ORIGINAL ARTICLE

Comparison of capillary blood ketone measurementby electrochemical method and urinary ketone in treatmentof diabetic ketosis and ketoacidosis in children

Serap Turan Æ Anjumanara Omar ÆAbdullah Bereket

Received: 25 October 2006 / Accepted: 14 January 2008 / Published online: 22 February 2008

� Springer-Verlag 2008

Abstract We aimed to compare the recent practical

method of capillary b-hydroxy butyrate (bOHB) mea-

surement with the widely used urinary ketone measurement

in monitoring metabolic status of the patient during treat-

ment of diabetic ketoacidosis (DKA) and diabetic ketosis

(DK). Patients with DKA and DK admitted to the hospital

were followed with simultaneous measurements of capil-

lary bOHB by electrochemical method (Medisense

Optium, Abbott), and urinary ketone by semi-quantitative

method. Blood gases were measured in 2–4 h intervals.

Fourteen patients with DKA/DK (7 males and 7 females,

age: 9.2 ± 4.2 years) were included with 50 simultaneous

measurements of capillary and urinary ketone. No corre-

lation was detected between urinary ketone and blood pH

(P = 0.06) and HCO3 (P = 0.79), whereas a significant

negative correlation was found between capillary bOHB

and blood pH (r = -0.41, P \ 0.05) and HCO3 (r =

-0.35, P \ 0.05). Capillary bOHB and urinary ketone

levels did not correlate at the beginning and 3.3 ± 1.4 h

after treatment, but did correlate in the third samples taken

7.8 ± 2.0 h after treatment (r = 0.8, P \ 0.05). Capillary

bOHB levels show good correlation with the degree of

acidosis (pH and HCO3). Capillary bOHB measurement is

more sensitive than urinary ketone measurement in

reflecting the patient’s metabolic status and improvement

during treatment.

Keywords Diabetic ketoacidosis � Type 1 diabetes

mellitus � b-hydroxy butyrate � Urinary ketone �Ketone measurement

Introduction

In diabetic ketosis (DK) and ketoacidosis (DKA), b-OH

butyrate (bOHB) is the major ketone body accumulated in

blood. bOHB is converted to acetoacetate and acetone

which are excreted in the urine [1,2]. Serum bOHB con-

centration is one of the best parameters reflecting the

metabolic status in DKA and DK. Nevertheless, quantita-

tive measurement of blood bOHB is not available in many

hospital laboratories. Therefore, the follow up of ketone

bodies is traditionally performed semi-quantitatively by

measuring ketone bodies in the urine. The use of urinary

ketone measurement is confusing because acetoacetate and

acetone but not bOHB is measured in urine by dipstick

method [1, 2]. Despite metabolic and clinical improve-

ments during the treatment of DKA, persistence of ketone

bodies in urine testing is commonly seen due to the pres-

ence of acetone or acetoacetic acid in the urine. It was

demonstrated that monitoring bOHB levels in children

with DKA reduces the cost of intensive care unit by

reflecting earlier normalization of metabolic status than

urine tests [3].

In recent years, ketone-meters have been manufactured

to measure capillary blood bOHB by electrochemical

method [4, 5]. New capillary ketone measurement method

is easy to perform with instant results and allows moni-

toring of blood bOHB at bedside and also at home by the

patients [6]. Additionally, blood ketone measurements

provide precise evaluation and action plan to the families,

like bOHB levels [ 1 mmol/l require further action by

S. Turan � A. Omar � A. Bereket

Faculty of Medicine, Department of Pediatric Endocrinology,

Marmara University, Istanbul, Turkey

S. Turan (&)

Cocuk Sagligi ve Hastaliklari, Marmara Universitesi Tip

Fakultesi Hastanesi, 8110 Altunizade, Uskudar, Istanbul, Turkey

e-mail: serapdemircioglu@yahoo.com

123

Acta Diabetol (2008) 45:83–85

DOI 10.1007/s00592-008-0026-y

retesting and giving additional short-acting insulin, and

levels [ 3 mmol/l necessitate medical review which is

unlikely in the use of urine ketone measurement [7]. A

previous study documented that bOHB threshold of

3 mmol/l detected DKA with a sensitivity of 100% and a

specificity of 88%, whereas which are 100 and 58% with

urine dipstick measurement, respectively [8].

This study was designed to compare the usefulness of

capillary blood bOHB measurement with traditional uri-

nary ketone measurement in the treatment of DKA and DK.

Patients and methods

The study was performed by reviewing hospital records of

patients with DKA and DK retrospectively. Patients with

newly diagnosed type 1 diabetes mellitus or known dia-

betics, admitted in the hospital for DKA or DK, were

included in the study. Patients with acidosis (pH \ 7.30 or

HCO3 \ 15 mEq/l) and/or capillary blood bOHB level

C 1 mmol/l and/or urinary ketone C 1 cross-cut off, fol-

lowed with simultaneous measurements of blood bOHB

and urine ketone until the capillary blood bOHB became

0 mmol/l as a standard hospital care of the DKA and/or DK

management (2 to 6 measurement per each patients).

Patients with DKA were treated with intravenous insulin

infusion (0.1 U/kg/h), and patients with DK were treated

with subcutaneous insulin injections (0.7–1 U/kg/day qid).

Capillary blood bOHB was measured by electrochemi-

cal method (Abott Laboratories, Medisense�, OptiumTM)

in 2–4 h interval. Urinary ketone measurements were

performed by automated analyzer which used sodium

nitroprusside reaction as test principle (URISYS 2400

urinanalysis analyzer, Roche Diagnostic Corporation,

Indianapolis, IN, USA) in each voiding and nearest 30 min

to capillary bOHB measurement was accepted as simulta-

neous measurements. Capillary or venous blood gases were

measured in 2–4 h intervals until the resolution of acidosis

(Venous pH [ 7.3) by a blood gas analyzer (Omni Mod-

ular System; AVL Medical Instruments, Staffordshire,

UK).

All statistical analyses were performed using Jandel

Sigmastat statistics program. The correlations between

urine ketone, blood ketone levels, blood pH and HCO3

were established by Pearson correlation analysis.

Results

Fourteen patients (9 with DKA, 5 with DK) who fulfilled the

criteria (7 males and 7 females, mean age 9.2 ± 4.2 years

with a range of 4–16 years) were included in the study. The

pH and HCO3 levels at the admission were 7.12–7.38 and

5.5–25.7 mEq/l in ranges, respectively. The meter’s work-

ing reference ranges 0–6 mmol/l, and only one patient had

bOHB levels higher than 6 mmol/l. Fifty simultaneous

measurements of blood ketone and urine ketone were

performed.

Urinary ketones were not correlated with blood pH

(P = 0.06) or HCO3 (P = 0.79) while a significant nega-

tive correlation was found between capillary blood ketone,

blood pH (r = -0.41, P \ 0.05) and HCO3 (r = -0.35,

P \ 0.05).

At the beginning and 3.3 ± 1.4 h after treatment there

was no significant correlation between blood ketone and

urinary ketone levels, while in the third blood sample taken

7.8 ± 2 h after treatment significant correlation was found

between blood and urinary ketone levels (r = 0.8, P \ 0.05)

(Fig. 1).

In addition to these findings, two of the patients have

negative urinary ketones while having high blood bOHB

concentrations at admission (one patient with DKA: pH

7.27, HCO3 13.8 mEq/l capillary bOHB 5.8 mmol/l; and

one patient with DK: pH 7.36, HCO3 13.4 mEq/l capillary

bOHB 5.4 mmol/l). The second sample taken 2 h after

admission showed urinary ketones in the patient with

DKA. However, urinary ketones of the patient with DK

remained negative in the second sample taken 3 h after

admission and became positive in the third sample taken

7.5 h after admission.

Discussion

This study shows that, during treatment of DKA, capillary

blood ketone level is directly related to the severity of

acidosis and is a more sensitive method in determining

metabolic status of the patient. Urinary ketone level did not

Fig. 1 Correlations between urinary and capillary blood ketones

shown as 1st samples (white quadrangle), 2nd samples (black square)

and 3rd samples (triangle) during DKA treatment

84 Acta Diabetol (2008) 45:83–85

123

correlate with the degree of acidosis. Urinary and capillary

ketone levels also did not correlate, especially at the

beginning of the treatment. The reason for this is the fact

that the concentration of bOHB is 4–10 times higher than

that of acetoacetate at initial presentation of DKA. During

insulin therapy, bOHB is oxidized to acetoacetate, which is

eventually detected in urine. As a direct consequence of this

finding, two patients in this study had initial negative urine

ketone despite the high capillary blood bOHB levels and

increased urine ketone later on with insulin treatment; a

similar patient was detected in cohort of Wallace et al. [7]. It

was demonstrated that blood levels of bOHB are correlated

with HbA1c and duration of symptoms in addition to arte-

rial pH and HCO3 levels similar to our study [9].

It has been demonstrated that capillary bOHB values are

strongly correlated with serum bOHB values [10]. Fur-

thermore, the meter also shows a high positive predictive

value, with meter measurements over the 1.5 mmol/l

indicating DKA [10]. Additionally, blood ketone mea-

surements provide precise evaluation and action plan for

families where bOHB levels [ 1 mmol/l require further

action by retesting and giving additional short-acting

insulin, and levels [ 3 mmol/l necessitate medical review

[7]. Ketone-meters are also useful for home monitorization

of ketone bodies in diabetic patients due to easy accessi-

bility and usage as gluco-ketone meters.

In conclusion, bedsides capillary blood bOHB mea-

surement is a practical method which shows good

correlation with the degree of acidosis (pH and HCO3)

during DKA management. Capillary bOHB is more reli-

able than urinary ketone measurement in reflecting the

patient’s metabolic status and response to treatment. Fur-

thermore, measurement of urinary ketone alone can cause

delay in diagnosis and treatment as in our two patients.

Capillary ketone measurement will not only facilitate

treatment of DKA, but will also help in preventing DKA

episodes by earlier and more reliable ketone determination

by the patients themselves during sick-day management.

References

1. Wildenhoff KE (1977) Tubular reabsorption and urinary excre-

tion of acetoacetate and 3-hydroxybutyrate in normal subjects

and juvenile diabetics. Acta Med Scand 201:63–67

2. Owen OE, Trapp VE, Skutches CL, Mozzoli MA, Hoeldtke RD,

Boden G, Reichard GA Jr (1982) Acetone metabolism during

diabetic ketoacidosis. Diabetes 31:242–248

3. Vanelli M, Chiari G, Capuano C, Iovane B, Bernardini A,

Giacalone T (2003) The direct measurement of 3-beta-hydroxy

butyrate enhances the management of diabetic ketoacidosis in

children and reduces time and costs of treatment. Diabetes Nutr

Metab 16:312–316

4. Harano Y, Suzuki M, Kojima H, Kashiwagi A, Hidaka H, Shigeta

Y (1984) Development of paper-strip test for 3-hydroxybutyrate

and its clinical application. Diabetes Care 7:481–485

5. McBride MO, Smye M, Nesbitt GS, Hadden DR (1991) Bedside

blood ketone body monitoring. Diabet Med 8:688–690

6. Guerci B, Benichou M, Floriot M, Bohme P, Fougnot S, Franck

P, Drouin P (2003) Accuracy of an electrochemical sensor for

measuring capillary blood ketones by fingerstick samples during

metabolic deterioration after continuous subcutaneous insulin

infusion interruption in type 1 diabetic patients. Diabetes Care

26:1137–1141

7. Wallace TM, Meston NM, Gardner SG, Matthews DR (2001)

The hospital and home use of a 30-second hand-held blood

ketone meter: guidelines for clinical practice. Diabet Med

18:640–645

8. Harris S, Ng R, Syed H, Hillson R (2005) Near patient blood

ketone measurement and their utility in predicting diabetic

ketoacidosis. Diabet Med 22:221–224

9. Rewers A, MacFann K, Chase HP (2006) Bedside mentoring of

blood beta-hydroxybutyrate levels in the management of diabetic

ketoacidosis in children. Diabetes Technol Ther 8:671–676

10. Ham MR, Okada P, White PC (2004) Bedside ketone determi-

nation in diabetic children with hyperglycemia and ketosis in the

acute care setting. Pediatr Diabetes 5:39–43

Acta Diabetol (2008) 45:83–85 85

123