Comparison of Multiple Equations for Low-Density Lipoprotein Cholesterol Calculation Against the Direct Homogeneous Method

Rawaa E K Alsadig,Adel N Morsi
DOI: https://doi.org/10.12997/jla.2024.13.3.348
Abstract:Objective: Several equations have been proposed as alternatives for the reference method of measuring low-density lipoprotein cholesterol (LDL-C). This study aimed to evaluate these alternatives in comparison to the homogeneous method and validate their clinical utility. Methods: Data on the lipid profiles of 1,006 Sudanese individuals were analyzed. The paired t-test was used to compare the results of direct and calculated LDL-C. Bland-Altman plots were used to demonstrate the differences between the measured and calculated LDL-C against the mean values. Linear regression was conducted, using the correlation coefficient (r) to quantify the relationship between methods. The bias between measured and calculated LDL-C was compared to the National Cholesterol Education Program Laboratory Standardization Panel criteria (i.e., accuracy within ±4% of expected values). Results: The Martin and Anandaraja equations showed no significant difference compared to directly measured LDL-C (p>0.05). The DeLong equation indicated an insignificant difference only with a 99% confidence interval (p>0.01). The Martin, DeLong, and Teerakanchana equations exhibited the smallest limits of agreement, with data points concentrated closely around the mean difference line. Linear regression analysis revealed strong positive correlations (r>0.8) for most equations, except for the Ahmadi equation. The DeLong, Rao, and Martin equations demonstrated superior performance for LDL cutoff points (bias within ± 4%). The DeLong formula also showed superior performance at different lipid levels, closely followed by the Martin equation (bias within ±4%). Conclusion: The DeLong and Martin equations outperformed others, such as the widely used Friedewald equation, in calculating LDL-C. Further validation studies are needed.
What problem does this paper attempt to address?