本研究旨在评估儿童和青少年1型糖尿病患者严重低血糖(SH)风险的决定因素和临床标志物。在EPI-Gluredia研究中,回顾性分析2017年7月至2022年6月儿童和青少年1型糖尿病患者的临床参数和连续血糖监测指标.根据他们是否经历严重低血糖,在传统和季度医疗咨询期间收集他们的临床参数。然后,我们分析了特定时期SH前后天数的连续血糖监测指标.根据血糖参数,与正常的三个月相比,SH前三个月的血糖血红蛋白和血糖平均值显着降低(p<0.05)。此外,在低血糖中花费的时间(低于该范围的时间,TBR<3.3)及其与SH频率的强相关性(R=0.9,p<0.001)代表了预测SH的敏感和特定的临床参数(截止值:9%,灵敏度:71%,特异性:63%)。GLUREDIA研究的第二个发现是,在我们的T1DM患者的血糖随访中,SH不是孤立的事件。的确,大多数血糖参数(即,血糖平均值,血糖变异性,低血糖的频率,和血糖目标)在SH前一个月变化很大(所有p<0.05),而这些研究的大多数血糖参数在没有严重急性并发症的情况下保持稳定(均p>0.05)。此外,ROC曲线的使用使我们能够为每个血糖参数确定能够更准确预测SH的敏感性或特异性阈值.例如,低血糖频率增加10%预示着接近SH的风险,同时具有良好的敏感性和特异性(敏感性:80%,特异性:60%)。GLUREDIA研究旨在以临床和血糖参数为目标,以预测有SH风险的患者。首先,我们确定TBR<3.3<9%是降低SH频率的敏感和特异性工具。此外,SH不是孤立事件,而是在SH发生前30天内伴有血糖紊乱。
This study aims to evaluate the determinants and clinical markers of patients at risk for severe hypoglycemia (SH) in children and adolescents with type 1 diabetes. In the EPI-GLUREDIA study, clinical parameters and continuous glucose monitoring metrics from children and adolescents with type 1 diabetes were retrospectively analyzed between July 2017 and June 2022. Their clinical parameters were collected during traditional and quarterly medical consultations according to whether they experienced severe hypoglycemia or not. Then, continuous glucose monitoring metrics were analyzed on days surrounding SH during specific periods. According to the glycemic parameters, glycemic hemoglobin and glycemic mean were significantly lower in the three months preceding a SH compared with during three normal months (p < 0.05). Moreover, the time spent in hypoglycemia(time below the range, TBR<3.3) and its strong correlation (R = 0.9, p < 0.001) with the frequency of SH represent a sensitive and specific clinical parameter to predict SH (cut-off: 9%, sensitivity: 71%, specificity: 63%). The second finding of the GLUREDIA study is that SH is not an isolated event in the glycemic follow-up of our T1DM patients. Indeed, most of the glycemic parameters (i.e., glycemic mean, glycemic variability, frequency of hypoglycemia, and glycemic targets) vary considerably in the month preceding an SH (all p < 0.05), whereas most of these studied glycemic parameters remain stable in the absence of a severe acute complication (all p > 0.05). Furthermore, the use of ROC curves allowed us to determine for each glycemic parameter a sensitive or specific threshold capable of more accurately predicting SH. For example, a 10% increase in the frequency of hypoglycemia predicts a risk of near SH with good combination of sensitivity and specificity (sensitivity: 80%, specificity: 60%). The GLUREDIA study aimed to target clinical and glycemic parameters to predict patients at risk for SH. First, we identified TBR<3.3 < 9% as a sensitive and specific tool to reduce the frequency of SH. In addition, SH was not an isolated event but rather it was accompanied by glycemic disturbances in the 30 days before SH.