Should we assume accuracy of point of care glucose meters? An observation from a tertiary health care centre

Abstract

Background: Self-monitoring of blood glucose is important in the management of patients with diabetes mellitus in the community as well as in the hospital. It has been used for calculation of insulin doses of individuals with dysglycaemia and monitoring of glucose control. Errors in the measurement of the blood glucose can lead poor management of a patient. There is therefore the need to ensure standardization of these meters in order to achieve accuracy and precision.

Objectives: To evaluate the precision and accuracy of four glucose meters commonly used for self-monitoring of blood glucose (SMBG) in a tertiary health care Centre compared to the reference laboratory method.

Materials and Methods: We analyzed blood glucose samples of 55 diabetic patients who came to diabetes clinic using 4 different glucose meters (Accucheck active, Novo max extra, One touch Ultra 2 and On call plus (accoson)). Capillary and Venous blood samples were taken simultaneously from each patient for analysis using four blood glucose meters and laboratory reference method respectively. The laboratory value was used as a tool for comparison. The accuracy and precision were evaluated by the ISO and ADA criteria. The results obtained were analysed using Bland Altman graphs, correlation coefficients, scatter plots and Clarke's error grid analysis.

Results: We observed good correlation between two glucose meters (Accucheck^TM and Novomax^TM) and laboratory analyzed values. Among the glucose meters Accucheck^TM, Novomax^TM, One touch Ultra^TM and On call plus^TM, the correlation coefficient was 0.97, 0.96, 0.88 and 0.69 respectively. The degree of agreement of the laboratory method and the Accucheck^TM, Novomax^TM, One touch ultra^TM and On call plus^TM glucose meters was 89.09, 80.00, 76.20 and 71.32% respectively. Accucheck and Novomax^TM were within ±20% accuracy (14.5% and 16.1%).

Conclusion: There is a need for adequate and appropriate evaluation of all glucose meters in our setting before that we deploy them for use. None of glucose meters met the ISO target. Only one glucose meter (Accucheck^TM) met the ADA guideline for accuracy.

Keywords: Accuracy; glucose meters; diabetes mellitus; self-monitoring of blood glucose