Bahira Shahim1, Barbro Kjellström1, Viveca Gyberg1,2, Catriona Jennings3, Stina Smetana1, Lars Rydén1. 1. 1 Cardiology Unit, Department of Medicine, Karolinska Institutet , Stockholm, Sweden . 2. 2 Department of Neurobiology, Care Sciences and Society, Centre for Family Medicine, Karolinska Institutet , Huddinge, Sweden . 3. 3 Faculty of Medicine, NHLI Imperial College London , Hammersmith Campus, London, United Kingdom .
Abstract
BACKGROUND: Point-of-care equipment for measuring glucose saves time and costs for both patients and professionals and minimizes preanalytic errors when screening for or managing dysglycemia, that is, impaired glucose tolerance and type 2 diabetes. The accuracy of such devices has, however, been questioned compared with analyses at an accredited hospital laboratory. OBJECTIVE: To investigate the agreement between glucose measurements made by the point-of-care HemoCue® Glucose 201 RT System (HemoCue, Ängelholm, Sweden) and local hospital laboratories. MATERIAL: Patients with established coronary artery disease (CAD) recruited in Sweden and the United Kingdom within the auspices of the European Action on Secondary and primary Prevention by Intervention to Reduce Events (EUROASPIRE) V survey (n = 87; 18-80 years) with or without previously known dysglycemia were investigated. Plasma glucose values collected in the fasting state (n = 85) and 60 (n = 57) and 120 (n = 72) min after a glucose load were analyzed both using HemoCue monitors and local hospital laboratories. The two measurement techniques were compared using a bias plot according to Bland-Altman, the surveillance error grid, and Spearman correlation test. RESULTS: The bias plot method showed small differences between the HemoCue and local hospital laboratory methods, the HemoCue and central hospital laboratory, and the local hospital laboratories and the central hospital laboratory. In the surveillance error grid, 98.6% of the values were in the deep green zone, indicating no risk and the remaining values (1.4%) were within the light green zone, indicating "slight lower risk." CONCLUSION: The HemoCue point-of-care system is accurate for dysglycemia screening in patients with CAD.
BACKGROUND: Point-of-care equipment for measuring glucose saves time and costs for both patients and professionals and minimizes preanalytic errors when screening for or managing dysglycemia, that is, impaired glucose tolerance and type 2 diabetes. The accuracy of such devices has, however, been questioned compared with analyses at an accredited hospital laboratory. OBJECTIVE: To investigate the agreement between glucose measurements made by the point-of-care HemoCue® Glucose 201 RT System (HemoCue, Ängelholm, Sweden) and local hospital laboratories. MATERIAL: Patients with established coronary artery disease (CAD) recruited in Sweden and the United Kingdom within the auspices of the European Action on Secondary and primary Prevention by Intervention to Reduce Events (EUROASPIRE) V survey (n = 87; 18-80 years) with or without previously known dysglycemia were investigated. Plasma glucose values collected in the fasting state (n = 85) and 60 (n = 57) and 120 (n = 72) min after a glucose load were analyzed both using HemoCue monitors and local hospital laboratories. The two measurement techniques were compared using a bias plot according to Bland-Altman, the surveillance error grid, and Spearman correlation test. RESULTS: The bias plot method showed small differences between the HemoCue and local hospital laboratory methods, the HemoCue and central hospital laboratory, and the local hospital laboratories and the central hospital laboratory. In the surveillance error grid, 98.6% of the values were in the deep green zone, indicating no risk and the remaining values (1.4%) were within the light green zone, indicating "slight lower risk." CONCLUSION: The HemoCue point-of-care system is accurate for dysglycemia screening in patients with CAD.
Authors: Giulia Ferrannini; Dirk De Bacquer; Pieter Vynckier; Guy De Backer; Viveca Gyberg; Kornelia Kotseva; Linda Mellbin; Anna Norhammar; Jaakko Tuomilehto; David Wood; Lars Rydén Journal: Cardiovasc Diabetol Date: 2021-02-11 Impact factor: 9.951