Association between continuous glucose monitoring-derived metrics and coronary plaque vulnerability: A retrospective exploratory analysis

Background: Impaired glucose homeostasis leads to numerous complications, with coronary artery disease (CAD) being a major contributor to healthcare costs worldwide. Because continuous glucose monitoring (CGM) captures multidimensional features of glucose regulation beyond average glycemia, we evaluated whether CGM-derived indices better predict coronary plaque vulnerability than conventional measures. Methods: We examined associations between CGM-derived indices and coronary plaque vulnerability assessed by virtual histology–intravascular ultrasound, focusing on the necrotic core (%NC) in humans. We analyzed 14 CGM-derived indices, including average daily risk ratio (ADRR) and autocorrelation-based metrics (AC_Mean and AC_Var), alongside commonly used measures, such as fasting blood glucose (FBG), hemoglobin A1c (HbA1c), and 120 min plasma glucose during oral glucose tolerance testing (PG120). Factor analysis was used to identify latent components underlying glucose dynamics and to relate these components to %NC. Findings were validated across independent datasets from Japan (n=64), the United States (n=53), and China (n=100). Results: CGM-derived indices, particularly ADRR and AC_Var, demonstrated stronger predictive capability for %NC than FBG, HbA1c, and PG120. Factor analysis identified three independent components of glucose dynamics: mean, variance, and autocorrelation, each showing an independent association with %NC. ADRR reflected both mean and variance components, whereas AC_Var primarily captured the autocorrelation component. In contrast, FBG, HbA1c, and PG120 primarily reflected the mean component alone and were, therefore, insufficient for %NC prediction. Conclusions: 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. Funding: 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.