Accuracy of blood lipids determination on express analyser CARDIOCHEK PA

Abstract

The aim of the study is to compare the accuracy of results of determination of the blood lipid spectrum in the capillary blood samples analyzed on a portable (CardioChek PA) and stationary (Roche Cobas c 702) biochemical analysers.
Methods. The study was conducted on 36 male and female volunteers aged from 18 to 65 years. The biomaterial was taken by highly qualified nurse. At first there was rapid testing and then capillary blood sampling from the same puncture into the standard micro tubes with clotting activator and separator gel for studies on stationary analyzer was performed. The paired difference in results obtained by different methods for each indicator was calculated in absolute and relative units according to the results of the statistical bootstrap and regression analysis. In addition, the comparison of average bias for each analysis was compared with the analytical variation (CVa) according to the formula reference change values (RCV).
Results. The following deviations between the results obtained by usage of point-of-care testing (CardioChek PA) and Roche Cobas c 702 were demonstrated: total cholesterol (TC) – the average offset based on the bootstrap estimates -3.70%, R2 = 0.6515; cholesterol high density lipoprotein (HDL) +9.16%, R2 = 0.9198; triglycerides (Tg) +8.99%, R2 = 0.8912; cholesterol low-density lipoprotein calculated according to the formula of Friedwald (LDL) -13.35%, R2 = 0.7176 and the result of direct measurement -23.96%, R2 = 0.6943. Due to difference of capillary and venous blood results, the offset for the cholesterol average was -1.08% and fits into the limit of CVa for the method. However, most other couples of the bias taking into account their 95% confidence intervals (CI) do not fit in the CVa for the considered methods.
Conclusion. The results of TC assays, obtained by point-of-care testing, are comparable with the results obtained on a stationary analyzer, and can be used for a preliminary assessment of hypercholesterolemia to determine the cardiovascular risk category. TG results can also be used for pre-screening of hypertriglyceridemia. Systematic biases of other studied biomarkers do not allow the use of rapid testing for more accurate diagnosis of dyslipidemia and monitoring of the effectiveness of lipid-lowering therapy.