A-239 A new method for estimating LDL-Cholesterol levels using calculated non-HDL Cholesterol

Abstract Background Low-density lipoprotein cholesterol (LDL-C) is a key target in clinical treatment. While ultracentrifugation is considered the gold standard for measuring LDL-C, its high cost and time demands have led to the development of alternative estimation methods, such as direct assays and calculation equations. This study aims to develop a novel equation for estimating LDL-C by establishing a mathematical correlation between calculated non-HDL cholesterol (non-HDL-C) and directly measured LDL-C. Methods This pilot study included 1,000 directly measured serum LDL-C results (range: 26–296 mg/dL) and calculated non-HDL-C results (range: 44–315 mg/dL), all performed at the central laboratory. The study was conducted in the State of Kuwait, with 877 results from Arabs and the remaining 123 from individuals of different ethnic groups. The test results were based on lipid profile samples received over a 6-week period, with no specific inclusion or exclusion criteria applied. All samples were analyzed using the Abbott Alinity chemistry analyzer. Assay methods for Total Cholesterol, direct HDL-C, and direct LDL-C employed enzymatic, accelerator selective detergent, and liquid selective detergent techniques, respectively. Non-HDL-C values were calculated by subtracting HDL-C from Total Cholesterol. The results of the calculated non-HDL-C and directly measured LDL-C were compared using linear regression. The LDL-C estimates derived from the regression equation were further validated by comparing them to 125 direct LDL-C results (range: 34–233 mg/dL) from prospectively collected samples. Results Linear regression analysis revealed a strong correlation between calculated non-HDL-C and direct LDL-C. The derived regression equation is, estimated LDL-C (mg/dL) = 0.98 × non-HDL-C (mg/dL) – 15.3, with R² = 0.89. The regression equations derived with and without the inclusion of non-Arab ethnic groups showed no significant difference. This equation enabled the estimation of LDL-C in 125 prospective samples, with the average absolute deviation between estimated LDL-C and direct LDL-C being just 3.6 mg/dL. Among these 125 samples, 73 were from patients being treated for dyslipidemia, with direct LDL-C levels ranging from 34 to 233 mg/dL. In this treated group, the average absolute deviation between estimated LDL-C and direct LDL-C was 4.4 mg/dL. Conclusion Using a regression equation to estimate LDL-C values could serve as an efficient alternative to direct LDL-C measurements. Laboratories should develop their own specific regression equations tailored to their direct assay methods. These equations can be validated through proficiency programs and, when implemented, have the potential to conserve resources.