Abstract
OBJECTIVE: This study was conducted to validate the Martin method in coronary atherosclerosis in comparison with the Friedewald equation. SUBJECTS AND METHODS: A total of 299 participants with a coronary artery calcium score (CACS) ≥300 and a serum triglyceride (TG) level <400 mg/dL at Seoul National University Hospital Healthcare System Gangnam Center, Seoul, Korea, were enrolled in this study. Low-density lipoprotein cholesterol (LDL-C) was directly measured with a homogeneous assay (DLDL) and estimated by both the Friedewald equation (FLDL) and Martin method (MLDL). Overall concordances between DLDL and LDL-C estimates were calculated as the percent agreement. The McNemar test was used to compare the rate of reclassification of participants with FLDL and MLDL, and to determine which differed significantly from each other. RESULTS: Overall concordance between DLDL and MLDL was slightly higher than that between DLDL and FLDL (73.2 vs. 70.9%, p < 0.001). The FLDL showed poor performance when the TG level was ≥200 mg/dL, mostly by underestimation, which represented a 64.7% discordance with DLDL. The reclassification rate by MLDL, however, did not exceed 35.3% in all of the TG groups. CONCLUSIONS: The Martin method to estimate LDL-C using the strata-specific TG:VLDL ratio showed a 2-fold better concordance with LDL-C measured with a direct homogeneous assay in coronary atherosclerosis compared to the Friedewald equation when the TG level was ≥200 mg/dL. This finding suggests that MLDL could be a better alternative for estimating LDL-C compared to FLDL when the TG level is ≥200 mg/dL in coronary atherosclerosis.