Evaluation of Lipid Profile Measurements Obtained From the Cholestech L??D??X Analyzer

Abstract

The Cholestech L·D·X analyzer has the capability of performing a lipid profile in approximately 5 minutes. The purpose of this study was to determine analytical performance capability of the L·D·X to perform lipid profile measurements. Forty subjects gave two finger capillary samples and one venous serum which were analyzed in duplicate by two technicians, on two different L·D·X analyzers. A local pathology laboratory was used as the standard for accuracy comparisons. Two controls with known values were provided by the manufacturer to assess within-day precision. Day-to-day precision was determined by analyzing high (240 mg/dL−1 total cholesterol (TC) and 60 mg/dL−1 high-density lipoprotein [HDL]) and low (200 mg/dL−1 TC and 35 mg/dL−1 HDL) human serum samples on 10 different days. Analysis of variance procedures revealed no significant differences between the two technicians or the two analyzers, nor among the fingerstick, venous serum, or reference measures for any of the analytes. The total error of measurement of TC, triglycerides, and HDL measurements were 6.3 to 16.2%, 14.8 to 30.8%, and 11.5 to 26.8%, respectively. In comparison to the National Cholesterol Education Program (NCEP) desirable, borderline-high, and high classifications obtained from the reference laboratory the L·D·X TC results were correctly classified in 92.5% of the cases. L·D·X triglyceride classifications of desirable, borderline hypertyglyceridemia, and distinct hypertriglyceridemia matched 100% with the classifications obtained from the reference laboratory. Although the Cholestech L·D·X analyzer did not consistently meet NCEP standards for acceptable total measurement error of TC, HDL, and triglyceride analyses, it seems capable of providing reasonable lipid profile measures in both a screening setting and a clinician office where the goal is correct classification of patients' results.