P3‐119: Higher blood glucose is associated with dementia among Chinese elderly independent of vascular risk factors and white matter hyperintensity volume: findings from SCOBHI
James Mortimer,Amy Borenstein,Ding Ding,Cathleen Copenhaver,Yougui Wu,Douglas Galasko,David Salmon,Qianhua Zhao,Qihao Guo,Charles DeCarli,Qi Dai,Ronald Petersen,Zhen Hong
DOI: https://doi.org/10.1016/j.jalz.2009.04.1094
2009-01-01
Abstract:Diabetes has been identified as a risk factor for AD as well as cognitive decline. The association between diabetes and AD is unclear, with both direct effects on beta amyloid and indirect effects through vascular disease possible. We report the findings of a cross-sectional study in Shanghai, China, comparing cases of dementia with cases of amnestic MCI (aMCI) and normal controls. An age- and sex-matched sample of 34 dementia patients, 30 aMCI patients and 32 normal controls (mean age =74.0, 49% female) was given MRI scans, a comprehensive neuropsychological battery, neurological exams, and an extensive risk factor evaluation. Logistic regression analyses were used to compare aMCI patients with normals as well as dementia patients with aMCI patients. Independent variables included age, sex, educational level, modified Hachinski ischemic score (a summary measure of vascular risk), WMHI volume, and fasting blood glucose. Additional linear regressions were run to assess the association of blood glucose with measures of memory, executive function, and psychomotor speed. Controlling for age, sex, education, Hachinski score and WMHI volume; higher blood glucose was associated with dementia vs. aMCI status (OR= 2.88; 95%CI: 1.17-7.10). Removal of 3 cases meeting NINDS-AIREN criteria for probable vascular dementia had little effect on the association (OR=3.16, 95%CI: 1.29-7.76). Similar analyses comparing aMCI vs. normal control status did not show a significant association (OR= 0.59, 95%CI: 0.28-1.24). Higher glucose was associated with Mattis Inititation Score (p=.02), Mattis Total Score (p=.03), Trails B time (p=.03) and Bell cancellation test score (p=.04), but not with other neuropsychological measures including scores on the Stroop test, WAIS-Similarities, Boston Naming Test, Clock Drawing Test, and several measures of verbal and visual memory. Higher blood glucose was associated with increased risk of dementia in comparison with aMCI, but was not associated with risk of aMCI vs. normal control status. Examination of neuropsychological data suggested that the increased risk may be related to poorer performance on tests of perceptual speed as previously reported (Arvanitakis et al., 2004). The results suggest that treatment of diabetes in individuals in early stages of AD may be useful in slowing progression.