Abstract
Self-monitoring of blood glucose (SMBG) is a cornerstone of diabetes care. However, the effectiveness of any glucose-monitoring programme depends on the ability to integrate SMBG into a program of self-care and therapeutic decision-making. Because the accuracy of SMBG is instrument dependent, we analysed 45 heparinized whole blood specimens using four marketed portable glucose meters to evaluate whether their precision and accuracy would be efficient and safe for clinical use. All measurements were standardized and performed by a single expert health-care professional at the same clinical chemistry laboratory. Results were compared with those obtained on the same plasma samples by the hexokinase method on a secondary reference analyser and further analysed according to the error tolerance criteria and the current American Diabetes Association (ADA), Clinical Laboratory Improvement Amendments (CLIA), and National Committee for Clinical Laboratory Standards (NCCLS) guidelines. The within-run imprecision ranged from 2.2% to 3.2%. Passing and Bablok regression analysis yielded slope values from 0.93 to 1.07 and correlation coefficients between 0.994 and 0.998. When compared with the secondary reference analyser, mean variations were between -4.9% and 14.1%, fulfilling in three out of four cases the 5.5% current desirable analytical quality specifications for total error. Nevertheless, when considering the two standard deviations level of this bias, several results exceeded this limit. Although three out of four devices tested achieved or came closer to the NCCLS C30-A2, CLIA and error tolerance targets, none of them met the current analytical ADA thresholds. Despite the acceptable analytical performances, we demonstrated that standardization and harmonization of results in SMBG have not been fully achieved.