BACKGROUND: Increased low density lipoprotein cholesterol (LDL-C) is an established risk factor for the development of coronary artery disease (CAD). Recent guidelines detail specific LDL-C cutpoints for therapeutic goals. In practice, LDL-C is usually derived from the Friedewald formula (FF). This calculation is known to be inaccurate with serum triglyceride (TG) concentrations >4.52 mmol/l, however, its accuracy among relatively healthy patient cohorts with TG concentrations < or =4.52 mmol/l is less well studied.

METHODS: We studied 661 ambulatory adults with TG concentrations < or =4.52 mmol/l and no overt CAD. Fasting venous lipid panels were obtained. LDL-C was calculated from the FF and also directly measured with the LipiDirect Magnetic LDL assay. Linear regression and paired t-test analyses were performed.

RESULTS: Calculated and directly measured LDL-C concentrations were significantly different (4.26+/-0.88 vs. 4.83+/-1.06 mmol/l respectively, p<0.0001). In 93% of measurements directly measured LDL-C exceeded calculated LDL-C. Although calculated and directly measured LDL-C concentrations were related (R=0.90), the discrepancy between them increased linearly with increasing TG concentrations (R=0.67) and clinically important differences existed at normal or slightly increased TG concentrations. Concordant results for NCEP ATP-III risk categories were present for only 48.1% of samples.

CONCLUSIONS: Significant differences between calculated and directly measured LDL-C using the LipiDirect Magnetic LDL assay exist in healthy subjects with TG < or =4.52 mmol/l. These differences are linearly related to TG concentrations and occur frequently at relatively low TG concentrations.