ObjectivesWe aimed to clarify the clinical utility of measuring serum pancreatic enzymes after endoscopic retrograde cholangiopancreatography (ERCP) for the purpose of predicting post‐ERCP pancreatitis (PEP) by a meta‐analysis of diagnostic test accuracy studies.MethodsStudies on the prediction accuracy of PEP by serum amylase or lipase measured at 2, 3, and 4 h after ERCP were collected. A literature search was performed in PubMed and the Cochrane Library database for studies published between January 1980 and March 2023. The quality of individual studies was evaluated using the Quality Assessment of Diagnostic Accuracy Studies‐2. Data were analyzed using Meta‐DiSc 2.0 software.ResultsWe searched the databases and identified 20 observational studies (12,313 participants). PEPs were defined according to criteria by Cotton or modified Cotton, revised Atlanta criteria, or the Japanese criteria. Meta‐analysis of eight studies (4389 participants) showed a pooled sensitivity of 71.1% (95% confidence interval [CI] 56.1–82.5) and pooled specificity of 91.2% (95% CI 85.9–94.6) for the serum amylase cut‐off value at 3 times the upper limit of normal (ULN). Another meta‐analysis of five studies (1970 participants) showed a pooled sensitivity of 85.8% (95% CI 61.9–95.7) and pooled specificity of 85.3% (95% CI 81.9–88.1) for the serum lipase cut‐off value at 3 times ULN.ConclusionDespite a high risk of bias due to various reference standards, this updated meta‐analysis and the utility assessment by a decision tree showed the utility of serum amylase or lipase levels more than 3 times ULN measured 2–4 h after ERCP for predicting PEP.