Introduction and Objective: Published data support use of 14 days of CGM to provide reliable estimates of CGM metrics, including the glucose management indicator (GMI) to estimate A1c. We examined relationships between A1c and GMI in AYA.Methods: AYA with T1D, aged 14-25 years, wore Dexcom CGM (G6 or G7) for up to 90 days prior to centralized A1c measurement. To be included in this analysis, ≥70% of CGM was required. A1c was compared with GMI from 14, 30 and 90 days of CGM data. Pearson correlations assessed relationship between A1c and GMI in all 3 time periods. Further analyses evaluated key CGM metrics when A1c and GMI differed by ≥0.5% with 14, 30 and 90 days of CGM.Results: There were 112 AYA (54% male, 82% NHW) with CGM data and A1c measurements; mean±SD age was 19.6±3.6 years, T1D duration 10.9±5.4 years and A1c 7.7±0.8% (range 6.2-10.1%). Correlations between A1c and GMI were r=0.75, 0.77 and 0.80 for 14, 30 and 90 days, respectively (all p<.0001). A1c was generally lower than GMI with mean differences of -0.16, -0.16 and -0.17% for 14, 30 and 90 days, respectively, revealing no clinically relevant differences. The A1c-GMI difference (∆A1c-GMI) ranged from -1.43 to +1.19 with 24 AYA (21%) having consistent absolute ∆A1c-GMI ≥0.5% at 14, 30 and 90 days. AYA with ∆A1c-GMI ≥0.5% had less optimal glycemic control than those with ∆A1c-GMI <0.5%. The 24 AYA with consistent discordance vs. those without discordance had higher A1c and GMI values, higher mean CGM glucose levels, more CGM time above range (>180, >250 mg/dL) and less time in range (70-180 mg/dL) and in tight range (70-140) with 14, 30 or 90 days of data; CV and time below range did not differ between the 2 groups.Conclusion: Correlation between A1c and GMI is high in AYA with 14, 30 or 90 days of data, confirming past observations. Even when A1c and GMI differ by ≥0.5%, CGM metrics provide salient data for optimization of glycemic control. Thus, GMI and CGM metrics remain good estimates of glycemic control with 14 days of CGM data in AYA with T1D, supporting use of remote visits and monitoring in the absence of A1c.
L.J. Tinsley: None. S.A. Alamarie: None. S. Hanes: None. K.K. Hood: Consultant; Sanofi. Advisory Panel; MannKind Corporation. Consultant; Havas Health. Research Support; embecta. G.E. Orloff: None. C.P. Raso: None. A.K. Schneider-Utaka: None. L.K. Volkening: None. L.M. Laffel: Advisory Panel; Boehringer-Ingelheim, Sanofi, MannKind Corporation, Medtronic, Vertex Pharmaceuticals Incorporated, Tandem Diabetes Care, Inc, Insulet Corporation. Research Support; Dexcom, Inc. Advisory Panel; Sequel Med Tech. Other Relationship; Janssen Pharmaceuticals, Inc. Consultant; Arbor Biotech.
National Institutes of Health (R01DK129479)
Source link
Leave a Reply