Performance of low-density lipoprotein cholesterol calculated using Martin’s formula compared with the Friedewald equation in a Japanese population

Abstract

Low-density lipoprotein cholesterol (LDL-C) is used for risk assessment of various arteriosclerotic diseases. In recent years, Martin’s formula, which is an improvement over the Friedewald equation, has been reported. We compared the homogenous method with LDL-C estimated using the Friedewald equation and Martin’s formula. Correlation analysis with the homogeneous method in healthy subjects showed that the Friedewald equation (Y = 0.956X + 9.690, r = 0.9183, Sy·x = 11.54) and Martin’s formula (Y = 1.019X + 2.425, r = 0.9757, Sy·x = 6.23) were obtained, and Martin’s formula showed a better correlation. When the correlation between the homogeneous method and the estimation formula for triglyceride (TG) level was analyzed, the regression equation for TG levels and LDL-C residuals estimated by the homogeneous method, and the Friedewald equation [Y = −0.0652X + 8.571 (TG < 400), Y = −0.0848X + 26.332 (TG ≥ 400)] and Martin’s formula [Y = 0.0059X + 1.3641 (TG < 400) and Y = −0.027X + 33.872 (TG ≥ 400)] were obtained. These results suggest that Martin’s formula is highly accurate and convenient, approximating the homogenous method, and is a better estimation method than the Friedewald equation.