"Mind the gap please…": estimated vs. measured A1c from continuous measurement of interstitial glucose over a 3-month period in patients with type 1 diabetes.

Introduction: Glycated hemoglobin (A1c) is the measurement of choice to estimate the glycemic exposure over the last 3 months prior to sampling. The Free Style Libre® is a continuous glucose monitoring device which provides an estimated A1c (eA1c) from average interstitial glucose using Nathan's ADAG equation. The objective of this study was to compare eA1c and A1c in type 1 diabetes patients (T1D) over a period of 3 months.Materials and methods: Data were collected from patient charts between July 2016 and October 2017. 3-months recordings with >70% of data available were analyzed. eA1c was recorded at each visit and the corresponding A1c value measured by high performance liquid chromatography in a single reference lab.Results: A total of 344 reports from 170 T1D were studied, 3 categories were identified: eA1c = A1c: 13% of reports. eA1c > A1c: 57% of reports, positive difference (eA1c - A1c) of +0.74 ± 0.5% (P < 0.0001). eA1c < A1c: 30% of reports, negative difference (eA1c - A1c) of -0.5 ± 0.3% (P < 0.0001).Conclusion: eA1c value was generally overestimated compared to measured A1c in this T1D cohort. This lesser concordance may result from differences in measured glucose source and/or frequency to calculate eA1c compared to ADAG, but also from using the reverse equation which is a source of potential bias. Another explanation could be a different rate of hypoglycemia between groups, or an asymmetric distribution of A1c patients' phenotypes with differential hyper- or hypoglycation intrinsic propensity.