The impact of periampullary diverticula on cannulation and adverse events in endoscopic retrograde cholangiopancreatography

Abstract

Background: Periampullary diverticulum (PAD) is commonly encountered in endoscopic retrograde cholangiopancreatography (ERCP) procedures.

Objectives: We sought to determine whether PADs are associated with a lower success rate of cannulation and an increased risk of adverse events.

Design: A retrospective cohort study was conducted using prospectively gathered nationwide registry data.

Methods: Using the Swedish registry for gallstone surgery and ERCP, we analyzed a cohort of 66,974 prospectively registered ERCP procedures performed in 2006-2021. The presence of PAD was divided into two groups based on the PAD type: Boix type 1 (the papilla located inside the PAD) and Boix types 2-3 (the papilla located either at the edge of the PAD or immediately adjacent to the PAD). The primary outcomes were the success rate of cannulation and overall adverse events within 30 days.

Results: PADs were registered in 8130 (12.1%) of ERCPs included in the study population. In total, 2114 (3.9%) patients had Boix type 1 PAD, while 5035 (8.2%) patients had Boix type 2 or 3 PAD. The chance of successful cannulation was lower in patients with type 1 PAD compared to no PAD (80.1% vs 88.7%; odds ratio: 0.42, 95% confidence interval: 0.38-0.46). No differences were seen in overall adverse events or post-ERCP pancreatitis. Adverse events occurred in 14.6% of patients with PAD type 1 and 16.0% of patients with PAD type 2 or 3, compared to 16.5% of patients without a PAD.

Conclusion: Cannulation appears less successful during ERCP when the papilla is located in the PAD (i.e., type 1). Adverse events seem not to increase with the presence of a PAD, but they could theoretically be influenced by the inability to cannulate.

Keywords: ERCP; adverse events; cannulation; duodenal diverticula; endoscopic retrograde cholangiopancreatography; periampullary diverticulum.

Figure 1.

This flow chart illustrates the total number of procedures with PAD in the registry, the exclusion of certain procedures, and the categorization of PAD subtypes. PAD, periampullary diverticulum.

Figure 2.

Odds ratio plot presenting the comparison of primary outcomes and specific adverse events between those with a PAD and those without. Furthermore, a separate comparison is displayed between various subtypes of PAD and no PAD. For details regarding the regression models see the Supplementary online material. ^aAdjusted for sex, age, ASA, indication, acute, or elective setting. ^bAdjusted for sex, age, ASA, precut, indication, acute, or elective setting, procedure time. ^cAdjusted for sex, age, ASA, precut, indication, acute, or elective setting, procedure time. ^dAdjusted for sex, age, precut, indication, procedure time. ^eAdjusted for sex, age, ASA, precut, indication, acute, or elective setting, procedure time. ^fAdjusted for sex, age, precut, indication, procedure time. CI, confidence interval; PAD, periampullary diverticulum.