Feasibility of continuous glucose monitoring in children with diabetic ketoacidosis: an exploratory observational study

Abstract

Diabetic ketoacidosis (DKA) is a life-threatening complication of diabetes and a leading cause of Pediatric Intensive Care Unit (PICU) admissions. The use of continuous glucose monitoring (CGM) during the acute and critical phase of DKA has been rarely explored and remains uncertain due to concerns about accuracy and utility in a setting where frequent capillary glucose measurements are standard practice. Data was collected from medical records of patients admitted to the PICU with new-onset DKA as the initial presentation of type 1 diabetes (T1D). Mean absolute relative difference (MARD) and Clarke Error Grid (CEG) analysis were used to assess CGM accuracy. Data from 19 patients (mean age 9.9 ± 3.4 years) were included. Within the first 48 h, 16 hypoglycemic episodes were recorded, with CGM detecting 14 episodes and capillary glucose detecting two. A total of 238 matched pairs of capillary and CGM interstitial glucose values were analyzed. Statistical analysis found capillary glucose values significantly higher than interstitial values (p < 0.001). The overall MARD was 14.5% and CEG analysis indicated 89.1% of matched pairs within zones A and B.

Conclusions: CGM might be a useful point-of-care tool that provides valuable information that may help clinicians to make timely management decisions. The ability of CGM to indicate trends in glucose fluctuations could be its main clinical advantage, particularly in anticipating and preventing potentially dangerous hypoglycemic events, thereby optimizing patient management and safety.