Concordance between screening method (Accutrend Plus ®) and traditional method for determining cholesterol, triglycerides and glucose screening population

Abstract

Objective: To evaluate the concordance of measurements of glucose, total cholesterol and triglycerides using capillary fresh whole blood sample with those using venous blood and the reference method in the laboratory. Method: Cross-sectional study carried out on samples of capillary and venous blood collected fasting adults of both sexes, patients. Statistical analysis: paired Wilcoxon Test for comparing the equality of medians between venous and capillary values. Concordance coefficients calculated using Lin Method, and Bland- Altman Graphics were built. MedCalc software 12.6.1.0 and R were used for exploratory analysis. Results: Observed concordance coefficients were 0.63 (95%CI: 0.53 to 0.71 ) for glucose, 0.77 (95%CI: 0.69 to 0.83 ) for total cholesterol and 0.95 (95%CI : 0.93 to 0.97 ) for triglycerides, revealing low concordance for the first two measures and good agreement for the latter. In analyzes according to levels defined by the authors, there were differences of magnitude, with lower values in capillary blood for glucose and total cholesterol. Triglycerides showed no statistically significant difference for values<150mg/dl and ≥300mg/dl. Conclusions: With the bias observed for glucose and total cholesterol, measured by Accutrend®, if it was used in population surveys, a large number of people would not be directed to health services for diagnosis.