Blood versus urine ketone monitoring in a pediatric cohort of patients with type 1 diabetes: a crossover study.

Abstract

The aim of our study was to determine the influence of routine ketone monitoring on hyperglycemic events (HE) and ketosis in youngsters with type 1 diabetes (T1D). Our single-site, controlled and randomized study was conducted on children and adolescents with T1D outside of remission phase. During two crossover periods of 6 months, patients (n = 22) experiencing HE tested ketones alternatively with a blood ketone meter or urine ketone test strips and gave their opinion on screening methods after completion of clinical trial. Moreover, we evaluated levels of awareness of ketone production in a series of 58 patients and sometimes parents via a multiple-choice questionnaire. Based on self-monitoring data, patients experienced a mean of 4.8 HE/month (range 0-9.3). Patients performed accurate ketone tests more frequently during urine (46%) than during blood-testing (29%) periods (p < 0.05); while globally, 50% of ketone tests were inaccurate (i.e. without HE). Ketosis occurred significantly more often during urine (46.4%) than during blood (14.8%) monitoring (p = 0.01), although no episodes of diabetic ketoacidosis (DKA) were noticed. Duration of hyperglycemia was not different whether patients measured ketones or not, suggesting that ketone monitoring did not affect correction of glycemia. Patients evaluated blood monitoring more frequently as being practical, reliable, and useful compared with urine testing. Scores in the awareness questionnaire were globally low (36.8%) without difference between patients and their parents. Although our study shows differences in outcomes (e.g. accurate use, detection of ketosis) of urine versus blood ketone monitoring, these did not affect the occurrence of HE. Whereas ketone monitoring is part of standardized diabetes education, its implementation in daily routine remains difficult, partly because patient awareness about mechanisms of ketosis is lacking.