Impact of Glucose Variability on Dynamic Cognitive Function in Youth with Type 1 Diabetes

项目摘要/摘要 1型糖尿病(T1D)是儿童最常见的慢性疾病之一，并有迅速上升的趋势 发病率和流行率。血糖的管理是具有挑战性的，并经常导致频繁和快速的 在正常、高血糖和低血糖之间波动。这种血糖变异性最近被确认为一种 糖尿病并发症的独立危险因素。最少被研究和了解的并发症之一 T1D是认知改变及其与血糖变异性的关系。因为大脑使用了大量的 身体的葡萄糖支持其功能，特别是在大脑发育的动态期，有 患有T1D的青少年潜在的脆弱性有很强的生物学原因。几项研究表明， 患有T1D的年轻人的认知分数与没有T1D的同龄人相比，显示出总体上更差 严重的血糖事件(如糖尿病酮症酸中毒)和慢性 高血糖，无论是急性还是长期，都与较低的认知分数有关。然而，这些研究 我们在受控的、优化的实验室环境中进行了单一时间点的实验。人们对此知之甚少。 现实生活中每天发生的典型的血糖波动会影响当下的认知功能， 尤其是对个人的环境或身体状况敏感的动态认知技能 (例如，睡眠不足)，因此在一天中波动(例如，工作记忆、处理速度)。 确定血糖波动和动态认知功能之间的关系至关重要 在年轻人中很重要，因为动态的认知技能是学习、整合和 在日常生活中使用新奇的信息。了解这些技能在日常生活中如何受到影响，如何 它们与血糖波动有关，任何促成因素都可以促进学业成绩的提高。 住宿和T1D治疗建议。重要的是，这项工作还可以扩展到 更好地了解血糖变异性对T1D期间大脑健康的影响，并可能 转化为其他形式的糖尿病(例如，2型糖尿病、妊娠期糖尿病)。这项研究的主要目标是 目的是确定实时、现实生活中的血糖变异性是否预测动态认知功能的波动 患有T1D的青年。第二个目标是确定患有T1D的年轻人的动态认知功能是否不同 从没有T1D的青年开始，独立于显著的血糖极端。为了解决这些问题，我将申请 持续健康和认知数据收集方法的独特和创新组合，多变量 时间序列分析和机器学习模型。这项工作，结合了量身定做的职业发展 计划，将成为我作为独立调查员未来工作的基础，这将扩大到如何解决 糖尿病相关的健康行为(如食物选择)会影响血糖波动，进而影响大脑健康 (即认知功能)，在不同类型的糖尿病中使用新的移动健康(MHealth)技术。

PROJECT SUMMARY/ABSTRACT Type 1 diabetes (T1D) is one of the most common chronic childhood diseases and has a rapidly rising incidence and prevalence. Management of blood glucose is challenging and often leads to frequent and rapid swings between normal, high, and low glucose. This glucose variability has recently been identified as an independent risk factor for diabetic complications. One of the least studied and understood complications of T1D is altered cognition and its relationship to glucose variability. As the brain uses a substantial amount of the body’s glucose to support its functions, especially during dynamic periods of brain development, there are strong biological reasons for the potential vulnerability of youth with T1D. Several studies have shown lower cognitive scores in youth with T1D compared to their peers without T1D, and shown that poorer overall glycemic control as indicated by severe glycemic events (e.g., diabetic ketoacidosis) and chronic hyperglycemia, are associated with lower cognitive scores both acutely and long-term. However, these studies we conducted in a controlled, optimized laboratory settings at a single time point. Little is still known about how typical glucose fluctuations that occur daily in real-life settings affect cognitive function in the moment, particularly dynamic cognitive skills that are sensitive to environmental or physical conditions of the individual (e.g. sleep deprivation) and thus fluctuate throughout the day (e.g., working memory, processing speed). Determining the relationship between glycemic fluctuations and dynamic cognitive function is critically important in youth given that dynamic cognitive skills are fundamental for learning, integrating, and using novel information in their daily lives. Understanding how these skills are affected in daily life, how they relate to glycemic fluctuations, and any contributing factors, could motivate improvements in academic accommodations and T1D treatment recommendations. Importantly, this work could also be expanded to better understand the impact of glycemic variability on brain health throughout the lifespan in T1D and could be translated to other forms of diabetes (e.g., type 2 diabetes, gestational diabetes). The primary goal of this study is to determine if real-time, real-life glycemic variability predicts fluctuations in dynamic cognitive function in youth with T1D. A secondary goal is to determine whether dynamic cognitive function in youth with T1D differs from youth without T1D, independent of significant glycemic extremes. To address these questions, I will apply a unique and innovative combination of continuous health and cognitive data collection methods, multivariate time-series analyses, and machine learning models. This work, combined with a tailored career development plan, will form the basis of my future work as an independent investigator, which will expand to address how diabetes-related health behaviors (e.g., food choice) impact glucose fluctuations, and subsequent brain health (i.e., cognitive function), using novel mobile health (mHealth) technologies across different types of diabetes.