Different Methods of Low-density Lipoprotein Cholesterol Estimation and the Impact on Lipid-lowering Therapy in...

Indirect estimation of low-density lipoprotein cholesterol (LDL-C) is a common clinical practice. The Friedewald equation is used most often but has inherent limitations. Clinical implications of such a practice have not been well defined, especially in the current era of targeting low (<50-70 mg/dL) or ultralow (<30-40 mg/dL) LDL-C levels.

Overall, 3,028 consecutive subjects with coronary artery disease (CAD) undergoing coronary revascularization were included. Four methods of LDL-C estimation were compared: direct estimation, the Friedewald, Martin, and Sampson equations.

The mean age of the subjects was 61.3 ± 10.2 years, and 2,525 (83.4%) were men. Mean direct LDL-C was 78.9 ± 32.9 mg/dL. Compared with the direct estimation, all three indirect methods significantly underestimated LDL-C, but the Martin equation had the least bias (mean differences of -10.5 ± 9.7 mg/dL, -5.2 ± 7.6 mg/dL, and -7.2 ± 8.3 mg/dL with the Friedewald, Martin, and Sampson equations, respectively; p-values <0.001 for all the comparisons). Among patients with LDL-C >70 mg/dL and >50 mg/dL, the Friedewald equation erroneously classified 24.6% and 19.9%, respectively, as having LDL-C below these thresholds. This error increased with increasing triglyceride levels. The Martin equation was the most accurate, whereas the Sampson equation had intermediate accuracy.

Our study shows that the Friedewald equation underestimates LDL-C and can potentially result in significant undertreatment in patients with CAD in whom aggressive LDL-C lowering is crucial. Direct estimation is the preferred method, but the Martin equation could be a reasonable alternative if the direct estimation is not feasible due to logistical constraints.