Bedside Monitoring of Blood β -Hydroxybutyrate Levels in the Management of Diabetic Ketoacidosis in Children

INTRODUCTION: Diabetic ketoacidosis (DKA) affects many children with type 1 diabetes. Insulin treatment of DKA is traditionally guided by changes in the blood glucose levels and blood gases, whereas beta-hydroxybutyrate (beta-OHB)--the main ketoacid causing acidosis--is rarely measured. The purpose of this study was to evaluate if bedside monitoring of blood beta-OHB levels can simplify management of DKA through elimination of superfluous laboratory monitoring. METHODS: Our emergency department treated 68 children with DKA using a standard protocol with monitoring of venous pH, partial pressure of CO(2) (pCO(2)), bicarbonate, glucose, blood urea nitrogen, and electrolytes (two to 10 time points per patient). Venous beta-OHB levels were measured using the Precision Xtra meter (MediSense/Abbott Diabetes Care, Abbott Park, IL) and, on duplicate batched serum samples, using a reference laboratory method (Cobas Mira Plus; Roche Diagnostics, Indianapolis, IN). Correlations between bedside meter beta-OHB and other parameters were evaluated in a series of general linear models with a time series covariance structure fit using spatial power law. RESULTS: The bedside meter beta-OHB levels were significantly correlated with pH (r = -0.63; P <0.0001), bicarbonate (r = -0.74; P <0.0001), and pCO(2) (r = -0.55; P <0.0001) at all points of measurement during the treatment (unadjusted Pearson correlations). The pH, bicarbonate, and pCO(2) were entered into separate time series analysis models with treatment duration as a measure of time. The results confirmed that bedside levels of beta-OHB correlated very closely with time-dependent levels of venous pH, bicarbonate, and pCO(2). Good agreement between the two methods of beta-OHB measurement (r = 0.92; P <0.0001) was confirmed using the Bland-Altman plot analysis. CONCLUSIONS: The Precision Xtra accurately measures blood beta-OHB levels, particularly at lower levels. While the initial measurement of pH and/or bicarbonates is warranted, real-time beta-OHB levels may replace repeat laboratory measurement of these parameters in the management of DKA. Future studies should evaluate safety and cost-effectiveness of such simplified DKA treatment protocol.