Clinicopathologic study of Alzheimer's disease: Alzheimer mimics

A definite diagnosis of Alzheimer's disease (AD) can only be made at autopsy. Even at expert research centers, diagnostic accuracy is relatively low. We conducted this study to examine the accuracy of clinical diagnosis of AD and present a list of clinical and neuropsychological findings that could render the clinical diagnosis difficult. Using the National Alzheimer's Coordinating Center database, the records of 533 patients who had been diagnosed clinically with AD, and later underwent autopsy, were reviewed retrospectively. Since the pathologic results of 119 subjects did not meet the criteria for definite AD, we labeled them as Alzheimer "mimics". The neuropathological diagnoses of Alzheimer mimics consisted of dementia with Lewy bodies (n = 35, 29%), insufficient AD (n = 22, 18%), vascular disease (n = 15, 13%), frontotemporal lobar degeneration (n = 14, 12%), and hippocampal sclerosis (n = 10, 8%). History of pacemaker insertion (10.92% versus 4.11%, p = 0.005), congestive heart failure (13.45% versus 6.04%, p = 0.007), hypertension (56.30% versus 47.83%, p = 0.037), and resting tremor (14.29% versus 10.87%, p = 0.170) was more prevalent in Alzheimer mimics. Clinical Dementia Rating score and frequency of Neuropsychiatric Inventory Questionnaire items reflecting delusions, agitation, depression, and motor disturbance were more severe in confirmed AD. In addition to Mini-Mental State Examination (16.97 ± 8.29 versus 12.74 ± 15.26, p < 0.001), Logical Memory, Animal Fluency, Boston Naming Test, and Digit Span scores showed more severe impairment in confirmed AD. Continuing systematic comparisons of the current criteria for the clinical and pathological dementia diagnoses are essential to clinical practice and research, and may also lead to further improvement of the diagnostic procedure.