Department of Paediatric Endocrinology and Diabetology, Jessa Hospital, Hasselt, Belgium.
Clinical Laboratory, Jessa Hospital, Hasselt, Belgium.
Horm Res Paediatr. 2018;89(3):189-199. doi: 10.1159/000487361. Epub 2018 Mar 27.
BACKGROUND/AIMS: The FreeStyle® Libre Flash Glucose Monitoring System (FGM, Abbott) measures glucose concentrations in the interstitial fluid for up to 14 days. It has been approved for use in children aged > 4 years in January 2016. Experience in children is still limited. We evaluated the accuracy and usability of the FGM in children with type 1 diabetes mellitus (DM).
67 children with type 1 DM (35 girls), aged 4-18 years, were included. Subjects wore a sensor on the back of their upper arm. For the first 14 days, they regularly measured capillary blood glucose (BG) with their usual BG meter (Accu-Chek® Mobile [ACM], Roche [n = 24]; Contour® Next Link [CNL], Bayer [n = 26]; OneTouch® Verio® IQ [OTV], LifeScan [n = 17]) followed by a sensor glucose (SG) scanning. SG readings were compared to BG measurements by consensus error grid (CEG) analysis; the mean difference (MD), the mean relative difference (MRD), the mean absolute difference (MAD), and the mean absolute relative difference (MARD) were calculated. After 14 days, subjects were asked to fill in a questionnaire on the usability of the FGM.
2,626 SG readings were paired with BG results. FGM readings were highly correlated with BG (r = 0.926, p < 0.001). 80.3% of the data pairs were in zone A (= no effect on clinical action) and 18.4% were in zone B (= altered clinical action with little or no effect on the clinical outcome) of the CEG. Overall MD was +7.5 mg/dL; MD varied with the BG meter: ACM +10.4 mg/dL, CNL +14.2 mg/dL, OTV -3.6 mg/dL (p < 0.001). Overall, MARD was 16.7%. We observed a large interindividual variability in the accuracy parameters. MD and MRD were inversely related to BMI (r = -0.261 [p < 0.05]; r = -0.266 [p < 0.05], respectively). MARD was inversely related to age (r = -0.266 [p < 0.05]). Twenty-nine patients (43.3%) reported sensor problems, mainly early detachment of the sensor. Nonetheless, the usability questionnaire indicated high levels of satisfaction.
Our results showed a reasonable agreement between the FGM SG readings and capillary BG measurements in children. There was, however, a large interindividual variability. The wearing of the sensor requires special attention. Further studies in children are imperative in order to document the accuracy and safety of the FGM in the paediatric population.
背景/目的：FreeStyle® Libre 闪光血糖监测系统（FGM，雅培）可测量间质液中的血糖浓度，最长可达 14 天。它已于 2016 年 1 月获准用于 4 岁以上儿童。目前儿童使用经验仍然有限。我们评估了 1 型糖尿病儿童使用 FGM 的准确性和可用性。
纳入 67 名年龄 4-18 岁的 1 型糖尿病儿童（35 名女孩）。受试者在上臂背部佩戴传感器。在前 14 天，他们使用常规血糖仪（罗氏 Accu-Chek® Mobile [ACM]，n = 24；拜耳 Contour® Next Link [CNL]，n = 26；强生 LifeScan OneTouch® Verio® IQ [OTV]，n = 17）定期测量毛细血管血糖（BG），随后进行传感器血糖（SG）扫描。使用一致性误差网格（CEG）分析比较 SG 读数和 BG 测量值；计算平均差值（MD）、平均相对差值（MRD）、平均绝对差值（MAD）和平均绝对相对差值（MARD）。14 天后，让受试者填写一份关于 FGM 可用性的问卷。
2626 个 SG 读数与 BG 结果配对。FGM 读数与 BG 高度相关（r = 0.926，p < 0.001）。80.3%的数据对位于 CEG 的 A 区（=对临床决策无影响），18.4%位于 B 区（=对临床结果影响不大或无影响，但改变临床决策）。总体 MD 为+7.5mg/dL；MD 随 BG 血糖仪而异：ACM 为+10.4mg/dL，CNL 为+14.2mg/dL，OTV 为-3.6mg/dL（p < 0.001）。总体 MARD 为 16.7%。我们观察到准确性参数存在较大的个体间差异。MD 和 MRD 与 BMI 呈负相关（r = -0.261 [p < 0.05]；r = -0.266 [p < 0.05]）。MARD 与年龄呈负相关（r = -0.266 [p < 0.05]）。29 名患者（43.3%）报告了传感器问题，主要是传感器早期脱落。然而，可用性问卷显示出较高的满意度。
我们的结果显示 FGM SG 读数与儿童毛细血管 BG 测量值之间具有合理的一致性。然而，存在较大的个体间差异。传感器的佩戴需要特别注意。需要进一步在儿童中开展研究，以记录 FGM 在儿科人群中的准确性和安全性。
