Relationships between glucose variability with white matter hyperintensity and cerebrovascular abnormalities.

OBJECTIVE

Epidemiological studies have revealed that glucose variability (GV) is a predictor of stroke, cognitive impairment, and dementia in patients with type 2 diabetes mellitus (T2DM). However, evidence on the associations of GV with white matter hyperintensity (WMH) and cerebrovascular abnormalities remains scarce. This study aimed to explore the relationships of GV with WMH and cerebrovascular abnormalities using epidemiological and Mendelian randomization (MR) approaches. The MR approach was used to assess the effects of genetic proxies for GV on MRI outcomes.

SUBJECTS AND

METHODS

This cross-sectional study was conducted at a medical center where patients with T2DM were recruited. The measures for fasting plasma glucose (FPG) and HbA1c variability included the standard deviation, coefficient of variation, average real variability (ARV), and variability independent of the mean (VIM). Brain magnetic resonance images were analyzed to assess WMHs and cerebrovascular abnormalities. For MR, instrumental variables were used to assess the causal relationships between glycemic variability and outcome based on two-stage regression analysis.

RESULTS

This study included 2,247 subjects, of whom 1,122 had WMH and 957 had cerebrovascular abnormalities. We found 80 independent single-nucleotide polymorphisms associated with GV but not with WMH or cerebrovascular abnormalities, which were subsequently used as genetic instruments. Genetically increased, unweighted FPG-VIM was linked with WMH (odds ratio 1.17 [95% CI 1.08, 1.27] per standard deviation). All genetically increased, unweighted and weighted GV measures were associated with cerebrovascular abnormalities, except FPG-ARV.

CONCLUSION

Our study provided evidence that genetically predicted GV was associated with WMH and cerebrovascular abnormalities, supporting a potential causal link under MR assumptions.