Neuroimaging in dementia: how best to use the guidelines?

Kuruvilla et al ¹ completed an audit cycle on neuroimaging practice after national and European guidance was adapted to local resource availability. The audit showed an improvement in the number of patients who have had at least one form of neuroimaging performed from 68 to 76%, and although this was not statistically significant, it seems to suggest a general improvement in the service provided, as reflected also in the improved documentation of the reason for not requesting neuroimaging and in having no significant impact on waiting times. Improvement in the service may also be reflected in a patient and relative satisfaction survey that could be carried out.

In a similar study (details available from the author on request), I audited the practice of a memory clinic in Southport, Merseyside, against 2006 National Institute for Health and Care Excellence (NICE) guidance on dementia, ² which stated that ‘structural imaging should be used in the assessment of people with suspected dementia’ and that magnetic resonance imaging (MRI) ‘is the preferred modality […] although computed tomography (CT) scanning could be used’. The audit included 75 patients and showed that 56 (75%) had at least one neuroimaging procedure performed: 53 (95%) of these had CT scans and only 1 patient had an MRI scan. My audit revealed a similar problem with documentation of reasons for not scanning patients, with 31% of patients who were not scanned lacking such documentation compared with 50% in Kuruvilla et al's initial audit. In my study a re-audit was not carried out.

An additional aim of my study was to look at whether the diagnosis of dementia subtype, provisionally made based on clinical interview and using scales such as MMSE and ACE-R, was changed following neuroimaging. This revealed that the diagnosis was changed following a scan in 45% of cases, mostly from Alzheimer's or vascular dementia into a mixed-type dementia. It also showed that no provisional diagnosis was documented in 38% of case notes reviewed, suggesting that clinicians were perhaps uncomfortable about making a diagnosis before a scan was performed.

Bearing in mind that NICE guidelines are driven partly by cost-effectiveness, studies such as Kuruvilla et al's provide good support for the usefulness of adapting these guidelines to the local availability of resources, which results in better care for patients with dementia.