Cardiovascular diseases (CVDs) represent the leading cause of global mortality, frequently exacerbated by the coexistence of type 2 diabetes mellitus. This study investigates glycemic variability through Continuous Glucose Monitoring (CGM) in a mixed cohort of cardiovascular patients (both diabetic and non-diabetic), aiming to define specific glucotypes and correlate them with the severity of vascular and inflammatory damage. Glycemic metrics were extracted from the CGM traces collected across the two phases (Home and Hospital), and multicollinearity was reduced using Variance Inflation Factor (VIF) analysis. Using the K-means clustering algorithm, the population was stratified into groups with distinct metabolic profiles, which were subsequently compared against angiographic scores (Gensini and SYNTAX) and inflammatory biomarkers (CRP and IL-18). The results demonstrate that the clustering approach effectively isolates markedly different glycemic profiles, highlighting that variability and acute fluctuations, while not showing a direct linear correlation with anatomical indices of coronary severity, are significantly associated with increased inflammatory markers.
Le malattie cardiovascolari (CVD) rappresentano la principale causa di mortalità globale, aggravate frequentemente dalla coesistenza del diabete mellito di tipo 2. Questo studio indaga la variabilità glicemica tramite il monitoraggio continuo del glucosio (CGM) in una coorte mista di pazienti cardiovascolari (diabetici e non), con l’obiettivo di definire specifici glucotipi e correlarli alla severità del danno vascolare e infiammatorio. A partire dai tracciati raccolti nelle due fasi (Home e Hospital), sono state estratte le metriche glicemiche e ridotta la multicollinearità mediante Variance Inflation Factor (VIF). Tramite l’algoritmo di clustering K-means, la popolazione è stata stratificata in gruppi con differente profilo metabolico, confrontati successivamente con gli score angiografici (Gensini e Syntax) e con i biomarcatori infiammatori (CRP e IL-18). I risultati dimostrano che l’approccio di clustering consente di isolare profili glicemici marcatamente differenti, evidenziando come la variabilità e le escursioni acute, pur non mostrando una correlazione lineare diretta con gli indici anatomici di gravità coronarica, si associno significativamente a un incremento dei marker infiammatori.
Esplorazione della relazione tra coronaropatia e complicanze tramite caratteristiche derivate dal monitoraggio continuo del glucosio attraverso metodologie di clustering.
DEL VECCHIO, DELFINA
2025/2026
Abstract
Cardiovascular diseases (CVDs) represent the leading cause of global mortality, frequently exacerbated by the coexistence of type 2 diabetes mellitus. This study investigates glycemic variability through Continuous Glucose Monitoring (CGM) in a mixed cohort of cardiovascular patients (both diabetic and non-diabetic), aiming to define specific glucotypes and correlate them with the severity of vascular and inflammatory damage. Glycemic metrics were extracted from the CGM traces collected across the two phases (Home and Hospital), and multicollinearity was reduced using Variance Inflation Factor (VIF) analysis. Using the K-means clustering algorithm, the population was stratified into groups with distinct metabolic profiles, which were subsequently compared against angiographic scores (Gensini and SYNTAX) and inflammatory biomarkers (CRP and IL-18). The results demonstrate that the clustering approach effectively isolates markedly different glycemic profiles, highlighting that variability and acute fluctuations, while not showing a direct linear correlation with anatomical indices of coronary severity, are significantly associated with increased inflammatory markers.| File | Dimensione | Formato | |
|---|---|---|---|
|
DelVecchio_Delfina.pdf
Accesso riservato
Dimensione
3.86 MB
Formato
Adobe PDF
|
3.86 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/116236