OBJECTIVE: The assay of dried blood spots on filter paper to determine blood glucose concentration has been used to detect hypoglycaemia in out patients. We assessed the accuracy of this approach in assaying blood glucose concentrations in the hypoglycaemic range. DESIGN: Volunteers were rendered hypoglycaemic by intravenous infusion of insulin. The glucose concentration in simultaneously taken blood samples was measured either fresh or after drying on filter paper. PATIENTS: Twenty-four healthy young volunteers and 9 patients with insulin-dependent diabetes were studied. MEASUREMENTS: Plasma glucose concentrations were measured using a standard auto analyser glucose oxidase method. Whole blood taken simultaneously was placed on prepared filter paper and allowed to dry; glucose concentration was then measured using a well-established technique. A correction factor was applied to convert the glucose concentration of plasma to that of whole blood. The relationship between glucose concentrations measured by the two methods was determined by regression coefficient. RESULTS: In the unequivocally hypoglycaemic range (plasma < or = 2.5 mmol/l), corrected dried blood spot glucose concentrations significantly correlated with standard plasma glucose concentrations (r = 0.81; P < 0.001). The dried blood spot method had a sensitivity of 91%. In the range designated probable hypoglycaemia (plasma < or = 3.3 mmol/l), there was also significant correlation (r = 0.90; P < 0.001) and the sensitivity was 96%. The specificity of the dried blood spot method was 100% in both ranges. CONCLUSIONS: Measurement of glucose concentrations in dried blood spots is specific and sensitive in the hypoglycaemic range. The present study indicates that hypoglycaemia may be excluded or confirmed respectively when levels in excess of 3.7 or below 2.8 mmol/l are found in uncorrected dried blood spot analysis.
OBJECTIVE: The assay of dried blood spots on filter paper to determine blood glucose concentration has been used to detect hypoglycaemia in out patients. We assessed the accuracy of this approach in assaying blood glucose concentrations in the hypoglycaemic range. DESIGN: Volunteers were rendered hypoglycaemic by intravenous infusion of insulin. The glucose concentration in simultaneously taken blood samples was measured either fresh or after drying on filter paper. PATIENTS: Twenty-four healthy young volunteers and 9 patients with insulin-dependent diabetes were studied. MEASUREMENTS: Plasma glucose concentrations were measured using a standard auto analyser glucose oxidase method. Whole blood taken simultaneously was placed on prepared filter paper and allowed to dry; glucose concentration was then measured using a well-established technique. A correction factor was applied to convert the glucose concentration of plasma to that of whole blood. The relationship between glucose concentrations measured by the two methods was determined by regression coefficient. RESULTS: In the unequivocally hypoglycaemic range (plasma < or = 2.5 mmol/l), corrected dried blood spot glucose concentrations significantly correlated with standard plasma glucose concentrations (r = 0.81; P < 0.001). The dried blood spot method had a sensitivity of 91%. In the range designated probable hypoglycaemia (plasma < or = 3.3 mmol/l), there was also significant correlation (r = 0.90; P < 0.001) and the sensitivity was 96%. The specificity of the dried blood spot method was 100% in both ranges. CONCLUSIONS: Measurement of glucose concentrations in dried blood spots is specific and sensitive in the hypoglycaemic range. The present study indicates that hypoglycaemia may be excluded or confirmed respectively when levels in excess of 3.7 or below 2.8 mmol/l are found in uncorrected dried blood spot analysis.