Wei Marcus Tong Zhen, Gallo Linda A, Hulme Katina D, Alzaid Fawaz, Julla Jean-Baptiste, Dorey Emily S, Morineau Gilles, Chew Keng Yih, Grant Emma J, Gras Stephanie, Barett Helen L, Riveline Jean-Pierre, Carney Meagan, Short Kirsty R
School of Chemistry and Molecular Biosciences, The University of Queensland, St Lucia, Australia.
School of Health, University of the Sunshine Coast, Petrie, Australia.
Cell Mol Life Sci. 2025 Feb 6;82(1):71. doi: 10.1007/s00018-024-05568-7.
Achieving an in-range glycated haemoglobin (HbA1c) is essential for managing diabetes mellitus (DM). However, this parameter provides an estimate of long-term blood glucose control rather than daily glycaemic variations. Glycaemic variability can be more predictive than HbA1c in terms of identifying those at risk for diabetes complications, including risk of severe respiratory virus infections and is usually measured via a continuous glucose monitor (CGM). For individuals for whom a CGM is not available, serum 1,5 anhydroglucitol (1,5-AG) level has shown potential as an alternative method for monitoring glycaemic variability. Despite this, at present 1,5-AG is not routinely used in the clinical assessment of DM. Here, we aim to determine whether assessing 1,5-AG, in addition to HbA1c, is of any potential clinical utility to the management of DM for patients.
Using machine learning and data derived from 78 patients with type I DM (for whom CGM data is available) we show that the combination of 1,5-AG and HbA1c improves the prediction of a patient's glycemia risk index (GRI) compared to HbA1c alone.
The GRI is an essential tool in the management of DM as it reflects both clinical priorities and patient centred outcomes. The inclusion of 1,5-AG in this prediction was particularly important for individuals who had very high or very low GRI. Furthermore, in the context of glycaemic variability and susceptibility to severe respiratory virus infections, we show that reduced 1,5-AG in the plasma is associated with reduced ex vivo CD4 + T cell cytokine responses to influenza virus in individuals with a matched HbA1c.
Taken together, these data argue for an increased monitoring of 1,5-AG in the clinic for individuals without a CGM to provide additional insights for diabetes management.
实现糖化血红蛋白（HbA1c）处于正常范围对于糖尿病（DM）的管理至关重要。然而，该参数提供的是长期血糖控制的估计值，而非每日血糖变化情况。在识别糖尿病并发症风险（包括严重呼吸道病毒感染风险）方面，血糖变异性可能比HbA1c更具预测性，通常通过连续血糖监测仪（CGM）进行测量。对于无法使用CGM的个体，血清1,5 - 脱水葡萄糖醇（1,5 - AG）水平已显示出作为监测血糖变异性的替代方法的潜力。尽管如此，目前1,5 - AG在DM的临床评估中并未常规使用。在此，我们旨在确定除HbA1c外，评估1,5 - AG对患者DM管理是否具有任何潜在的临床效用。
利用机器学习和来自78例1型糖尿病患者（可获取CGM数据）的数据，我们表明与单独使用HbA1c相比，1,5 - AG和HbA1c的组合可改善对患者血糖风险指数（GRI）的预测。
GRI是DM管理中的重要工具，因为它既反映临床重点又反映以患者为中心的结果。在该预测中纳入1,5 - AG对于GRI非常高或非常低的个体尤为重要。此外，在血糖变异性和对严重呼吸道病毒感染的易感性背景下，我们表明在HbA1c匹配的个体中，血浆中1, AG的降低与体外CD4 + T细胞对流感病毒的细胞因子反应降低相关。
综上所述，这些数据支持在临床中对没有CGM的个体增加对1,5 - AG的监测，以为糖尿病管理提供更多见解。
