Factors associated with the severity of ERCP-related complications: a large retrospective single-centre study

doi:10.21203/rs.3.rs-4695526/v1

Download PDF

Research Article

Factors associated with the severity of ERCP-related complications: a large retrospective single-centre study

https://doi.org/10.21203/rs.3.rs-4695526/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Prospective studies have identified a number of patient- and procedure-related risk factors of post-ERCP complications. Many of these complications are mild and self-limiting.

Aims

The aim of this study is to identify the patients at risk of severe, early post-ERCP complications.

Methods

A retrospective cohort study was conducted using data from 2810 ERCP procedures performed at Ghent University Hospital from 2016 to 2022. Data were extracted from electronic patient records and a prospectively maintained ERCP registry. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for severe complications.

Results

Among the 2810 ERCP procedures, 223 cases (7,9%) with post-ERCP complications were analyzed. Severe complications were observed in 20.3% of these cases. Haemorrhage was the most reported severe complication (22/49 severe complications, 44.9%), while a perforation had the most chance to result in a severe complication (10/15 perforation cases, 67%).

Anticoagulative therapy (OR 6.3, 95% CI 1.4–28.3, P = 0.016) and high procedural difficulty (Shutz category 3: OR 11.5, 95% CI 2.4–54.6, P = 0.002; Shutz category 4: OR 5.9, 95% CI 1.4–23.5, P = 0.012) emerged as independent predictors of severe complications.

Conclusions

In patients with an early post-ERCP complication, those under anticoagulant therapy and those who underwent difficult ERCP procedures (Schutz 3 or 4) have the highest risk of a severe course.

ERCP

Risk Factors

Adverse effects

Since its introduction in 1968, ERCP (endoscopic retrograde cholangiopancreatography) has become an indispensable endoscopic technique of the diagnosis and treatment of pancreaticobiliary diseases (1). Despite its minimally invasive nature, the technique has been associated with several complications, including pancreatitis, infections (cholangitis, cholecystitis,…), bleeding, perforation, and anaesthesia-related complications (2–10). Over the past two decades, numerous pro-and retrospective studies have identified various patient- and procedure-related risk factors of post-ERCP complications (11–19).

Luckily, the majority of post-ERCP complications are mild and self-limiting (2, 5).

Little is known about the risk factors associated with the severity of post-ERCP complications. This study aims to identify the risk factors of severity in patients with post-ERCP complications.

2.1. Patients

We performed a retrospective cohort study. Patient data from all ECRP procedures at the Ghent University Hospital between 2016 and 2022 was collected in a prospectively collected ERCP register, consisting of a variety of data, although the extent of prospectively collected data varied each year. Retrospective analysis of electronic patient files (EPD UZ Gent) and/or the prospective registry was performed to gather patient characteristics, procedural details, technical procedures, diagnosis of complications, and their severity. Procedures that deviated from a classic ERCP procedure such as direct EUS (endoscopic ultrasound) -guided pancreaticobiliary drainage (eg. choledochoduodenostomy or hepaticogastrostomy) or any other interventional EUS procedures and ampullectomy were excluded from the analysis. However, EUS-guided rendezvous procedures were included. When multiple complications arose from a single procedure, only the most severe one was included in the analysis. For patients that underwent multiple ERCPs between 2016 and 2022, only the first recorded procedure with complication for each patient was withheld for analysis. The study was approved by the local ethical committee (June 6, 2023).

2.2. ERCP procedure

Prior to the procedure, each patient provided written informed consent. The ERCPs were performed by a team of 6 endoscopists with varying experience levels (ranging from 1–30 years) and 3 trainees (with less than 1 year of experience) under the guidance of one of the 6 more experienced endoscopists. All patients were given detailed instructions on their antithrombotic therapy regimen, which included specific timelines for interrupting various medications before their procedure, based on consensus guidelines (20). New oral anticoagulants (NOACs) were to be interrupted 48 hours prior, Vitamin K antagonists (VKAs) 5 days prior, Low-Molecular-Weight Heparin (LMWHs) 24 hours prior, and P2Y12 inhibitors 5 days prior. Acetylsalicylic acid was not interrupted. Additionally, all patients received preprocedural prophylactic therapy as per standing order, which included IV antibiotics to prevent infectious complications and rectal NSAIDs aimed at mitigating the risk of ERCP pancreatitis, unless strictly contraindicated.

All procedures were performed under general anaesthesia with intravenous propofol. If needed, the endoscopist requested hyoscine-n-butyl (Buscopan) as a smooth muscle cell relaxant. Automatic devices monitored oxygen saturation, heart rate, and blood pressure throughout the procedure. Standard of care in the department is first to attempt a guidewire-assisted cannulation of the CBD (common bile duct) or PD (pancreatic duct). In case of unintentional PD cannulation, the guidewire was left in place and a second wire was used to cannulate the CBD (double-wire method). In case of failure, a pre-cut technique was attempted. In case of failure despite precut with an urgent ERCP indication, an EUS-guided rendezvous procedure was performed during the same procedure. In less urgent cases, a reattempt was performed after an interval of at least 2 days, with an EUS-guided rendezvous procedure in case of a second failure. The endoscopist wrote the procedural report, always detailing the pre-procedural indication, the difficulty of cannulation, the findings on cholangiography/pancreatography, the materials used, the interventions performed, the technical success of the procedure, and the immediate complications (such as perforation or bleeding) with their management.

The morning after the procedure, each patient underwent a standard laboratory check which included complete blood count, renal function, liver set, bilirubin, lipases and CRP.

2.3. Definitions

We utilized the definitions of complications outlined in the 2020 ESGE (European Society of Gastrointestinal Endoscopy) guideline on ERCP-related adverse events (6). Sepsis was defined as life-threatening organ dysfunction caused by a dysregulated host response to infection (21). Recently, Nass et al. introduced the AGREE (Adverse events in gastrointestinal endoscopy) classification, a standardized method for assessing adverse events that result from endoscopic procedures (22).

All complications were classified based on this classification system, with those falling into AGREE 0–2 being deemed 'non-severe', and those in AGREE 3–5 being classified as 'severe'. The details of this classification system can be found in Supplemental Table 1.

2.4. Follow-up

In the absence of complications, the patient was discharged upon approval of the endoscopist. The hospitalization report included the ERCP procedural report and provided information on the post-procedural course of the patient. Upon discharge, the patients were advised to seek medical help in case new symptoms (such as abdominal pain or fever) would occur. While there was no active standardized follow-up for all patients, the majority were scheduled for consultation based on their individual needs and the success of the procedure. Any late complications following discharge were only detected if the patient was readmitted to our hospital.

2.5. Study aim

The primary study aim was to identify factors, both patient- and procedure-related, associated with severity in patients experiencing post-ERCP complications.

2.6. Statistical analysis

To determine potential factors related to the severity of post-ERCP complications, we utilized the well-known risk factors outlined in the recent ESGE guideline (6), which were based on multiple meta-analyses and prospective studies. Additionally, we included risk factors for 'severe' post-ERCP complications identified in three previous studies: two retrospective studies by Cotten et al. (17) and Kwak et al. (23) and one prospective study by Glomsaker et al. (24) (see Supplemental Table 2 for study details). These factors included general health status (according to the ASA classification 1–5 (25)) and the degree of difficulty/complexity of the procedure (according to the Modified Shutz criteria 1–4 (26)). To expand upon this set of risk factors, we also analysed less frequently studied factors such as time to cannulation, immunosuppressive therapy, years of experience of the ERCP endoscopist, and type of intervention performed (biliary stone extraction, biliary stricture treatment, pancreatic stricture treatment, pancreatic stone treatment, EUS-assisted biliary drainage, EUS-assisted pancreatic drainage, biliary leak, pancreatic leak and other). In total, we analysed 30 potential risk factors, 17 of which were patient-related and 13 were procedure-related (see Table 5 for all included variables). Chi-squared and Fisher's exact tests were used to analyse categorical variables, while continuous variables were analysed using the Mann-Whitney U test and expressed as medians with IQR (interquartile range). Variables with a P-value < 0.20 in the univariate analysis, along with identified risk factors related to 'severe' complications from three previously mentioned studies (17, 23–24), were included in a stepwise multivariate logistical regression analysis to identify independent risk factors of severe post-ERCP complications. We evaluated the goodness-of-fit using the Hosmer-Lemeshow test. The calculated risk of severe versus non-severe complications was expressed as an OR (odds ratio) with a 95% CI (confidence interval), and statistical significance was determined as a P-value < 0.05. The Statistical Package for Social Sciences (SPSS, version 29.0, Chicago, IL, USA) was used for the statistical analysis.

3.1. Dataset

Between January 2016 and December 2022, a total of 2810 ERCP procedures were performed at UZ Ghent, of which 260 (9,2%) were complicated by a post-ERCP complication. In case a patient had received multiple ERCP procedures, only the first procedure that led to a complication was included to avoid selection bias. A total of 223 (7,9%) procedures were ultimately included for analysis. See Fig. 1.

< position Fig. 1 >

3.2. Patient and procedural characteristics

Hundred and thirty-six (60,9%) patients had major co-morbidities and 97 (43,4%) procedures were defined as complex (Shutz 3 or 4). The procedures were carried out by a total of 9 endoscopists, including 3 trainees, and a vast majority of them (91.5%, or 204 procedures) were performed by the three most practising endoscopists. All patient and procedural characteristics are displayed in Table 1.

Table 1

Patient and procedural characteristics
Patient demografics and characteristics		N	%
Sex	Man Woman	114 109	51,1 48,9
Age	< 35y 35-64y 65-79y > 80y	16 83 82 42	7,1 37,2 36,7 18,8
ASA	1 2 3 4	11 76 122 14	49,3 34,0 54,7 6,2
Comorbidities	Chronic pancreatitis ESKD Thrombocytopenia Cirrhosis PSC	24 2 13 6 6	10,8 0,9 5,8 2,7 2,7
Immunosuppressive therapy	yes	40	17,9
Antithrombotic therapy	None Anticoagulant Antiplatelet	138 35 50	61,8 15,7 22,4
Previous ERCP	yes	96	43,0
Previous PEP	yes	10	4,5
Procedural characteristics		N	%
Difficult cannulation	yes	104	46,1
Time to cannulation (min)	< 5 min 5–10 min > 10 min	88 34 69	46,1 17,8 36,1
Precut sphincterotomy	CBD Duct of Wirsung	32 7	14,3 3,1
Incomplete billiary drainage	yes	31	15,9
Incomplete stone extraction	yes	12	12,7
Bilirubin < 1 mg/dL	yes	97	44,3
Non-dilated CBD	yes	62	29,9
Degree of difficulty (Shutz)	1 2 3 4	5 121 55 42	2,2 54,2 24,7 18,8
Endoscopist’s experience	< 1y 1-5y > 5y	39 60 124	17,4 26,9 55,6

< position Table 1 >

3.3. Indications and interventions performed

Table 2 provides an overview of the ERCP indications. The primary indication was related to biliary obstruction (71.8%).

Table 2

Indications and type of interventions performed
Indication	number(n)	n/N	%
Bile duct stones	84	84/223	37,7%
(History of) pancreatitis	24	24/223	10,8%
Biliary cancer	25	25/223	11,2%
Bile duct leak	3	3/223	1,3%
Benign biliary stricture	51	51/223	22,9%
Pancreatic cancer	17	17/223	7,6%
Sphincter of Oddi Dysfunction	1	1/223	0,4%
Ampullary stenosis	8	8/223	3,6%
Other	10	10/223	4,5%
Type intervention	number(n)	n/N	%
Biliary stone extraction	70	70/223	31,4%
Biliary stricture treatment	101	101/223	45,3%
Pancreatic stricture treatment	15	15/223	6,7%
Pancreatic stone treatment	1	1/223	0,4%
EUS assisted biliary drainage	25	25/223	11,2%
EUS assisted pancreatic drainage	4	4/223	1,8%
Biliary leak	3	3/223	1,3%
Pancreatic leak	1	1/223	0,4%
Other	3	3/223	1,3%
Total (N)	223		100,0%

< position Table 2 >

3.4. Complications

Table 3 provides an overview of the post-ERCP complications and their severity. A total of 241 cases with complication occurred of which 223 cases were withheld for analysis. The most frequently reported complication was pancreatitis, followed by haemorrhage, infectious complication, perforation, and cardiopulmonary failure. Complications were classified as ‘severe’ in approximately 1 out of 5 complicated cases (49/241, 20.3%). Among all the complications, perforation had the highest likelihood of resulting in a severe complication (10 out of 15 perforations were severe, which accounts for 66,7%). When considering the absolute number of severe complications reported, haemorrhage tops the list (22 out of 49 severe complications, accounting for 44.9% of severe complications). In 2.7% of cases analysed, the complication led to death (6/223) resulting in an overall mortality rate of 0.21% (6/2810) for the whole population. From the information available, it was determined that 5 patients died from ERCP-related causes, with 4 being classified as probable and 1 as possible, while 1 patient died from circulatory arrest after induction anaesthesia at the start of the ERCP.

Table 3

Incidence of all post-ERCP complications and ‘severe’ post-ERCP complications per type of complication
Complication type	n	n/N	% n/N	A3 ⁽⁴⁾	A4	A5	n_severe ⁽³⁾ (3–5)	% n_severe /N	n_severe /N⁽¹⁾	n_severe /n⁽²⁾	% n_severe/n
Pancreatitis	82	82/223	36,8%	3	1	0	4	1,8%	4/223	4/82	5%
Infection (cholecystitis/ cholangitis/sepsis)	70	70/223	31,4%	5	5	2	12	5,4%	12/223	12/55	22%
Haemorrhage	55	55/223	24,7%	17	3	2	22	9,9%	22/223	22/70	31%
Perforation	15	15/223	6,7%	6	3	1	10	4,5%	10/223	10/15	67%
Cardiopulmonary failure	1	1/223	0,4%	0	0	1	1	0,4%	1/223	1/1	100%
Total (N)	223		100.0%	31	12	6	49	22,0%	49/223	49/223	22%
⁽¹⁾ N = total number of complicated procedures analyzed (= 223)
⁽²⁾ n = number of cases per type of complication
⁽³⁾ n_severe = number of ‘severe’ cases per type of complication (AGREE 3, 4 of 5)
⁽⁴⁾ A = AGREE classification

< position Table 3 >

3.5. Predictors of severe post-ERCP complications

3.5.1. Univariate analysis

Univariate analysis identified four risk factors that increase the likelihood of experiencing a severe post-ERCP complication. Two of these risk factors are patient-related, including anticoagulative therapy (P = 0.055) and an age range of 65–79 years (P = 0.045). The other two risk factors are related to the procedure itself, including degree of difficulty (P = 0.002) and incomplete biliary drainage (P = 0.041). For a more detailed look into the results of the univariate analysis, please refer to Table 4a + b and Table 5

Table 4

a: Univariate analysis of patient-related risk factors for severe complications. Categorical variables.
Variable	category	n_severe ⁽²⁾/N⁽¹⁾	%	P value⁽³⁾
Sex				0,507
	male	23/114	20,2%
	female	26/109	23,9%
Age				0,113
	< 35 y	5/16	31,3%	0,354
	35–64 y	13/83	15,7%	0,080
	65–79 y	24/82	29,3%	0,045
	> 80 y	7/42	16,7%	0,357
ASA classification				0,497
	1	1/11	9,1%	0,463
	2	14/76	18,4%	0,357
	3	30/122	24,6%	0,300
	4/5	4/14	28,6%	0,514
Previous pancreatitis				0,136
	yes	6/44	13,6%
	no	43/179	24,0%
Previous post-ERCP pancreatitis (PEP)				0,695
	yes	1/10	10,0%
	no	48/213	22,5%
No chronic pancreatitis				0,886
	yes	5/24	20,8%
	no	44/199	22,1%
Previous ERCP				0,721
	yes	20/96	20,8%
	no	29/127	22,8%
ESKD				0,000
	yes	0/2	0,0%
	no	49/221	22,2%
Thrombocytopenia < 50 000/mm3				1,000
	yes	3/13	23,1%
	no	46/210	21,9%
Liver cirrhosis				0,343
	yes	0/6	0,0%
	no	49/217	22,6%
PSC				0,122
	yes	3/6	50,0%
	no	46/217	21,2%
Immunosuppressive therapy				0,176
	yes	12/40	30,0%
	no	37/183	20,2%
Suspected SOD				1,000
	yes	0/2	0,0%
	no	49/221	22,2%
Normal bilirubin < 1mg/dL				0,227
	yes	18/97	18,6%
	no	31/122	25,4%
Non-dilated CBD				0,105
	yes	9/62	14,5%
	no	40/152	26,3%
Hilar obstruction				0,136
	yes	10/31	32,3%
	no	39/192	20,3%
Antithrombotic therapy				0,159
	No	27/138	19,6%	0,269
	Anticoagulation	12/35	34,3%	0,055
	Antiplatelet	10/50	20,0%	0,703

Table 4

b: Univariate analysis of procedure-related risk factors for severe complications. Categorical variables.
Variable	category	n_severe ⁽²⁾/N⁽¹⁾	%	P value⁽³⁾
Degree of difficulty according to Shutz criteria	Category			0,002
	1	0/5	0,0%	0,589
	2	16/121	13,2%	< 0,001
	3	18/55	32,7%	0,026
	4	15/42	35,7%	0,017
Difficult cannulation				0,669
	yes	24/104	23,1%
	no	24/116	20,7%
Time to cannulation (in min)				0,909
	< 5min	18/88	20,5%	0,658
	5-10min	7/34	20,6%	0,831
	> 10min	16/69	23,2%	0,769
Pancreatic injection				0,279
	yes	9/54	16,7%
	no	40/169	23,7%
Biliary precut sphincterotomy				0,988
	yes	7/32	21,9%
	no	42/191	22,0%
Pancreatic sphincterotomy				0,352
	yes	0/7	0,0%
	no	49/216	22,7%
Incomplete stone extraction				0,131
	yes	5/12	41,7%
	no	16/82	19,5%
Incomplete biliary drainage				0,041
	yes	11/31	35,5%
	no	31/163	19,0%
Endoscopist’s experience (in years)				0,276
	< 1y	5/39	12,8%	0,129
	1-5y	13/60	21,7%	0,947
	> 5y	31/124	25,0%	0,222
Cholangioscopy				0,142
	yes	5/12	41,7%
	no	44/211	20,9%
Balloon sphincterotomy				1,000
	yes	3/15	20,0%
	no	46/208	22,1%
Intervention type				0,261
	Biliary stone extraction	16/70	22,9%	0,829
	Biliary stricture treatment	22/101	21,8%	0,950
	Pancreatic stricture treatment	1/15	6,7%	0,200
	Pancreatic stone treatment	0/1	0,0%	1,000
	EUS assisted biliary drainage	9/25	36,0%	0,072
	EUS assisted pancreatic drainage	0/4	0,0%	0,578
	Biliary leak	0/3	0,0%	1,000
	Pancreatic leak	1/1	100,0%	0,220
	Other	0/3	0,0%	1,000
Rendezvous procedure (EUS-assisted procedure)				0,260
	yes	9/29	31,0%
	no	40/194	20,6%
⁽¹⁾ N = number of analysed procedures
⁽²⁾n_severe = number of ‘severe’ complications
⁽³⁾ Statistical test used for categorical variables: Chi square test (or Fisher’s exact test where > 20% of cells in contingency table have expected frequencies < 5)

Table 5: Univariate analysis of risk factors for severe post-ERCP complication. Continuous variables.

Variable	Severe complication	Non-severe complication	P value⁽²⁾
	Median (IQR⁽¹⁾)	Median (IQR⁽¹⁾)
Age	69,0 (56,0 -76,5)	68,0 (54,0 – 76,3)	0,865
Time to cannulation in min⁽³⁾	8,0 (2,0 – 18,5)	6,5 (3,0 -13,0)	0,983

⁽¹⁾ IQR = interquartile range

⁽²⁾ Statistical test used for continuous variables = Mann-Whitney U test

⁽³⁾ Analysis based on 41/49 severe cases and 150/174 non-severe cases (due to missing data)

< position Table 4a + b and Table 5 >

3.5.2. Multivariate analysis

Variables with a P-value < 0.20 in the univariate analysis and risk factors of ‘severe’ post-ERCP complication as identified in 3 previous studies (17, 23–24) were included in a stepwise multivariate logistical regression analysis. This multivariate analysis identified two variables as independent risk factors of a severe post-ERCP complication: one patient-related risk factor, i.e. anticoagulative therapy (OR 6.3, 95% CI 1.4–28.3, P = 0.016), and one procedure-related risk factor, specifically a high degree of difficulty according to Shutz's categories 3 (OR 11.5, 95% CI 2.4–54.6, P = 0.002) and 4 (OR 5.9, 95% CI 1.4–23.5, P = 0.012).

< position Table 6 >

Table 6

Multivariate logistical regression analysis of patient- and procedure related risk factors for severe complications
Variable ⁽¹⁾	category	OR (95% Cl)	P value
Antithrombotic therapy			0,051
	No	1,00 (Reference)
	Anticoagulation	6,3 (1,4;28,3)	0,016
	Antiplatelet		0,260
Age			0,995
	< 35y		0,916
	35-64y		0,939
	65-79y		0,833
	> 80 y	1,00 (Reference)
Degree of difficulty according to Shutz criteria			0,004
	1
	2	1.00 (Reference)
	3	11,5 (2,4;54,6)	0,002
	4	5,9 (1,4;23,5)	0,012
Incomplete biliary drainage	Yes/no		0,291
(1) Selected variables for multivariate analysis = variables with a P value < 0.20 in the univariate analysis and identified risk factors related to severe complication from 3 previous studies (17, 23–24)

In this large single-centre cohort study, we identified anticoagulative therapy and high degree of difficulty of the procedure as independent risk factors associated with severity in patients with post-ERCP complication.

According to our study, the total complication rate was 7.9%, which is comparable to the rate (6.8%) cited in a comprehensive systematic review conducted by Andriulli et al. (5). Moreover, our study found that 20.3% of post-ERCP complications were severe, which is in line with the results of Andriulli et al.'s review (24%), though the incidence of severe post-ERCP pancreatitis cases was different (11% compared to 5% in our study). This is likely because we did not consider prolonged hospitalisation as a severe adverse event, whereas most of the studies incorporated in the aforementioned review adopted the definition of Cotten et al.(17) which defines a prolonged stay in hospital exceeding 10 days as 'severe’. Many of our patients undergoing ERCP require prolonged hospital stays for various reasons, often unrelated to the severity of pancreatitis, including age, comorbidities, and performance status.

The percentage of ERCP-related deaths was very low at 0.21% (6 out of 2180), which is in line with the expectations (5). Of the deceased patients, half (3 out of 6) had a malignancy affecting their pancreatic or biliary tract, which is consistent with the findings of Glomsaker T et al. (24).

Although extensive research has been conducted to identify the risk factors associated with post-ERCP complications (11–19), only limited information is available regarding risk factors of severe ERCP-related adverse events. We identified only three studies that previously investigated these risk factors (17, 23–24, Table 2). Interestingly, a poor general medical condition expressed as ASA classification 3–5 was the only common risk factor identified in these studies. Although a higher percentage of severe complications was observed in patients with ASA classification of 3 or higher, it did not prove to be statistically significant in either the univariate or multivariate analysis conducted in our study. Our analysis found that procedures with a higher level of technical difficulty (Shutz 3 + 4) have a significantly increased risk of severe post-ERCP complications (OR 11.5 and OR 5.9, respectively). This finding was also supported by Cotton et al.'s study (OR 2.86)(17). This emphasises the need for careful consideration before performing complex procedures, as well as the importance of providing patients with clear information about the increased risk of severe complications. As a tertiary referral centre, we frequently receive cases for EUS-guided rendezvous procedures when peripheral centres are unable to cannulate via papilla. Although we did observe a relatively higher incidence of severe complications for these procedures (9/29, or 31% versus the overall 20,3%), this difference was not statistically significant (P = 0.260). The second independent predictor of severe complications in our multivariate analysis was anticoagulative therapy (NOAC/VKA/LMWH). The majority of severe complications in patients under anticoagulation were indeed severe bleedings (7 severe bleedings out of 12 severe complications in patients under anticoagulation). Interestingly, antiplatelet therapy (Acetylsalicylic acid/P2Y12 inhibitors) did not emerge as a risk factor in either univariate or multivariate analysis. Previous research did not explore the impact of anticoagulative therapy on severe post-ERCP complications, but a 2018 study conducted by So Nakaji et al. (27) did recognize anticoagulative therapy as a risk factor in multivariate analysis for all complications. Similarly, antiplatelet therapy was not found to be a significant risk factor. This implies that even with careful adherence to guidelines for halting anticoagulative therapy before the procedure, patients on this therapy still face a higher risk of a severe post-ERCP complication, especially severe bleeding.

Incomplete biliary drainage has been linked to a higher risk of infectious complications (28, 29). Our study found that patients with incomplete drainage were more likely to experience severe complications, but this relationship did not hold up in multivariate analysis.

It is reasonable to assume that endoscopists with more experience would have a lower risk of (severe) complications when performing ERCPs, given the procedure's steep learning curve. While some studies (30–32) have confirmed this, others (33) have not. Our analysis of risk factors of severe complications did not find a correlation with experience, which is consistent with Cotton et al.'s study which found no link between training and severe complications (17). Notably, our study found a low incidence of severe complications (5/39, 12.8%) among endoscopists with less than one year of experience, although this was not statistically different from other groups. A possible explanation is that less experienced endoscopists tend to perform less complex procedures. Also, in our centre an experienced staff member takes over if the trainee fails to cannulate the bile duct in 7 minutes.

Our study had several limitations. Although we could utilise our prospective registry of ERCP data, the quantity of prospectively gathered information varied annually and the dataset had to be completed by retrospective analysis of the patient records. As such, certain patient- or procedure-related variables could sometimes not be retrieved from the ERCP and hospitalisation reports, leading to missing data and hence, a less robust analysis.

Secondly, our study was conducted at a tertiary referral centre, with selection bias towards a greater proportion of challenging and high-risk cases. For example, we found that 60.9% of patients had significant comorbidities, classified as ASA 3 or higher, which is twice the prevalence found in three comparable, large-scale European studies (24, 34–35).

Thirdly, we only included our 223 patients with post-ERCP complications in our analysis, since we were mainly interested in predictors of a severe course in this subgroup of patients rather than in all patients undergoing ERCP. We compared 174 cases with a non-severe post-ERCP complication with 49 patients who had a severe post-ERCP complication. Due to this limited sample size, we refrained from conducting sub-analyses of particular complications such as severe pancreatitis. Despite the presence of various types of complications, some of which may be influenced by a specific risk factor while others are not, the logistic regression model still demonstrated a satisfactory level of predictability.

Finally, our analysis did not include any unreported or untreated complications that may have occurred outside of our hospital, as we did not conduct a systematic follow-up 30 days after ERCP. This could have resulted in an underestimation of the total number of complications and an overestimation of the proportion of severe complications.

In conclusion, our study shows that the overall likelihood of a severe complication after ERCP is relatively low, even in a tertiary centre with more complex procedures often performed in patients with significant comorbidities. The procedure's level of difficulty and anticoagulative therapy were identified as independent risk factors of a severe course of post-ERCP complications. Our findings should be validated in large prospective registries that include both academic and non-academic centres.

Competing Interests

Prof. Dr. Hindryckx Pieter reports speaker and consultancy fees from Boston Scientific, consultancy fees from Taewoong Medical and speaker fees from DuomedDr. Goubert Kristel, Dr. Degroote Helena, Prof. Dr. De Vos Martine, Mr. Khalenkow Maxim have no conflicts of interest to disclose.

Funding information:

none

Conflict of interest:

Dr. Goubert Kristel, Dr. Degroote Helena, Prof. Dr. De Vos Martine, Mr. Khalenkow Maxim have no conflicts of interest to disclose. Prof. Dr. Hindryckx Pieter reports speaker and consultancy fees from Boston Scientific, consultancy fees from Taewoong Medical and speaker fees from Duomed

Author Contribution

K.G. and P.H. wrote the main manuscript text. K.G. prepared tables 1-6 en figure 1. K.G. en M.K. collected the data. K.G. and P.H. did the analysis and interpretation of the data. H.G. and M.D. did critical revision of the article. All authors reviewed the manuscript.

Chandrasekhara V, Chathadi K V., Acosta RD, Decker GA, Early DS, Eloubeidi MA, et al. The role of endoscopy in benign pancreatic disease. Gastrointest Endosc [Internet]. 2015 Aug 1 [cited 2023 Dec 3];82(2):203–14. Available from: https://pubmed.ncbi.nlm.nih.gov/26077456/
Bilbao MK, Dotter CT, Lee TG, et al. Complications of endoscopic retrograde cholangiopancreatography (ERCP). A study of 10,000 cases. Gastroenterology. 1976;70:314–20.
Loperfido S AGBG et al. Major early complications from diagnostic and therapeutic ERCP: a prospective multicenter study. Gastrointest Endosc. 1998;48:1–10.
Masci E, Toti G, Mariani A, Curioni S, Lomazzi A, Dinelli M, et al. Complications of diagnostic and therapeutic ERCP: a prospective multicenter study. Am J Gastroenterol [Internet]. 2001 Feb [cited 2023 Nov 19];96(2):417–23. Available from: https://pubmed.ncbi.nlm.nih.gov/11232684/
Andriulli A, Loperfido S, Napolitano G, Niro G, Valvano MR, Spirito F, et al. Incidence rates of post-ERCP complications: a systematic survey of prospective studies. Am J Gastroenterol [Internet]. 2007 Aug [cited 2023 Nov 19];102(8):1781–8. Available from: https://pubmed.ncbi.nlm.nih.gov/17509029/
Dumonceau JM, Kapral C, Aabakken L, Papanikolaou IS, Tringali A, Vanbiervliet G, et al. ERCP-related adverse events: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy [Internet]. 2020 [cited 2023 Nov 19];52(2):127–49. Available from: https://pubmed.ncbi.nlm.nih.gov/31863440/
Christensen M, Matzen P, Schulze S, Rosenberg J. Complications of ERCP: a prospective study. Gastrointest Endosc [Internet]. 2004 Nov [cited 2023 Nov 19];60(5):721–31. Available from: https://pubmed.ncbi.nlm.nih.gov/15557948/
Kochar B, Akshintala VS, Afghani E, Joseph Elmunzer B, Kim KJ, Lennon AM, et al. Incidence, severity, and mortality of post-ERCP pancreatitis: a systematic review by using randomized, controlled trials. Gastrointest Endosc [Internet]. 2015 Jan 1 [cited 2023 Nov 19];81(1):143-149.e9. Available from: https://pubmed.ncbi.nlm.nih.gov/25088919/
Cirocchi R, Denis Kelly M, Griffiths EA, Tabola R, Sartelli M, Carlini L, et al. A systematic review of the management and outcome of ERCP related duodenal perforations using a standardized classification system *. 2017 [cited 2023 Nov 19]; Available from: http://dx.doi.org/10.1016/j.surge.2017.05.004
Cohen S, Bacon BR, Berlin JA, Fleischer D, Hecht GA, Loehrer PJ, et al. National Institutes of Health State-of-the-Science Conference Statement: ERCP for diagnosis and therapy, January 14-16, 2002. Gastrointest Endosc [Internet]. 2002 Dec [cited 2023 Nov 26];56(6):803–9. Available from: https://pubmed.ncbi.nlm.nih.gov/12447289/
Freeman ML, DiSario JA, Nelson DB, Fennerty MB, Lee JG, Bjorkman DJ, et al. Risk factors for post-ERCP pancreatitis: a prospective, multicenter study. Gastrointest Endosc [Internet]. 2001 [cited 2023 Nov 19];54(4):425–34. Available from: https://pubmed.ncbi.nlm.nih.gov/11577302/
Vandervoort J, Soetikno RM, Tham TCK, Wong RCK, Ferrari AP, Montes H, et al. Risk factors for complications after performance of ERCP. Gastrointest Endosc [Internet]. 2002 Nov [cited 2023 Nov 19];56(5):652–6. Available from: https://pubmed.ncbi.nlm.nih.gov/12397271/
Mehta SN, Pavone E, Barkun JS, Bouchard S, Barkun AN. Predictors of post-ERCP complications in patients with suspected choledocholithiasis. Endoscopy [Internet]. 1998 [cited 2023 Nov 19];30(5):457–63. Available from: https://pubmed.ncbi.nlm.nih.gov/9693893/
Williams EJ, Taylor S, Fairclough P, Hamlyn A, Logan RF, Martin D, et al. Risk factors for complication following ERCP; results of a large-scale, prospective multicenter study. Endoscopy [Internet]. 2007 Sep [cited 2023 Nov 19];39(9):793–801. Available from: https://pubmed.ncbi.nlm.nih.gov/17703388/
Cheng CL, Sherman S, Watkins JL, Barnett J, Freeman M, Geenen J, et al. Risk factors for post-ERCP pancreatitis: a prospective multicenter study. Am J Gastroenterol [Internet]. 2006 Jan [cited 2023 Nov 19];101(1):139–47. Available from: https://pubmed.ncbi.nlm.nih.gov/16405547/
Nakeeb A El, Hanafy E El, Salah T, Atef E, Hamed H, Sultan AM, et al. Post-endoscopic retrograde cholangiopancreatography pancreatitis: Risk factors and predictors of severity. World J Gastrointest Endosc [Internet]. 2016 Nov 11 [cited 2023 Nov 19];8(19):709. Available from: /pmc/articles/PMC5114460/
Cotton PB, Garrow DA, Gallagher J, Romagnuolo J. Risk factors for complications after ERCP: a multivariate analysis of 11,497 procedures over 12 years. Gastrointest Endosc [Internet]. 2009 Jul [cited 2023 Nov 19];70(1):80–8. Available from: https://pubmed.ncbi.nlm.nih.gov/19286178/
Chen JJ, Wang XM, Liu XQ, Li W, Dong M, Suo ZW, et al. Risk factors for post-ERCP pancreatitis: A systematic review of clinical trials with a large sample size in the past 10 years. Eur J Med Res [Internet]. 2014 May 15 [cited 2023 Nov 19];19(1):1–7. Available from: https://eurjmedres.biomedcentral.com/articles/10.1186/2047-783X-19-26
Ding X, Zhang FC, Wang YJ. Risk factors for post-ERCP pancreatitis: A systematic review and meta-analysis. Surgeon [Internet]. 2015 [cited 2023 Nov 19];13(4):218–29. Available from: https://pubmed.ncbi.nlm.nih.gov/25547802/
Veitch AM, Radaelli F, Alikhan R, Dumonceau JM, Eaton D, Jerrome J, Lester W, Nylander D, Thoufeeq M, Vanbiervliet G, Wilkinson JR, Van Hooft JE. Endoscopy in patients on antiplatelet or anticoagulant therapy: British Society of Gastroenterology (BSG) and European Society of Gastrointestinal Endoscopy (ESGE) guideline update. Gut. 2021 Sep;70(9):1611-1628. doi: 10.1136/gutjnl-2021-325184.
Singer M, Deutschman CS, Seymour C, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA [Internet]. 2016 Feb 2 [cited 2023 Nov 24];315(8):801. Available from: /pmc/articles/PMC4968574/
Nass KJ, Zwager LW, van der Vlugt M, Dekker E, Bossuyt PMM, Ravindran S, et al. Novel classification for adverse events in GI endoscopy: the AGREE classification. Gastrointest Endosc [Internet]. 2022 Jun 1 [cited 2023 Nov 19];95(6):1078-1085.e8. Available from: https://pubmed.ncbi.nlm.nih.gov/34890695/
Kwak N, Yeoun D, Arroyo-Mercado F, Mubarak G, Cheung D, Vignesh S. Outcomes and risk factors for ERCP-related complications in a predominantly black urban population. BMJ Open Gastroenterol [Internet]. 2020 Sep 17 [cited 2023 Nov 24];7(1). Available from: https://pubmed.ncbi.nlm.nih.gov/32943462/
Glomsaker T, Hoff G, Kvaløy JT, Søreide K, Aabakken L, Søreide JA. Patterns and predictive factors of complications after endoscopic retrograde cholangiopancreatography. Br J Surg [Internet]. 2013 Feb [cited 2023 Nov 24];100(3):373–80. Available from: https://pubmed.ncbi.nlm.nih.gov/23225493/
Doyle DJ, Hendrix JM, Garmon EH. American Society of Anesthesiologists Classification. StatPearls [Internet]. 2023 Aug 17 [cited 2023 Nov 26]; Available from: https://www.ncbi.nlm.nih.gov/books/NBK441940/
Schutz SM, Boise Gastroenterology Associates Boise F. Grading the Degree of Difficulty of ERCP Procedures. Gastroenterol Hepatol (N Y) [Internet]. 2011 Oct [cited 2023 Nov 24];7(10):674. Available from: /pmc/articles/PMC3265009/
Nakaji S, Hirata N, Matsui H, Shiratori T, Kobayashi M, Yoshimura S, et al. Hemodialysis is a strong risk factor for post-endoscopic sphincterotomy bleeding in patients with choledocholithiasis. Endosc Int Open [Internet]. 2018 May [cited 2023 Dec 2];6(5):E568–74. Available from: https://pubmed.ncbi.nlm.nih.gov/29756014/
Cotton PB, Connor P, Rawls E, Romagnuolo J. Infection after ERCP, and antibiotic prophylaxis: a sequential quality-improvement approach over 11 years. Gastrointest Endosc [Internet]. 2008 Mar [cited 2023 Dec 2];67(3):471–5. Available from: https://pubmed.ncbi.nlm.nih.gov/18061594/
Motte S, Deviere J, Dumonceau JM, Serruys E, Thys JP, Cremer M. Risk factors for septicemia following endoscopic biliary stenting. Gastroenterology [Internet]. 1991 [cited 2023 Dec 2];101(5):1374–81. Available from: https://pubmed.ncbi.nlm.nih.gov/1936809/
Endoscopic retrograde cholangiopancreatography (ERCP) outcomes can improve after further training for an individual already experienced in ERCP - Minerva Gastroenterologica e Dietologica 2015 September;61(3):117-20 - Minerva Medica - Journals [Internet]. [cited 2023 Dec 2]. Available from: https://www.minervamedica.it/en/journals/gastroenterology/article.php?cod=R08Y2015N03A0117
Petersen BT. ERCP outcomes: defining the operators, experience, and environments. Gastrointest Endosc [Internet]. 2002 [cited 2023 Dec 2];55(7):953–8. Available from: https://pubmed.ncbi.nlm.nih.gov/12024166/
Jowell PS, Baillie J, Branch MS, Affronti J, Browning CL, Bute BP. Quantitative assessment of procedural competence: A prospective study of training in endoscopic retrograde cholangiopancreatography. Ann Intern Med. 1996;125(12):983–9.
Freeman ML, DiSario JA, Nelson DB, Fennerty MB, Lee JG, Bjorkman DJ, et al. Risk factors for post-ercp pancreatitis: A prospective, multicenter study. Gastrointest Endosc [Internet]. 2001 Oct 1 [cited 2023 Dec 2];54(4):425–34. Available from: http://www.giejournal.org/article/S0016510701702003/fulltext
Williams EJ, Taylor S, Fairclough P, Hamlyn A, Logan RF, Martin D, et al. Risk factors for complication following ERCP; results of a large-scale, prospective multicenter study. Endoscopy [Internet]. 2007 Sep [cited 2023 Nov 26];39(9):793–801. Available from: https://pubmed.ncbi.nlm.nih.gov/17703388/
Enochsson L, Swahn F, Arnelo U, Nilsson M, Lhr M, Persson G. Nationwide, population-based data from 11,074 ERCP procedures from the Swedish Registry for Gallstone Surgery and ERCP. Gastrointest Endosc [Internet]. 2010 Dec [cited 2023 Nov 26];72(6). Available from: https://pubmed.ncbi.nlm.nih.gov/20970787/

Competing interest reported. Prof. Dr. Hindryckx Pieter reports speaker and consultancy fees from Boston Scientific, consultancy fees from Taewoong Medical and speaker fees from Duomed

Dr. Goubert Kristel, Dr. Degroote Helena, Prof. Dr. De Vos Martine, Mr. Khalenkow Maxim have no conflicts of interest to disclose.

Appendices.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Factors associated with the severity of ERCP-related complications: a large retrospective single-centre study

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. MATERIAL AND METHODS

2.1. Patients

2.2. ERCP procedure

2.3. Definitions

2.4. Follow-up

2.5. Study aim

2.6. Statistical analysis

3. RESULTS

3.1. Dataset

3.2. Patient and procedural characteristics

3.3. Indications and interventions performed

3.4. Complications

3.5. Predictors of severe post-ERCP complications

3.5.1. Univariate analysis

3.5.2. Multivariate analysis

4. DISCUSSION

Declarations

Funding information:

Author Contribution

References

Additional Declarations

Supplementary Files

Status:

Version 1