BACKGROUND: Several studies demonstrate significant bias in analytical methods used to measure glycohemoglobin. The clinical importance of that fact is evident when HbA1c overestimation leads to aggressive glucose management, resulting in more frequent hypoglycaemic episodes. Our study was aimed to compare two automated instruments (Integra 400 and D-10) in the evaluation of HbA1c in the Tunisian population. METHODS: Samples of 205 Tunisian diabetic patients were collected. The HbA1c assay was done simultaneously with a first generation immunoturbidimetric assay on an INTEGRA 400 (ROCHE) and using ionic exchange high pressure liquid chromatography (HPLC) on a D-10 system (BIO-RAD). RESULTS: Correlation is determined by linear regression analysis: D-10 = 0.921*(Integra 400) +1.125; coefficient of correlation (r) = 0.946. This r increases to 0.973 when samples of carriers for HbS and HbC (n = 9) are filtered out. For the carrier patients, significant differences in the percentage of HbA1c were observed relating to the methodology used. CONCLUSIONS: Laboratories must be aware of hemoglobin variant interferences on their methods of assessment of glycated hemoglobin. Using ion-exchange HPLC to control glycated hemoglobin seems to be essential to prevent mis-management in diabetic patients and to permit the diagnosis of the presence of HbS in patients.
BACKGROUND: Several studies demonstrate significant bias in analytical methods used to measure glycohemoglobin. The clinical importance of that fact is evident when HbA1c overestimation leads to aggressive glucose management, resulting in more frequent hypoglycaemic episodes. Our study was aimed to compare two automated instruments (Integra 400 and D-10) in the evaluation of HbA1c in the Tunisian population. METHODS: Samples of 205 Tunisian diabeticpatients were collected. The HbA1c assay was done simultaneously with a first generation immunoturbidimetric assay on an INTEGRA 400 (ROCHE) and using ionic exchange high pressure liquid chromatography (HPLC) on a D-10 system (BIO-RAD). RESULTS: Correlation is determined by linear regression analysis: D-10 = 0.921*(Integra 400) +1.125; coefficient of correlation (r) = 0.946. This r increases to 0.973 when samples of carriers for HbS and HbC (n = 9) are filtered out. For the carrier patients, significant differences in the percentage of HbA1c were observed relating to the methodology used. CONCLUSIONS: Laboratories must be aware of hemoglobin variant interferences on their methods of assessment of glycated hemoglobin. Using ion-exchange HPLC to control glycated hemoglobin seems to be essential to prevent mis-management in diabeticpatients and to permit the diagnosis of the presence of HbS in patients.