Continuous glucose monitoring-derived time in range and CV are associated with altered tissue characteristics of the carotid artery wall in people with type 2 diabetes

Aims/hypothesis Previous studies have suggested that glucose variability may accelerate atherosclerosis progression in people with type 2 diabetes. Current guidelines recommend assessing glycaemic control using continuous glucose monitoring (CGM), which provides a comprehensive glycaemic profile to supplement HbA 1c measurement. However, the association between CGM-derived metrics and atherosclerosis progression is not entirely clear. Methods This exploratory study used baseline data and data obtained after 104 weeks from an ongoing prospective, multicentre, observational study. Six hundred study participants with type 2 diabetes and no apparent history of symptomatic cardiovascular disease underwent CGM and ultrasonographic atherosclerosis measurements of the carotid arteries, including the intima–media thickness (IMT) and grey-scale median (GSM), at baseline and 104 weeks. Non-invasive ultrasonic tissue characterisation of the carotid artery wall or plaque using the GSM reflects vascular composition. Multivariate regression models were used to analyse the association between CGM-derived indices, mainly time in range (TIR) and CV, and changes in carotid atherosclerosis index values. Results Over the 104-week study period, there were modest increases in mean IMT (from 0.759±0.153 to 0.773±0.152 mm, p <0.001) and thickened-lesion GSM (from 43.5±19.5 to 53.9±23.5 units, p <0.001), but no significant changes in common carotid artery maximum-IMT (from 1.109±0.442 to 1.116±0.469 mm, p =0.453) or mean GSM (from 48.7±19.3 to 49.8±20.8 units, p =0.092). In a linear regression model with adjustment for possible atherosclerotic risk factors, including HbA 1c , TIR and CV at baseline were significantly associated with the annual change in mean GSM (regression coefficient per 10% increase in TIR 0.52; 95% CI 0.06, 0.98; Hochberg-adjusted p value 0.038; regression coefficient per 1% increase in CV −0.12; 95% CI −0.22, −0.02; Hochberg-adjusted p value 0.038). TIR and CV at baseline were also significantly associated with the annual change in thickened-lesion GSM (regression coefficient per 10% increase in TIR 0.95; 95% CI 0.12, 1.79; Hochberg-adjusted p value 0.038; regression coefficient per 1% increase in CV −0.19; 95% CI −0.36, −0.01; Hochberg-adjusted p value 0.038). Participants who achieved target CGM-derived metrics at baseline, as proposed by an international consensus, showed significant annual changes in mean GSM compared with those who did not (0.94±6.88 vs −0.21±6.19 units/year, p =0.007). Conclusions/interpretation TIR and CV were significantly associated with changes in the tissue characteristics of the carotid artery wall. Trial registration University Hospital Medical Information Network Clinical Trials Registry, number UMIN000032325 Graphical Abstract