ERCP remains a high-risk endoscopic procedure, and many studies have demonstrated its efficacy and safety in the elderly. However, few have assessed these parameters in an inpatient setting, and some data has shown a higher mortality rate, thereby suggesting the physiological insult combined with the risks of the procedure deem this intervention highly dangerous in this cohort. We therefore aimed to review our experience as a single tertiary centre, postulating that ERCP remains a safe procedure in this group, with a complication rate within acceptable limits as well as facilitating discharge from hospital, especially in cases with a palliative indication.

The endoscopic database was searched from the dates 01/01/2022 to 24/11/2025 in a tertiary centre for all ERCP procedures. Records were filtered to exclude outpatient procedures and patients below the age of 80. Data was gathered regarding the age demographics, ASA grade, overall and post-procedure length of stay (LOS), indications and details of the procedures. A successful procedure was defined as having dealt with the acute indication to facilitate discharge from hospital. Any procedure-related complications were noted, alongside 30- and 90-day mortalities, and any re-admissions within 30 days.

186 procedures were included. The median age was 85 (IQR 82-88), and the eldest patient was 99. 87% of patients were ASA grade 3 or 4. The indications were related to malignancy in 21.5%, and non-malignant in 78.5%; predominantly choledocholithiasis associated with jaundice, cholangitis or pancreatitis. 30% of procedures were Schutz complexity level 3 or 4. The aim of the procedure was successfully achieved in 87.6% (including biliary decompression with stenting where stones could not be completely removed). Median total LOS was 10.5 days (IQR 6-19.5), and median post-procedure LOS was 4 days (IQR 1-9). 7 patients (3.8%) experienced complications directly related to the procedure (3 episodes of cholangitis, 2 of pancreatitis, 1 of bleeding and 1 related to stent migration). This would even be consistent with British Society of Gastroenterology (BSG) guidance of a complication rate of < 6% for Schutz level 1-2. All-cause mortality at 30 days was 6.5%, and at 90 days 12.4%, with published data reporting 2-5.3% in general populations. Readmission rate within 30 days for all indications was 16.1% (30 patients), though only 2 of these were due to complications of the procedure, and neither were fatal. 

The interaction of age and frailty with acute illness understandably provokes concern regarding the risks of ERCP. We identified no excess procedure-specific complication rate among our cohort. The 30- and 90-day mortality rates are acceptable given that 20% were performed for malignancy, and in fact suggest benefit in the context of palliation. The slight excess rate reported is likely to reflect a higher incidence of malignant pathology in the elderly. Reported literature has shown mortality rates exclusively in this group of up to 20%. Our results suggest that with careful peri-procedural decision-making, in-patient ERCP can be safely offered to this cohort.