The aim of our study was to determine the utility of point-of-care blood ketone testing in diabetic patients presenting to the emergency department. In this prospective, observational clinical study, patients with known or newly diagnosed diabetes mellitus presenting to our tertiary care university emergency department with any nontrauma related medical complaint and a high fingerstick glucose (> or =200 mg/dL) were eligible for inclusion. Capillary blood beta-hydroxybutyrate (beta-HBA), venous blood beta-HBA level, venous blood glucose level, arterial blood gas analysis, and urine ketone dipstickstick were measured in each patient as primary outcome measures. Of the 479 diabetic patients presenting during the study period, a total of 139 diabetic patients with high capillary blood glucose level (> or =200 mg/dL) and a positive capillary blood beta-HBA (> or =0.1 mmol/L) were included in the study. Hyperketonemia (> or =0.42 mmol/L) was found in 48 of these patients by Sigma Diagnostics reference testing (diabetic ketosis in 35%). The calculated blood pH was less than 7.3 in 18 of these 48 patients (ketoacidosis in 31%). Capillary and venous blood beta-hydroxybutyrate levels were not statistically different from each other (P = 0.824). There was a positive correlation between capillary and venous blood beta-HBA levels (r = 0.488, P < 0.001). The sensitivity and specificity of urine ketone dipstick testing and capillary blood ketone testing in determining diabetic ketoacidosis were 66% and 78%, and 72% and 82%; and in determining hyperketonemia (both in diabetic ketosis and diabetic ketoacidosis) were 82% and 54%, and 91% and 56%, respectively. A rapid, bedside capillary blood ketone test for beta-HBA can accurately measure blood concentrations of beta-HBA in diabetic patients in an emergency department setting. This device can be used as a reliable diagnostic test to detect emergency metabolic problems in diabetic patients, such as diabetic ketosis or ketoacidosis.