Three components of glucose dynamics – value, variability, and autocorrelation – are independently associated with coronary plaque vulnerability ==================================================================================================================================================== * Hikaru Sugimoto * Ken-ichi Hironaka * Tomoko Yamada * Natsu Otowa-Suematsu * Yushi Hirota * Hiromasa Otake * Ken-Ichi Hirata * Kazuhiko Sakaguchi * Wataru Ogawa * Shinya Kuroda ## ABSTRACT Impaired glucose homeostasis leads to numerous complications, with coronary artery disease (CAD) being a major contributor to healthcare costs worldwide. Given the limited efficacy of current CAD screening methods, we investigated the association between glucose dynamics and a predictor of coronary events measured by virtual histology-intravascular ultrasound (%NC), with the aim of predicting CAD using easy-to-measure indices. We found that continuous glucose monitoring (CGM)-derived indices, particularly average daily risk ratio (ADRR) and AC_Var, exhibited stronger predictive capabilities for %NC compared to commonly used indices such as fasting blood glucose (FBG), hemoglobin A1C (HbA1c), and plasma glucose level at 120 min during oral glucose tolerance tests (PG120). Factor analysis identified three distinct components underlying glucose dynamics – value, variability, and autocorrelation – each independently associated with %NC. ADRR was influenced by the first two components and AC_Var by the third. FBG, HbA1c, and PG120 were influenced only by the value component, making them insufficient for %NC prediction. Our results were validated using data sets from Japan (n=64), America (n=53), and China (n=100). CGM-derived indices reflecting the three components of glucose dynamics can serve as more effective screening tools for CAD risk assessment, complementing or possibly replacing traditional diabetes diagnostic methods. ## INTRODUCTION Diabetes mellitus (DM) is a metabolic disorder affecting more than 400 million people worldwide. Among its complications, coronary artery disease (CAD) accounts for a significant proportion of morbidity, mortality and healthcare costs in patients with type 2 DM (T2DM) (Bax et al., 2007). Various prognostic models (Fiarni et al., 2019; Ravaut et al., 2021) and diagnostic markers (Bax et al., 2007) have been developed to predict CAD; however, screening of CAD can be ineffective, costly, or laborious (Bax et al., 2007; Young et al., 2009). More effective approaches for identifying individuals at high risk for complications using readily available clinical variables are warranted. Blood glucose levels are among the readily obtained predictors of the complications (Psoma et al., 2022). The disrupted conditions of glucose dynamics seen in impaired glucose tolerance (IGT) and T2DM are partly characterized by high concentrations of blood glucose levels (Monnier et al., 2008). High concentrations of blood glucose levels have been defined as having high hemoglobin A1c (HbA1c) levels, fasting blood glucose (FBG) levels, and plasma glucose concentration at 120 min during the oral glucose tolerance test (OGTT) (PG120) (Monnier et al., 2008). These indices, especially HbA1c, are associated with complications of T2DM (Selvin et al., 2010). Recent studies have shown that glucose variability, in addition to absolute glucose concentration, significantly contributes to the prognosis of complications (Gerbaud et al., 2019; Gorst et al., 2015; Monnier et al., 2008; Psoma et al., 2022; Su et al., 2011; Zhou et al., 2018) and all-cause mortality (Cai et al., 2022). Continuous glucose monitoring (CGM) can estimate short-term glycemic variability (Service, 2013), and has been reported to predict T2DM complications (Tang et al., 2016). Standard deviation (Std) of glucose levels (CGM_Std), mean amplitude of glycemic excursion (MAGE), mean of daily difference (MODD), and continuous overlapping net glycemic action (CONGA) are established indices of glycemic variability, of which CGM_Std and MAGE are more highly correlated with coronary plaque properties (Otowa-Suematsu et al., 2018). Among glucose level-related indices, including HbA1c and FBG, MAGE is an independent determinant of coronary plaque instability (Okada et al., 2015). Other CGM-derived indices such as average daily risk ratio (ADRR), lability index (LI), J-index, mean absolute glucose (MAG), and glycemic risk assessment in diabetes equation (GRADE) have also been developed (Hill et al., 2011). We recently showed that AC_Mean and AC_Var, which are calculated from the autocorrelation function of glucose levels measured by CGM, can detect decreased abilities in glucose regulation that cannot be captured by FBG, HbA1c, or the other conventional CGM-derived indices (Sugimoto et al., 2023). The characteristics of glucose dynamics can also be estimated from insulin concentrations. The disposition index (DI), which is the product of insulin sensitivity and insulin secretion, reflects and predicts glycemic disability beyond FBG (Utzschneider et al., 2009). Several other glucose-related indices and the relationship between the indices have also been reported (Fabris et al., 2015, 2014; Keshet et al., 2023). Despite these advances, a comprehensive understanding of how these various indices can be optimally combined to predict T2DM complications, particularly CAD, remains elusive. Furthermore, the underlying factors that these indices represent and their individual associations with CAD remain to be fully elucidated. This study aims to address these knowledge gaps through three objectives: (i) to determine which clinical parameters are effective predictors of coronary plaque vulnerability; (ii) to identify the factors underlying these indices; and (iii) to elucidate how these factors are associated with coronary plaque vulnerability. We investigated the characteristics of 14 CGM-derived indices: 12 relatively well-known CGM-derived indices (Hill et al., 2011) and 2 indices (AC_Mean and AC_Var) as well as OGTT-derived indices, and investigated the relationship between these parameters and coronary plaque vulnerability assessed by virtual histology-intravascular ultrasound (VH-IVAS). We showed that three factors, namely, value, variability, and autocorrelation, underly blood glucose level-related indices, and that the three are independently associated with coronary plaque vulnerability. ## RESULTS ### Mean, standard deviation, and autocorrelation function of glucose levels independently contribute to the prediction of coronary plaque vulnerability We previously reported that AC\_Var, calculated from the autocorrelation function of glucose levels, can capture decreased glucose handling capacities that cannot be captured by the mean (CGM\_Mean) and standard deviation (CGM_Std) of glucose levels measured by CGM (Sugimoto et al., 2023). Based on the study, we hypothesized that AC\_Var could identify individuals with high %NC, a widely used parameter of plaque vulnerability, independently from CGM\_Mean and CGM\_Std. To test this hypothesis, we conducted multiple regression analysis with CGM_Mean, CGM_Std, and AC_Var as input variables and %NC as the objective variable (Fig. 1A). For comparison, we also performed multiple regression analysis with established diagnostic markers of diabetes (FBG, HbA1c, and PG120) as input variables (Fig. 1B). We performed this analysis using a previously described cohort consisting of 8 individuals with NGT, 16 with IGT, and 29 with T2DM (Otowa-Suematsu et al., 2018). ![Figure 1.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/02/2023.11.21.23298816/F1.medium.gif) [Figure 1.](http://medrxiv.org/content/early/2024/09/02/2023.11.21.23298816/F1) Figure 1. Multiple regression analyses for predicting %NC. Multiple regression analysis between %NC and CGM\_Mean, CGM\_Std, and AC_Var (A). That between %NC and FBG, HbA1c, and PG120 (B). Scatter plots for predicted %NC versus measured %NC (the left). Each point corresponds to the values for a single subject. Bars represent the 95% CIs of the coefficients of the regression models (the right). The variance inflation factor (VIF) for CGM\_Mean, CGM\_Std, and AC_Var were 1.1, 1.1, and 1.0, respectively, indicating low multicollinearity among these variables. The *R*2 of the model that predicted %NC from the three indices was 0.36. CGM\_Mean, CGM_Std, and AC_Var had statistically significant independent positive correlations with %NC (Fig. 1A), suggesting that CGM_Mean, CGM_Std, and AC_Var are independently associated with %NC. In contrast, the *R*2 of the model that predicted %NC from FBG, HbA1c, and PG120 was only 0.05 (Fig. 1B). Univariate and multivariate analyses including other indices further confirmed that several CGM-derived indices were significantly correlated with %NC, even after adjustment for multiple testing (Supplementary text). ### CGM-derived indices, particularly ADRR, AC_Var, MAGE, and LI, contribute to the prediction of coronary plaque vulnerability To address the challenge of unstable results when dealing with numerous variables and to investigate which variables are particularly useful in estimating %NC, we used two statistical techniques: Least Absolute Shrinkage and Selection Operator (LASSO) regression and Partial Least Squares (PLS) regression (Fig. 2) (Tibshirani, 1996; Wold et al., 2001). These regression models have been used for studies where the number of input variables is large relative to the sample size (Pei et al., 2023; Wang et al., 2005). ![Figure 2.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/02/2023.11.21.23298816/F2.medium.gif) [Figure 2.](http://medrxiv.org/content/early/2024/09/02/2023.11.21.23298816/F2) Figure 2. LASSO and PLS regression analyses for predicting %NC. (A) Relationship between regularization coefficients (lambda) and the MSE based on the leave-one-out cross-validation in predicting %NC. Dotted vertical line indicates the optimal lambda, which provides the least MSE. The optimal lambda was 0.849. (B) LASSO regularization paths along the lambda in predicting %NC. Cyan, magenta, and gray lines indicate the estimated coefficients of AC\_Mean, AC\_Var, and the other input variables, respectively. Dotted vertical line indicates the optimal lambda. (C) Estimated coefficients with the optimal lambda. Only variables with non-zero coefficients are shown. Input variables include the following 21 variables: BMI, FBG, HbA1c, PG120, I.I., composite index, oral DI, CGM\_Mean, CGM\_Std, CONGA, LI, JINDEX, HBGI, GRADE, MODD, MAGE, ADRR, MVALUE, MAG, AC\_Mean, and AC\_Var. (D) VIP generated from the PLS regression predicting %NC. Variables with a VIPLJ≥LJ1 (the dotted line) were considered to significantly contribute to the prediction. LASSO uses L1 regularization to produce models with fewer parameters and has been widely used for feature selection in predictive modeling (Wei et al., 2022). We included BMI, FBG, HbA1c, OGTT-derived indices, and CGM-derived indices as the input variables. The leave-one-out cross-validation identified the optimal regularization coefficient, lambda, as 0.849 (Fig. 2A). At the lambda, the coefficients of ADRR, AC_Var, MAGE, and LI were estimated to be non-zero coefficients (Fig. 2B, C), suggesting that CGM-derived indices, particularly ADRR, AC_Var, MAGE, and LI, contribute to the prediction of %NC. Even with the inclusion of SBP, DBP, TG, LDL-C, and HDL-C as additional input variables, the results remained consistent, with the coefficients of ADRR, AC_Var, MAGE, and LI still estimated as non-zero coefficients (Fig. S4). To further validate the LASSO results and address potential instability, we performed PLS regression and examined the Variable Importance in Projection (VIP) scores (Fig. 2D). PLS regression is particularly useful when dealing with many input variables that may be highly collinear (Wold et al., 2001). The VIP scores of ADRR, AC_Var, MAGE, and LI, which were estimated to be non-zero coefficients by LASSO, were higher than 1, indicating that these four variables especially contribute to the prediction of %NC. ### Three components of glucose dynamics –value, variability, and autocorrelation – are associated with coronary plaque vulnerability To elucidate the underlying factors of clinical parameters and their association with %NC, we performed an exploratory factor analysis. Factor analysis reduces interrelated indices into a smaller set of underlying common factors, and has been employed to examine the interdependencies among various clinical parameters (Augstein et al., 2015; Cappelleri et al., 2000; Oh et al., 2004) and DM complications (Guo et al., 2020). The optimal number of underlying factors was determined using Bayesian information criterion (BIC) and minimum average partial (MAP) methods, which indicated that five and six factors were appropriate, respectively. We first set the number of underlying factors as five. Figure 3A shows that FBG, HbA1c, PG120, I.I., oral DI, CGM\_Mean, CONGA, HBGI, MVALUE, GRADE, JINDEX, and ADRR were included in the first factor. Given that most of these indices are related to the value of blood glucose concentration, factor 1 was labeled “value.” CGM\_Std, MAGE, LI, MAG, MODD, JINDEX, and ADRR were included in the second factor. Given that these indices are related to glucose variability, factor 2 was labeled “variability.” Given that the definition of JINDEX is based on the sum of CGM\_Mean and CGM_Std, and that of ADRR is based on both high and low values of glucose, the result that JINDEX and ADRR clustered in both factors 1 and 2 is plausible. Given that autocorrelation-derived indices, AC_Mean and AC_Var, were included in the third factor, factor 3 was labeled “autocorrelation.” BMI, PG120, composite index, and oral DI were included in the fourth factor. Factor 4 did not include any CGM-derived indices. Given that this combination of indices indicates a decrease in oral DI and associated increase in blood glucose due to decreased insulin sensitivity, factor 4 was labeled “sensitivity (without CGM)”. PG120, I.I., oral DI, and MAG were included in the fifth factor. Factor 5 did not have positive loadings of any CGM-derived indices. Given that this combination of the indices indicates a decrease in oral DI and associated increase in blood glucose due to decreased insulin secretion (I.I.), factor 5 was labeled “secretion (without CGM).” The cumulative percentages of the total variance of the factors were 39%, 60%, 70%, 75%, and 80%, respectively. ![Figure 3.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/02/2023.11.21.23298816/F3.medium.gif) [Figure 3.](http://medrxiv.org/content/early/2024/09/02/2023.11.21.23298816/F3) Figure 3. Factor analysis of the clinical parameters. (A) Factor analysis after orthogonal rotation. The values and colors were based on the factor loadings. The columns represent each factor. The rows represent input indices. (B) Cronbach’s α for each factor. Bars represent the 95% CI. (C) Scatter plots and fitted linear regression lines for factor scores versus %NC. Each point corresponds to the values for a single subject. *r* is Spearman’s correlation coefficient, and the value in parentheses is the 95% CI. The validity of the factor analysis was assessed according to previous studies (Cappelleri et al., 2000; Guo et al., 2020). To evaluate the applicability of the factor analysis, the Kaiser-Meyer-Olkin (KMO) and Bartlett’s spherical test were performed. The KMO test indicated that the value of the measure of sampling adequacy for this data was 0.64, and Bartlett’s spherical test indicated that the variables were statistically significantly intercorrelated (*P* < 0.01), suggesting that this dataset was applicable for the factor analysis. To evaluate internal consistency, Cronbach’s α (Fig. 3B) and item–total correlations were calculated for each factor. Cronbach’s α was 0.97 for factor 1, 0.93 for factor 2, 0.90 for factor 3, 0.72 for factor 4, and 0.66 for factor 5; these values were larger than 0.65 (Fig. 3B), suggesting that the internal consistency was satisfactory. While Cronbach’s α of factor 5 was relatively low, exclusion of MAG increased the Cronbach’s α to 0.84, indicating that the association between factor 5 and decrease in oral DI and associated increase in blood glucose due to decreased insulin secretion could be considered reliable. Item–total correlations ranged from 0.63 to 0.97 for factor 1, 0.72 to 0.94 for factor 2, 0.82 for factor 3, 0.54 to 0.76 for factor 4, and 0.37 to 0.86 for factor 5. With the exception of MAG, item–total correlations ranged from 0.84 to 0.91 for factor 5. The correlation coefficient of MAG was 0.37, which can be considered a modest correlation (Cappelleri et al., 2000), and the item–total correlations were generally reasonably strong in demonstrating reliability. We also investigated a 6-factor solution (Fig. S5A). Factors 1, 2, and 3 could be interpreted as value, variability, and autocorrelation, respectively, similar to the 5-factor solution. Given that factor 6 had no factor loadings ≥ 0.3, we applied the 5-factor solution in the subsequent analysis. Furthermore, the inclusion of SBP, DBP, TG, LDL-C, and HDL-C into the input variables did not change the presence of the three components (value, variability, and autocorrelation) in glucose dynamics (Fig. S5B). Since we only included only individuals with well-controlled serum cholesterol and BP levels in this study, we applied the 5-factor solution without these indices (Fig. 3A) to the following analysis. To further examine the stability of the results of the factor analysis, we also conducted hierarchical clustering analysis (Fig. S6). The optimal number of clusters was determined based on silhouette analysis. A large positive silhouette coefficient indicates that each cluster is compact and distinct from the others. The analysis indicated that the four clusters were appropriate (Fig. S6A). Clusters 1, 2, and 3 can be interpreted as value, variability, and autocorrelation, respectively (Fig. S6B). To investigate the association between these underlying factors and %NC, we investigated the correlation between the factor scores and %NC (Fig. 3C). The factor value and variability showed significant positive correlations with %NC, whereas autocorrelation showed a significant negative correlation. Factors 4 and 5, which were less related to the CGM-derived indices, showed weaker correlations with %NC. Collectively, we conclude that glucose dynamics has three components – value, variability, and autocorrelation – and that these three components are associated with %NC. To assess the robustness and generalizability of our factor analysis results, we performed similar analyses using previously published datasets from diverse populations (Figs. S7, 8). Factors that could be interpreted as representing the value, variability, and autocorrelation of glucose dynamics were consistently observed across diverse populations, including Japanese (Sugimoto et al., 2023) (Fig. S7A), American (Hall et al., 2018) (Fig. S7B), and Chinese (Zhao et al., 2023) (Fig. S8A) cohorts. In the Chinese dataset, all three factors were significantly different between individuals with and without diabetic macrovascular complications (Fig. S8B). Taken together, these results support the reproducibility and cross-cultural validity of our three-component model of glucose dynamics. ### Overview of the three components of glucose dynamics –value, variability, and autocorrelation – We have shown that three components of glucose dynamics – value, variability, and autocorrelation – are associated with %NC. To overview the characteristics of glucose dynamics with different values of the components, we simulated glucose fluctuations using a previously reported mathematical model (De Gaetano and Arino, 2000) (Figs. 4, S9). ![Figure 4.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/02/2023.11.21.23298816/F4.medium.gif) [Figure 4.](http://medrxiv.org/content/early/2024/09/02/2023.11.21.23298816/F4) Figure 4. Overview of the three components of glucose dynamics. (A) 240 min simulated glucose concentration. The colors of the line are based on the mean value (Mean), Std, and AC_Var of the simulated blood glucose. Red and gray dotted horizontal lines indicate the minimum or maximum values of blood glucose, respectively. (B) Previously reported patterns of blood glucose during the OGTT (Hulman et al., 2018). Green, class 1; light blue, class 2; dark blue, class 3; red, class 4. (C) Mean, Std, and AC_Var of the glucose during the OGTT. Colors are based on the class shown in Figure 4B. We could generate glucose fluctuations with almost the same standard deviation (Std) and AC_Var but with a different mean (Fig. 4A). Similarly, we could also simulate glucose fluctuations with almost the same mean and AC\_Var but different Std, and with almost the same mean and Std but different AC_Var. These three components were independently adjustable by changing parameters within the range of values for NGT individuals (Fig. S9B). Individuals with higher AC_Var tended to have higher %NC (Fig. 1); however, comparing the glucose dynamics with higher and lower AC_Var, the maximum value of blood glucose was lower in individuals with higher AC_Var (Fig. 4A). To facilitate a more comprehensive exploration of the three components of glucose dynamics, we developed a web-based application ([https://simulator-glucose.streamlit.app/](https://simulator-glucose.streamlit.app/)) using a more detailed mathematical model (Dalla Man et al., 2007). This model incorporates additional physiological parameters and allows for a more comprehensive simulation of glucose-insulin interactions. The web application allows users to manipulate a wide range of parameters, including insulin sensitivity, beta cell function, and insulin clearance. By systematically varying these parameters, users can examine how the Mean, Std, and AC_Var of glucose concentrations change under different conditions. We also investigated the relationship between the three components and shapes of the glucose response curve after OGTT. Patterns of the glucose response curve after OGTT were heterogeneous, and four distinct patterns, denoted class 1–4, were previously identified by analyzing the glucose dynamics of 5861 individuals in Denmark (Fig. 4B) (Hulman et al., 2018). The classes with a high mean did not necessarily have high Std or AC_Var (Fig. 4C). The classes with high Std did not necessarily have high mean and AC_Var (Fig. 4C), consistent with the result that mean, Std, and AC\_Var had low multicollinearity with each other. Compared to class 2, only class 1 was lower in mean and Std, and was higher in AC\_Var. Compared to class 2, only class 3 was higher in mean, Std, and AC\_Var. Compared to class 2, only class 4 was higher in mean and Std, and lower in AC\_Var. Collectively, the three components could characterize the previously reported four distinct patterns during OGTT. Class 3 was characterized by normal FBG and PG120 values, but is reportedly associated with increased risk of diabetes and higher all-cause mortality rate, suggesting that subgroups at high risk may not be identified by investigating only FBG and PG120 (Hulman et al., 2018). Std and AC_Var were high in class 3 (Fig. 4C), suggesting that high Std and high AC_Var indicate glycemic disability independent of FBG and PG120. ## DISCUSSION Here, we identified three distinct components of glucose dynamics: value, variability, and autocorrelation, each independently associated with coronary plaque vulnerability. We previously reported that AC\_Var, an index reflecting autocorrelation, can detect decreased abilities in glucose regulation independently of other CGM-derived indices including CGM\_Mean and CGM_Std, which reflect value and variability components, respectively (Sugimoto et al., 2023). Diabetes diagnosis has been based on elevated FBG, PG120, and HbA1c levels. However, these indices primarily reflect only the value component of glucose dynamics, and consequently the predictive performance of the prediction model for %NC using FBG, PG120, and HbA1c was relatively modest compared to that of the model using all three components of glucose dynamics. This result is partially consistent with a previous notion that glucose dynamics include two components: amplitude and timing (Cobelli and Facchinetti, 2018). Collectively, CGM-derived indices reflecting the three components of glucose dynamics – value, variability, and autocorrelation – can outperform indices used to diagnose diabetes reflecting only the value component in predicting glycemic control capacity and coronary plaque vulnerability. We also showed that CGM-derived indices, especially ADRR and AC_Var, contribute to the prediction of %NC by using LASSO and PLS (Fig. 2). Given that the definition equation for ADRR is affected by both high and low concentrations of blood glucose (Hill et al., 2011), it is likely affected by both glucose concentration values and glycemic variability. Factor analysis (Fig. 3) also showed that ADRR was included in both factor 1 (value) and factor 2 (variability). Since three factors, value, variability, and autocorrelation, contribute independently to the prediction of the complication, it would be useful to examine ADRR, which is influenced by both value and variability, and AC\_Var, which is influenced by autocorrelation, in predicting %NC with a minimal number of variables. Therefore, the result of the LASSO showing that ADRR and AC\_Var are particularly effective in predicting %NC is consistent with the results of the factor analysis that the three components contribute to the prediction. This study also provided evidence that autocorrelation can vary independently from the value and variability components by using simulated data. As shown in Figure 4, these three components could be varied independently by simply changing the parameters related to glucose regulation within the range of NGT individuals. In addition, simulated glucose dynamics indicated that even subjects with high AC_Var did not necessarily have high maximum and minimum blood glucose levels. This study also indicated that these three components qualitatively corresponded to the four distinct glucose patterns observed after glucose administration, which were identified in a previous study (Hulman et al., 2018). Glycemic variability is involved in T2DM complications by oxidative stress and endothelial dysfunction (Ceriello et al., 2008; Monnier et al., 2008); however, the reasons why the three components, especially autocorrelation, independently contribute to the prediction remain unknown. The underlying biological mechanisms and effects of the three components on living systems need to be investigated in future studies. The current study had several limitations. LASSO and factor analysis indicated that CGM-related features were particularly important in predicting %NC. However, these results do not mean that other clinical parameters do not associate with T2DM complications, because we only included subjects with well-controlled serum cholesterol and blood pressure levels in this study. A previous study identified components of interday variability and hypoglycemia in CGM-derived indices (Augstein et al., 2015) that were not observed in our analysis. This discrepancy may be due to the relatively small number of T2DM subjects in our study. We acknowledge that factor analyses of data from longer measurement periods, including more patients with T1DM and T2DM, could potentially yield different results. However, it is noteworthy that our analysis of longer-term CGM data sets from Japanese and American populations confirmed the existence of the same three factors - value, variability, and autocorrelation - in glucose dynamics. Moreover, even with the short measurement period, CGM-derived indices reflecting these three factors demonstrated superior predictive accuracy for %NC compared to traditional indices such as FBG, HbA1c, and PG120, underscoring the potential utility of CGM. Although time in range (TIR) was not included in the main analyses due to the small number of T2DM patients and the large number of patients with TIR greater than 70%, the CGM-derived indices still outperformed FBG, HbA1c, and PG120 in predicting %NC. The multiple regression analysis between factor scores and TIR showed that only factors 1 (value) and 2 (variability) were significantly associated with TIR (Fig. S10), further supporting the existence of three components in glucose dynamics and the potential value of examining AC_Var in addition to other CGM-derived indices. We acknowledge the potential concern of multiple testing in our study. However, even after adjusting for multiple comparisons, the CGM-derived indices retained significant correlations with %NC (Fig. S1). The consistency of our findings across different analytical approaches (Lasso, PLS, and factor analysis) and different data sets further supports the robustness of our conclusions regarding the characteristics of glucose dynamics. While we used methods that assume linearity, such as LASSO, we also examined nonlinear relationships using Spearman’s correlation for index relationships and factor loadings with %NC, and found significant associations. We acknowledge that some significant correlations appear to be relatively small. However, these findings, combined with our predictive models showing improved accuracy using CGM compared to traditional diabetes diagnostic indices, and the theoretical framework showing that conventional markers only consider the “value” component of glucose dynamics, can demonstrate the clinical significance. Finally, although we analyzed three different datasets with a total of 270 subjects, the sample size may still be considered relatively small to comprehensively examine the relationships between the variables in this study. Larger, prospective studies are needed to provide a more accurate assessment of these variables in predicting abnormalities and to further validate our findings. In conclusion, glucose dynamics has three components: value, variability, and autocorrelation. These three components are associated with coronary plaque vulnerability. CGM-derived indices reflecting these three components can be valuable predictive tools for T2DM complications, compared to conventional diabetes diagnostic markers reflecting only the value component. This new predictive model has the potential to improve the diagnosis and management of diabetes worldwide. To facilitate this CGM-derived prediction, we created a web application that performs a multiple regression model with these three components as input variables ([https://cgm-basedregression.streamlit.app/](https://cgm-basedregression.streamlit.app/)). ## METHODS ### Key Resources Table View this table: [Table1](http://medrxiv.org/content/early/2024/09/02/2023.11.21.23298816/T1) ### Subjects and measurements This was a retrospective observational study approved by the ethics committee of Kobe University Graduate School of Medicine (UMIN000018326; Kobe, Japan), as described previously (Otowa-Suematsu et al., 2018). The study included 53 participants who underwent a 75-g oral glucose tolerance test (OGTT), continuous glucose monitoring (CGM) with the use of an iPro2 CGM system (Medtronic, Northridge, CA, USA), and percutaneous coronary intervention (PCI). During PCI, VH-IVUS was carried out to assess the plaque components. Among the 53 participants, eight, 16, and 29 individuals were categorized as having normal glucose tolerance (NGT), impaired glucose tolerance (IGT) and T2DM, respectively. Of note, with a type I error of 0.05, a power of 0.8, and an expected correlation coefficient of 0.4, a sample size of 47 was required to detect a significant difference from zero in the correlation coefficient. Detailed participant characteristics have been reported in the previous study (Otowa-Suematsu et al., 2018). Briefly, participants aged 20–80 years with LDL-C levels < 120 mg/dL under statin administration or < 100 mg/dL under other treatments for dyslipidemia, including lifestyle intervention, were included in this study. Participants with acute coronary syndrome, unsuitable anatomy for virtual VH-IVUS, poor imaging by VH-IVUS, hemodialysis, inflammatory disease, shock, low cardiac output, or concurrent malignant disease were excluded from this study. To validate the findings on glucose dynamics, the present study analyzed a previously reported data set from Japan (Sugimoto et al., 2023). The study was conducted in accordance with the Declaration of Helsinki and approved by the Kobe University Hospital Ethics Committee (Approval No. 1834). Briefly, individuals aged ≥ 20 years with no previous diagnosis of diabetes were recruited from Kobe University Hospital between January 2016 and March 2018. Exclusion criteria included use of medications that affect glucose metabolism (e.g., steroids, β-blockers), psychiatric disorders, pregnancy or lactation, and ineligibility as determined by treating physicians. Participants wore a CGM device (iPro; Medtronic, USA). The study included 52 individuals with NGT, nine with IGT, and three with T2DM. Further validation of the glucose dynamics findings was performed using a previously reported CGM dataset (Dexcom G4 CGM System; Dexcom, Fort Lauderdale, FL, USA) obtained in the United States (Hall et al., 2018). The study included 53 individuals with no previous diagnosis of diabetes. In addition, the present study analyzed a previously reported CGM (FreeStyle Libre H, Abbott Diabetes Care, Witney, UK) dataset from China (Zhao et al., 2023). Participants were recruited from the DiaDRIL registry at Shanghai East Hospital (September 2019 to March 2021) and Shanghai Fourth People’s Hospital (June 2021 to November 2021). Inclusion criteria for this dataset were: diagnosis of diabetes according to the 1999 World Health Organization (WHO) criteria, age 18 years or older, willingness to provide informed consent, and CGM recording for at least three days. Exclusion criteria included reported alcohol or drug abuse, inability to comply with study protocols, or inability to participate as determined by the investigators. We extracted and analyzed glucose profile characteristics from the first three days of CGM data for each participant. 100 individuals with T2DM were analyzed in this study. Our primary objective was to explore the relationship between the CGM-derived indices and the presence of diabetic macrovascular complications. ### Calculation of clinical indices #### CGM-derived indices The CGM data collected in this study were obtained at different sampling frequencies in the different data sets. For the two Japanese cohorts and the American cohort, glucose measurements were collected at 5-minute intervals. In contrast, the Chinese cohort collected glucose measurements at 15-minute intervals. Fourteen CGM-derived indices were evaluated in this study: twelve well-established CGM-derived indices (Hill et al., 2011) and two indices (AC\_Mean and AC\_Var) that have been reported to capture glucose handling capacity independently of the aforementioned twelve indices (Sugimoto et al., 2023). For the datasets with 5-minute sampling intervals (the two Japanese datasets and the American dataset), AC\_Mean and AC\_Var were calculated as the mean and variance of the autocorrelation functions at lags 1-30 of the glucose levels, respectively. For the data set with 15-minute sampling intervals (the Chinese data set), AC\_Mean and AC_Var were calculated as the mean and variance of the autocorrelation functions at lags 1-10 of the glucose levels, respectively. These calculations correspond to a time window of 150 minutes. CGM_Mean and CGM_Std indicate the mean value and standard deviation of glucose levels measured by CGM, respectively. CONGA, LI, JINDEX, HBGI, GRADE, MODD, MAGE, ADRR, MVALUE, and MAG were calculated using EasyGV software (Hill et al., 2011). The calculating formulae of these indices are shown in Table S1. In the two Japanese datasets and the American dataset, there was a relatively small proportion of T2DM patients and a substantial number of subjects with time in range (TIR) (Battelino et al., 2023; Larkin et al., 2019) values (glucose levels between 70 and 180 mg/dL) greater than 70% (Hall et al., 2018; Otowa-Suematsu et al., 2018; Sugimoto et al., 2023). Given this distribution, we decided not to include TIR in our primary analyses. However, given the potential interest of this metric, we provided correlations between TIR and other indices as supplementary information. #### OGTT-derived indices Three OGTT-derived indices were calculated as previously described (Matsuda and DeFronzo, 1999; Otowa-Suematsu et al., 2018). The insulinogenic index (I.I.) indicates insulin secretion and is calculated from the ratio of the increment of immunoreactive insulin (IRI) to that of plasma glucose at 30 min after onset of the OGTT. The composite index indicates insulin sensitivity, which can be calculated from fasting plasma glucose, fasting IRI, mean blood glucose levels, and mean serum IRI concentrations during the OGTT. The oral disposition index (Oral DI) was calculated from the product of the composite index and the ratio of the area under the insulin concentration curve from 0 to 120 minutes to that for plasma glucose from 0 to 120 minutes, without using the data measured at 90 min, in the OGTT. #### VH-IVUS-derived index VH-IVUS was carried out using the Eagle Eye Platinum 3.5-Fr 20-MHz catheter (Volcano, Rancho Cordova, CA, USA), as previously described (Otowa-Suematsu et al., 2018). The intraclass correlation coefficients for interobserver and intraobserver reliability of external elastic membrane volume were 0.95 and 0.97, respectively (Otowa-Suematsu et al., 2018), indicating high reproducibility. The VH-IVUS categorized plaque into four components: fibrous, fibrofatty, necrotic core, and dense calcium. Following the previous study, our investigation focused specifically on the ratio of necrotic core to total plaque volume (%NC), a widely used parameter of plaque vulnerability. For patients with multiple plaques, the mean %NC was calculated. ### Prediction models and statistical analyses In this study, we conducted multiple linear regression, LASSO regression, and PLS regression. The input variables in these models included the following 26 variables: BMI, SBP, DBP, TGs, LDL-C, HDL-C, FBG, HbA1c, PG120, I.I., composite index, oral DI, CGM\_Mean, CGM\_Std, CONGA, LI, JINDEX, HBGI, GRADE, MODD, MAGE, ADRR, MVALUE, MAG, AC\_Mean, and AC\_Var. In conducting these models, z-score normalization on each input variable was performed. The predictive performance of multiple linear regression was evaluated by the coefficient of determination (*R*2), the adjusted coefficient of determination (Adj *R*2), or AIC. The multicollinearity of the input variables was estimated by VIF. LASSO regression is a kind of linear regression with L1 regularization (Tibshirani, 1996). The optimal regularization coefficient, lambda, was based on leave-one-out cross-validation and mean-squared error (MSE). The importance of the input variables in predicting %NC was evaluated by the VIP scores (Wold et al., 2001) that were generated from PLS regression. These models were conducted using scikit-learn v1.0.2, a Python-based toolkit ([https://scikit-learn.org/stable/](https://scikit-learn.org/stable/)). Relationships among indices were also evaluated using Spearman’s correlation coefficients (*r*), and the correlation coefficients were reported with 95% CIs through bootstrap resampling. The number of resamples performed to form the distribution was set at 10000. Benjamini–Hochberg’s multiple comparison test was also performed with a significance threshold of Q < 0.05. ### Factor analysis and hierarchical clustering analysis The intercorrelations of the clinical parameters and their associations with %NC were assessed using exploratory factor analyses and hierarchical clustering analyses. We followed the previously reported approach (Cappelleri et al., 2000; Guo et al., 2020; Lakka et al., 2002; Oh et al., 2004) with some modifications in conducting our exploratory factor analyses. BIC and MAP methods were used to determine the number of underlying factors. Variables with factor loadings of ≥ 0.30 were used in interpretation. To improve the interpretation, orthogonal (varimax) rotation was used. To evaluate the applicability of the factor analysis, KMO and Bartlett’s spherical test were performed. To evaluate internal consistency of each factor, Cronbach’s α and item–total correlations were calculated. The association of the factor scores with %NC was assessed using Spearman’s correlation. Hierarchical clustering analysis was conducted using a method that combines a Euclidean distance measure and Ward linkage. I.I., composite index, oral DI, and AC_Mean were inverted negatively so that the value of indices increased in subjects with abnormalities. The quality of the hierarchical clustering analysis was evaluated based on silhouette analysis (Rousseeuw, 1987). These analyses were performed after Z score normalization using scikit-learn v1.0.2, a Python-based toolkit ([https://scikit-learn.org/stable/](https://scikit-learn.org/stable/)). ### Mathematical models used for simulating the characteristics of glucose dynamics In simulating the characteristics of glucose dynamics, we used a simple and stable model (De Gaetano and Arino, 2000), which can be written as follows: ![Formula][1] where the variables *G* and *I* denote blood glucose and insulin concentrations, respectively. We simulated 240-minutes profiles of *G*, and calculated the mean, Std, and AC_Var of *G*. The parameters were changed within the range participants could take (De Gaetano and Arino, 2000). Five mg/dL/min glucose was applied for 10 minutes at 30 minutes as the external input of glucose *f*. For more detailed simulations of the postprandial response, we used a complex, fine-grained model (Dalla Man et al., 2007). Briefly, the model consists of two interconnected subsystems: the glucose subsystem and the insulin subsystem. The glucose subsystem is represented by a two-compartment model. The first compartment (*G*p) represents glucose in plasma and rapidly equilibrating tissues, while the second compartment (*G*t) represents glucose in slowly equilibrating tissues. The dynamics of this subsystem is described by the differential equations, as follows: ![Formula][2] In these equations, *k*1 and *k*2 are rate constants, EGP is endogenous glucose production, Ra is the rate of glucose appearance from meal absorption, *U*ii is insulin-independent glucose utilization, *E* is renal excretion, *U*id is insulin-dependent glucose utilization, *G* is plasma glucose concentration, and *V*G is the distribution volume of glucose. The insulin subsystem is also modeled with two compartments, one for plasma insulin (*I*p) and another for liver insulin (*I*l). The dynamics of this subsystem is described by the differential equations, as follows: ![Formula][3] where, S is the insulin secretion rate, *m*1, *m*2, and *m*4 are rate parameters, *m*4 is peripheral insulin clearance. For simplicity, we assumed that insulin sensitivity and beta cell responsivity to glucose are constant. We simulated 48-hour profiles of *G*, and calculated the Mean, Std, and AC_Var of *G*. The code that simulates the glucose dynamics is available from the repository ([https://github.com/HikaruSugimoto/Simulator](https://github.com/HikaruSugimoto/Simulator)) and the web application ([https://simulator-glucose.streamlit.app/](https://simulator-glucose.streamlit.app/)). The simulations were conducted using SciPy v1.10.1 (Virtanen et al., 2020). ### Characterization of glucose patterns during the OGTT We investigated the characteristics of previously reported glucose patterns during the OGTT (Hulman et al., 2018). In the study, 5861 subjects without diabetes in Denmark underwent the OGTT with measurements of glucose levels at three time points (0, 30, and 120 min), and four distinct glucose patterns associated with long-term outcomes including diabetes onset, cardiovascular disease, and all-cause mortality rate were identified. For the calculation of mean, Std, and AC\_Var of glucose levels, each time point was linearly imputed. Here, AC\_Var was calculated from the autocorrelation function at lags 1–20, as we had glucose data available for only 2 h after the OGTT. ## Supporting information Supplementary file [[supplements/298816_file07.pdf]](pending:yes) ## Data Availability The CGM data that support the findings of this study are available from the GitHub repository ([https://github.com/HikaruSugimoto/CGM\_regression_app](https://github.com/HikaruSugimoto/CGM_regression_app)). ## Author Contributions H.S. analyzed the data. H.S., K.H., T.Y., N.O.S., Y.H., H.O., K.H., K.S., W.O., and S.K. wrote the manuscript. W.O. and S.K. supervised the study. ## Conflict of Interest The authors have no conflicts of interest to declare. ## Funding and Assistance This study was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (JP21H04759), CREST, the Japan Science and Technology Agency (JST) (JPMJCR2123), The Uehara Memorial Foundation, and The Takeda Science Foundation. ## Data and code availability We used only previously published data sets. The code that calculates AC\_Mean and AC\_Var and that performs regression analysis with CGM-derived indices as input variables are available from the repository ([https://github.com/HikaruSugimoto/CGM\_regression_app](https://github.com/HikaruSugimoto/CGM_regression_app)) and the web application ([https://cgm-basedregression.streamlit.app/](https://cgm-basedregression.streamlit.app/)). ## Acknowledgments ## Footnotes * 5 Lead contact * Supplementary files and Methods section updated. * Received November 21, 2023. * Revision received September 1, 2024. * Accepted September 2, 2024. * © 2024, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution 4.0 International), CC BY 4.0, as described at [http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/) ## References 1. Augstein P, Heinke P, Vogt L, Vogt R, Rackow C, Kohnert K-D, Salzsieder E. 2015. Q-Score: development of a new metric for continuous glucose monitoring that enables stratification of antihyperglycaemic therapies. BMC Endocr Disord 15:22. 2. Battelino T, Alexander CM, Amiel SA, Arreaza-Rubin G, Beck RW, Bergenstal RM, Buckingham BA, Carroll J, Ceriello A, Chow E, Choudhary P, Close K, Danne T, Dutta S, Gabbay R, Garg S, Heverly J, Hirsch IB, Kader T, Kenney J, Kovatchev B, Laffel L, Maahs D, Mathieu C, Mauricio D, Nimri R, Nishimura R, Scharf M, Del Prato S, Renard E, Rosenstock J, Saboo B, Ueki K, Umpierrez GE, Weinzimer SA, Phillip M. 2023. Continuous glucose monitoring and metrics for clinical trials: an international consensus statement. Lancet Diabetes Endocrinol 11:42–57. 3. Bax JJ, Young LH, Frye RL, Bonow RO, Steinberg HO, Barrett EJ, ADA. 2007. Screening for coronary artery disease in patients with diabetes. Diabetes Care 30:2729–2736. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czoxMDoiMzAvMTAvMjcyOSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 4. Cai J, Yang Q, Lu J, Shen Y, Wang C, Chen L, Zhang L, Lu W, Zhu W, Xia T, Zhou J. 2022. Impact of the complexity of glucose time series on all-cause mortality in patients with type 2 diabetes. J Clin Endocrinol Metab. doi:10.1210/clinem/dgac692 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1210/clinem/dgac692&link_type=DOI) 5. Cappelleri JC, Gerber RA, Kourides IA, Gelfand RA. 2000. Development and factor analysis of a questionnaire to measure patient satisfaction with injected and inhaled insulin for type 1 diabetes. Diabetes Care 23:1799–1803. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czoxMDoiMjMvMTIvMTc5OSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 6. Ceriello A, Esposito K, Piconi L, Ihnat MA, Thorpe JE, Testa R, Boemi M, Giugliano D. 2008. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes 57:1349–1354. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiZGlhYmV0ZXMiO3M6NToicmVzaWQiO3M6OToiNTcvNS8xMzQ5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDkvMDIvMjAyMy4xMS4yMS4yMzI5ODgxNi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 7. Cobelli C, Facchinetti A. 2018. Yet Another Glucose Variability Index: Time for a Paradigm Change? Diabetes Technol Ther 20:1–3. 8. Dalla Man C, Rizza RA, Cobelli C. 2007. Meal simulation model of the glucose-insulin system. IEEE Trans Biomed Eng 54:1740–1749. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1109/TBME.2007.893506&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17926672&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000249840500002&link_type=ISI) 9. De Gaetano A, Arino O. 2000. Mathematical modelling of the intravenous glucose tolerance test. J Math Biol 40:136–168. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s002850050007&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=10743599&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000085695600002&link_type=ISI) 10. Fabris C, Facchinetti A, Fico G, Sambo F, Arredondo MT, Cobelli C, MOSAIC EU Project Consortium. 2015. Parsimonious Description of Glucose Variability in Type 2 Diabetes by Sparse Principal Component Analysis. J Diabetes Sci Technol 10:119–124. 11. Fabris C, Facchinetti A, Sparacino G, Zanon M, Guerra S, Maran A, Cobelli C. 2014. Glucose variability indices in type 1 diabetes: parsimonious set of indices revealed by sparse principal component analysis. Diabetes Technol Ther 16:644–652. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1089/dia.2013.0252&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24956070&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 12. Fiarni C, Sipayung EM, Maemunah S. 2019. Analysis and Prediction of Diabetes Complication Disease using Data Mining Algorithm. Procedia Comput Sci 161:449–457. 13. Gerbaud E, Darier R, Montaudon M, Beauvieux M-C, Coffin-Boutreux C, Coste P, Douard H, Ouattara A, Catargi B. 2019. Glycemic Variability Is a Powerful Independent Predictive Factor of Midterm Major Adverse Cardiac Events in Patients With Diabetes With Acute Coronary Syndrome. Diabetes Care 42:674–681. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo4OiI0Mi80LzY3NCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 14. Gorst C, Kwok CS, Aslam S, Buchan I, Kontopantelis E, Myint PK, Heatlie G, Loke Y, Rutter MK, Mamas MA. 2015. Long-term Glycemic Variability and Risk of Adverse Outcomes: A Systematic Review and Meta-analysis. Diabetes Care 38:2354–2369. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czoxMDoiMzgvMTIvMjM1NCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 15. Guo W, Zhou Q, Jia Y, Xu J. 2020. Cluster and Factor Analysis of Elements in Serum and Urine of Diabetic Patients with Peripheral Neuropathy and Healthy People. Biol Trace Elem Res 194:48–57. 16. Hall H, Perelman D, Breschi A, Limcaoco P, Kellogg R, McLaughlin T, Snyder M. 2018. Glucotypes reveal new patterns of glucose dysregulation. PLoS Biol 16:e2005143. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 17. Hill NR, Oliver NS, Choudhary P, Levy JC, Hindmarsh P, Matthews DR. 2011. Normal reference range for mean tissue glucose and glycemic variability derived from continuous glucose monitoring for subjects without diabetes in different ethnic groups. Diabetes Technol Ther 13:921–928. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1089/dia.2010.0247&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21714681&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000294222400007&link_type=ISI) 18. Hulman A, Vistisen D, Glümer C, Bergman M, Witte DR, Færch K. 2018. Glucose patterns during an oral glucose tolerance test and associations with future diabetes, cardiovascular disease and all-cause mortality rate. Diabetologia 61:101–107. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00125-017-4468-z&link_type=DOI) 19. Keshet A, Shilo S, Godneva A, Talmor-Barkan Y, Aviv Y, Segal E, Rossman H. 2023. CGMap: Characterizing continuous glucose monitor data in thousands of non-diabetic individuals. Cell Metab 35:758–769.e3. 20. Lakka H-M, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, Salonen JT. 2002. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 288:2709–2716. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.288.21.2709&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=12460094&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000179549800030&link_type=ISI) 21. Larkin ME, Nathan DM, Bebu I, Krause-Steinrauf H, Herman WH, Higgins JM, Tiktin M, Cohen RM, Lund C, Bergenstal RM, Johnson ML, Arends V, GRADE Research Group. 2019. Rationale and Design for a GRADE Substudy of Continuous Glucose Monitoring. Diabetes Technol Ther 21:682–690. 22. Matsuda M, DeFronzo RA. 1999. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 22:1462–1470. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo5OiIyMi85LzE0NjIiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyNC8wOS8wMi8yMDIzLjExLjIxLjIzMjk4ODE2LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 23. Monnier L, Colette C, Owens DR. 2008. Glycemic variability: the third component of the dysglycemia in diabetes. Is it important? How to measure it? J Diabetes Sci Technol 2:1094–1100. 24. Oh J-Y, Hong YS, Sung Y-A, Barrett-Connor E. 2004. Prevalence and factor analysis of metabolic syndrome in an urban Korean population. Diabetes Care 27:2027–2032. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo5OiIyNy84LzIwMjciO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyNC8wOS8wMi8yMDIzLjExLjIxLjIzMjk4ODE2LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 25. Okada K, Hibi K, Gohbara M, Kataoka S, Takano K, Akiyama E, Matsuzawa Y, Saka K, Maejima N, Endo M, Iwahashi N, Tsukahara K, Kosuge M, Ebina T, Fitzgerald PJ, Honda Y, Umemura S, Kimura K. 2015. Association between blood glucose variability and coronary plaque instability in patients with acute coronary syndromes. Cardiovasc Diabetol 14:111. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s12933-015-0275-3&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26289581&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 26. Otowa-Suematsu N, Sakaguchi K, Komada H, Nakamura T, Sou A, Hirota Y, Kuroda M, Shinke T, Hirata K-I, Ogawa W. 2018. Comparison of the relationship between multiple parameters of glycemic variability and coronary plaque vulnerability assessed by virtual histology-intravascular ultrasound. J Diabetes Investig 9:610–615. 27. Pei X, Qi D, Liu J, Si H, Huang S, Zou S, Lu D, Li Z. 2023. Screening marker genes of type 2 diabetes mellitus in mouse lacrimal gland by LASSO regression. Sci Rep 13:6862. 28. Psoma O, Makris M, Tselepis A, Tsimihodimos V. 2022. Short-term Glycemic Variability and Its Association With Macrovascular and Microvascular Complications in Patients With Diabetes. J Diabetes Sci Technol 19322968221146810. 29. Ravaut M, Sadeghi H, Leung KK, Volkovs M, Kornas K, Harish V, Watson T, Lewis GF, Weisman A, Poutanen T, Rosella L. 2021. Predicting adverse outcomes due to diabetes complications with machine learning using administrative health data. NPJ Digit Med 4:24. 30. Rousseeuw PJ. 1987. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math 20:53–65. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/0377-0427(87)90125-7&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=WOS:A1987L11&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1987L111800005&link_type=ISI) 31. Selvin E, Steffes MW, Zhu H, Matsushita K, Wagenknecht L, Pankow J, Coresh J, Brancati FL. 2010. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 362:800–811. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1056/NEJMoa0908359&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20200384&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000275108300007&link_type=ISI) 32. Service FJ. 2013. Glucose variability. Diabetes 62:1398–1404. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiZGlhYmV0ZXMiO3M6NToicmVzaWQiO3M6OToiNjIvNS8xMzk4IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDkvMDIvMjAyMy4xMS4yMS4yMzI5ODgxNi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 33. Su G, Mi S, Tao H, Li Z, Yang H, Zheng H, Zhou Y, Ma C. 2011. Association of glycemic variability and the presence and severity of coronary artery disease in patients with type 2 diabetes. Cardiovasc Diabetol 10:19. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/1475-2840-10-19&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21349201&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 34. Sugimoto H, Hironaka K-I, Nakamura T, Yamada T, Miura H, Otowa-Suematsu N, Fujii M, Hirota Y, Sakaguchi K, Ogawa W, Others. 2023. Improved Detection of Decreased Glucose Handling Capacities via Novel Continuous Glucose Monitoring-Derived Indices: AC\_Mean and AC\_Var. medRxiv 2023–2009. 35. Tang X, Li S, Wang Y, Wang M, Yin Q, Mu P, Lin S, Qian X, Ye X, Chen Y. 2016. Glycemic variability evaluated by continuous glucose monitoring system is associated with the 10-y cardiovascular risk of diabetic patients with well-controlled HbA1c. Clin Chim Acta 461:146–150. 36. Tibshirani R. 1996. Regression shrinkage and selection via the lasso. J R Stat Soc 58:267–288. 37. Utzschneider KM, Prigeon RL, Faulenbach MV, Tong J, Carr DB, Boyko EJ, Leonetti DL, McNeely MJ, Fujimoto WY, Kahn SE. 2009. Oral disposition index predicts the development of future diabetes above and beyond fasting and 2-h glucose levels. Diabetes Care 32:335–341. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo4OiIzMi8yLzMzNSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 38. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, Burovski E, Peterson P, Weckesser W, Bright J, van der Walt SJ, Brett M, Wilson J, Millman KJ, Mayorov N, Nelson ARJ, Jones E, Kern R, Larson E, Carey CJ, Polat İ, Feng Y, Moore EW, VanderPlas J, Laxalde D, Perktold J, Cimrman R, Henriksen I, Quintero EA, Harris CR, Archibald AM, Ribeiro AH, Pedregosa F, van Mulbregt P, SciPy 1.0 Contributors. 2020. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 17:261–272. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41592-019-0686-2&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=32015543&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 39. Wang C, Kong H, Guan Y, Yang J, Gu J, Yang S, Xu G. 2005. Plasma phospholipid metabolic profiling and biomarkers of type 2 diabetes mellitus based on high-performance liquid chromatography/electrospray mass spectrometry and multivariate statistical analysis. Anal Chem 77:4108–4116. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1021/ac0481001&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15987116&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) 40. Wei H, Sun J, Shan W, Xiao W, Wang B, Ma X, Hu W, Wang X, Xia Y. 2022. Environmental chemical exposure dynamics and machine learning-based prediction of diabetes mellitus. Sci Total Environ 806:150674. 41. Wold S, Sjöström M, Eriksson L. 2001. PLS-regression: a basic tool of chemometrics. Chemometrics Intellig Lab Syst 58:109–130. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0169-7439(01)00155-1&link_type=DOI) 42. Young LH, Wackers FJT, Chyun DA, Davey JA, Barrett EJ, Taillefer R, Heller GV, Iskandrian AE, Wittlin SD, Filipchuk N, Ratner RE, Inzucchi SE, DIAD Investigators. 2009. Cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial. JAMA 301:1547–1555. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.2009.476&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19366774&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F09%2F02%2F2023.11.21.23298816.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000265137900025&link_type=ISI) 43. Zhao Q, Zhu J, Shen X, Lin C, Zhang Y, Liang Y, Cao B, Li J, Liu X, Rao W, Wang C. 2023. Chinese diabetes datasets for data-driven machine learning. Sci Data 10:35. 44. Zhou JJ, Schwenke DC, Bahn G, Reaven P, VADT Investigators. 2018. Glycemic Variation and Cardiovascular Risk in the Veterans Affairs Diabetes Trial. Diabetes Care 41:2187–2194. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czoxMDoiNDEvMTAvMjE4NyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA5LzAyLzIwMjMuMTEuMjEuMjMyOTg4MTYuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) [1]: /embed/graphic-6.gif [2]: /embed/graphic-7.gif [3]: /embed/graphic-8.gif