Adverse Events Associated with Endoscopic Retrograde Cholangiopancreatography: Systematic Review and Meta-analysis

Background & Aims

Endoscopic retrograde cholangiopancreatography (ERCP)-related adverse events (AEs) are associated with morbidity, mortality, and healthcare expenditure. We aimed to assess incidences and comparisons of ERCP AEs.

Methods

We included studies performed after 2000 reporting on ERCP AEs from database inception through March 12, 2024. Outcomes included pancreatitis, bleeding, cholangitis, cholecystitis, perforation, and death. DerSimonian and Laird random effects meta-analyses were performed to calculate incidences of AEs. Subgroup and pairwise meta-analyses were performed. Meta-regression was performed on median recruitment year to assess temporal trends in pancreatitis incidence.

Results

A total of 380 studies were included. The incidence of death attributable to ERCP was 0.2% (95% confidence intervals, CI, 0.1-0.3%, I² 44%, n=47,258) in all-comers. The overall incidence of pancreatitis was 4.6% (95% CI 4.0-5.1%, I² 96%, n=293,378) among all-comers and 6.5% (95% CI 5.9-7.1%, I² 89%, n=88,809) among first-time patients. Pancreatitis incidence remained stable between 2000-2023 (average annual percent change 0.06, 95%CI -0.27 to 0.39). The overall incidences of the following AEs for all-comers were: bleeding (1.5%, 95% CI 1.2-1.7%, I² 93%, n=229,655), cholangitis (2.5%, 95%CI 1.9-3.3%, I² 96%, n=121,619), cholecystitis (0.8%, 95%CI 0.5-1.2%, I² 39%, n=7,799), and perforation (0.5%, 95%CI 0.4-0.6%, I² 90%, n=306,378).

Conclusions

ERCP-associated AEs remain common. Incidence of post-ERCP pancreatitis remained static despite improvements in techniques, prevention, and recognition. These results are important to patients, endoscopists, and policy makers to inform consent and to encourage implementation of available risk mitigation strategies.