Predictors of Post-endoscopic Retrograde Cholangiopancreatography Pancreatitis: A Comprehensive Systematic Review and Meta-analysis
Date
Language
Embargo Lift Date
Department
Committee Members
Degree
Degree Year
Department
Grantor
Journal Title
Journal ISSN
Volume Title
Found At
Abstract
Background & Aims Pancreatitis is the most common serious adverse event associated with endoscopic retrograde cholangiopancreatography (ERCP). This meta-analysis aimed to precisely assess the risk factors for post-ERCP pancreatitis (PEP). Methods We searched electronic databases for studies that assessed risk factors for PEP after adjusting for ≥3 risk factors, including at least one pre-specified patient-related and one procedure-related risk factor, and reported the data as adjusted odds ratios (ORs) with 95% confidence intervals. Meta-analyses were conducted using a random-effects model, and pooled adjusted ORs for risk factors reported in ≥3 studies were constructed. Results A total of 159 studies with 315,580 ERCPs were included, assessing 31 unique risk factors (20 patient-related and 11 procedure-related). Key patient-related predictors of PEP were age ≤60 years (OR, 1.81; high credibility), prior acute pancreatitis (OR, 2.59; moderate), age ≤40 years (OR, 2.33; moderate), asymptomatic choledocholithiasis (OR, 4.76; low), prior PEP (OR, 4.40; low), sphincter of Oddi dysfunction (OR, 3.11; low), and female gender (OR, 1.70; low). Key procedure-related predictors of PEP were any guidewire passage into the pancreatic duct (PD) (OR, 2.18; high), first ERCP with a native papilla (OR, 1.91; high), endoscopic papillary balloon dilation of an intact papilla (OR, 2.91; moderate), pancreatic acinarization (OR, 4.23; low), any PD cannulation (OR, 2.73; low), pancreatic sphincterotomy (OR, 2.64; low), difficult cannulation (OR, 2.60; low), any pancreatogram (OR, 2.40; low), and precut sphincterotomy (OR, 1.98; low). Conclusions Our meta-analysis focused on adjusted risk factors to provide precise estimates of the most important risk factors for PEP. Incorporating our results into a prediction model may reliably help identify high-risk patients, optimize informed consent, and guide prevention and management strategies for PEP.