Taking into account the lower distances travelled, drivers aged eighty and above have an accident risk that approaches the high levels seen in teenagers. When people who die in car crashes are compared with those dying from other causes, they are twice as likely to have neurotic plaques. And the presence of symptomatic Alzheimer’s Disease (AD) is a known risk factor: one study found that the accident rate in the first year after diagnosis is double that of age-matched controls.

Driving recruits an extensive brain network, and we can expect an impact from cognitive deficits. So there is good reason for the functional MRI studies Megan Hird and colleagues (from St Michael’s Hospital, Toronto, Canada) have been conducting to assess connectivity between spatially distinct brain regions while subjects use a driving simulator.

In an initial study, 20 people with mild cognitive impairment (MCI) were compared against 20 matched controls. Steering was significantly worse in those with MCI, and there was a trend towards a higher rate of driving errors and poorer maintenance of lane discipline.

On fMRI, there is less information sharing in the brains of drivers with MCI

This was followed by the fMRI study which matched 15 people with MCI against 15 cognitively unimpaired controls. Motor, visual and frontal areas were engaged in all subjects. But among people with MCI, there was decreased connectivity between posterior visual areas and frontal brain regions involved in error monitoring. So it looks as if there is less information sharing going on in the brains of people with MCI.

Pre-dementia linked to poorer driving

Catherine Roe (Washington University School of Medicine, St Louis, USA) reported a study of 121 participants from the Knight Alzheimer’s Disease Research Center who were positive for AD biomarkers. Subjects were given a forty minute road test during which their performance was assessed by professional driving examiners. Subjects’ mean age was 72 years.

The more abnormal the subject’s CSF tau result, the shorter the time to the examiner’s judgement that the candidate’s performance was marginal or would justify failing the test. Much against expectations, higher levels of education in this study were associated with poorer driving performance. The search for potentially confounding variables has failed to find an explanation.

A robust model decides who to test for AD biomarkers

Predicting amyloid positivity using simple measures

Large numbers of people have cognitive symptoms or complaints, and not all of them can be screened for β-amyloidosis by CSF or PET. Sebastian Palmqvist (from the Clinical Memory Research Unit, Lund, Sweden) and colleagues have developed a risk score that identifies those most in need of biomarker testing.   

The score was developed using data from 474 memory clinic subjects enrolled in the Swedish BioFINDER cohort. Fifty-seven percent had MCI and 43% subjective cognitive decline. Their mean age was 71 years and 49% were amyloid positive. Factors significantly predictive of amyloid positivity were age, Apo-E4 genotype, and performance on the orientation and memory tasks of the MMSE, the 10-word recall test, the Alzheimer’s Disease Assessment Scale-Cognition, and delayed recall.

The algorithm derived from the training cohort was then validated on 661 people from the Alzheimer’s Disease Neuroimaging Initiative. It proved highly robust. Dr Palmqvist foresees a future in which people with subjective or objective cognitive symptoms provide a blood sample which is used to establish Apo-E status. They are given a battery of quick and simple tests. And then a score is calculated which, if high, leads to measurement of AD markers in CSF.

Characterizing people with subjective cognitive decline

There is increasing interest in people with subjective cognitive decline (SCD) who present to memory clinics but are assessed as cognitively normal. Are they the “worried well”, or are they a group at higher than average risk of subsequent problems?

Are people with subjective cognitive decline the “worried well” or at real increased risk?

Rosalinde Slot (VU Medical Center, Amsterdam, the Netherlands) described the characteristics of the first 151 people who have entered the longitudinal Subjective Cognitive Impairment Cohort since its start in 2014.

Their mean age was 64 years, 76% had a family history of AD, and 34% were Apo-E4 carriers. Twenty five percent were amyloid positive according to CSF or PET, while 38% had evidence of sub-threshold psychiatric problems. The presence of such problems correlated with poorer quality of life and worse performance on attention and executive function tasks.

People with subjective cognitive decline are a mixed group, but there is some evidence that severity of SCD is associated with poorer objective cognitive performance, and worse quality of life. The cohort is being followed to assess the longer-term implications.