Search tips
Search criteria 


Logo of canvetjReference to the Publisher site.Journal Web siteJournal Web siteHow to Submit
Can Vet J. 2010 September; 51(9): 1000–1002.
PMCID: PMC2920155

Language: English | French

Use of a point-of-care beta-hydroxybutyrate sensor for detection of ketonemia in dogs


The urine test strip is the most common test used to detect ketones in veterinary patients, but it can underestimate the degree of ketonuria and hence, ketonemia. Additionally, adequate urine samples for analysis may be difficult to obtain from dehydrated animals. The standard method used to detect and monitor ketonemia in human medicine is measurement of serum or whole blood beta-hydroxybutyrate (βHOB). A point-of-care (POC) analyzer has been validated for this purpose in humans. This study compared the accuracy of the POC device to an enzymatic reaction laboratory method for measurement of βHOB in dogs. Although the POC sensor tended to overestimate βHOB concentrations, there was good correlation (R2 = 0.96) and good agreement between the 2 methods with a bias +/− precision of 0.0860 +/− 0.3410 mmol/L βHOB. The POC βHOB sensor can be useful for assessing ketonemia in dogs.


Usage d’un capteur au bêta-hydroxybutyrate pour la détection de la cétonémie chez les chiens. La bandelette réactive d’urine est le test le plus communément utilisé pour détecter les cétones chez les patients vétérinaires, mais elle peut sous-estimer le degré de cétonémie et, par conséquent, la cétonémie. De plus, il peut être difficile de se procurer des échantillons d’urine adéquats auprès d’animaux déshydratés. La méthode standard utilisée pour détecter et surveiller la cétonémie en médecine humaine est la mesure du sérum ou le bêta-hydroxybutrate de sang total (βHOB). Un analyseur au point de service a été validé à cette fin chez les humains. Cette étude a comparé l’exactitude du dispositif au point de service à une méthode de réaction enzymatique en laboratoire pour l’évaluation du βHOB chez les chiens. Même si le capteur au point de service avait tendance à surestimer les concentrations de βHOB, il y avait une bonne corrélation (R2 = 0,96) et une bonne concordance entre les 2 méthodes avec un biais +/− de précision de 0,0860 +/− 0,3410 mmol/L βHOB. Le capteur de βHOB au point de service peut être utile pour l’évaluation de la cétonémie chez les chiens.

(Traduit par Isabelle Vallières)


Urine test strips have traditionally been used for detecting and monitoring ketonemia in dogs. The nitroprusside reaction with these test strips, however, only measures one of the major ketoacids, acetoacetate, in a semi-quantitative manner (1). The other major ketoacid, beta-hydroxybutyrate (βHOB) (2), does not react with the urine test strips (3). Use of test strips, therefore, could underestimate the degree of ketonuria. Additionally, it may be difficult to obtain a urine sample from small or dehydrated patients.

The current recommendation of the American Diabetic Association for detecting ketonemia in humans is the measurement of serum βHOB levels (4). A point-of-care (POC) sensor has been validated for this use (57). Some human intensive care units, especially with pediatric patients, prefer this sensor for monitoring ketonemia instead of the reference laboratory analyzer because of its accuracy, small sample requirement, and rapid results (810). Additionally, measurement of POC βHOB levels may also be used to monitor resolution of metabolic acidosis caused by ketonemia (11).

A previously published report on the use of this POC βHOB sensor in a small population of 17 dogs and 3 cats showed good correlation between sensor values and reference laboratory values (12). The purpose of the study reported here was to compare the use of this POC βHOB sensor with a reference laboratory analyzer method for detection of βHOB in a larger population of dogs.

Materials and methods


Data from 46 dogs presenting to a large private veterinary hospital were included in the study. All blood samples were collected after obtaining owner consent. Patients were selected at random from the general hospital population and included diabetic ketoacidotic (DKA) patients (n = 8), stable diabetics (n = 11), healthy patients presented for wellness examination or vaccination (n = 14), and patients presented for illness unrelated to diabetes (n = 13). These groups were included to provide a wide range of βHOB concentrations.

Jugular venous blood samples were collected into a syringe via direct venipuncture and analyzed within 24 h of collection. For the laboratory analyzer βHOB levels, blood was immediately transferred to serum separator tubes and centrifuged within 30 min. These samples were submitted to the reference laboratory and analysis for βHOB was performed by an automated analyzer with the use of a standard liquid reagent (Dade Dimension Chemistry Analyzer, Beta-hydroxybutyrate Kit; Pointe Scientific, Chicago, Illinois, USA). This method is based on the oxidation of βHOB to acetoacetate by the enzyme β-hydroxybutyrate dehydrogenase. During this reaction, an equimolar amount of nicotinamide adenine dinucleotide (NAD+) is reduced to nicotine adenine dinucleotide (NADH). The NADH absorbs light at 340 nm, and the increase in absorbance is directly proportional to the βHOB concentration in the sample. The POC βHOB sensor (Medisense Presision Xtra; Abbott Laboratories, Bedford, Massachusetts, USA) analysis was performed immediately after venipuncture with 5.0 μL of whole blood from the same sample. The POC sensor uses the same reaction as the serum chemistry analyzer, but an electrochemical strip is used instead of spectrophotometry to detect βHOB concentration (7). Results are displayed in 30 s.

Statistical analysis

The POC sensor measures βHOB values between 0.00 mmol/L and 6.0 mmol/L. Values > 6 mmol/L are displayed as “HI” by the sensor. Therefore, data from patients with reference laboratory analyzer values > 6 mmol/L (n = 4) or POC sensor βHOB values recorded as “HI” (n = 1) were not included in the evaluation.

Agreement between the POC βHOB meter and the laboratory analyzer was determined using a modified Bland-Altman method and a commercially available software program (Analyse-It for Microsoft Excel-Method Evaluation Edition; Analyze-It Software, Leeds, United Kingdom). The Bland-Altman method was chosen because it has been used to evaluate this sensor in human medicine (57) and it has also been used to evaluate different methods of measuring other parameters in veterinary medicine (1315). The modified method was chosen because it seemed to be easier to understand. In the traditional Bland-Altman method (16), mean difference (bias) and standard of the differences are calculated by subtracting the value obtained by the new technique from the value obtained by the gold standard (gold standard-new). In the modified method (17), mean difference is calculated by subtracting the gold standard from the new technique (new-gold standard). With the modified method, a positive bias results if the new technique overestimates the gold standard, while a negative bias results if the new technique underestimates the gold standard. Limits of agreement were calculated from the bias and precision (standard deviation of the differences), and the correlation coefficient (R) was calculated using Passing-Bablok regression analysis with a commercially available software program (Med Calc Software bvba Version 10.4; Broekstraat 52, 9030 Mariakerke, Belgium).


The modified Bland-Altman analysis showed that the 95% limits of agreement between the 2 methods ranged from − 0.7560 to 0.9280 mmol/L βHOB. The POC sensor tended to overestimate the βHOB value when compared with the laboratory enzymatic method (positive bias) (Figure 1).

Figure 1
Modified Bland-Altman plot of the differences (POC − laboratory method, mmol/L βHOB) against the mean (POC + laboratory/2, mmol/L βHOB) in 46 dogs. Limits of agreement and confidence intervals are shown. There is close approximation ...

Good linear correlation was observed between the 2 techniques with an R2 value of 0.96 (Figure 2). The 95% confidence intervals (95% CI) for the intercept were −0.1100 to − 0.03957. The 95% CI for the slope were 0.6957 to 1.100.

Figure 2
Passing-Bablok regression plot of βHOB (mmol/L) measurements using the POC sensor and a laboratory method in 46 dogs. (βHOB — beta-hydroxybutyrate, POC — point-of-care). Slope = 0.9471. Intercept = −0.08471. The ...

Table 1 illustrates bias, precision, confidence intervals, and correlation of the POC βHOB sensor compared with the laboratory enzymatic method. βHOB concentration was overestimated by the POC sensor in 67.4% of the measurements, underestimated in 21.7%, and was the same in 10.8%.

Table 1
Bias, precision, limits of agreement, and correlation coefficient of point-of-care sensor and laboratory enzymatic method of measuring beta-hydroxybutyrate


This study demonstrates that the POC βHOB sensor compares favorably with the laboratory method for measuring βHOB concentration. There is good correlation between the 2 methods, making the POC sensor acceptable for monitoring trends in βHOB concentrations. The positive bias displayed by the POC sensor is small. In fact, as a quick screening test, the overestimation of the βHOB could be considered beneficial because, if POC βHOB concentration is normal, it is unlikely the patient has elevated serum ketones.

This sensor is easy to use; it functions similarly to the hand-held glucose sensors already in widespread use in veterinary hospitals and homes of many owners of diabetic pets. In human medicine, βHOB levels, in conjunction with blood glucose readings, are used by diabetics at home to determine if medical intervention for impending DKA might be necessary. This results in rapid attention to a potentially serious state of DKA and may avoid costly therapy and long hospital stays. The sensor has also been used in the pediatric intensive care unit to monitor treatment of DKA. This sensor is advantageous compared to the laboratory analyzer tests because it costs about 80% less and a much smaller sample size is required.

One veterinary study (18) has shown that a βHOB value of 3.8 mmol/L is most consistent with DKA in the dog, while normal ketone levels range from 0.00 to 0.32 mmol/L (19). If values for acceptable levels of βHOB in diabetic dogs were known, then this sensor could have the same applications for veterinary patients as it does for humans. More investigation into these potential uses is needed.


The authors thank Drs. William Cox and G. Neal Mauldin for statistical analysis, Dr. Nigel Caulkett for statistical advice and for critical review of the manuscript, and Shelly Myshaniuk and Thierry Locati for help in collecting and compiling the data. CVJ


Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (gro.vmca-amvc@nothguorbh) for additional copies or permission to use this material elsewhere.


1. Osborne CA, Stevens JB. Urinalysis: A Clinical Guide to Compassionate Care. 1st ed. Shawnee Mission, Kansas: Veterinary Learning systems and Bayer Corporation; 1999.
2. Laffel L. Ketone bodies: A review of the physiology, pathophysiology and application in monitoring diabetes. Diabetes Metab Res Rev. 1999;15:412–426. [PubMed]
3. Goldstein DE. Tests of glycemia in diabetes. Diabetes Care. 1995;18:896–909. [PubMed]
4. The American Diabetic Association. Tests of glycemia in diabetes. Diabetes Care. 2003;26:5106–5108.
5. Chiu RWK, Ho CS, Tong SF, Ng KF, Lam CW. Evaluation of a new handheld biosensor for point-of-care testing of whole blood beta-hydroxybutyrate concentration. Hong Kong Med J. 2002;8:172–176. [PubMed]
6. Khan ASA, Talbot JA, Tieszen KL, Gardener JM, Gibson, New JP. Evaluation of a bedside blood ketone sensor: The effects of acidosis, hyperglycemia and acetoacetate on sensor performance. Diabet Med. 2004;21:782–785. [PubMed]
7. Bryne HA, Tieszen KL, Hollis S, Dornan TL, New JP. Evaluation of an electrochemical sensor for measuring blood ketones. Diabetes Care. 2000;23:500–503. [PubMed]
8. Noyes KJ, Crofton P, Bath L, et al. Hydroxybutyrate near-patient testing to evaluate a new end-point for intravenous Insulin therapy in the treatment of diabetic ketoacidosis in children. Ped Diab. 2007;8:15–156. [PubMed]
9. Rewars A, McFann K, Chase P. Bedside monitoring of blood beta-hydroxy-butyrate levels in the management of diabetic ketoacidosis in children. Diab Tech and Ther. 2006;8:671–676. [PubMed]
10. Ham MR, Okada P, White P. Bedside ketone determination in diabetic children with hyperglycemia and ketosis in the acute care setting. Ped Diab. 2004;5:39–43. [PubMed]
11. Wallace TM, Meston NM, Gardnert SG, Matthews DR. The hospital and home use of a 30-second hand-held blood ketone meter: Guidelines for clinical practice. Diabetic Med. 2001;18:640–645. [PubMed]
12. Hoenig M, Dorfman M, Koenig A. Use of a hand-held meter for the measurement of blood beta-hydroxybutyrate in dogs and cats. J Vet Emerg Crit Care. 2008;18:86–87.
13. Caulkett NA, Cantwell SL, Houston DM. A comparison of indirect blood pressure monitoring techniques in the anesthetized cat. Vet Surg. 1998;27:370–377. [PubMed]
14. Acierno MJ, Mitchell MA. Evaluation of four point-of-care meters for reapid determination of blood lactate concentrations in dogs. J Am Vet Med Assoc. 2007;230:1315–1318. [PubMed]
15. Tappin S, Rizzo F, Dodkin S, Papasouliotis K, Tasker S, Murphy K. Measurement of ionized calcium in canine blood samples collected in prefilled and self-filled heparinized syringes using the I-STAT point-of-care analyzer. Vet Clin Path. 2008;37:66–72. [PubMed]
16. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res. 1999;8:135–160. [PubMed]
17. Lee S, Tremper KK, Barker SJ. Effects of anemia on pulse oximetry and continuous mixed venous hemoglobin saturation monitoring in dogs. Anesthesiology. 1991;75:118–122. [PubMed]
18. Duarte R, Simoes D, Franchini ML, Marquezi M, Ikesaki JH, Kogika M. Accuracy of serum beta-hydroxybutyrate measurements for the diagnosis of diabetic ketoacidosis in 116 dogs. J Vet Int Med. 2002;16:411–17. [PubMed]
19. Lester S. Laboratory Manual Central Laboratory for Veterinarians; Langley, BC: 2005. p. 32.

Articles from The Canadian Veterinary Journal are provided here courtesy of Canadian Veterinary Medical Association