Abstract
Hypothesizing that non-significant cerebrovascular lesions on structural brain imaging lead to overdiagnosis of a vascular etiology of dementia as compared to autopsy-confirmed diagnosis, we set up a study including 71 patients with autopsy-confirmed diagnoses. Forty-two patients in the population (59%) appeared to have definite Alzheimer's disease (AD), whereas 29 (41%) had a non-AD dementia form. The panel clinically diagnosed possible or probable vascular dementia (VaD) in 27 (38%) patients, whereas only five (19%) patients (p = 0.017) had an autopsy-confirmed diagnosis of VaD. Patients with vascular lesions on structural brain imaging were often misdiagnosed as possible or probable VaD as compared to autopsy-confirmed diagnosis.