Introduction The 2004 NCEPOD report “Scoping our Practice”¹ had been highly critical of certain aspects of ERCP practise in UK, raising specific concerns about case selection and sedation practise. We analysed our own ERCP practise in a medium sized district hospital with a moderate case workload and a growing proportion of elderly population.

Methods Retrospective data was collected from 263 ERCPs performed between 2009–2011. Comprehensive information regarding demographics, indications, success and complication rates was recorded from ERCP reports and case notes and our practise was compared to NCEPOD recommendations.

Results 263 (n) ERCPs were included in this study. Median age was 72 (range = 16–98), 63% were females. 55% of patients were ASA grade 3–4. 84% of ERCPs were of grade 1 difficulty. All ERCP referrals were reviewed and authorised by a consultant gastroenterologist. Indications for ERCP were choledocholithiasis (63%), pancreatic or biliary malignancy with obstructive jaundice (18%), stent removal/replacement (10%), dilatation of biliary ducts with abnormal liver function tests (10%) and others (4%). > 90% of ERCPs were performed with a therapeutic intent and success was achieved in 86% of ERCPs at first attempt. Our successful cannulation rate was of 92%. Only 9.1% of cases were referred to tertiary centres for further management. Prophylactic oral ciprofloxacin was used in 60% of patients. Patients received a combination of midazolam and pethidine with a mean dose (±SD) of 3.2 mg (± 2.03) and 44.3 mg (±16.05) respectively. Reversal with flumazenil or naloxone was not required in any of the patients included in this study. Biliary sphincterotomy was performed in 60%(156), pre-cut sphincterotomy in 2.6%(7), stricture dilatation in 9.5%(25), biliary stenting 30.4%(80), balloon sphincteroplasty 3%(8), balloon trawl 67%(177) and mechanical lithotripsy 8.7%(23). 78.7% of malignant strictures were successfully stented (37). Overall complication rate was 5.7% - moderate haemorrhage requiring blood transfusion in 1.5%(4), post ERCP pancreatitis in 2%(6), sepsis 1.9%(5), duodenal perforation 0.7%(1) and respiratory arrest in 0.7%(1). 30 day mortality rate was 0.76%(2).

Conclusion In contrast to NCEPOD report, our audit demonstrates that ERCP practise is effective, safe and of high quality in a district general hospital setting. Complication and mortality rates are minimal and comparable to national standards, even in the elderly population. Post ERCP very low sepsis rate is most likely due to use of prophylactic antibiotics (Ciprofloxacin).

Disclosure of Interest None Declared.

Reference

1. NCEPOD. Scoping our practise; the 2004 report of the National Confidential Enquiry into patient Outcome and Death. London. NCEPOD 2004. Available from: URL: http://www. ncepod.org.uk/2004.htm