haemogram and INR values were studied for all of our patients included in this study. After the haematology consultation, thrombocyte suspension was given to the patients with low platelet counts. Patients whose INR values were within the normal limits were included in the process. In our practice, the patients were taken to the ERCP procedure by keeping the INR value below 1.5. Fresh frozen plasma was transfused and processed in two patients with high INR (in the range of 1.5–4.3). Anticoagulant drugs were discontinued 5 days before the procedure. It was started again 3 days after the procedure. Although there are many studies recommending discontinuation of aspirin (especially if papillectomy/resection is to be performed), there are also studies in the literature claiming that this is not necessary in recent years.[43] In our practice, we stop aspirin 5 days before the procedure. Although perforations are rarely observed after ERCP, they can cause serious complications. The mortality rate is higher than 37.5%.[44] ERCP is routinely applied with ES, and perforation develops in this procedure at rates from 0.3% to 1%. It was observed that patients with perforation were also mortal with a rate of 16%–18%.[45] Conservative treatment is emphasised in perforations that develop as a complication of ERCP, unlike standard gastrointestinal perforations, and the surgical decision should be applied to highly selected patients. In the study conducted by Wu et al. on 30 perforation cases who occurred in the 6620 ERCP procedures, it was reported that surgery was applied in 10 of these patients, 5 had periampullary perforation, 3 had free duodenal perforation and the perforation area could not be detected by laparotomy in 2 patients.[46] The most important reason for perforation is uncontrolled incision and incision by keeping most of the sphincterotomy wire in the papilla. The incision angle should be well adjusted and the guidewire passing through the sphincterotome should be in the common bile duct. In this study series, there was no perforation due to sphincterotomy in the ERCP procedure. The most common site of diverticula in the duodenum is the 2.5 cm circumference of the ampulla of Vater. The presence of peri-ampullary diverticulum reported in the literature is between 6% and 20%.[47] Cannulation failure resulting from the change of papilla position is also accepted as another morbidity caused by peri-ampullary diverticulum, but data on this subject are contradictory. In addition, the presence of this diverticulum increases the risk of stone formation in the common bile duct after ERCP. Tyagi et al. showed that the size of the diverticulum and the location of the papilla had no effect on cannulation success.[48] Peri-ampullary diverticulum was present in 96 (9.6%) of our patients in this study. Of these cases, 70 were female and 26 were male. Our cannulation time was longer in patients with peri-ampullary diverticulum. The complication rate was not different from the cases without diverticulum. Among the late complications of ERCP are 5%–10% restenosis and cholecystitis after 5 years. Papillary stenosis after an adequate sphincterotomy is clinically rare. Since this study period was not long enough to evaluate late complications, information on this subject was not presented.

CONCLUSION

With the introduction of imaging methods such as MRCP into clinical practice, ERCP rapidly lost its diagnostic function. Despite the possibility of serious complications, we believe that performing ERCP in appropriate indications and early diagnosis of complications are the most important rules in preventing morbidity and mortality. We also believe that it is appropriate for a team to evaluate the indications and complications of the ERCP procedure. Consultation of indications and possible complications by general surgeon, gastroenterology and radiology specialist is very important for patient and physician safety.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.