Prediction of longitudinal cognitive decline in normal elderly with subjective complaints using electrophysiological...

An extensive literature reports changes in quantitative electroencephalogram (QEEG) with aging and a relationship between magnitude of changes and degree of clinical deterioration in progressive dementia. Longitudinal studies have demonstrated QEEG differences between mild cognitively impaired (MCI) elderly who go on to decline and those who do not. This study focuses on normal elderly with subjective cognitive complaints to assess the utility of QEEG in predicting future decline within 7 years. Forty-four normal elderly received extensive clinical, neurocognitive and QEEG examinations at baseline. All study subjects (N = 44) had only subjective complaints but no objective evidence of cognitive deficit (evaluated using the Global Deterioration Scale [GDS] score, GDS stage = 2) at baseline and were re-evaluated during 7-9 year follow-up. Baseline QEEGs of Decliners differed significantly (p < 0.0001, by MANOVA) from Non-Decliners, characterized by increases in theta power, slowing of mean frequency, and changes in covariance among regions, especially on the right hemisphere. Using logistic regression, an R2 of 0.93 (p < 0.001) was obtained between baseline QEEG features and probability of future decline, with an overall predictive accuracy of 90%. These data indicate high sensitivity and specificity for baseline QEEG as a differential predictor of future cognitive state in normal, subjectively impaired elderly.