Audit of amylase and lipase requests in suspected acute pancreatitis and cost implications, South Africa

Abstract

Background: The internationally accepted criteria for the diagnosis of acute pancreatitis (AP) requires two of the three following features to be present: characteristic abdominal pain, elevated serum amylase and/or lipase enzymes, or consistent imaging results. However, sensitivity and specificity can vary depending on the population and cut-off values used. Objective: This study evaluated the suitability of amylase and lipase as first-line diagnostic biomarkers of suspected AP for the local population served by Tygerberg Hospital, South Africa. Methods: This retrospective analysis was conducted in June 2019 using all amylase and/or lipase request data from December 2018. Patient clinical data were included in sensitivity and specificity analyses of amylase, lipase or dual requests for diagnosis of AP. Cost per test data were obtained from the National Health Laboratory Service and used to calculate the total cost of the tests and potential savings. Results: Sensitivity for lipase was 90.0% compared to 50.0% for amylase. Specificity was similar for singular measurements of lipase and amylase. Dual measurement of amylase and lipase showed no improvement in sensitivity (83.3%) and only a minor increase in specificity (97.4%) compared with measurement of lipase alone. The estimated savings was R2522.85 ($174.98 USD), with a potential annual cost saving of R84 423.74 ($5855.69 USD). Conclusion: Lipase was shown to be a more sensitive biomarker compared to amylase for the screening of AP, providing evidence for laboratories to educate local staff and promote improved requesting practices by clinicians. Additionally, preventing unnecessary dual requests may reduce costs.

Keywords

• audit
• pancreatitis
• amylase
• lipase
• cost