Interlaboratory variability in serum lipid measurements

Identification of elevated blood cholesterol has become a priority in the effort to reduce coronary heart disease. Previous data indicate that considerable laboratory variability in lipid testing exists in the United States. To determine the interlaboratory variability of serum lipid measurements in the Little Rock area, split serum samples from three subjects were sent to ten area laboratories. The average coefficient of variation among the ten local laboratories was 3.20% for total cholesterol, 9.46% for HDL-cholesterol, 7.73% for triglycerides, and 5.95% for calculated LDL-cholesterol. The results indicate that clinicians cannot assume that adequate standardization for lipid determinations exists for all laboratories. Without accuracy and precision data for the specific laboratory, caution should be used in the strict application of guidelines for classification and treatment of patients with abnormal serum-lipid measurements.