Endoscopic retrograde cholangiopancreatography (ERCP) includes the simultaneous application of two methods: upper endoscopy and X-ray technology. During the examination, a special side-viewing upper endoscope is inserted into the duodenum, allowing instruments to be passed through the hepatopancreatic ampulla and into the biliary and pancreatic ducts. The ducts are examined by injecting a contrast agent, enabling radiographic examination and therapeutic interventions. Owing to advances in technology, this method is effective for the endoscopic therapy of gallstones, bile duct strictures, and pancreatic and liver diseases. Periodically, this manipulation is repeated for several reasons. ERCP, a complicated procedure, requires special training and experience. It can cause more serious problems than other procedures such as upper endoscopy or colonoscopy.¹

The incidence of adverse events associated with ERCP has been reported to be 5%–12%, while the mortality rate is 0.1%–1.4%.²

In a systematic review of 21 studies involving 16,855 patients undergoing ERCP, the overall complication rate was 2%–7%.³

Some risk factors associated with the side effects of ERCP include difficult cannulation, precut sphincterotomy, surgically altered anatomy, periampullary diverticulum (PAD), and older age.⁴

In the medical literature, the most common complications of ERCP included bleeding, perforation, acute postERCP pancreatitis, and secondary infection. Acute postERCP pancreatitis can be caused by mechanical injury, contrast media injection, or guidewire manipulation through the pancreatic duct, while bleeding may occur during sphincterotomy. Moreover, perforation can occur during cannulation or therapeutic manipulation, such as bile stone extraction using a special balloon or stone retrieval basket. Infections occurring after ERCP may be due to incomplete drainage of the infected biliary system, cystic duct obstruction, collection of infected pancreatic fluid, or, in rare cases, contamination of endoscopic equipment.^5,6

Currently, ERCP is mostly used for therapeutic intervention. The most common therapeutic manipulations are associated with bile duct or pancreatic duct stones, strictures (malign, benign) or stenosis, bile leaks, cytology, and biopsy. Choledocholitiasis management starts with biliary papillotomy with or without precut sphincterotomy to ease stone passage through the bile duct. Then, a special balloon or stone retrieval basket may be passed over the guidewire into the bile duct to assist in retrieving the stone. Another common treatment modality during ERCP is biliary stent placement in the bile duct because of secondary obstruction to a benign or malignant biliary stricture. Several stents are available that vary in design, material, and size; the choice of a suitable stent depends on the clinical situation.⁷

ERCP is an advanced complex procedure, during which events may occur that require a repeat procedure. Herein, we aimed to determine the extent to which ERCP is performed repeatedly and to investigate the primary diagnosis, possible causes of repeat ERCP, associated factors, 30-day mortality, and overall death rate.

In this retrospective study, we evaluated ERCP procedures performed in Pauls Stradins Clinical University Hospital from 2021 to February 2024. Fifty patients (28 women and 22 men) with a history of repeat ERCP were enrolled, undergoing a total of 122 ERCP procedures.

The inclusion criteria were as follows: the patient underwent ≥ 2 ERCP procedures; patients were men or women and older than 18 years.

The exclusion criteria were: patients aged < 18 years and did not undergo ≥ 2 ERCP procedures.

The results of the patient’s demographic data (sex, age, obesity, history of cholecystectomy, jaundice), 30-day postprocedural mortality, overall death rate, and laboratory markers such as hemoglobin, leucocytes, C-reactive protein, bilirubin, alkaline phosphatase, and creatinine were analyzed. We analyzed data pertaining to the primary diagnosis of the patient, therapeutic intervention used during ERCP, diameter of the common bile duct (CBD), and causes of repeat ERCP.

The data about 30-day postprocedural mortality were obtained from The Office of Citizenship and Migration Affairs of the Republic of Latvia.

All procedures were performed by the same team of endoscopists with compatible levels of ERCP experience (> 10 years).

Statistical evaluation was performed by evaluating the first ERCP procedures and records of repeat procedures.

IBM SPSS Statistics 29.0 software (IBM Corp.) was used for statistical data analysis. A P-value < 0.05 was considered significant. The statistical reliability of the research data was analyzed using the independent samples t-test, Fisher’s exact test, and Pearson’s chi-square test.

The research protocol was approved by the Ethics Committee for Clinical Research at the Development Society of Pauls Stradins Clinical University Hospital (Protocol No. 280923–6L).

Written informed consent was obtained from all participants.

The study included 50 patients: 28 women (56%) and 22 men (44%). The age of the patients was 31–90 years (mean [M] = 69, mediana [Me] = 72, mode [Mo] = 77, standard deviation [SD] = 13.3). Twenty-one patients had a history of cholecystectomy (42%), 25 were obese (50%), and 17 (34%) had PAD.

The independent samples t-test revealed no significant difference between the age of patients and PAD existence (P = 0.35).

The total number of ERCP procedures was 122. From all cases – 35% were Grade 1 (American Society for Gastrointestinal Endoscopy Grading System), 25% - Grade 2, 38% - Grade 3, and 2% - Grade 4. The number of repeat ERCPs were 1–5 (M = 1.44, Me = 1, Mo = 1, SD = 0.76).

The postprocedural 30-day mortality of repeat ERCP was 3.3%; four of 122 procedures ended with a fatal outcome within 30 days after the last ERCP. The causes of death after ERCP were: post-ERCP pancreatitis (n = 2), cholangitis leading to subsequent sepsis (n = 1) and common bile duct perforation (n = 1). The overall death rate was 8%, and a fatal outcome was found in four out of 50 patients.

The independent samples t-test revealed a significant difference between the number of repeat ERCP and exitus letalis (P = 0.025). The mean number of repeat ERCPs for patients with exitus letalis outcome was 2.25, but for those without exitus letalis outcome was 1.37.

We evaluated the laboratory markers before the first and repeat ERCP (Table 1).

Table 1 . Laboratory Markers Before Endoscopic Retrograde Cholangiopancreatography.

Laboratory marker	Mean value	Standard deviation	Minimum value	Maximum value
Hemoglobin (g/L)	122	1.75	78	190
Leucocytes (× 109)	6.70	0.21	2.3	15.5
Bilirubin (µmol/L)	75.50	11.65	3	722
Creatinine (µmol/L)	84.60	3.10	39	300
Alkaline phosphatase (IU/L)	306	29.24	46	1,517
C-reactive protein (mg/dL)	17.72	2.42	1	150

The independent samples t-test revealed a significant difference between all laboratory markers before ERCP and the 30-day mortality outcome of the procedure (Table 2).

Table 2 . Laboratory Markers Depending on the 30-Day Post-Endoscopic Retrograde Cholangiopancreatography Mortality.

Laboratory marker	30-day postprocedural mortality	Mean value	Standard deviation	P-value
Hemoglobin (g/L)	Yes	99.50	14.29	0.018
	No	122.82	19.15
Leucocytes (× 109)	Yes	8.92	2.15	0.028
	No	6.63	2.32
Bilirubin (µmol/L)	Yes	516.50	220.31	0.025
	No	60.63	94.85
Creatinine (µmol/L)	Yes	138.00	36.57	0.001
	No	82.86	32.86
Alkaline phosphatase (IU/L)	Yes	927.50	220.53	0.001
	No	285.08	304.89
C-reactive protein (mg/dL)	Yes	51.50	32.18	0.010
	No	16.58	25.93

The most common primary diagnosis of the patient cohort who underwent repeat ERCP was choledocholithiasis, at 64% (n = 32). Other primary diagnoses were: pancreatic head tumor, 20% (n = 10), papillary stenosis, 6% (n = 3), and hepatic echinococcosis, 6% (n = 3), while distal cholangiocarcinoma was a less common primary diagnosis, at 4% (n = 2).

Using Fisher’s exact test, there was a significant relationship between choledocholithiasis and a history of cholecystectomy (P < 0.001) and obesity (P < 0.001). These anamnestic data increased the risk of choledocholithiasis.

According to our data, the rate of successful cannulation for ERCP reached 88.5%. The independent samples t-test revealed a significant difference between the success of cannulation and bilirubin level (205.64 ± 234.42 µmol/L vs. 58.71 ± 97.65 µmol/L, P = 0.037). This indicates that the higher the bilirubin level, the lower the chance of cannulation of the CBD. Moreover, Pearson’s chi-square test showed a significant relationship between the presence of jaundice and success of cannulation (P = 0.014) (Fig. 1).

Figure 1. Relationship between successful cannulation of the common bile duct and the presence of jaundice.

Based of ESGE criteria, prevalence of difficult cannulation cases (5 minutes, 5 times or 2 pancreatic duct cannulation) were 20%.

Interval ERCP was applied as a salvage maneuver in all cases with initial cannulation failure.

Inadvertent pancreatic duct guidewire passage was noticed in 23% of all ERCP procedures (n = 28). Multiple cannulation (≥ 3 times) of the pancreatic duct was observed in 3.3% of procedures.

The most common invasive intervention used during the ERCP procedure was papillotomy in 96.7% of endoscopic procedures (n = 118). Precut sphincterotomy was used in 12.3% of all ERCP procedures (n = 15), while other interventions used during the ERCP procedure included extraction of bile stone/stones using balloon/basket (49.2%; n = 60), biliary stent removal (38.5%; n = 47), biliary stent implantation (51.6%; n = 63), and pancreatic stent implantation (4.1%; n = 5). During the cholangiopancreatography manipulation, the mean CBD diameter was 13.4 mm (SD = 4.52, minimum = 6, maximum = 27).

The independent samples t-test revealed a significant difference between the CBD diameter and 30-day mortality outcome (P = 0.021) (Fig. 2). PAD was found in 34.4% of all procedures (n = 42). Fisher’s exact test revealed a significant relationship between the success of cannulation and the presence of PAD (P = 0.017) (Fig. 3).

Figure 2. The CBD diameter difference depending on the 30-day post-ERCP mortality outcome. CBD, common bile duct; ERCP, endoscopic retrograde cholangiopancreatography.

Figure 3. Relationship between successful cannulation of the common bile duct and the presence of periampullary diverticulum.

The most common indication of repeat ERCP was biliary stent change (36.1% of 72 procedures (n = 26), while other causes included incomplete evacuation of bile stones (33.3%; n = 24), unsuccessful cannulation of the CBD (12.5%; n = 9), and bleeding after papillotomy (2.8%; n = 2). The least common indications for repeat ERCP were laboratory disturbance (dyscoagulation) and multiple cannulations of the pancreatic duct, each at 1.4% (n = 1).

ERCP procedure-related complications occurred in 10% of all cases (n = 12). The most common complication was postERCP pancreatitis (5.9%; n = 7), while other complications included bleeding (1.6%; n = 2), cholangitis leading to subsequent sepsis (1.6%; n = 2), and common bile duct perforation (0.9%; n = 1).

Currently, ERCP is the primary minimally invasive endoscopic procedure for the diagnosis and treatment of many pancreaticobiliary diseases. In certain cases, patients require at least two ERCP procedures.

Herein, we evaluated the risk factors, outcomes, primary diagnosis, laboratory markers, therapeutic interventions, and indications associated with repeat ERCP.

According to our data, 28 women (56%) and 22 men (44%) required repeat ERCP, with no significant sex difference. Another study showed similar results (52.8% and 47.2%).⁸

Choledocholitiasis was the most common primary diagnosis of patients who received at least two ERCP manipulations (64%; n = 32), which is similar to a previous study, showing that the most common indication (primary diagnosis) for ERCP was choledocholithiasis (49.17%).⁹

We concluded (using Fisher’s exact test) that there is a significant relationship between choledocholithiasis and a history of cholecystectomy (P < 0.001) and obesity (P < 0.001). In 2016, Kharga et al¹⁰ published research that determined the same idea about obesity: the risk of symptomatic cholelithiasis increases with every increase in body mass index. Therefore, the presence of these anamnestic data could be a risk factor for repeat ERCP, considering that the most frequent cause of recurrent ERCP in our study was choledocholithiasis.

ERCP is considered a minimally invasive endoscopic procedure with a lower risk of complications, 30-day postprocedural mortality, and overall death outcome than surgery. Our study showed 30-day postprocedural mortality and overall death outcome of repeat ERCP: 3.3% and 8%, respectively. However, two studies with a larger number of participants reported an overall mortality rate of 0.3%–1.1%.^11,12 The higher mortality rates in our research may be associated with the fact that we analyzed repeat ERCP. Our study showed a significant difference between the number of repeat ERCP and exitus letalis (M = 2.25 vs. M = 1.37) (P = 0.025).

Routinely, before each endoscopic retrograde cholangiopancreatography, certain laboratory tests are performed to assess the severity of the disease and exclude contraindications. Our observations show that fatal procedures within 30 days had more pronounced disturbances in laboratory tests before ERCP. This can be considered a risk factor for negative postprocedural outcomes.

Successful cannulation of the CBD is the most important step in ERCP with rates known to vary among countries and centers (82.6%–98%).^13-15 Our center’s rate of successful cannulation for ERCP reached 88.5%, which is a good indicator. In our research, we noticed several patterns. First, there was a significant difference between the success of cannulation and bilirubin level using the independent samples t-test (P = 0.037): unsuccessful cannulation, 205.64 ± 234.42 µmol/L and successful cannulation, 58.71 ± 97.65 µmol/L. A similar pattern was identified in another study (304.38 ± 126.54 µmol/L vs. 179.55 ± 114.57 µmol/L, P = 0.0009).¹⁶ We found a significant relationship between the presence of jaundice and the success of cannulation (P = 0.014). Elevated bilirubin and jaundice could be risk factors for repeat ERCP because, in 12.5% of cases, unsuccessful cannulation was the causative factor.

Second, there was a significant relationship between the presence of the PAD and the success of cannulation (P = 0.017). Sometimes this structure prevents the endoscope from adequately reaching the papilla of Vater, complicating cannulation. Therefore, PAD could be a risk factor for repeating ERCP.

According to an investigation that enrolled 3,564 patients with and without PAD, patients in the PAD group had a greater average age (65 ± 13 vs. 58 ± 16, P < 0.001).¹⁷ However, in compliance with our calculations, there was no significant difference between the age of patients and PAD existence (P = 0.35). The different numbers of patients involved in the study can explain this difference.

None.

All data generated or analyzed during this study are included in this published article and readily available for share.

No potential conflict of interest relevant to this article was reported.