Renata Paleari1, Ferruccio Ceriotti2, Cornelis L Harteveld3, Marta Strollo4, Greet Bakker-Verweij3, Jeanet Ter Huurne3, Sharda Bisoen3, Andrea Mosca5. 1. Dip. di Fisiopatologia Medico-Chirurgica e dei Trapianti, Centro per la Riferibilità Metrologica in Medicina di Laboratorio (CIRME), Università degli Studi di Milano, Milano, Italy. 2. Clincal Laboratory, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milano, Italy. 3. Dept. of Clinical Genetics/LDGA, Leiden University Medical Center, Leiden, The Netherlands. 4. Laboratorio di Standardizzazione, Servizio Medicina di Laboratorio, Ospedale San Raffaele, Milano, Italy. 5. Dip. di Fisiopatologia Medico-Chirurgica e dei Trapianti, Centro per la Riferibilità Metrologica in Medicina di Laboratorio (CIRME), Università degli Studi di Milano, Milano, Italy. Electronic address: andrea.mosca@unimi.it.
Abstract
BACKGROUND: Most of the current methods used for the determination of HbA2 seem not well aligned. A comparison among the best performing techniques and the commutability of some control materials currently available and under development has been evaluated. METHODS: Forty blood samples were analyzed in duplicate over two separate days by different HPLC and capillary electrophoresis systems. The commutabilities of different control materials (NIBSC WHO reagent, Bio-Rad Lyphochek, and home prepared lyophilized controls RP1-3) have been assessed by analyzing the controls in quadruplicate over two consecutive days together with the blood samples. RESULTS: The mean within-run imprecision of HbA2 measurement on blood samples (CV, %) was between 0.6% and 10.1% for HbA2 values <3.5%, and between 1.1 and 3.1 for HbA2≥3.5%. The different methods were highly correlated (r between 0.9941 and 0.9995) although biased each other. The NIBSC WHO reagent was found not commutable in 15 over 28 comparisons, the Lyphochek 2 in 18/28, and RP3 in 4/28. Recalibration of all methods by RP1 and RP2 materials was able to reduce the overall variability from 6.8% to 3.4% at HbA2≤3.0% and from 6.7% to 3.0% at HbA2≥4.6%. CONCLUSION: The use of adequate commutable control materials as calibrators may reduce the inter-method variability of routine methods to an extent closer to the current analytical goals of bias based on biological variability.
BACKGROUND: Most of the current methods used for the determination of HbA2 seem not well aligned. A comparison among the best performing techniques and the commutability of some control materials currently available and under development has been evaluated. METHODS: Forty blood samples were analyzed in duplicate over two separate days by different HPLC and capillary electrophoresis systems. The commutabilities of different control materials (NIBSC WHO reagent, Bio-Rad Lyphochek, and home prepared lyophilized controls RP1-3) have been assessed by analyzing the controls in quadruplicate over two consecutive days together with the blood samples. RESULTS: The mean within-run imprecision of HbA2 measurement on blood samples (CV, %) was between 0.6% and 10.1% for HbA2 values <3.5%, and between 1.1 and 3.1 for HbA2≥3.5%. The different methods were highly correlated (r between 0.9941 and 0.9995) although biased each other. The NIBSC WHO reagent was found not commutable in 15 over 28 comparisons, the Lyphochek 2 in 18/28, and RP3 in 4/28. Recalibration of all methods by RP1 and RP2 materials was able to reduce the overall variability from 6.8% to 3.4% at HbA2≤3.0% and from 6.7% to 3.0% at HbA2≥4.6%. CONCLUSION: The use of adequate commutable control materials as calibrators may reduce the inter-method variability of routine methods to an extent closer to the current analytical goals of bias based on biological variability.