'Hospital hyperglycaemia'; perception or reality?

Patients' perceptions that clinic fasting blood glucose measurements in Type 2 diabetes are artefactually high were investigated. Eighteen men and 14 women in the Salford cohort of the United Kingdom Prospective Diabetes Study (UKPDS) with a median age of 51 (range 37-69) years measured their fasting blood glucose concentration at home with visually read sticks or meters; they then collected capillary blood into fluoride tubes both at home and in clinic after a median lapse of 77 (range 25-173) min. The clinic samples were measured immediately and after a time lapse equivalent to the time taken to reach clinic to control for continued glycolysis in the home sample. Mean fasting blood glucose was 5.9 +/- 1.9 mmol 1(-1) by the patients' own measurement and 6.0 +/- 1.8 mmol 1(-1) on the home collected sample compared with 6.2 +/- 2.0 mmol 1(-1) and 6.1 +/- 2.0 mmol 1(-1) on the hospital immediate and time-lapsed measurements, respectively (p > 0.05 for all comparisons between home and hospital measurements); although there were no systematic differences, occasional measurements differed more than 2 mmol 1(-1). These data support the use of the fasting blood glucose level as a stable, patient-independent measure of glycaemic control in Type 2 diabetes.