You have to be registered and logged in for purchasing articles.

Abstract

LDL Cholesterol: Don't guess. Measure it. A critical examination of the Friedewald formula by Johannes Aufenanger and Bernd Zawta

The Friedewald formula is still used today to estimate the concentration of LDL cholesterol in a patient sample, even though modern, direct methods of determining LDL cholesterol are available. The limits of the Friedewald formula have been known for some time: falsely low results in the presence of chylomicrons and remnants and increasing triglyceride values. Due to the altered triglyceride composition of lipoprotein particles in primary lipid metabolic disorders (e.g., dysbeta-lipoproteinemia) and secondary diseases (diabetes mellitus, liver cirrhosis, kidney disease, haemodialysis), the formula often delivers incorrect results. The formula does not take Lp(a) into consideration. Error assessments carried out according to Gauss's law of error propagation show that, despite a low analytical scattering of individual determinations of total cholesterol, triglycerides and HDL cholesterol, the relative total error (CV) for LDL cholesterol can increase greatly depending on the ratio of cholesterol to triglyceride values. The Friedewald formula usually does not meet the level of precision for LDL cholesterol determination (CV ≤ 4%) recommended by the National Cholesterol Education Program (NCEP) Laboratory Standardisation Panel. This makes it difficult or impossible to categorise coronary heart disease, assess a patient's risk of developing it, or effectively monitor therapy. Conclusion: in today's world of precise, reliable, and automated methods for performing direct determinations of LDL cholesterol by using a homogeneous assay, for example, there is no reason to use the Friedewald formula to calculate LDL cholesterol.

DOI: Clin. Lab. 1999;45:617-622