Survey on common reference intervals for general chemistry analytes in Hong Kong

Abstract

Introduction: Reference intervals (RIs) are essential tool for proper interpretation of results. There is a global trend towards implementing common RIs to avoid confusion and enhance patient management across different laboratories. However, local practices with respect to RIs lack harmonisation. Methods: We have conducted the first local survey regarding RIs for 14 general chemistry analytes in 10 chemical pathology laboratories that employ four different analytical platforms (Abbott Architect, Beckman Coulter AU, Roche Cobas, and Siemens Dimension EXL). Analytical bias was assessed by an inter-laboratory results comparison of external quality assurance programmes. Results: Sufficient inter-laboratory and interplatform agreement regarding the 10 analytes (albumin, alanine aminotransferase, aspartate aminotransferase, chloride, gamma-glutamyl transferase, phosphate, potassium, sodium, total protein, and urea) were demonstrated. However, the RIs were heterogeneous across all laboratories, with percentage differences of the upper RI value of up to 47% for aspartate aminotransferase (absolute difference of 16 U/L), 29% for urea (1.8 mmol/L), and 18% for potassium (0.8 mmol/L). The percentage difference between lower RI values was up to 24% for urea (0.6 mmol/L), 22% for phosphate (0.16 mmol/L), and 8% for total protein (5 g/L). The coefficients of variation of the upper RI values of potassium and sodium were 1.2 times and 1.0 times of their corresponding between-subject biological variation, respectively, representing unnecessary variations that are overlooked and unchecked in current practice. Conclusions: We recommend the use of common RIs for general chemistry analytes in Hong Kong to prevent interpreter confusion, improve electronic data transfer, and unite laboratory practice. This is the first local study on this topic, and our data can lay the groundwork for increasing harmonisation of RIs across more laboratory tests.