Problems with the Old Method

Currently, most clinical laboratories utilize the equation known as the Friedewald Formula to estimate a patient's LDL cholesterol concentration. The formula uses the following calculation:

Estimation of LDL Cholesterol = Total Cholesterol - HDL Cholesterol - VLDL Cholesterol*

* Estimation of VLDL Cholesterol = Trigs/5 (Trigs/2.2 for mmol/L)

The formula estimates the LDL cholesterol concentration by subtracting the cholesterol associated with the other classes of lipoproteins from total cholesterol. This involves three independent lipid analyses, each contributing a potential source of error. It also involves a potentially inaccurate estimate of VLDL cholesterol. Since no direct VLDL cholesterol assay is available, it is calculated from the triglyceride value divided by a factor of 5 (triglycerides/2.2 for mmol/L). This divisor can also add error to all LDL cholesterol estimates, but is especially inappropriate for individuals with elevated triglyceride levels. In addition, clinical laboratories utilize automated enzymatic analyses for cholesterol and triglyceride quantitation within serum or plasma, and HDL cholesterol is measured after precipitation of other lipoproteins in serum or plasma with either heparin manganese chloride, dextran magnesium sulfate, or phosphotungstic acid.

The drawbacks of using the Friedewald Formula for determining levels of LDL cholesterol are: it is estimated by calucation; it requires multiple assays and multiple steps each adding a potential source of error; it is inaccurate as triglyceride levels increase; it requires that patients fast for 12 to 14 hours prior to specimen collection to avoid a triglyceride bias; and it is not standardized.

Moreover, LDL cholesterol concentrations cannot be reported in individuals with elevated triglyceride levels (>400 mg/dL or 4.5 mmol/L). In addition, it has been reported that the formula becomes increasingly inaccurate at borderline triglyceride levels (200-400 mg/dL or 2.3-4.5 mmol/L).

Friedewald Formula Drawbacks

Indirect measurement: An estimation

Multiple assays and multiple sample steps: Increased imprecision and inaccuracy

Triglyceride assay method variation

Triglycerides / 5 algorithm is inaccurate with increasing triglyceride levels: With triglycerides >400 mg/dL (4.5 mmol/L), LDL cannot be reported

Sample must be fasting to avoid triglyceride bias

HDL reagent/method variation