A comparison of the clinical utility, patient preference and cost benefit of SPECT and PET brain imaging in the diagnosis of dementia

Award Number
Research for Patient Benefit
Status / Stage
1 March 2010 -
28 February 2013
Duration (calculated)
02 years 11 months
Funding Amount
Funder/Grant study page
Contracted Centre
Cumbria, Northumberland, Tyne and Wear NHS Foundation Trust
Contracted Centre Webpage
Principal Investigator
Professor John O'Brien
PI Contact
WHO Catergories
Economic Impact of Dementia
Tools and methodologies for interventions
Disease Type
Alzheimer's Disease (AD)
Lewy body dementia (LBD)

CPEC Review Info
Reference ID114
ResearcherReside Team


Award NumberPB-PG-1207-13105
Status / StageCompleted
Start Date20100301
End Date20130228
Duration (calculated) 02 years 11 months
Funder/Grant study pageNIHR
Contracted CentreCumbria, Northumberland, Tyne and Wear NHS Foundation Trust
Contracted Centre Webpage
Funding Amount£249,437.00


Neuroimaging is regularly used to assist clinicians in the diagnosis of dementia. Current NICE guidelines for dementia (2006) recommend the use of brain SPECT or PET imaging when the diagnosis is in doubt. However, it is not known which is the best modality to use, nor which patients prefer if given the choice. While SPECT is a widely available investigation of relatively modest cost (£250), PET is limited to specialist centres and is 3-4 times more expensive, though availability is rising rapidly. Limited evidence to date suggests that PET has superior diagnostic accuracy in dementia compared to SPECT, but very few comparative studies have been undertaken. There is a need for a systematic comparison of SPECT and PET imaging in terms of clinical utility for the diagnosis of dementia. Plan of Investigation We will recruit 100 subjects aged over 60 (30 control subjects, 40 with Alzheimer’s disease and 30 with dementia with Lewy bodies). Groups will be matched for age and (for dementia groups) severity of dementia. Subjects will undergo clinical and cognitive assessment and SPECT and PET scanning according to standard clinical protocols. SPECT and PET scans will be analysed blind to grouping by (a) visual report and assessment as per usual clinical routine, and (b) quantitative assessment using region of interest analysis and voxel based methods (statistical parametric mapping). The accuracy of each technique (sensitivity, specificity and overall accuracy) will be assessed and SPECT and PET will be compared. Information will also be collected regarding patient and carer preferences. Potential Impact Brain PET is a much more expensive and less widely available clinical tool than SPECT, though there is limited evidence that it may be superior in the assessment of patients with dementia. However, this has not been systematically studied in an older population, yet with the increasing prevalence of dementia determining the optimal use of expensive imaging technologies is essential for efficient resource allocation within the NHS. This study will provide key information appropriate to NICE and other agencies (e.g. healthcare commissioners) in determining whether one imaging technology is more useful, and if so at what cost, compared to another. We will also determine whether there are any major differences in patient preferences.


To determine whether PET has superior diagnostic accuracy to SPECT in (a) differentiation of neurodegenerative dementia (Alzheimer’s disease and dementia with Lewy bodies) from controls; (b) the distinction of dementia subtype (Alzheimer’s disease from dementia with Lewy bodies). Secondary aims are (a) to determine the cost benefit of using PET imaging over SPECT, in terms of the additional cost per extra correct diagnosis made, and (b) to investigate patient views on which imaging procedure, if any, they prefer and reasons for this.