Abstract Background Beta-Hydroxybutyrate (BHB) is a blood-based biomarker of ketosis and plays an important role in the assessment of life-threatening metabolic acidotic states, including diabetic ketoacidosis (DKA). Traditionally, DKA has been informed by the semi-quantitative detection of urine ketones, which is performed as part of routine urinalysis (UA) testing, and this testing does not detect BHB. Currently, professional guidelines recommend BHB testing for DKA diagnosis and monitoring, as BHB is the primary ketone body produced in ketotic states and demonstrates improved analytical sensitivity and specificity. The aim of this study was to evaluate BHB utilization, with a focus on its comparison to contemporaneous urine ketones, within our healthcare system. Methods Within the Johns Hopkins Hospital Core Laboratories, serum BHB was measured using a Cardinal LiquiColor® reagent pack on a Roche cobas® c502 automated analyzer (reference interval: 0.00-0.27 mmol/L). Urine ketones were measured using Siemens Mutlistix® 10SG reagent test strips on a Clinitek Status®+ Connect System and a Clinitek Novus® Automated Urine Chemistry Analyzer. Data on ordering practices from November 2023 to October 2024 were obtained using Epic SlicerDicer, a data analytics tool within our electronic health record system. To examine the relationship between BHB and urine ketones, data from contemporaneous BHB and UA orders were extracted using Epic Beaker reports. Contemporaneous ordering was defined as serum BHB and UA specimens collected within one hour of each other. Serum BHB results 0.27 mmol/L were considered abnormal. Urine ketones were classified as normal if the result was negative or trace, and abnormal if the result was small, moderate, or large. An in-depth chart review was conducted in cases where urine ketones and BHB results were discordant. Results Over a one-year evaluable period, 3,235 BHB tests were performed from 2,011 unique patients across our health system, with an average volume 270 ± 26 tests per month. BHB was most frequently ordered in older adults, with 42.3% serum BHB orders coming from patients aged 60 or older. Serum BHB was most frequently ordered from patients receiving care in surgical intensive care units (22.3%; 722/3,235) and emergency department (20.8%; 673/3,235) settings. Further, 83% (1,671/2,011) of patients with BHB testing ordered had clinical symptoms (e.g., hyperglycemia, acidosis, dehydration) associated with DKA. While urine ketones were frequently ordered with BHB, 13.9% (280/2,011) of patients had BHB and urine ketones ordered within one hour of each other (321 specimens). Of these contemporaneously collected specimens, 76% (245/321) agreed while 24% (76/321) were discordant. A chart review of the discordant results revealed a higher frequency of elevated BHB measurements with negative or trace ketones (89.5%; 68/76). Conclusion BHB is utilized across our health system, particularly in settings in which metabolic acidosis is commonly on the differential. Although BHB and urine ketones were not frequently ordered together, in situations when they were, nearly one quarter yielded disparate results, with serum BHB concentrations more consistent with clinical presentation. These results suggest the need for increased utilization of BHB during initial clinical workups.
Harsha et al. (Wed,) studied this question.