Research evaluation and diagnosis of probable Alzheimer's disease over the last two decades: I

Objective: To describe the experience of a research clinic diagnosing AD during the last two decades, with special emphasis on patients who meet the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association criteria for probable AD, their patterns of clinical presentation, and neuropathologic outcomes. Background: Probable AD has a heterogeneous clinical presentation, and can occur in the context of complicating factors. There are few reports, and none with this large of a sample, about the pattern of presentation, the nature of comorbidities, and the sensitivity and specificity of diagnosis. Results: The AD Research Center of Pittsburgh examined 1139 patients with probable AD between April 1983 and February 2000. Of these 1139 probable AD patients, 29 (2.5%) had slow progression, 27 (2%) had rapid progression, 70 (6%) had an atypical presentation, and 85 (7%) had coexistent cerebrovascular disease. Confluent periventricular white matter lesions were found in 348 (30.5%) patients with probable AD. The overall sensitivity for the diagnosis of AD was 97% and the specificity 80%. However, the accuracy for the diagnosis of AD varied over the years: from 1983 to 1989, the sensitivity was 94% and specificity 52%, and from 1990 to 2000, the sensitivity was 98% and specificity 88%. Conclusion: Although the diagnosis of probable AD has been used to indicate the presence of a homogeneous clinical entity, these patients can vary in presentation, onset, or clinical course. This finding is of particular importance for the understanding of the pathophysiologic basis of the disease, and for the better identification of responders to dementia treatments. Although the sensitivity for the diagnosis of AD remained above 90% over the last two decades, the specificity increased, reflecting progressive improvement in the diagnosis of other dementing disorders.