Evaluation of the point-of-care devices KetoSure^TM and StatStrip Express^® blood ketone tests using β-hydroxybutyrate spiked samples

Abstract

Measurement of β-hydroxybutyrate (BHB) in blood is used clinically as a measure of ketosis when diabetic ketoacidosis is expected. With the introduction of point-of-care testing (POCT) for blood-BHB, nonproductive time is reduced to a minimum in a potential critical situation; however, studies have observed inferior quality of POCT-BHB. Recently, the POCT device KetoSure (Roche) has been introduced to the clinic. In this study, we evaluated the imprecision and linearity of KetoSure and compared this to the established StatStrip Express (Nova Biomedical) based on spiked full blood samples. We found comparable imprecision for KetoSure and StatStrip Express. However, linearity was only observed in the lower part of the measuring range for both devices. In a method comparison, higher values of BHB were measured by KetoSure than by an enzymatic endpoint spectrophotometric reference method (mean bias: 21% (95% confidence interval (CI): 4%–37%)). Conversely, StatStrip Express returned lower values of BHB (mean bias: −16% (95% CI: −38%–7%), while the widely applied POCT device FreeStyle Precision Neo (Abbott) returned values equivalent with the reference method (mean bias: 5% (95% CI: −14%–24%). In samples with concentrations of BHB above the measuring range, the POCT devices could be provoked to return falsely low results. In conclusion, the quality of KetoSure is in line with other established POCT devices; however, the KetoSure measures higher concentrations than other POCT devices. As, linearity only was observed in the lower part of the measuring range and as falsely low measures could be provoked, we advise users to interpret results with precaution.

Keywords:

• Diabetic ketoacidosis
• point-of-care systems
• blood
• ketone bodies
• 3-hydroxybutyric acid
• metabolic diseases

Acknowledgments

The authors thank the laboratory technologists at the Department of Biochemistry and Immunology, Kolding Sygehus and the Department of Clinical Biochemistry, Viborg Regional Hospital for excellent technical assistance.

Disclosure statement

The authors report there are no competing interests to declare.