RESUMO
BACKGROUND: within hospitals, bedside blood glucose measurements are widely used for investigating suspected hyper- or hypoglycemia, monitoring diabetes, and adjusting glucose-lowering medication. Numerous point-of-care glucose meters are available, but for large hospitals using the same meter in all settings has practical and economic advantages. This investigation sought to identify a meter that was accurate, precise, and free from interferences, making it suitable for use across all ages and diseases. METHODS: lithium-heparinized whole blood was analyzed, under various conditions, on the HemoCue Glucose 201 (Hemocue AB, Ängelhom, Sweden), Accu-Chek Performa (Roche Diagnostics, Basel Switzerland) (using the newly reformulated maltose-insensitive strips), and Optium (Abbott Diabetes, Alameda, CA, USA) glucose meters and compared with plasma glucose measurements on the Vitros 5,1 FS analyzer (Ortho Clinical Diagnostics, Neckargemund, Germany). RESULTS: biases of 3.2%, -5.8%, and -8% were found with Accu-Chek, Optium, and HemoCue, respectively. Within-run imprecision was 2.5-5.8%. Between-run imprecision was 3.1-6.8%, with the Accu-Chek performing best. All meters measured to 1.3 mmol/L with acceptable precision (coefficient of variation, <14%). Varying hematocrits between 0.2 and 0.7 L/L affected results of all meters. Interference at clinically relevant concentrations of galactose and possibly maltose was demonstrated with the Accu-Chek. CONCLUSIONS: all three meters are sufficiently accurate and precise for in-hospital use. Because of possible interference by galactosemia or high hematocrit, the Accu-Chek is not the safest option for neonatal use. Patients receiving high doses of maltose in therapeutic infusions may still be at risk of being falsely classified as euglycemic or hyperglycemic with the reformulated Accu-Chek strips, and clinical evaluation of these strips in patients receiving maltose-containing infusions is urgently needed.