Given the importance of LDL cholesterol as a focus of coronary heart disease (CHD) risk reduction, it is an irony of contemporary medicine generally not recognized by clinicians that LDL cholesterol values are based on an estimate rather than a direct measurement. The limitations of the Friedewald estimate are well documented but, unfortunately, not widely appreciated. Direct measurement of LDL cholesterol obviates the limitations of the Friedewald formula, including the intrinsic variability in its multiple components. Among those potential sources of error are patient noncompliance with 12-hour fasting and the potential interference from elevated triglyceride levels.
Estimated vs. measured LDL values
Most low-density lipoprotein (LDL) cholesterol values provided to clinicians are not a direct measure of low-density lipoprotein particles. Instead, they are estimated by the Friedewald formula, a linear calculation that subtracts two variables — (1) measured high-density lipoprotein (HDL) cholesterol and (2) an estimate of very low-density lipoprotein (VLDL) cholesterol from total cholesterol.
The Friedewald formula is accurate only when applied to blood samples from patients who have fasted for at least 12 hours. Moreover, the ratio between total cholesterol and triglycerides in the Friedewald formula used to estimate VLDL cholesterol is not constant, and may lead to misclassification.
The Friedewald formula.
LDL-C = TC – HDL-C – TG/5, where VLDL-C = TG/5
LDL-C: low-density lipoprotein cholesterol
TC: total cholesterol
HDL-C: high-density lipoprotein cholesterol
TG: triglycerides
The NCEP Lipoprotein Measurement Working Group has recommended that the Freidewald formula not be used as a reference method for LDL cholesterol measurement because there are too many circumstances where Friedewald formula LDL cholesterol values are known to be inaccurate. NCEP, the National Cholesterol Education Program, is an expert panel working under the auspices of the National Institute of Health (NIH) in Bethesda, Md.
According to NCEP, even if a patient’s total cholesterol, HDL cholesterol and triglyceride values are within the range of acceptable total-error performance, there is no guarantee that LDL cholesterol values would meet the group’s total-error performance goal.
Practical problems with the Friedewald formula
Because transient postprandial increases in triglyceride concentrations can significantly affect estimated LDL cholesterol values, the NCEP recommends that the Friedewald formula only be used when direct total cholesterol, HDL cholesterol and triglyceride measurements are made on fasting samples.
The need to use fasting samples to estimate LDL cholesterol values underscores a basic fact about the Friedewald formula. The relation between VLDL cholesterol and triglycerides is not linear for non-fasting patients or for fasting patients with elevated triglyceride levels. With the Friedewald formula, the higher the triglyceride levels, the greater the overestimation of VLDL cholesterol and the greater the corresponding underestimation of LDL cholesterol. In fact, the miscalculations are so huge when triglyceride levels are Ž400mg/dL that the NCEP recommends that non-fasting Friedewald formula LDL cholesterol values be discarded. However, even when triglyceride levels are <400 mg/dL (4.52 mmol/L), the Friedewald formula might still considerably underestimate LDL cholesterol.
Because diabetes Mellitus is associated with triglyceride abnormalities, the extent to which the Friedewald formula should be relied upon for the management of lipoprotein disorders in diabetic patients has been questioned.
Given the high number of patients for whom Friedewald formula LDL cholesterol values is problematic, it is little wonder that the NCEP has called for the development of new methods for the direct quantifation of LDL cholesterol.
The Friedewald formula is problematic for estimating LDL cholesterol values in the following populations: those who have trouble with the fasting requirement; geriatric patients; patients taking multiple medications; diabetic patients; pediatric patients; patients with high triglyceride levels; diabetic patients; patients with dysbetalipoproteinemia (type III hyperlipoproteinemia).
Direct-measurement LDL cholesterol values have shown a very high correlation with LDL cholesterol values obtained by beta quantification with correlation coefficients of 0.97, 30.98,4 0.97 with triglyceride levels ( 400 mg/L (4.52 mmol/L) and 0.94 with triglyceride levels Ž400 mg/L (4.52 mmol/L),5 and 0.96.6
One advantage of the direct measurement LDL cholesterol is that it can be applied to both fed and fasting samples. Another advantage is that LDL cholesterol values are not affected by high triglyceride levels. Directly measured LDL cholesterol values also appear to be more accurate than Friedewald-formula values for diabetic patients.
Reimbursement
Medicare reimbursement is available for the direct measurement of LDL cholesterol under Health Care Financing Administration (HCFA) rules. The CPT code is 83721, and the Medicare reimbursement rate currently averages $12.
One benefit of direct measurement of LDL cholesterol is the increased confidence with which clinicians and patients can regard LDL cholesterol values. An unexpected rise in the LDL cholesterol value at a regular eight- to 12-week follow-up visit is usually the source of a lot of questions from both the patient and the clinician. Is the LDL value increased because of medication noncompliance? Has the lipid-lowering agent lost its potency?
Unlike other aspects of medicine, lipid-lowering treatment is focused on a goal represented by a single value, the LDL cholesterol level. Patients expect and deserve the most accurate laboratory values available. Unreliable or questionable laboratory values may have negative psychological effects for patients, and adherence to treatment might suffer as a consequence.
An estimate calculated with the Friedewald formula should no longer be acceptable by clinicians as the only option for obtaining LDL cholesterol values. Because direct measurement of LDL cholesterol is still a recent concept for clinical laboratories, clinicians must regularly specify this test when sending patient samples out for processing.
Daniel Wohlgelernter, M.D., is assistant clinical professor of medicine at the UCLA School of Medicine, and staff cardiologist at Santa Monica UCLA Medical Center and St. John’s Hospital and Health Center.
For further information about Equal Diagnostics’ Automated Direct LDL Reagent for Chemistry Analyzers, call Equal Diagnostics at 1-800-999-6578.
