Comparaison des techniques de dosage de L'HBA1C utilisées au chu de Sétif

Abstract

HbA1c is a crucial indicator for evaluating blood glucose in diabetic patients. Its measurement is performed using rigorously controlled techniques, with advantages and limitations that must be taken into account when interpreting results in clinical practice. In our study, we aimed to compare two methods of HbA1c measurement: an immunoturbidimetric method (using the Cobas Integra 400 Plus analyzer) versus a chromatographic method (HPLC Bio-rad D-10). In our study, the percentage of men is equal to that of women (50%), with no predominance of one gender over the other. The mean age of our population is 50.9 years with a standard deviation of 13.88 years, and a predominance in the age group of 40-60 years. The HbA1c measurements from the patient series using the Bio-rad D-10 show an average of 6.47% with a standard deviation of 1.63%. However, the same patient series analyzed on the Cobas Integra 400 gives an average of 6.72% with a standard deviation of 1.59%. The comparison of means between the two systems using the Student's t-test confirms that there is a statistically significant difference between the two readings, with a p-value of 0.000. These results are further supported by Passing-Bablok regressions, Bland-Altman plots, and the discordance in patient classification for one-third of the study population, which show that the D-10 method is much more accurate than the Cobas method. This raises questions about the possibility of replacing HbA1c measurements on the D-10 with immunoturbidimetry on the Cobas under the current conditions of our laboratory.