Voxel based morphometry features and follow-up of amnestic patients at high risk for Alzheimer's disease conversion.

OBJECTIVE: Neuroimaging techniques are able to mark distinct structural and metabolic changes in patients at risk for Alzheimer's disease (AD). The objectives of the study were to compare regional grey matter density in prodromal Alzheimer's disease (Prd-AD), amnestic mild cognitive impairment (aMCI), mild AD patients and healthy aged controls, to study prospectively their clinical and neuropsychological evolution and to evaluate the accuracy of proposed Prd-AD criteria to detect AD conversion.

METHODS: Twenty-seven controls, 16 aMCI, 32 Prd-AD and 34 probable mild AD were included. Evaluations were performed at baseline and annually during a 2-year prospective follow-up period. Focal grey matter density loss was calculated through voxel based morphometry analysis at baseline. Sensitivity, specificity, positive and negative predictive values of Prd-AD criteria were calculated.

RESULTS: Pr-AD, compared to aMCI, had decreased grey matter density mainly in both hippocampi and inferior temporal cortex (p < 0.001). AD patients compared with Prd-AD, presented grey matter loss in the right posterior and lateral temporal, inferior frontal and parietal cortex and left posterior temporal (p < 0.001). After 2-year follow-up, Prd-AD patients presented higher cognitive decline and conversion rate to AD (83.3%) than aMCI (21.4%; p < 0.0001). The sensitivity of Prd-AD criteria to predict AD conversion in the group of amnestic patients was 89.3; specificity 68.7; positive predictive value 83.3 and negative predictive value 78.6.

CONCLUSION: Magnetic resonance imaging provides evidence of more severe temporal grey matter loss in Prd-AD group than in a-MCI. The proposed criteria present good accuracy to predict AD conversion among amnestic patients.