Introduction
Endoscopic retrograde cholangiopancreatography (ERCP) is one of the most demanding and high-risk endoscopic procedures.
Failures and complications are more common than we realise, they lead to low satisfaction for the patient, high frustration for the endoscopist, higher risks of more instrumentalization in further re-attempts.
Keeping a meticulous updated data logging is fundamental to understand how we perform, what specific factors predict our failures, and identifying areas of improvement, mistakes or learning opportunities.
Aims & Methods
This prospective cohort study was conducted over a 3-year period (September 2020 to January 2024) at a tertiary care centre.
We defined all the adverse events as per ESGE guidelines1 The 30- mortality was analysed and stratified based on relation to ERCP.
We also identified patients’ factors (Age, sex, frailty, NEWS>4, anticoagulation, antiplatelets, palliative) and technique (First pass cannulation, double wire, precut sphincterotomy, sphincteroplasty). Descriptive statistics, chi-square or t-tests were used to evaluate associations between clinical characteristics and outcomes.
Results
A total of 438 patients underwent ERCP by the single operator during the study period. Procedural success was achieved in 389 cases (88.9%) and in 100% of cases of non-native ampulla. Success rates were lower in malignant cases (80.6%) compared to benign cases (90.05%) with an odds ratio (OR) of 2.21 (CI 1.05–4.63, p = 0.03).
Complications included: Post-ERCP pancreatitis: 12/438 (2.74%) Infections: 4/438 (0.91%) (3 cholangitis, 1 cholecystitis) Perforation: 3/438 (0.68%) and Bleeding: 2/438 (0.45%)
Mortality within 1 month was 17/438 (3.88%) mostly unrelated to the procedure in palliative and frail patients. Procedure-related mortality occurred in 4 cases (0.91%),due to pancreatitis (n=2),perforation (n=1) and sepsis secondary to cholecystitis (n=1).
Of all conscious sedation, only 204 /423 (48.2%) were deemed comfortable. 174/423 (41.1%) had mild discomfort, 32/423 (7.5%) had significant discomfort and 13/423 ( 3.07%) severe or very poor. Female patients demonstrated higher risk of poor tolerance (OR 2.09, CI 1.05–4.18 p = 0.04) but no higher procedural failure.
In difficult cannulation, double-wire succeeded in 19/36(52.8%). When combined with additional techniques after initial failure, success increased to 77.8%. Pre-cut sphincterotomy succeeded only in 13/25( 52%) lower than data published (>75%), thought to be due to learning curve and limited patient tolerance. However, pre-cut helped cannulation in most follow-up ERCPs .
Double wire associated risk of post-ERCP pancreatitis -but exclusively when prophylactic pancreatic duct stent was not placed-OR 3.96 (CI 1.02-15.3) p=0.04. None other AEs were significantly associated with any advanced cannulation techniques.
Conclusion
Overall failure rates and AEs were consistent with published literature. Statistically significant risks of failure included malignant aetiology (OR 2.21) and poor tolerance (OR 2.20). Female sex associated higher intolerance (OR 2.09). Both double-wire and pre-cut sphincterotomy showed failure rates of approximately 50%. Frailty was a significant predictor of 30-day mortality (OR 4.83).
| AUDIT RESULTS | STRATEGY IMPLEMENTED |
| Success rates lower in malignant cases (OR 2.21) | Consent form adapted in malignant cases (cannulation rate of 80% instead of 90%) |
| Poor tolerance (10.57%) → higher in females (OR 2.09) | Increasing the availability of deep sedation lists. Pre-identifying cases of possible low tolerance/anxiety |
| Post-ERCP Pancreatitis in double wire when prophylactic pancreatic duct stent was not placed-(OR 3.96 ) | Any unintentional cannulation pancreatic duct ≥2 times, early PD stent placement |
| Needle knife success only in 52% | Enhancing mentorship and training in needle-knife |
| Frailty significant predictor of 30-day mortality (OR 4.83) | Involving palliative care in early discussion with patient. Reevaluate ERCP indication in jaundice without pruritus. |
References
1. Dumonceau JM, Kapral C, Aabakken L, Papanikolaou IS, Tringali A, Vanbiervliet G, Beyna T, Dinis-Ribeiro M, Hritz I, Mariani A, Paspatis G, Radaelli F, Lakhtakia S, Veitch AM, van Hooft JE. ERCP-related adverse events: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2020 Feb;52(2):127-149.
2. Banks PA, BollenTL, Dervenis C et al. Classification of acute pancreatitis– 2012: revision of the Atlanta classification and definitions by international consensus. Gut 2013; 62: 102–111
3. Enochsson L., Swahn F., Arnelo U., Nilsson M., Löhr M., Persson G. Nationwide, population-based data from 11,074 ERCP procedures from the Swedish Registry for Gallstone Surgery and ERCP. Gastrointest. Endosc. 2010;72:1175–1184.
4. Spadaccini M, Binda C, Fugazza A, Repici A, Tarantino I, Fabbri C, Cugia L, Anderloni A, On Behalf Of The Interventional Endoscopy Amp Ultra Sound I-Eus Group. Informed Consent for Endoscopic Biliary Drainage: Time for a New Paradigm. Medicina (Kaunas). 2022 Feb 22;58(3):331.
5.Chen J, Wan JH, Wu DY, Shu WQ, Xia L, Lu NH. Assessing Quality of Precut Sphincterotomy in Patients With Difficult Biliary Access: An Updated Meta-analysis of Randomized Controlled Trials. J Clin Gastroenterol. 2018 Aug;52(7):573-578
