Clinical significance of an elevated on‐admission beta‐hydroxybutyrate in acutely ill adult patients without diabetes

Abstract

ObjectiveTo determine the relationship between point‐of‐care β‐hydroxybutyrate (BHB) concentration and outcomes in adult patients without diabetes admitted through ED.MethodsThis was a prospective study from 10 March to 2 July 2021. Admitted patients without diabetes had capillary BHB sampled in ED. Outcomes of length‐of‐stay (LOS), composite mortality/ICU admission rates and clinical severity scores (Quick Sepsis Organ Failure Assessment score/National Early Warning Score [qSOFA/NEWS]) were measured. BHB was assessed as a continuous variable and between those with BHB above and equal to 1.0 mmol/L and those below 1.0 mmol/L.ResultsA total of 311 patients were included from 2377 admissions. Median length‐of‐stay was 4.1 days (IQR 2.1–9.8), 18 (5.8%) died and 37 (11.8%) were admitted to ICU. Median BHB was 0.2 mmol/L (IQR 0.1–0.4). Twenty‐five patients had BHB ≥1.0 mmol/L and five were >3.0 mmol/L. There was no significant difference in median LOS for patients with BHB ≥1.0 mmol/L compared to non‐ketotic patients, 5.3 days (IQR 2.2–7.5) versus 4.1 days, respectively (IQR 2.0–9.8) (P = 0.69). BHB did not correlate with LOS (Spearman ρ = 0.116, 95% confidence interval: 0.006–0.223). qSOFA and NEWS also did not differ between these cohorts. For those 25 patients with BHB ≥1.0 mmol/L, an infective/inflammatory diagnosis was present in 11 (44%), at least 2 days of fasting in 10 (40%) and ethanol intake >40 g within 48 h in 4 (16%).ConclusionsRoutine BHB measurement in patients without diabetes does not add to clinical bedside assessment and use should be limited to when required to confirm a clinical impression.