Diagnostic guidelines for Alzheimer’s disease

Overview

These recommendations from the National Institute on Aging-Alzheimer’s Association workgroups, published in Alzheimer’s & Dementia in 2011, revise the 1984 NINCDS-ADRDA criteria for dementia due to Alzheimer’s disease (AD). A revision was needed because AD pathology is now known to occur across a spectrum from normal cognition to dementia, other dementias have been better characterized, biomarkers have emerged and nonamnestic presentations are recognized. The criteria are designed to be usable both by general clinicians without advanced testing and by researchers with access to imaging and cerebrospinal fluid (CSF) measures.

Criteria for all-cause dementia

Dementia is diagnosed when cognitive or behavioral symptoms interfere with work or usual activities, represent a decline from previous functioning and are not explained by delirium or a major psychiatric disorder. Impairment is established through history from the patient and an informant plus an objective assessment, and it must involve at least two domains: learning and memory, reasoning and judgment, visuospatial ability, language, or personality and behavior. Whether daily function is significantly affected is what separates dementia from mild cognitive impairment.

Probable and possible AD dementia

Probable AD dementia requires insidious onset, a clear history of worsening and prominent deficits in either an amnestic presentation or a nonamnestic one (language, visuospatial or executive). It should not be diagnosed when there is substantial cerebrovascular disease, core features of dementia with Lewy bodies, prominent features of frontotemporal dementia or primary progressive aphasia, or another condition or medication that could explain the cognitive changes. Documented decline or a causative mutation in APP, PSEN1 or PSEN2 increases certainty, while APOE ε4 is not specific enough. Possible AD dementia covers an atypical course or an etiologically mixed presentation.

Role of biomarkers

Biomarkers fall into two groups: markers of amyloid deposition (low CSF amyloid-beta 42, positive amyloid PET) and markers of downstream neuronal injury (raised CSF tau, reduced FDG-PET uptake in temporoparietal cortex and disproportionate atrophy on MRI). They can increase confidence that dementia is due to AD, but the workgroup does not advise their routine use, citing the good accuracy of the clinical criteria and limited standardization and access. For now they belong in research, clinical trials and optional clinical use. The document also defines pathophysiologically proved AD dementia and dementia unlikely to be due to AD.