Rawlings Ra, Yuan L, Shi H, Brehm W, Pop-Busui R, Nelson Pw
Center for Computational Medicine and Bioinformatics, University of Michigan, USA ; Departments of Biophysics, University of Michigan, USA.
Departments of Mathematics, University of Michigan, USA ; University of Michigan Program in Informatics, University of Michigan, USA.
J Diabetes Metab. 2012 Feb 28;3:177. doi: 10.4172/2155-6156.1000177.
Hemoglobin A1c (HbA1c) is the current standard used in the clinical treatment of patients with diabetes. However, it has been shown that patients with similar HbA1c values may have widely different fluctuations in blood glucose values over the same period of time, including time spent in hyper- and/or hypo-glycemia. Hence, there exists a need for quantitative measures that can supplement HbA1c in managing patients with diabetes. We introduce and compare the Dynamic Stress Factor, DySF, a newly developed metric that quantifies glycemic volatility based on patient-specific glucose transition density profiles with HbA1c and with currently used glucose variability metrics in predicting severe hypoglycemia in children with type 1 diabetes. DySF, the daily weighted number of large monotonic glycemic transitions that occur within one hour, was calculated for 441 total subjects with type 1 diabetes (146 children, aged 8-14 yrs) to assess the magnitude and frequency of glucose transitions per day. Severe hypoglycemic episodes (HE) were quantified for all subjects and evaluated against HbA1c and existing measures of glucose variability, including SD, MAGE, MODD, and CONGA using logistic regression models. DySF was found to be a predictor of severe HE in children (p = 0.018) with the likelihood of a child, aged 8-14 yrs, experiencing severe hypoglycemia increasing by up to 20% with decreasing values of up to 60% of DySF. Patients of any age who had one or multiple severe hypoglycemic episodes had on average a lower DySF when compared to those with no HE. Additionally, when considering mean glucose levels, DySF/mean was a preliminary predictor of severe HE in patients with HbA1c ≤ 6.5% (p = 0.062). DySF is a dynamic, quantitative, measure of daily glucose "volatility" that separates patients, within the same strata of HbA1c, into visually distinct patient profiles. DySF can be used as a preliminary predictor of clinically severe hypoglycemia in children and "well-controlled" patients with HbA1c ≤ 6.5%.
糖化血红蛋白(HbA1c)是目前糖尿病患者临床治疗中使用的标准指标。然而,研究表明,具有相似HbA1c值的患者在同一时间段内血糖值可能存在广泛差异,包括高血糖和/或低血糖持续时间。因此,需要一些定量指标来补充HbA1c,以更好地管理糖尿病患者。我们引入并比较了动态应激因子(DySF),这是一种新开发的指标,它基于患者特定的血糖转换密度曲线来量化血糖波动,并将其与HbA1c以及目前用于预测1型糖尿病儿童严重低血糖的血糖变异性指标进行比较。DySF是指一小时内发生的每日大的单调血糖转换的加权数量,我们对441名1型糖尿病患者(146名8至14岁儿童)计算了DySF,以评估每天血糖转换的幅度和频率。对所有受试者的严重低血糖事件(HE)进行量化,并使用逻辑回归模型,根据HbA1c和现有的血糖变异性指标(包括标准差(SD)、平均血糖波动幅度(MAGE)、平均血糖波动离散度(MODD)和血糖波动连续性指标(CONGA))进行评估。结果发现,DySF是儿童严重HE的一个预测指标(p = 0.018),8至14岁儿童发生严重低血糖的可能性随着DySF值降低高达60%而增加多达20%。与没有发生HE的患者相比,任何年龄有一次或多次严重低血糖事件的患者平均DySF较低。此外,在考虑平均血糖水平时,DySF/平均值是HbA1c≤6.5%患者严重HE的一个初步预测指标(p = 0.062)。DySF是一种动态的、定量的每日血糖“波动”测量指标,它能将HbA1c处于同一分层的患者分为视觉上明显不同的患者类型。DySF可作为儿童和HbA1c≤6.5%的“血糖控制良好”患者临床严重低血糖的初步预测指标。