Jingzhen Li1, Jingyi Lu2, Igbe Tobore1, Yuhang Liu1, Abhishek Kandwal1, Lei Wang1, Xiaojing Ma2, Wei Lu2, Yuqian Bao2, Jian Zhou3, Zedong Nie4. 1. Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, 518055, Shenzhen, China. 2. Department of Endocrinology and Metabolism, Shanghai Clinical Center for Diabetes, Shanghai Diabetes Institute, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, 200233, Shanghai, China. 3. Department of Endocrinology and Metabolism, Shanghai Clinical Center for Diabetes, Shanghai Diabetes Institute, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, 200233, Shanghai, China. zhoujian@sjtu.edu.cn. 4. Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, 518055, Shenzhen, China. zd.nie@siat.ac.cn.
Abstract
OBJECTIVE: Despite the clinical importance of glycemic variability and hypoglycemia, thus far, there is no consensus on the optimum method for assessing glycemic variability and risk of hypoglycemia simultaneously. RESEARCH DESIGN AND METHODS: A novel metric, the gradient variability coefficient (GVC), was proposed for characterizing glycemic variability and risk of hypoglycemia. A total of 208 daily records of CGM encompassing 104 patients with T1DM and 2380 daily records from 1190 patients with T2DM were obtained in our study. Simulated CGM waveforms were used to assess the ability of GVC and other metrics to capture the amplitude and frequency of glucose fluctuations. In addition, the association between GVC and the risk of hypoglycemia was evaluated by receiver operating characteristic (ROC) curve. RESULTS: The results of simulated CGM waveforms indicated that, compared with the widely used metrics of glycemic variability including standard deviation of sensor glucose (SD), coefficient of variation (CV), and mean amplitude of glycemic excursion (MAGE), GVC could reflect both the amplitude and frequency of glucose oscillations. In addition, the area under the curve (AUC) of ROC was 0.827 in T1DM and 0.873 in T2DM, indicating good performance in predicting hypoglycemia. CONCLUSIONS: The proposed GVC might be a clinically useful tool in characterizing glycemic variability and the assessment of hypoglycemia risk in patients with diabetes.
OBJECTIVE: Despite the clinical importance of glycemic variability and hypoglycemia, thus far, there is no consensus on the optimum method for assessing glycemic variability and risk of hypoglycemia simultaneously. RESEARCH DESIGN AND METHODS: A novel metric, the gradient variability coefficient (GVC), was proposed for characterizing glycemic variability and risk of hypoglycemia. A total of 208 daily records of CGM encompassing 104 patients with T1DM and 2380 daily records from 1190 patients with T2DM were obtained in our study. Simulated CGM waveforms were used to assess the ability of GVC and other metrics to capture the amplitude and frequency of glucose fluctuations. In addition, the association between GVC and the risk of hypoglycemia was evaluated by receiver operating characteristic (ROC) curve. RESULTS: The results of simulated CGM waveforms indicated that, compared with the widely used metrics of glycemic variability including standard deviation of sensor glucose (SD), coefficient of variation (CV), and mean amplitude of glycemic excursion (MAGE), GVC could reflect both the amplitude and frequency of glucose oscillations. In addition, the area under the curve (AUC) of ROC was 0.827 in T1DM and 0.873 in T2DM, indicating good performance in predicting hypoglycemia. CONCLUSIONS: The proposed GVC might be a clinically useful tool in characterizing glycemic variability and the assessment of hypoglycemia risk in patients with diabetes.
Authors: Gong Su; Tao Zhang; Hongxia Yang; Wenlong Dai; Lei Tian; Hong Tao; Tao Wang; Shuhua Mi Journal: Anatol J Cardiol Date: 2018-06 Impact factor: 1.596