A discrepancy between laboratory-measured HbA1c and Glucose Management Indicator (GMI), estimated from continuous glucose monitoring, is frequently encountered in clinical practice. However, its evolution over time is not yet known. Methodology: We conducted a multicenter retrospective study (9 centers) that collected pairs of HbA1c and GMI (calculated over 90 days) at T0, T1 year, T2 years of follow-up in patients with diabetes, all users of FreeStyleLibre®. The primary study endpoint was the analysis of the mean HbA1c-GMI differences at the 3 time points. Glucose data, clinical parameters, and complications were also analyzed. Patients were classified based on the HbA1c-GMI discrepancy: positive (PosD, HbA1c-GMI>+0.3%), neutral (NullD, HbA1c-GMI from -0.3 to +0.3%), negative (NegD, HbA1c-GMI< -0.3%) at each time point, and with the average differences over the 3 time points. Group comparisons were assessed using ANOVA. Result: We included 347 patients (82% type 1 diabetes), mean age of 51±17 years, diabetes duration 20±13 years, HbA1c 7.6±1.0%, 90±9% CGM data collected, Time in Range 70-180 mg/dl (TIR) 57±17%, GMI 7.4±0.8%. The mean HbA1c-GMI differed over time (T0: 0.27%, T1 year: 0.16%, T2 years: 0.04%, P<0.0001). Considering the mean HbA1c-GMI differences over the 3 time points for all patients, PosD individuals were statistically older, had higher BMI and HbA1c compared to NegD patients. At T0, the patients were distributed as follows: 168 PosD (48.4%), 129 NullD (37.2%), 50 NegD (14.4%). The 121 patients (only 34.8% of the cohort) who stayed in the same group at the three time-points were 44.6% PosD, 38% NullD and 17.4% NegD. Conclusion: In only 1/3 of patients does the difference between HbA1c and GMI appear to be stable over time. This should be taken into account when analyzing the supposed poor prognosis associated with PosD. Disclosure J. Riveline: Board Member; Abbott, Novo Nordisk A/S, Sanofi, Eli Lilly and Company, Medtronic, Dexcom, Inc., Insulet Corporation, Air Liquide, AstraZeneca. G. Prevost: Board Member; Abbott. A. Andrieu: None. M. Joubert: Consultant; Abbott, Medtronic, Dexcom, Inc. P. Oriot: Research Support; Abbott. A. Penfornis: Speaker's Bureau; Sanofi, Dexcom, Inc., Diabeloop SA. Board Member; AstraZeneca. Speaker's Bureau; Novo Nordisk, Lilly Diabetes. Board Member; Novo Nordisk, Bayer Inc. Advisory Panel; Abbott, Sanofi. J. Philips: Consultant; Sanofi, Novo Nordisk, Abbott, Avazzia, Boehringer-Ingelheim, Eli Lilly and Company. J. Julla: Speaker's Bureau; Lilly Diabetes, Novo Nordisk. Board Member; Sanofi. E. Cosson: Advisory Panel; Abbott, AstraZeneca, Lilly Diabetes, Novo Nordisk, Sanofi, Roche Diagnostics, Novartis AG, Amgen Inc. Funding Abbott Diabetes Care

中文翻译：

---

1020-P：HbA1c 和血糖管理指标之间的差异随时间的演变 - 来自法国-比利时 347 名患者队列的研究结果


临床实践中经常会遇到实验室测量的 HbA1c 与通过连续血糖监测估算的血糖管理指标 (GMI) 之间的差异。然而，其随时间的演变尚不清楚。方法：我们进行了一项多中心回顾性研究（9 个中心），收集了糖尿病患者（所有 FreeStyleLibre® 用户）在 T0、T1 年、T2 年随访时的 HbA1c 和 GMI（计算超过 90 天）对。主要研究终点是分析 3 个时间点的平均 HbA1c-GMI 差异。还分析了血糖数据、临床参数和并发症。根据 HbA1c-GMI 差异对患者进行分类：阳性（PosD，HbA1c-GMI>+0.3%）、中性（NullD，HbA1c-GMI 从 -0.3 至 +0.3%）、阴性（NegD，HbA1c-GMI< -0.3） %)在每个时间点，以及3个时间点的平均差异。使用方差分析评估组比较。结果：我们纳入了 347 名患者（82% 1 型糖尿病），平均年龄 51±17 岁，糖尿病病程 20±13 年，HbA1c 7.6±1.0%，收集 90±9% CGM 数据，时间范围 70-180 mg /dl (TIR) 57±17%，GMI 7.4±0.8%。平均 HbA1c-GMI 随时间变化（T0：0.27%，T1 年：0.16%，T2 年：0.04%，P<0.0001）。考虑到所有患者在 3 个时间点的平均 HbA1c-GMI 差异，与 NegD 患者相比，PosD 患者在统计上年龄更大，BMI 和 HbA1c 更高。 T0 时，患者分布如下：168 名 PosD（48.4%）、129 名 NullD（37.2%）、50 名 NegD（14.4%）。在三个时间点留在同一组的 121 名患者（仅占队列的 34.8%）分别为 44.6% PosD、38% NullD 和 17.4% NegD。结论：只有 1/3 的患者 HbA1c 和 GMI 之间的差异随着时间的推移似乎是稳定的。 在分析与 PosD 相关的假定不良预后时，应考虑到这一点。披露 J. Riveline：董事会成员；雅培、诺和诺德公司、赛诺菲、礼来公司、美敦力、Dexcom, Inc.、Insulet Corporation、液化空气集团、阿斯利康。 G. Prevost：董事会成员；雅培。 A.安德鲁：没有。 M. Joubert：顾问； Abbott、Medtronic、Dexcom, Inc. P. Oriot：研究支持；雅培。 A. Penfornis：议长局；赛诺菲、Dexcom, Inc.、Diabeloop SA。董事会成员;阿斯利康。议长局；诺和诺德、礼来糖尿病。董事会成员;诺和诺德、拜耳公司顾问小组；雅培、赛诺菲。 J.飞利浦：顾问；赛诺菲、诺和诺德、雅培、Avazzia、勃林格殷格翰、礼来公司。 J. Julla：议长局；礼来糖尿病、诺和诺德。董事会成员;赛诺菲。 E. Cosson：顾问小组；雅培、阿斯利康、礼来糖尿病、诺和诺德、赛诺菲、罗氏诊断、诺华公司、安进公司 资助雅培糖尿病护理