Background: One of the strongest risk factors of cardiovascular disease is a high concentration of low-density lipoprotein cholesterol (LDL-C); thus, the accurate measurement of LDL-C concentration in children is important. The aim of this study was to compare the concentration of LDL-C measured by direct method and the LDL-C concentration estimated by different formulas with the modified β-quantification method in children with type 1 diabetes.
Methods: Thirty-one serum samples received from diabetic children (15/18; M/F) and 26 serum samples taken from nondiabetic children (18/14; M/F) were used in the study. LDL-C concentrations were determined by direct enzymatic method (LDL-CD) and the modified β-quantification method (LDL-CmBQ). The concentration of LDL-C was also calculated by different formulas.
Results: The results of LDL-CmBQ and LDL-CD correlated. However, the correlation coefficient obtain in the nondiabetic samples was much higher (r=0.927; p<0.001) than the correlation coefficient for LDL-CmBQ and LDL-CD concentration obtained in the diabetic samples (r=0.691; p<0.001). Additionally, the coefficients of correlation between the LDL-CmBQ concentration and the LDL-C concentrations calculated by different formulas were lower in diabetic (range: 0.514-0.693) than in nondiabetic samples (range: 0.834-0.937). Bland-Altman plots showed much higher confidence intervals for 95% limits of agreement for the differences between LDL-CmBQ and LDL-CD as well as for the differences between LDL-CmBQ and LDL-C estimated by different formulas for samples from the diabetic samples compared with nondiabetic samples.
Conclusions: The limitations of methods measurement and of each of the formulas should be taken into account when a medical decision to lower LDL-C is made in the treatment of diabetic children.
Keywords: LDL-cholesterol estimation; children; diabetes mellitus type 1; modified β-quantification method.