Background: This study aimed to determine Consortium to Establish a Registry for Alzheimer's Disease (CERAD) Neuropsychological Assessment Battery total score diagnostic accuracy in the diagnosis of mild cognitive impairment (MCI) and dementia in familial Alzheimer's disease (FAD) with E280A mutation on presenilin-1 gene (PSEN1).
Methods: A cross-sectional study was conducted in a cohort of PSEN1 E280A carriers and non-carriers assessed between January 1995 and February 2013. During the first neuropsychological assessment, 76 were having dementia, 46 had MCI, and 1,576 were asymptomatic. CERAD cut-off points were established for MCI and dementia using a Receiver Operating Characteristics (ROC) analysis, and were further analyzed according to education level in two groups: low education level (eight years or less), and high education level (over eight years).
Results: The area under curve-ROC CERAD total score for dementia was 0.994 (95% CI = 0.989-0.999), and that for MCI was 0.862 (95% CI = 0.816-0.908). The dementia diagnosis cut-off point for the low education group was 54, (98.4% sensitivity, 92.6% specificity), and that for the high education group was 67 (100% sensitivity, 94.1% specificity). The MCI diagnosis cut-off point for the low education group was 66 (91.2% sensitivity, 56.4% specificity), and that for the high education group was 72 (91.7% sensitivity, 76.3% specificity).
Conclusions: The CERAD total score is a useful screening tool for dementia and MCI in a population at risk of FAD.
Keywords: Alzheimer's disease; CERAD; mild cognitive impairment; sensitivity; validity.