Ke-ke Jia1, Jie Zhang. 1. Department of Clinical Laboratory, Peking University Third Hospital, Beijing, China.
Abstract
BACKGROUND: In China, two common methods are used to detect serum glucose, glucose oxidase (GOD) and hexokinase (HK) methods. We evaluated five commercially available glucose reagent kits with the Center of Disease Control (CDC) HK reference method. METHODS: Five routine methods were compared with the HK reference method according to Clinical and Laboratory Standards Institute (CLSI) guideline (EP9-A2). These five routine methods were performed on an "open channel" Olympus 5400 analyzer. RESULTS: All five evaluated methods were closely correlated to the reference method (r > 0.9998). Compared to the HK reference method, the Olympus-HK method and Roche-GOD method showed a very small positive bias (mean +/- SD: 0.08 +/- 0.09 mmol/L and 0.03 +/- 0.08 mmol/L); while the Biosino-HK method, Osborne-GOD method and Baiding-GOD method showed a very small negative bias (mean +/- SD: -0.07 +/- 0.06 mmol/L, -0.09 +/- 0.07 mmol/L and -0.01 +/- 0.04 mmol/L). CONCLUSIONS: All five reagents we evaluated using matching calibrators showed good correlation to the reference method and small bias. These five glucose reagents are acceptable for use in clinical laboratories.
BACKGROUND: In China, two common methods are used to detect serum glucose, glucose oxidase (GOD) and hexokinase (HK) methods. We evaluated five commercially available glucose reagent kits with the Center of Disease Control (CDC) HK reference method. METHODS: Five routine methods were compared with the HK reference method according to Clinical and Laboratory Standards Institute (CLSI) guideline (EP9-A2). These five routine methods were performed on an "open channel" Olympus 5400 analyzer. RESULTS: All five evaluated methods were closely correlated to the reference method (r > 0.9998). Compared to the HK reference method, the Olympus-HK method and Roche-GOD method showed a very small positive bias (mean +/- SD: 0.08 +/- 0.09 mmol/L and 0.03 +/- 0.08 mmol/L); while the Biosino-HK method, Osborne-GOD method and Baiding-GOD method showed a very small negative bias (mean +/- SD: -0.07 +/- 0.06 mmol/L, -0.09 +/- 0.07 mmol/L and -0.01 +/- 0.04 mmol/L). CONCLUSIONS: All five reagents we evaluated using matching calibrators showed good correlation to the reference method and small bias. These five glucose reagents are acceptable for use in clinical laboratories.