P2-096: Ischemic lesion determines dementia outcome in Alzheimer’s disease with mixed pathology

Abstract

Alzheimer's disease (AD) and vascular dementia (VaD) behave differently in neuropsychology and disease progression. However, cerebral ischemic pathology commonly coexists in AD, and studies suggest the cognitive profile in mixed dementia resembles VaD rather than AD. The impact of ischemic lesion on the clinical progression of AD with mixed pathology is unclear. To define the role of ischemic lesion in dementia progression in AD patients with mixed pathology. A retrospective observational study on consecutive AD, mixed dementia and VaD patients seen in a multidisciplinary memory clinic of a tertiary hospital from 1 Jan 1999 to 30 Jun 2004. Mixed dementia and AD patients were reclassified according to the published radiology operational NINDS–AIREN criteria for VaD into AD–N, no ischemic lesion on CT or MRI; AD–I, presence of ischemic lesion but not fulfilled the operational criteria; and AD–V, fulfilled operational criteria. A radiologist blinded to clinical diagnosis reviewed the imaging findings. Patients followed for more than 1 year were included in the study. Need for institution care and loss of self independence as indicated by score 3 in clinical dementia rating (CDR) were used as composite endpoints for poor outcome. One hundred seventy–three patients were included in the study. Their mean age was 76.3, 38.7% was male. The initial MMSE was 15.8±0.4, and the average duration of follow up was 29.7 months. Forty–three patients were diagnosed VaD, 53, 42 and 35 patients had AD–N, AD–I and AD–V respectively. There was no difference in the rate of change in MMSE among 4 groups of patients. Grouping patients with and without operational VaD criteria showed a significant worse performance on outcome endpoint in those with operational VaD criteria (VaD + AD–V), regardless on the clinical diagnosis of AD or VaD (figure 1). Presence of operational VaD criteria remained the only significant independent risk factor in the Cox–regression proportional hazard model after controlling for age, sex and ChEI treatment, the hazard ratio was 1.9 (1.1–3.3). Kaplan–Meier curve after regrouping patients according to the presence of radiological criteria for VaD. Log rank χ2 5.28, df=1, p value– 0.02.