ESPE Abstracts (2015) 84 P-3-693

ESPE2015 Poster Category 3 Diabetes (94 abstracts)

Blood vs Urine Ketone Monitoring in a Pediatric Cohort of Patients with Type 1 Diabetes: a Crossover Study

Line Goffinet , Thierry Barrea , Valérie Vandooren & Philippe Lysy


Pediatric Endocrinology Unit, Cliniques Universitaires Saint Luc, Université Catholique de Louvain, Brussels, Belgium


Background: Diabetes ketoacidosis (DKA) is the most severe complication in type 1 diabetes (T1D) but patient education and ketone monitoring may help decrease its frequency. However, the influence on glucose homeostasis of systematic ketone monitoring and of the nature of monitoring (urine vs blood) is unclear.

Objective and hypotheses: To determine whether the use of blood ketone monitoring, as compared to urine ketone testing, decreases the duration of hyperglycemia, the occurrence of ketosis events and of DKA in the daily management of T1D in paediatrics.

Method: Our single-site, controlled and randomized study was done on prepubertal patients with T1D outside of remission phase. Over 118 patients screened, 28 were actively enrolled. Patients were asked to test blood ketones during hyperglycemic episodes (HE), being two consecutive (but 1 h delayed) capillary glycaemia ≥250 mg/dl, during two periods of 6 months alternatively with a blood ketone meter and urine ketone test strips.

Results: Our cohort was homogeneous for gender (F/M ratio: 1), age (mean 10.4 years) and diabetes duration (≥2 years). During the study period, no episodes of DKA or severe hypoglycemia were noticed. Patients experienced a mean of 4.6 HE/month (range 0–9). They adequately controlled ketones 1.6 times/month (36% of HEs) and inadequately (when no HE occurred) 1.5 times/month. Missed ketone bodies measurements (2.9/month) tended to be higher with blood meter (3.7/month) than with urine strips (2.1/month) (P=0.05). Duration of hyperglycaemia during HEs was not different when patients measured or not ketones (4.3±1.9 h vs 4.3±2.0 h respectively), meaning that ketone monitoring did not allow rapid normalisation of glycaemia. The proportion of severe ketosis (>3 mM for blood and >28 mM for urine tests) was similar between the two groups (4% for urine and 5% for blood monitoring). Also, no difference was noticed in mean blood glucose or HbA1c levels before and after the study period.

Conclusions: Whereas ketone monitoring is part of standardized diabetes education, its implementation in the daily routine remains difficult. Although, we did not observe DKA, ketone monitoring did not impact glycemic control during our study. Further research is necessary to evaluate how ketone monitoring may impact long-term diabetes control and complications.

Article tools

My recent searches

No recent searches.