OBJECTIVES: To verify the clinical utility of serum amylase and lipase in differentiating acute pancreatitis (AP) from extrapancreatic acute abdominal diseases and in predicting post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP).

METHODS: A comprehensive search was carried out using MEDLINE, EMBASE, Cochrane Library, and three Korean databases for clinical studies published until July 2023. We included all studies that reported the diagnostic accuracy of serum amylase and lipase in screening AP and PEP, which were defined according to Atlanta consensus criteria. The quality of individual studies was assessed using Quality Assessment of Diagnostic Accuracy Studies-2. The pooled sensitivity, specificity, and area under the curve (AUC) were calculated with a bivariate model due to heterogeneity among studies.

RESULTS: A total of 11 cohort studies was included for analysis, 5 for AP and 6 for PEP. Pooled results for AP showed that serum lipase had higher diagnostic value than serum amylase, with sensitivity and AUC of 0.75 vs. 0.69 and 0.92 vs. 0.77, respectively; the specificity was the same for the two at 0.95. In contrast, pooled results for PEP showed that serum amylase at 3-4h post-ERCP had higher pooled specificity and AUC of 0.91 vs. 0.87 and 0.90 vs. 0.88, respectively, while the sensitivity was lower than that of serum lipase (0.71 vs. 0.84).

CONCLUSIONS: Our meta-analysis shows that serum lipase level greater than three times the normal value is better than that of amylase to differentiate AP from extrapancreatic acute abdominal diseases. For early prediction of PEP, serum amylase greater than three times normal at 3-4h post-ERCP can be used as a good predictor to rule out PEP.