RESUMO
BACKGROUND: Alzhiemeranjx disease (AD) and vascular dementia (VD) are common types of dementia. As a result of the development of new specific agents for AD, and because vascular dementia is a potentially preventable dementia, differentiating these diseases is of great importance. The role of EEG spectral analysis in the differential diagnosis between Alzheimer type and vascular dementia is still controversial. Since there have been few studies concerning the differential diagnosis of dementia by EEG, the present study has focused on this aspect. Usefulness of EEG in differen-tial diagnosis of dementia will be elucidated by clarifying relationship between type of dementia and spectral profile of EEG. METHODS: We analyzed the power spectra taken from 16 derivations and spectral profile was constructed by averaging twenty 2 sec epochs in three study groups (normal controls, AD and VD). Spectral profile was divided into three groups; (I) type A, showing a dominant 6.5-12 Hz peak (ii) type B, lacking a dominant peak in the 6.5-12 Hz (iii) type C, corresponding to a flat, low voltage, spectrum. To elucidate the relationship between spectral profile and other factors including diagnosis, statistical test was done. RESULTS: (1) In AD, type C profile was statistically more prevalent than in VD and type A profile was reversed. (2) In AD, Mini-Mental State Examination (MMSE) score was statistically lower in type C profile. (3) Spectral profile was not associated with age, age of symptom onset, and symptom duration. CONCLUSIONS: This study suggested that spectral profile is a useful tool for the differential diagnosis of dementia (AD and VD) and correlated with the severity of disease in AD.