Endoscopic retrograde cholangiopancreatography in pediatric and adult populations: 17-year experience from Pakistan with trends, complications, and global comparison

Abstract

BACKGROUND

Endoscopic retrograde cholangiopancreatography (ERCP) is an essential diagnostic and therapeutic procedure for pancreatobiliary disorders. However, few large-scale studies from South Asia have examined long-term ERCP outcomes, particularly using established quality benchmarks.

AIM

To evaluate ERCP indications, success rates, complications, and quality performance at a high-volume tertiary care center in Pakistan over a 17-year period.

METHODS

This retrospective study analyzed 13215 ERCP procedures performed between 2006 and 2023. Data included demographics, indications, cannulation rates, complications, and pediatric cases. Findings were assessed against American Society of Gastroenterology/European Society of Gastrointestinal Endoscopy quality indicators.

RESULTS

Biliary ERCP accounted for 93.1% of procedures; choledocholithiasis was the most common indication (40%). Cannulation success was 93.9% for biliary and 94.2% for pancreatic ERCP. Pediatric ERCP comprised 4% of cases, mostly for stones and chronic pancreatitis. Bleeding (1.7%) and post-ERCP pancreatitis (2.3%) were the most frequent complications. Performance met or exceeded most American Society of Benchmarks.

CONCLUSION

This study offers insight into nearly two decades of ERCP practice within a public sector hospital. Our experience echoes the quality and efficiency of ERCP not previously available in Pakistan. As healthcare systems in resource-limited sectors expand, our findings serve as a reference point. Continued training and quality improvement studies can further enhance ERCP effectiveness in the region and beyond.

Key Words: Endoscopic retrograde cholangiopancreatography; Pediatric; Adult populations; Global comparison; Biliary; Pancreatic; Pakistan

Core Tip: We performed a retrospective analysis of 13215 endoscopic retrograde cholangiopancreatography (ERCP) procedures at a single tertiary care hospital in Karachi, Pakistan. We aimed to identify the quality, safety, and efficacy of ERCP in South Asia. The findings demonstrated the dependability of ERCP in a resource-constrained environment with excellent cannulation success rates (over 93%) and a low incidence of significant problems. The findings highlighted the critical need to increase ERCP access and specialized training throughout the region to improve outcomes for individuals with pancreatobiliary illnesses.

INTRODUCTION

Endoscopic retrograde cholangiopancreatography (ERCP) is a minimally invasive therapeutic endoscopic procedure for the management of pancreatobiliary disorders. It is the first-line therapeutic modality for choledocholithiasis, periampullary tumors causing obstructive jaundice, management of periprocedural biliary adverse effects, and a spectrum of pancreatic disorders such as chronic pancreatitis[1]. A skilled endoscopist with the capacity to guide case selection based on high-quality evidence is essential to the success of ERCP. Therefore, a thorough benefit-risk assessment of ERCP must be undertaken for each patient. The most common risks of ERCP include pancreatitis (1.6%-15.7%), hemorrhage (1.3%), perforation (0.1%-0.6%), and cholangitis (1.0%)[2].

ERCP is an advanced procedure under intense scrutiny in training hospitals and credentialing bodies[3]. In Pakistan, although a large proportion of patients come from low socioeconomic backgrounds[4], the ERCP procedure is conducted free-of-charge in only a few facilities. Currently, there are limited nationwide, population-based ERCP audits that have been published. In the published studies the cohorts have been relatively small (3544 patients). Healthcare authorities remain focused on improving and standardizing the quality of endoscopy procedures. The American Society of Gastroenterology (ASGE) has provided specific training criteria for ERCP, and these guidelines detail the high-quality skills required to perform the procedure[5]. Herein we describe our retrospective analysis of ERCP procedures performed in the previous 17 years to evaluate the indications, procedure success, and safety of ERCP.

MATERIALS AND METHODS

Operational definitions

The registry defines successful ductal cannulation as the injection of contrast material into the duct, whereas failure occurs when the endoscopist fails to cannulate the papilla following visualization. We classified cases as unspecified jaundice if imaging revealed a dilated common bile duct (CBD) and laboratory results indicated elevated bilirubin levels.

Study design

Our endoscopy unit recently transitioned into an institute by the name of the Sindh Institute of Advanced Endoscopy and Gastroenterology. It is in the Dr Ruth K.M. Pfau Civil Hospital, which is one of the largest tertiary care public hospitals in Pakistan. Over 13000 ERCP procedures have been conducted free of cost from January 2006 to December 2023 at our hospital. The current unit is a 40-bed endoscopy facility staffed by a team of expert gastroenterologists and surgeons. The endoscopists were fully trained, having conducted more than 500 procedures annually. We are financially supported by the government and philanthropic endeavors. Our unit receives referrals from all over the country because our hospital is one of the busiest tertiary care facilities in Pakistan.

Inpatient and outpatient ERCP procedures are performed when the endoscopist deems the procedure necessary. A registry of all procedures is recorded on an in-house database. The database includes a record of patient identification, relevant lab investigations, and recent imaging findings along with procedure findings and outcomes. All records, including follow-up procedures, can be retrieved on demand, allowing easy analysis of procedure outcomes, including adverse effects. Key quality indicators set by the ASGE were mostly met by our unit, reflecting the good practice of ERCP (Table 1)[6]. Data was collected and analyzed retrospectively. To the best of our understanding, this was the largest retrospective ERCP registry to date in South Asia.

Table 1 Proposed quality indicators for endoscopic retrograde cholangiopancreatography.

Quality indicators	Grade of recommendation	Measure type	Performance target, %	Target achieved
Pre-procedure
Frequency with which ERCP is performed for an indication that is included in a published standard list of appropriate indications, and the indication is documented (priority indicator)	1C+	Process	> 90.0	Yes
Frequency with which informed consent is obtained, including specific discussions of risks associated with ERCP, and fully documented medical consent is obtained and fully documented	1C	Process	> 98.0	Yes
Frequency with which appropriate antibiotics for ERCP are administered for settings in which they are indicated	2B	Process	> 98.0	Yes
Frequency with which ERCP is performed by an endoscopist who is fully trained and credentialed to perform ERCP	3	Process	> 98.0	Yes
Frequency with which the volume of ERCP procedures performed per year is recorded per endoscopist	1C	Process	> 98.0	Yes
Procedure
Frequency with which deep cannulation of the ducts of interest is documented	1C	Process	> 98.0	Yes
Frequency with which deep cannulation of the ducts of interest in patients with native papillae without surgically altered anatomy is achieved and documented (priority indicator)	1C	Process	> 90.0	Yes
Frequency with which fluoroscopy time and radiation dose are measured and documented	2C	Process	> 98.0	No
Frequency with which CBD stones (1 cm) in patients with normal bile duct anatomy are extracted successfully and documented (priority indicator)	1C	Outcome	> 90.0	Yes
Frequency with which stent placement for biliary obstruction in patients with normal anatomy, whose obstruction is below the bifurcation, is successfully achieved and documented (priority indicator)	1C	Outcome	> 90.0	Yes
Post-procedure
Frequency with which a complete ERCP report that details the specific techniques performed, particular accessories used, and all intended outcomes is prepared	3	Process	> 98.0	Yes
Frequency with which acute adverse events and hospital transfers are documented	3	Process	> 98.0	Yes
Rate of post-ERCP pancreatitis (priority indicator)	1C	Outcome	N/A	Yes
Rate and type of perforation	2C	Outcome	< 0.2	Yes
Rate of clinically significant hemorrhage after sphincterotomy or sphincteroplasty in patients undergoing ERCP	1C	Outcome	< 1.0	Yes
Frequency with which patients are contacted at 14 days or greater to detect and record the occurrence of delayed adverse events after ERCP	3	Process	> 90.0	Yes

CBD: Common bile duct; ERCP: Endoscopic retrograde cholangiopancreatography; N/A: Not available.

Inclusion and exclusion criteria

All patients who underwent ERCP in the stated timeline were included. Incomplete or records with missing data were excluded.

Study aim

The primary outcome was to evaluate the success of biliary or pancreatic cannulation in a large retrospective cohort. Secondary outcomes included intra-procedural and immediate adverse events, clinically significant bleeding, perforation, and death.

Statistical analysis

The collected data were uploaded to Microsoft Excel and then imported to IBM SPSS statistics software, version 25 (IBM Corp., Armonk, NY, United States). In addition to descriptive statistical techniques such as means, percentages, and standard deviations, the study data were analyzed using the Student’s t-test for independent univariate analysis of measurement data and the χ² test for independent univariate analysis of qualitative data. To identify independent predictors of successful cannulation, a multivariate regression analysis was also performed. This allowed us to adjust for potential confounding variables and gain a better understanding of the factors influencing procedural outcomes. The significance level was set at P < 0.05, and the results were assessed within the 95% confidence interval.

RESULTS

Overview of study data

This study reviewed 17 years of retrospective data from 13215 ERCP records. The average age of the patients was 46.5 years with 4% aged under 18 years. Females comprised of 59.2% cases. For expert anesthesiologists, midazolam was the sedation of choice in 95.1% of cases followed by injection of propofol (6.3%) in both pediatric and adult patients. Demographics of patients along with pre-procedure investigations are presented in Table 2.

Table 2 Demographics and procedural parameters of the included endoscopic retrograde cholangiopancreatography procedures.

Characteristics	n (%)/mean ± SD
Sex
Male	5392 (40.8)
Female	7823 (59.2)
Age in years
< 18 (pediatric)	533 (4.0)
0-3 (toddlers)	6
4-11 (childhood)	126
12-18 (adolescence)	401
> 18 (adult)	12682 (96.0)
Mean age, years	46.50 ± 16.03
Clinical symptoms
Jaundice	6308 (47.7)
Abdominal pain	5911 (44.7)
Other	996 (7.6)
Laboratory parameters
Increased total bilirubin	8515 (64.4)
Mean total bilirubin in mg/dL	9.50 ± 48.86
Increased ALP	10256 (77.6)
Mean ALP in IU/L	605.6 ± 1218.4
Post-ERCP pancreatitis	356 (2.7)
Mild pancreatitis	164 (1.2)
Moderate pancreatitis	136 (1.0)
Severe pancreatitis	56 (0.4)
Clinically significant bleeding after EST/EPLBD	230 (1.7)
Hemostasis with balloon tamponade	35 (0.2)
Hemostasis with adrenaline spray	195 (1.5)
Cardio pulmonary adverse effects	31 (0.2)
Perforation	26 (0.2)
Death	5 (0.0)

ALP: Alkaline phosphatase; ERCP: Endoscopic retrograde cholangiopancreatography; EST: Endoscopic sphincterotomy; EPLBD: Endoscopic papillary large balloon dilation.

Procedural type and access

A total of 12298 procedures were biliary ERCP, and 567 procedures were pancreatic ERCP. The ampulla was inaccessible due to gastric or duodenal outlet obstruction in 350 patients. Successful cannulation during index ERCP for biliary cases was 93.9% and 94.2% for index ERCP for pancreatic cases (Figure 1 and Table 3).

Figure 1 The consolidated standards of reporting trials. ¹Access to the ampulla.

Table 3 Endoscopic retrograde cholangiopancreatography cases.

Type	n	%
Biliary cases	12298	100.0
Native	9196	74.8
Follow-up	3102	25.2
Pancreatic cases	567	100.0
Native1	346	61.0
Follow-up	221	38.9

¹Inadvertent pancreatic duct cannulation was reported in 202 cases, they were not included as pancreatic cases.

Indications and outcomes of ERCP

Indications for ERCP referral are presented in Table 4.

Table 4 Imaging findings in native cases (indications).

Types	n	%
Benign biliary cases
Choledocholithiasis	3682	40.0
Bile leak	466	5.1
Choledochal cyst	22	0.2
CBD worms	21	0.2
Unspecified jaundice1	3046	33.1
Malignant cases
Pancreatic mass	389	4.2
Perihilar	145	1.6
Periampullary mass	1307	14.2
Gallbladder mass	118	1.3
Pancreatic cases
Chronic pancreatitis	201	58.1
Other	145	41.9

¹Unspecified obstructive jaundice was defined as increased total bilirubin (> 1.3 mg/dL) and alkaline phosphatase (> 125 U/L) along with evidence of biliary dilatation on imaging.

CBD: Common bile duct.

Unspecified obstructive jaundice

Of the 3046 unspecified obstructive jaundice cases, 54.1% had CBD stones, 38.4% had biliary strictures, 7.3% had normal examination findings.

Malignant strictures

Index biliary ERCP was performed in 1959 cases for malignant indications with a cannulation success rate of 92.7% (Table 5)[7]. Males were most frequently found to have peri-ampullary mass (23.9%) followed by pancreatic malignancy (14.3%). In females’ perihilar cholangiocarcinoma (36.0%) was the most common malignant etiology for biliary obstruction followed by peri-ampullary mass (30.6%; Figure 2).

Figure 2 Distribution of malignant strictures by sex. CA: Cancer.

Table 5 Endoscopic retrograde cholangiopancreatography findings of native biliary and pancreatic cases.

ERCP findings	n	%
Benign cases
CBD stones	4901	53.2
Mirizzi syndrome	53	0.6
Normal CBD	1545	16.8
Bile leak	740	8.0
Post-laparoscopic cholecystectomy stricture	55	0.6
CBD worms	28	0.3
Choledochal cyst	61	0.7
Cannulation successful in benign biliary cases	6955	94.2
Cannulation unsuccessful in benign biliary cases	428	5.8
Malignant strictures
Periampullary cancer	477	5.2
Gallbladder mass	66	0.7
Perihilar stricture	329	3.6
Pancreatic mass	262	2.9
Indeterminate stricture	679	7.4
Total malignant strictures	1813	19.7
Gastric outlet obstruction, ampulla not accessed	350
Cannulation successful in malignant strictures	1681	92.7
Cannulation unsuccessful in malignant stricture cases	132	7.3
Pancreatic cases
Normal PD	27	7.8
Dilated PD	160	46.2
DPDS	22	6.4
PD stones	83	24.0
PD stricture	80	23.1
Pancreatic divisum	75	21.7

CBD: Common bile duct; PD: Pancreatic duct; DPDS: Disconnected pancreatic duct syndrome.

Biliary stones

Table 6 highlights the characteristics of cases that underwent ERCP for CBD stones. Of the 6323 cases of biliary stones, majority were females. Ductal clearance was achieved in 77.5%. However, 1422 (22.5%) patients needed repeat procedure. Majority of stones were retrieved after sphincterotomy (67.8%) and sphincteroplasty (8.3%). Multivariate logistic regression analysis was performed to assess factors associated with successful CBD and pancreatic duct (PD) cannulation. For CBD cannulation, the variables included presence of Mirizzi’s syndrome (P = 0.998), biliary stricture (P = 0.993), CBD stone (P = 0.989), gender (P = 0.721), and age group (P = 1.000). None of these demonstrated a statistically significant association with CBD cannulation success. Similarly, for PD cannulation, the factors analyzed were age group (P = 0.518), gender (P = 0.523), PD stones (P = 0.987), PD strictures (P = 0.985), and disconnected PD syndrome (P = 0.994). None of these variables were significantly associated with PD cannulation either. These findings indicate that the assessed clinical and demographic factors did not independently influence the outcomes of either CBD or PD cannulation in this study.

Table 6 Characteristics of cases with common bile duct stones that underwent endoscopic retrograde cholangiopancreatography.

CBD stones	n (%)
Biliary stones	6323 (52.3)
Duct clearance
Index duct cleared	4901 (77.5)
Follow-up duct clearance	1422 (22.5)
Interventions
Sphincterotomy	4883 (39.7)
Sphincteroplasty	965 (7.8)
Basket	418 (3.4)
Spyglass EHL	201 (1.6)
Maneuvers
Biliary sphincterotomy	8347 (67.8)
Balloon sphincteroplasty	1024 (8.3)
Needle-knife papillotomy	714 (5.8)
Cholangioscopy and EHL	214 (1.7)
Trapezoid basket	432 (3.5)
Soehendra dilator	94 (0.8)
Biliary stenting	5381 (43.8)
CBD cleared in one session	1683 (13.7)

All the percentages were calculated from the total biliary cases. CBD: Common bile duct; EHL: Electrohydraulic lithotripsy.

Pancreatic ERCP

Pancreatic ERCP procedures were performed in 567 cases, of which 61.0% were with native papilla. Chronic pancreatitis accounted for 58.1% cases. With ductal stones and strictures being predominant findings. In 75 cases, pancreatic divisum was diagnosed via pancreatogram performed during pancreatic ERCP.

Pediatric ERCP

Pediatric ERCP procedures were performed in 533 patients with an average age of patients being 14 years. This included 401 adolescence, 126 patients between 4-11 years, and 6 patients under 4 years. Choledocholithiasis was the indication for 41% of the pediatric cases, followed by chronic pancreatitis in 35%. The imaging findings of native cases are presented in Table 4. No major complications were reported in pediatric patients. Sedation was tailored according to age and weight.

Peri-procedural and immediate adverse effects

Peri-procedure bleeding was observed in 1.7% of all performed procedures. However, significant bleeding occurred in 2 patients. These patients required blood transfusion, whereas the remaining cases achieved hemostasis during the procedure. They were labelled as minor bleeds per the Cotton et al[8] classification (Table 2). Post-ERCP pancreatitis was observed in 2.3% of patients. Based on established guidelines[8], we defined pancreatitis as amylase > 3 times the upper limit of normal and further classified based on severity as mild, moderate, and severe. Around 68.3% of patients were discharged within 24 hours (mild), 31.3% were discharged within 48 hours (moderate), and 0.5% required hospitalization for more than 4 days (severe). The 30-day mortality after ERCP was 5 deaths out of 13,215 procedures.

DISCUSSION

ERCP is an ever-evolving advanced procedure that demands standard quality and safety. Meticulous training and adherence to protocol are of the utmost importance to achieve the best outcomes. The ASGE guidelines have guided ERCP practice in our unit, which is located in one of the busiest tertiary care centers in the 12^th largest city in the world. Our unit comprises a team of expert endoscopists, surgeons, and anesthesiologists. The objective for this study was to assess the quality of ERCP procedures and report the success rate, complications, and findings for ERCP. This seventeen-year-old retrospective study is the largest recorded dataset for ERCP from South Asia to date. High cannulation success was observed (biliary ERCP 93.9%) and (pancreatic ERCP 94.2%) which aligned with ASGE benchmarks[9].

Park et al[10] conducted a nationwide survey in Korea about ERCP trends. They reported that ERCP was performed more frequently with increasing age of patients, likely linked to gallstones and pancreatobiliary malignancy. This is a stark contrast from our study findings where the average age of patients was mid-forties. Another study conducted by Moffatt et al[11] stated that ERCPs were more commonly done in women than in men (62%). The results tally with our study, where women more commonly had ERCP. A study by Parveen et al[12] reported a much higher incidence of perihilar carcinoma in both male and female populations. In our cohort, female predominance of perihilar cholangiocarcinoma. This suggests a possible biological difference that warrants for further investigation.

Our descriptive data revealed that 22% of patients achieved ductal clearance in the index ERCP. 59.64% of patients required only a single ERCP procedure (Table 7). Since 2013, Spyglass cholangioscopy has been occasionally used for selective cases (1.6%). Due to cost of equipment, the impact of new technology is yet to be further explored. Keil et al[13] retrospective study corroborated our findings with a cannulation success rate of 96% in peadiatric ERCP with choledochal cyst, choledocholithiasis, and pancreatic pathology as the most common findings. In contrast, the primary diagnosis in our unit was biliary stones, followed by pancreatic pathologies. A study conducted in India quoted similar findings to our unit[14]. Post-ERCP pancreatitis was the most frequent adverse effect. Significant gastrointestinal bleeding was infrequent. The rate of bleeding was below 1% and aligns with the standard quality indicators of ERCP recommended by both European and American societies of gastroenterology.

Table 7 Number of endoscopic retrograde cholangiopancreatography procedures per patient.

Number of procedures	n	%
18	1	0.007
17	1	0.007
16	1	0.007
15	1	0.007
14	1	0.007
13	1	0.007
12	2	0.015
11	3	0.022
10	5	0.037
9	9	0.068
8	20	0.151
7	61	0.461
6	109	0.824
5	255	1.929
4	631	4.774
3	2116	16.01
2	2116	16.01
1	7882	59.64
Total	13215	100

This study has several notable strengths. It was the largest single-center investigation conducted within a tertiary care facility that was offered entirely at no cost to participants. Importantly, it marked the first endeavor of this scale in South Asia and delivered meaningful data regarding the etiologies of both biliary and pancreatic ERCP procedures. Moreover, it comprises the most extensive dataset of pediatric pancreatic ERCP procedures (Figure 3) reported in Asia to date. Nevertheless, several limitations should be acknowledged. The retrospective design may introduce certain biases that are difficult to eliminate. Definitive histopathological confirmation was unavailable for several malignant strictures, and endoscopic ultrasound data were not incorporated as it was beyond the objective of this article. Additionally, post-procedural follow-up for delayed complications was suboptimal including cholangitis, potentially impacting the reliability of the study outcomes.

Figure 3 Distribution of etiologies by life stage. A: Pediatric population; B: Adult population. CBD: Common bile duct; DPD: Disconnected pancreatic duct syndrome.

CONCLUSION

This study offers insight of nearly 2 decades of ERCP practice within a public sector hospital. Beyond the numbers, our experience echoes the quality and efficiency of ERCP that was otherwise not available for Pakistan. As healthcare systems in resource limited sectors move to expand, our findings can serve as a reference point. Moving forward, continued training measures, and further studies with quality improvements can help to further the effectiveness of ERCP within our region and beyond.