Abstract MP40: Title: Investigating the link between hypoglycemia assessed by continuous glucose monitoring with cerebral small vessel disease

Introduction: Continuous glucose monitoring (CGM) is a minimally invasive technology that can assess glucose levels every few minutes, allowing for the characterization of nuanced glucose patterns such as subclinical hypoglycemia. However, few studies have examined whether subclinical episodes of hypoglycemia, as assessed by CGMs, are associated with cerebrovascular alterations in older adults with diabetes. Hypothesis: CGM-assessed hypoglycemia is associated with a greater prevalence of lobar microhemorrhage, subcortical infarction, cortical infarction, and increased volume of white matter hyperintensities. Methods: We conducted a cross-sectional study of participants with type 2 diabetes in the Atherosclerosis Risk in Communities (ARIC) Study. Participants wore a CGM (Abbott Libre Pro) sensor for up to two weeks during Visit 9 (2021-2022). A 3 Tesla brain MRI scan was obtained during Visits 8 to 11 (2020-2024). We evaluated the prevalence of any lobar microhemorrhage, subcortical infarction, cortical infarction, and median white matter hyperintensity volume according to the presence of hypoglycemia, defined as any episode of sustained glucose (≥ 30 minutes) below 54 mg/dL (level 2 hypoglycemia) during the two weeks of CGM wear. Prevalence estimates were generated using logistic regression. Median white matter hyperintensity volume was estimated using quantile regression. Estimates were adjusted for age, sex, race, and HbA1c. White matter hyperintensity was additionally adjusted for total intracranial volume. Results: We included 138 participants with diabetes (mean age 82 years, 59% White adults, 51% Male). Among participants, 15%, 20%, and 32% had a cortical infraction, lobar microhemorrhage, and subcortical microhemorrhage, respectively. A total of 40 participants experienced a hypoglycemic episode. The adjusted prevalence of subcortical infarction was significantly higher among persons experiencing one or more hypoglycemic episodes (p=0.021) (Figure). The adjusted prevalences of lobar microhemorrhage, cortical infarction, and white matter hyperintensity volume were also higher among persons experiencing hypoglycemia, but these differences were not significant. Conclusions: CGM-assessed hypoglycemia was associated with an increased burden of cerebrovascular alterations. Our findings indicate that these subclinical hypoglycemic episodes could play a role in linking diabetes to changes in the brain.