The assessment of clinical factors influencing Glucose Management Indicator and Hba1c discordance in children with type one diabetes - one-year real-world data observation.

Abstract

Objectives: Published data still highlighted discordances between GMI (parameter estimating HbA1c from CGM report) and laboratory HbA1c, with reasons yet to be explored. This study aimed: to identify potential clinical factors contributing to these discordances.

Methods: A retrospective study of 99 children aged 12.92 (SD 4.03) using CGM devices (Dexcom G6-31, Libre 2-30, Guardian 3-38) was conducted. Inclusion criteria were: type 1 diabetes, continuous use of one type of CGM (with >70% sensor activity) over the last year, quarterly visits. At each visit, we collected data: age, sex, BMI, diabetes duration, daily insulin dose, CGM report (14/90 days) and laboratory HbA1c.

Results: We confirmed linear dependency between HbA1c and GMI - higher HbA1c led to more HbA1c-GMI differences. The HbA1c-GMI 90 days discordance was categorized into four thresholds: 48.7% <0.25, 20.1% in the range 0.25-0.5, 22.4% in 0.5-0.75, and 8.7% >0.75. Children with HbA1c-GMI 90 discordance <0.5% had significantly lower HbA1c (6.80 vs 7.59%), shorter T1D duration (<5 years) and more stable HbA1c (differences <0.4 between results). The analysis of participants stability, based on comparing HbA1c-GMI 90 discordances at subsequent follow-up visits confirmed the individual variability <0.25 in two-thirds of participants. Other factors were not associated with the HbA1c-GMI discordance.

Conclusion: One-year real-world data showed that clinically significant discordances (HbA1c-GMI 90 >0.5%) occurred in less than 30% children. Higher difference is more likely in individuals with higher HbA1c values, longer diabetes duration, and less stable glycemic control. Individual discordance HbA1c-GMI 90 is rather stable, although with varying degrees of difference.