The low-density lipoprotein debate: indirect methods vs direct measurement in the Nepalese population.

Abstract

Introduction: Many clinical laboratories use indirect estimation methods for low-density lipoprotein cholesterol (LDL-C) because of cost and practicality. This study compared the accuracy of indirect LDL-C estimates derived from the Friedewald, Puavilai, Hatta, and Martin equations in a Nepalese population.

Methods: This retrospective analysis included 7750 patients who underwent lipid profile testing at the Norvic International Hospital, Kathmandu, from February 1 to July 31, 2021. Participants were categorized based on triglyceride levels (<400 mg/dL and ≥400 mg/dL). Pearson correlation, paired t tests, and analysis of variance were employed to assess relationships and differences between directly measured and estimated LDL-C values.

Results: The Puavilai equation demonstrated the highest concordance with directly measured LDL-C (92%), followed by the Martin (89%), Friedewald (88%), and Hatta (81%) equations. The Puavilai equation exhibited the smallest discrepancies (mean difference = 8.9 mg/dL) for triglyceride values below 150 mg/dL. The Martin equation was most accurate for triglyceride values above 150 mg/dL (mean difference = 10.0 mg/dL) and remained reliable, even for triglyceride values above 400 mg/dL. The Hatta equation showed the largest estimation errors.

Discussion: The Puavilai equation is recommended for estimating LDL-C when triglyceride levels are below 150 mg/dL, while the Martin equation is preferred for levels above this threshold in the Nepalese population.