Capillary versus Serum β-hydroxybutyrate in Pediatric Diabetic Ketoacidosis

Analysis of common biomarkers in capillary blood in routine clinical laboratory. Preanalytical and analytical comparison with venous blood

OBJECTIVES

Remote self-collected capillary blood samples have been proposed as alternative to venous blood samples as an aid in telemedicine. The aim of this work is to compare the preanalytical and analytical performance of these two types of samples and to study the stability of common measurands in capillary blood.

METHODS

Capillary and venous blood samples were collected in parallel from 296 patients in serum tubes to analyze 22 common biochemistry magnitudes after centrifugation and in EDTA tubes to analyze 15 hematologic magnitudes. Quality of the preanalytical process was assessed applying the model of quality indicator. 24 h stability at room temperature was studied by obtaining paired capillary samples. A questionnaire of assessment was conducted.

RESULTS

Mean hemolysis index was higher in capillary samples compared to venous blood samples (p<0.001). Regression analysis and difference analysis showed no bias for all studied biochemistry parameters and hematologic parameters, except mean corpuscular volume (MCV), between capillary and venous blood samples. Regarding sample stability, percentage deviation was higher than the corresponding minimum analytical performance specification for ferritin, vitamin D, hematocrit, MCV, mean corpuscular hemoglobin concentration, platelets distribution wide, mean platelet volume and basophils. Finger pricking was perceived as less painful (p<0.05) than venipuncture in participants who undergo more than one blood test per year.

CONCLUSIONS

Capillary blood can be used as an alternative to venous blood for the studied parameters in automated common clinical analyzers. Cautious should be taken if samples are not analyzed within 24 h from the collection.

Fasting ketone levels vary by age: implications for differentiating physiologic from pathologic ketotic hypoglycemia

OBJECTIVES

Ketone production is a physiological phenomenon that occurs during beta-oxidation of free fatty acids. Distinguishing physiologic ketosis from pathologic over-production/underutilization of ketones is critical as part of the diagnostic evaluation of disorders of carbohydrate metabolism, but there is limited literature on normal ketone production with fasting. Our aim is to measure fasting serum beta-hydroxybutyrate (BHB) concentrations in healthy children after an overnight fast.

METHODS

Children ≤18 years of age were prospectively recruited from elective procedures through our surgery centers. Exclusion criteria included a history of diabetes, hypopituitarism, adrenal, metabolic or inflammatory disorders, dietary restrictions, trauma, or use of medications that might affect blood glucose. Serum glucose, cortisol, and BHB were assessed after an overnight fast.

RESULTS

Data from 94 participants (mean 8.3 ± 5.7 years, 54 % male, 46 % female, were analyzed. Children ≤3 years of age (19) have significantly higher mean (0.40 ± 0.06 mmol/L) and median (0.4, IQR 0.2-0.6 mmol/L) BHB concentrations compared to children >3 years of age (75) with mean (0.21 ± 0.02 mmol/L) and median BHB (0.1, IQR 0.1-0.2 mmol/L) (p<0.0001). Fasting BHB levels of >1.0 mmol/L was rare (2 %, N=2) and 74 % (N=70) of participants had BHB levels <0.3 mmol/L.

CONCLUSIONS

BHB concentrations are significantly higher in young children (≤3 years of age) compared to older children. Fasting BHB levels >1.0 mmol/L are rare within our population and therefore may identify a value above which there may a greater concern for pathologic ketotic hypoglycemia. It is imperative to establish the normative range in children to differentiate physiological from pathological ketotic hypoglycemia.

Relationship Between Bedside Ketone Levels and Time to Resolution of Diabetic Ketoacidosis: A Retrospective Cohort Study

INTRODUCTION

There is no information on the factors that influence the time required to induce resolution of diabetic ketoacidosis (DKA). New methods are currently available for bedside measurement of serum 3-hydroxybutyrate (3HB). The aim of this study was to determine the relationship between serum 3HB and the time to DKA resolution.

METHODS

We reviewed the medical records of patients with type 1 diabetes (T1D) and a history of DKA who were admitted to the Department of Pediatrics, Osaka City University Hospital, between November 2008 and October 2018. DKA resolution was defined as 3HB below 1.0 mmol/L as measured by a bedside ketone meter.

RESULTS

Data of 52 T1D-DKA episodes were analyzed (median age, 8.0 years; 20 male patients; 32 female patients; new T1D diagnosis, n = 13; established diagnosis, n = 39). In all cases, correction of serum 3HB was an important aspect of T1D management. The median time to DKA resolution (defined as the time from the start of insulin infusion until the fall of 3HB level to below 1.0 mmol/L) was 11 and 10 h in new and established T1D cases, respectively. 3HB on admission and the required insulin infusion dose per body weight, but not blood pH level on admission, correlated with time to DKA resolution. There was no relationship between blood pH level and 3HB on admission.

CONCLUSIONS

Our results showed that DKA resolution could be achieved within 10-11 h when DKA treatment is guided by bedside 3HB monitoring without any severe complications. Blood 3HB level is a potentially suitable marker for the severity and resolution of DKA.

Method comparison of beta-hydroxybutyrate point-of-care testing to serum in healthy children

Ketone production is a physiological phenomenon that occurs to avoid irreversible neurological damage from hypoglycemia, thereby serving as a marker of metabolic stress. The primary ketone body, beta-hydroxybutyrate (BHB), guides the diagnostic evaluation and management of many hypoglycemic disorders. Serum and point-of-care (POC) BHB values were not been compared in children without diabetes or metabolic disorders. We aim at comparing the serum and point-of-care BHB values in healthy children after an overnight fast. Eligible participants were ≤18 years of age prospectively recruited from elective procedures through our surgery centers. Exclusion criteria included a history of diabetes, hypopituitarism, adrenal, metabolic or inflammatory disorders, dietary restrictions, trauma, or use of medications that might affect blood glucose. The main outcome measure was comparing serum and POC BHB levels after a period of fasting. Data from 94 participants (mean age 8.29 ± 5.68 years, 54.3% male, 45.7% female, BMI mean 19.28 ± 5.25 kg/m2) were analyzed. There was a strong correlation between serum BHB (mean 0.25 ± 0.23 mmol/L) and POC BHB (mean 0.18 ± 0.20 mmol/L) (r s = 0.803, p < 0.0001). The majority (96.81%) of values for serum BHB compared with POC BHB fell within 0.1 ± 0.1 mmol/L. The average of difference between serum and POC BHB (the bias) was 0.064 mmol/L (95% CI 0.047-0.081), and percentage error was 3.19%. Point-of-care BHB is accurate and comparable to serum BHB levels in our cohort of children after an overnight fast.

SYNOPSIS

Point-of-care BHB agrees with serum values in healthy children.