Interlaboratory standardization of glycated hemoglobin determinations

• Published in: Clinical chemistry (Baltimore, Md.) Vol. 32; no. 2; pp. 358 - 360
• Main Authors: Little, RR, England, JD, Wiedmeyer, HM, McKenzie, EM, Mitra, R, Erhart, PM, Durham, JB, Goldstein, DE
• Format: Journal Article
• Language: English
• Published: Washington, DC Am Assoc Clin Chem 01.02.1986; American Association for Clinical Chemistry
• Subjects: Analytical, structural and metabolic biochemistry; Biological and medical sciences; Chromatography, Affinity; Chromatography, High Pressure Liquid; Chromatography, Ion Exchange; Diabetes Mellitus - blood; Fundamental and applied biological sciences. Psychology; Glycated Hemoglobin A - analysis; Glycated Hemoglobin A - standards; Hemoproteins; Humans; Laboratories - organization & administration; Laboratories - standards; Metalloproteins; Proteins; Quality Control; Reference Standards; Assay; Human; Hemoprotein; Clinical biology; Quality control; Standardization; Hemoglobin; Hemoglobin A1; Chromatography; Blood
• ISSN: 0009-9147; 1530-8561
• DOI: 10.1093/clinchem/32.2.358

As the clinical utility of glycated hemoglobin (gHb) measurement increases, so does the need for standardization of values between different methods and different laboratories. Using three different methods, we examined the feasibility of interlaboratory standardization of gHb measurement. A liquid-chromatographic (HPLC) system from our research laboratory was designated the reference method. For gHb standards we used erythrocyte hemolysates prepared from blood samples from nondiabetic and diabetic subjects. Values assigned to each standard were based on the mean of multiple gHb determinations by the HPLC method. A clinical laboratory routinely prepared hemolysates and assayed gHb by commercially available ion-exchange ("mini column") and affinity chromatographic methods. For each assay a standard curve was constructed and gHb values were derived from these curves. Samples analyzed in the clinical laboratory were also analyzed in the research laboratory and the curve-derived values were compared with the HPLC-measured values, to determine the accuracy of our interlaboratory standardization procedure. Correlations were excellent (r = 0.99). The lack of significant differences between calculated and HPLC-measured values indicates that interlaboratory standardization is feasible.