This patient should have an MRI of the brain. He most likely has mild cognitive impairment (MCI), as evidenced by cognitive difficulties that are greater than those typical of normal aging but do not impair activities of daily living and have minimal effect on instrumental activities of daily living. MCI is a clinical diagnosis made exclusively on the basis of history and the results of cognitive testing. Although neuroimaging cannot be used to detect whether or not a cognitive disorder is present, structural neuroimaging studies, specifically head CT and brain MRI, can play a key role in the routine diagnostic evaluation of patients with established cognitive impairment to exclude structural lesions of the brain, such as strokes, hematomas, brain tumors, or other mass lesions.

Amyloid PET imaging allows for in vivo detection of amyloid plaques, which are a core pathologic feature of Alzheimer disease. However, positive results are not synonymous with the presence of Alzheimer disease because cognitively normal persons can have abnormal scans. Amyloid PET imaging may provide prognostic information in patients with MCI by identifying those whose cognitive impairment may be related to underlying Alzheimer disease pathology, but much more research is still required before the routine use of this technology in clinical practice.

Screening for carotid stenosis with ultrasonography or any imaging modality has no role in the routine evaluation of cognitive impairment in an asymptomatic patient. An asymptomatic patient in this context is one without previous hemispheric neurologic symptoms—such as transient ischemic attack, stroke, or amaurosis fugax—or the presence of a carotid bruit on examination.

The apolipoprotein E gene (APOE ε4) located on chromosome 19 has been identified as a genetic risk factor for late-onset Alzheimer disease. APOE ε4 genotyping has marginal additive value over clinical diagnoses but is neither necessary nor sufficient to predict who will develop Alzheimer disease and thus is not recommended for broad clinical use.

A fluorodeoxyglucose-PET scan of the brain or any other metabolic scan of the brain cannot determine if a patient's cognitive symptoms indicate cognitive impairment. Abnormal results of fluorodeoxyglucose-PET scanning also occur in persons who are cognitively normal. Although these abnormal results may indicate a future risk of cognitive impairment, this relationship has not been established.