ABSTRACT
Dementia encompasses many illnesses of which Alzheimer’s disease (AD) is the most common form over the age of 60 years.1 The clinical diagnosis of AD can be made using clinical criteria.2-4 Based on clinicopathological correlation, the accuracy of these criteria is limited, with a specificity ranging between 76% and 88% and a sensitivity between 65% and 53%.5 Varma et al6 found an even lower specificity of 0.23 when using the NINCDS-ADRDA criteria to distinguish patients with AD from those with fronto-temporal dementia (FTD). For FTD, vascular dementia (VaD) and dementia with Lewy bodies (DLB) clinical criteria are also available,7-9 but the diagnostic accuracy has not yet been defined. In conjunction with the clinical criteria other diagnostic tests are needed and sometimes specified in the criteria. Neuroimaging is the most commonly used ancillary investigation for this purpose, but its value in clinical practice (i.e. not in research settings) has still not been fully evaluated. Traditionally, neuroimaging was used for the exclusion of treatable causes of dementia, which reflected the general clinical attitude towards dementia. In view of the potential treatments for AD, and possibly other dementias, this traditional view can no longer be held and likewise, the use of neuroimaging has changed substantially.
