Glycated hemoglobin measurements from dried blood spots: reliability and relation to results obtained from whole blood samples

Abstract

Background: Main objective was to measure glycated hemoglobin (HbA1c) in dried blood spots on paper filter and in whole blood samples in diabetic patients to evaluate relationship between two methods and their respective reliability.
Methods: The 20×10 µl of venous blood samples of 33 diabetics were blotted onto the filter paper allowed to dry at room temperature and then stored at 25˚C and 4˚C. HbA1c was measured via the Turbidimetric Inhibition Immunoassay Technique. The relation was evaluated with correlation and linear regression tests using STATA software and SPSS. Agreement between the results obtained from the dried blood spots and others was evaluated using the Bland and Altman. The pitman's permutation test was also employed to compare the difference in variance.
Results: A high positive correlation was detected between whole blood samples and dried blood spots stored at 4˚C (r² =0.90) and at 25˚C (r² = 0.95). The Bland and Altman graphs, as well as the Pitman tests, showed statistically significant differences in variability between the values obtained from whole blood samples and those derived from dried spots stored at 4˚C (p=0.05) or 25˚C (p=0.004).
Conclusion: HbA1c measurements from dried blood spots on the filter paper yielded reliable results. That the Hitachi autoanalyzer is available in most countries renders this assay less costly than the High Performance Liquid Chromatography Method (HPLC). In addition, the filter paper method for Immuno-turbidimetric estimations of HbA1c at different temperatures is reliable and may be particularly useful in outpatient diabetes clinic.