The National Cholesterol Education Program (NCEP) emphasizes that clinicians should be aware of the limitations ofthe Friedewald equation that we have noted (12). The ideal use of the approximation involves a 12-h fast; clinicians should be aware that even a 9-h fast will underestimate LDL-C by at least 2-4% (e.g., 3-6 mg/dL less in a patient with a measured LDL-C of 150 mg/dL). To establish an ~LDL-C within a 10% error limit, the
NCEP estimated that it would be necessary to collect four samples on different days. It also estimated that by averaging two ~LDL-C readings and using a cut point of 130 mg/ dL, a result of >145 mg/dL is reliably above the cutoff, and a result of <115 mg/dL is reliably below the cutoff (12). Thus, if a patient's goal is an LDL-C of <130 mg/dL, and the average of two samples on different days shows that medication lowered the ~LDL-C to 115 mg/dL or less, by the NCEP analysis the clinician can be reasonably confident that the patient has reached the goal. If only one reading is available, it is not certain whether the patient has reached the goal, because the error interval is larger. This wide range of error betrays a conceptual problem with using ~LDL-C in practice: because the cutoffs in the ATP-III guidelines are spaced in 30-mg/dL intervals (i.e., 70, 100, 130, and 160 mg/dL) (2), and the margin of error is at least 15 mg/dL with repeated sampling, the approximation may be poorly suited to pursue such fine distinctions in the clinical setting. The problem is further compounded by the reliance on solitary samples to assess efficacy.
The multitude of errors related to the approximation led the NCEP to recommend the development of direct measures of LDL-C, with the ultimate goal of phasing out the approximation (12). Several assays directly measure LDL-C without ultracentrifugation and are superior to the approximation when TGs exceed 400 mg/dL; unfortunately, below this level, these assays do not improve on the performance of the approximation (14).
Was this article helpful?