Background: Accurate venous plasma glucose estimation is essential for both high quality diabetes research and optimal clinical practice. Glucose loss secondary to glycolysis following delayed separation and extraction of plasma from the cellular components of blood is reported as ≤7% per hour, however previous studies were across a narrow glucose range, with few participants. A high rate of glucose loss has implications for the interpretation of results from some epidemiological studies.
Aim: We aimed to determine pre-analytical glucose loss due to glycolysis from a larger number of participants, across a wide range of glucose concentrations.
Methods: Antecubital venous samples were collected from diabetes outpatients and placed into PST (plasma separator tube) ‘green top’ lithium heparin and also fluoride ‘grey top’ Vacutainers™. Samples were either prepared using ideal methodology (ice slurry, immediate refrigerated centrifugation and plasma extraction) or plasma extraction was delayed by ≤ 1 hour. Laboratory plasma glucose was analysed using the hexokinase method.
Results: Baseline glucose from 63 participants ranged from 3.4 to 31.1mmol/L. Glucose loss averaged 0.24mmol/L/hr for both types of Vacutainer™and was; a) similar at 0’- 30’ and 30’- 60’(minutes), b) independent of baseline glucose concentration and c) independent of individual sample’s cellular count (red cells, leucocytes, platelets). Immediate room temperature plasma separation using PST tubes, followed by delayed (≤4 hours) laboratory plasma extraction, resulted in a glucose loss of 0.02±0.23mmol/L.
Conclusion and recommendation: Pre-analytical glucose loss was reassuringly low and was independent of both baseline glucose and also cellular count. These findings differ from our currently accepted understanding of the determinants of pre-analytical glycolysis. Sample collection into a PST tube followed by immediate plasma separation (but not extraction) resulted in minimal pre-analytical glucose loss, thus this preparation method represents a pragmatic method of minimising glycolysis, in settings where a centrifuge is available. Pre-analytical glucose loss should be expressed as an absolute value i.e. 0.24 mmol/L/hr, rather than as a percentage.