Table 2.
Overview of included studies.
| Author Year, Country | Aim | Participants | Measurement Method and Operationalisation of β-hydroxybutyrate (β-OHB) Levels and Hyperglycaemia | Results | Conclusion |
|---|---|---|---|---|---|
| Age (Mean ± Standard Deviation (SD) | |||||
| Gender, Diabetic Status | |||||
| Tantiwong et al. 2005 [6], Thailand | Comparison of capillary β-OHB measurements with urine ketone testing in diagnosing DKA. | 19 DKA (45.6 ± 16.95 years) | Capillary blood ketones meter (MediSense OptiumTM) β-OHB ≥ 3.0 mmol/L diagnostic of DKA Blood glucose ≥ 250 mg/dL (13.9 mmol/L) |
The β-OHB value of >3.0 mmol/L had a sensitivity of 90% and a specificity of 100%, whereas urine ketones had a sensitivity of 95% and a specificity of 100% in diagnosing DKA. | β-OHB and urine ketones are effective to confirm DKA in uncomplicated cases. |
| 42.1% type 1 diabetes | |||||
| 38.8% type 2 diabetes | |||||
| 19.1% unclassified type of diabetes | |||||
| Naunheim et al. 2006 [23], USA | Comparison of capillary β-OHB measurements with standard measures (arterial blood gases, anion gap and carbon dioxide) for accuracy in predicting DKA in the ED. | 160 (41 ± 15 years) | Capillary blood ketones meter (Precision Xtra, Abbott Laboraotries) No pre-operationalisation of β-OHB levels Blood glucose ≥ 250 mg/dL |
High correlation between β-OHB levels and anion gap (r = 0.66, p < 0.001), carbon dioxide (r = −0.69, p < 0.001), and blood glucose (r = 0.31, p < 0.001). | A β-OHB test can accurately identify patients with DKA, adding this at triage is likely to decrease delays in recognition of DKA in ED. |
| 46.8% female | |||||
| 103 Non-DKA (38 years) | The β-OHB value of >1.5 mmol/L had a sensitivity of 98% (95% CI: 91–100), a specificity of 85% (95% CI: 78–91), with a positive likelihood ratio of 6.7 (95% CI: 4.22–10.78), and negative likelihood ratio of 0.021 (95% CI: 0.003–0.144) of DKA. | ||||
| 57 DKA (38 years) | |||||
| Raised blood glucose | |||||
| Charles et al. 2007 [24], Singapore | Comparison of capillary β-OHB measurements with clinical diagnosis, venous bicarbonate levels and urine ketone testing in assessing DKA. | 111 (median 60 years) | Capillary blood ketones meter (MediSense OptiumTM) No pre-operationalisation of β-OHB levels Blood glucose ≥ 14.0 mmol/L |
The β-OHB value of 1.5 mmol/L compared with other diagnostic methods of DKA had a sensitivity of 100% (95% CI: 59.0–100), a specificity of 93.3% (95% CI: 86.6–97.2), a positive predictive value of 50% (95% CI: 23.0–77.0), and a negative predictive value of 100% (95% CI: 96.3–100). | β-OHB testing is a simple tool that can support early identification of DKA. |
| 45.5% female | |||||
| 4.5% type 1 diabetes | |||||
| The β-OHB value of 3.5 mmol/L, compared with other diagnostic measures of DKA had a sensitivity of 100% (95% CI: 59.0–100), a specificity of 100% (95% CI: 96.5–100), a positive predictive value of 100% (95% CI: 59.0–100) and a negative predictive value of 100% (95% CI: 96.5–100). | |||||
| 95.5% type 2 diabetes | |||||
| Taboulet et al. 2007 [14], France | Comparison of capillary β-OHB measurements with urine ketone testing in hyperglycaemic patients in the ED. | 529 (53 ± 17 years) | Capillary blood ketones meter (Optium, Abbott Laboratories) β-OHB ≥ 1.0 mmol/L Blood glucose ≥ 250 mg/dL (13.9 mmol/L) |
Urine ketones scored as +, ++ and +++ corresponded to medium capillary β-OHB levels of 0.5 mmol/L (IQR: 0.1–0.9), 0.7 mmol/L (IQR: 0.2–1.8) and 3 mmol/L (IQR: 1.4–5.2) respectively. | β-OHB is more accurate than urine ketones to confirm ketoacidosis in the ED. |
| 36% female | |||||
| Raised blood glucose | The β-OHB value of ≥ 3.0 mmol/L was associated with the relative risk of DKA or hospitalization 74 (95% CI: 44–88) and 2.9 (95% CI: 2.5–3) respectively than ketones scored as +++ ketones 31 (95% CI: 18–45) and 2 (95% CI: 1.7–2.1) respectively. | ||||
| Confirmed evaluation of blood ketones beyond 2.9 mmol/L is associated with higher risk of DKA than +++ urine ketones. | |||||
| Voulgari and Tentolouris 2010 [25], Greece | Comparison of serum with capillary β-OHB and urine ketones in detecting DKA in ED. | 400 Non-DKA (58.5 ± 9.7 years) | Capillaroy blood ketones (Precision-Xtra device Abbott Laboratories) β-OHB > 1.0 mmol/L Blood glucose >13.9 mmol/L |
Capillary and serum β-OHB were highly significantly correlated (r = 0.99, p < 0.001). | Capillary β-OHB of >3.0 mmol/L offers the best combination of sensitivity and specificity for diagnosis of DKA. |
| The capillary β-OHB value of >3.0 mmol/L had a sensitivity of 99.87% and specificity of 92.89% with positive predictive value of 92.89% for the diagnosis of DKA. Urine ketones had a sensitivity of 89.89% and specificity of 52.73% with positive predictive value of 41.87% for the diagnosis of DKA. | |||||
| 48% female | |||||
| 50 DKA (60.2 ± 8.2 years) | |||||
| 48% female | |||||
| Insulin-treated type 2 diabetes | |||||
| Arora et al. 2011 [26], USA | Comparison of capillary β-OHB measurement with urine ketone testing in assessment of DKA in the ED. | 462 Non-DKA (48, age range 40–57 years) | Capillary blood ketones meter (Precision Xtra, Abbott Laboraotries) β-OHB ≤ 1.5 mmol/L insignificant β-OHB > 1.5 mmol/L raised Blood glucose ≥ 250 mg/dL |
Urine dipstick sensitivity of 98.1% (95% CI: 90.1–100), a specificity of 35.1% (30.7–39.6), a positive predictive value of 15% (11.5–19.2) and a negative predictive value of 99.4% (96.6–100) for DKA. | β-OHB is more specific than urine ketone testing, therefore β-OHB testing could significantly reduce unnecessary DKA work-ups in the ED. |
| 35.3% female | |||||
| 44.5% insulin-requiring | The β-OHB value of >1.5 mmol/L had a sensitivity of 98.1% (95% CI: 90.1–100), a specificity of 78.6% (74.5–82.2), a positive predictive value of 34.9% (27.3–43) and a negative predictive value of 99.7% (95% CI: 98.5–100) of DKA. | ||||
| 54 DKA (41, age range 30–48 years) | |||||
| 27.8% female | |||||
| 56.6% insulin requiring | |||||
| Kinsella et al. 2012 [9], Australia | The use of capillary β-OHB measurement in early assessment of hyperglycaemia in the Emergency Department (ED) | 72 (60.97 ± 22.07 years) | Capillary blood ketones meter (Optium Xceed—Abbott Laboratories, Illinois) β-OHB ≥ 1.0 mmol/L elevated Blood glucose ≥ 10.0 mmol/L |
Modest correlation between capillary β-OHB levels and blood glucose levels on presentation (r = 0.29, p = 0.02) and over the ED stay (r = 0.21, p = 0.01) | β-OHB testing can support early identification of DKA in the ED. |
| 56.9% female | |||||
| Insulin-requiring diabetics | |||||
| Rashid et al. 2013 [27], Bangladesh | Comparison of capillary β-OHB levels with serum electrolytes, urea and creatinine, plasma glucose, and urine ketone testing in assessing and managing DKA. | 121 (39 ± 15 years) | Biosensor method (Medisense) β-OHB ≥ 3.0 mmol/L diagnostic of DKA Blood glucose ≥ 12.0 mmol/L |
The relative frequencies of DKA, using urinary ketone and capillary β-OHB were 15.6% and 13.9%. Using capillary β-OHB as the reference method, the sensitivity of urinary ketones were 32.6% and the specificity was 93.7%, the positive predictive value, and negative predictive value of urine ketones against blood ketones were 73.68% and 71.84% respectively. | Urine ketone testing has severe limitations in assessing for DKA in patients with type 2 diabetes, with an error of 25%–30%. |
| 36% female | |||||
| 100% type 2 diabetes | |||||
| Coetzee et al. 2015 [28], South Africa | Comparison of capillary β-OHB measurement and the gold-standard manual enzymatic method in assessment of DKA | 41 (33, age range 17–52 years) | Capillary blood ketones meter (Optium Xceed, Medisense/Abbott) β-OHB < 1.0 mmol/L insignificant β-OHB ≥ 3.0 mmol/L diagnostic of DKA Blood glucose > 13.9 mmol/L |
High correlation between capillary β-OHB levels and the manual enzymatic method (r = 0.95) | β-OHB capillary testing is as accurate and precise as the manual enzymatic method in confirming DKA. |
| 58.5% female | Capillary β-OHB levels when compared to the manual enzymatic method demonstrated a sensitivity of 100% and a specificity of 89.5% for diagnosing DKA and a sensitivity of 100% and a specificity of 87.3% for excluding DKA. | ||||
| 70.7% type 1 diabetes | |||||
| 29.3% type 2 diabetes |