Lipid profile in ambulatory patients using 3 point-of-care devices and comparison with reference methods

Plasma lipid measurements are widely used for the screening and control of dyslipemias. Point-of-care devices could facilitate these analyses, but results should be comparable to those of reference methods. Analytical performance of total cholesterol (TC) and triglyceride (TG) levels has been evaluated in 3 different point-of-care systems: CardioChek (Polymer Technology Systems, Indianapolis, Ind), Accutrend GCT (existing version; Roche Diagnostics, Basel, Switzerland), and Accutrend Plus (new version; Roche Diagnostics). As CardioChek allows high- (HDL-C) measurement and low-density lipoprotein cholesterol (LDL-C) calculation, both magnitudes were also evaluated. One hundred one samples for TC values and 102 for TG values were analyzed using the Accutrend devices and 109 using the CardioChek. For TC samples, total error based on control material was 8.5% for Accutrend Plus, 8.9% for Accutrend GCT, and 25.1% for CardioChek. Biases at 5.17 mmol/L based on patient samples were -7.5%, -6.0%, and 20.0%, respectively. For TG samples, total error was 10.3% for Accutrend Plus, 6.9% for Accutrend GCT, and 14% for CardioChek, whereas biases at 1.69 mmol/L were 5.4%, 11.5%, and -8.5%, respectively. For HDL-C samples, total error of CardioChek was 18.4%, and bias at 1.16 mmol/L was -3.4%. For LDL-C samples, bias for CardioChek was 32.5% at 2.59 mmol/L. All 3 devices correctly classified a high percentage of patients with dyslipemia. However, CardioChek classified 48% of patients with normolipemia as having hypercholesterolemia. In conclusion, although CardioChek could be valid for TG and HDL-C value determinations, it clearly overestimates TC and LDL-C values. On the other hand, Accutrend devices are useful for TC and TG level measurements and may be suitable for screening and dyslipemia control. © 2009 by Lippincott Williams & Wilkins.