Introduction & Objectives: The adoption of continuous glucose monitors (CGM) in inpatient settings in the pediatric population has remained slow due to a scarcity of data on their reliability in hospitalized children.
Methods: We retrospectively reviewed the accuracy of the Dexcom G6 CGM system in pediatric patients with diabetes admitted to our academic children’s hospital from March 2018-September 2023. We cross referenced our institutional Dexcom Clarity database against an internal database of inpatient admissions to identify all children admitted to the hospital with CGM data. We then recorded sensor glucose readings from Clarity and values for blood glucose (BG), blood urea nitrogen, and pH from the electronic medical record.
Results: There were 3,200 admissions of children with diabetes in this period, of which 295 (from 211 patients 2-18 years old) had associated CGM data. A total of 3,952 paired CGM and BG measurements (serum and point-of-care, POC) (40-400 mg/dl) were compared resulting in a mean absolute relative difference (MARD) of 15.6%, with 96.5% of the values in zones A+B of Clarke error grid analysis. The MARD for POC BG vs CGM (n=3,518 pairs) was 15.5%, and for serum BG vs CGM (n=434 pairs), 17%. MARD varied by degree of hyperglycemia (Glucose 70-180 mg/dL: n=1866 pairs, MARD 16.4%; Glucose 181-250: n=1023, MARD 13.2%; Glucose 251-400: n=831, MARD 12.7%, p <0.0001), but not by degree of acidosis (pH 7.3-7.5: n=611, MARD 15.8%, pH 7.1-7.3: n=330, MARD 16.1%, pH <7.1: n=12, MARD 14.8%). We performed linear regression, and pH, sodium, and BUN had no impact on CGM accuracy.
Conclusion: Based on our retrospective data, CGMs demonstrated high clinical reliability in hospitalized children with diabetes.
N. Garg: None. P.C. White: Consultant; Provention Bio, Inc. Research Support; MannKind Corporation, National Institute of Diabetes and Digestive and Kidney Diseases. S. Adhikari: None.
