Abstract
The distribution of pathology related to Alzheimer's disease (AD) is not uniform throughout the brain. Sites which have a predilection for the development of Alzheimer-type pathology are the limbic regions and neocortical association areas. The changes in these areas of the brain develop gradually, following a well-determined sequence that allows a pathological staging of the disease process. According to the staging hypothesis, the first pathological alterations develop in the transentorhinal and entorhinal regions. The neurofibrillary pathology then spreads into the hippocampus, but not until the final stages does it affect the neocortex. In this study we analyse the relationship between the pathological stages of AD, according ot the staging hypothesis, and the clinical diagnosis in a prospectively assessed patient group. Prediction of any given pathological stage from the clinical diagnosis was found to be poor. This may be partly due to the fact that additional pathologies can alter the clinical picture and severity of dementia in patients who are only in the initial stages of AD. Nevertheless, the NINCDS-ADRDA clinical criteria had a high sensitivity for detection of AD-related pathology: the 'probable AD' category included 22/38 (57.9%) of those in the late isocortical stage, while the 'possible AD' category included 19/23 (82.6%) of those in the limbic stage. Using proposed neuro-imaging protocols for improved identification of patients with AD-related pathology, we largely identified subjects in whom the extent of pathology had spread to the neocortex.