News · Brain & Mental Health
A brain structure few have heard of tracked with dementia
A Molecular Psychiatry study of 45,306 UK Biobank brain scans found a larger choroid plexus tracked with 61% higher dementia risk, and a brighter one with much lower risk.
Based on a peer-reviewed cohort study in Molecular Psychiatry
- Researchers analyzed brain scans from 45,306 UK Biobank participants, mean age 64, publishing in Molecular Psychiatry.
- The choroid plexus makes cerebrospinal fluid and helps keep the brain's chemical environment stable.
- Its volume and signal intensity were measured automatically from the scans, then linked to later dementia diagnoses.
- Greater volume tracked with 61% higher risk of all-cause dementia.
- Higher signal intensity tracked the other way, with less than half the risk.
- Age itself moved both measures: older brains had larger, dimmer choroid plexus.
- Brain changes including smaller hippocampus and more white matter damage partly explained the link.
- Observational imaging. A structure that changes early in dementia can be a marker of the process rather than a cause of it.
Dementia research spends most of its attention on the hippocampus and on amyloid. A study in Molecular Psychiatry looked at a structure most people have never heard of, and found it moving in step with who later developed the disease.
Greater choroid plexus volume was associated with an increased risk of all-cause dementia, while a brighter signal from the same tissue went the other way.
The tissue in question
The choroid plexus sits inside the brain’s ventricles and does unglamorous maintenance work. It plays a crucial role in cerebrospinal fluid secretion and the maintenance of brain homeostasis.
Its involvement has been suspected before without being pinned down. Its structure and function have been implicated in the pathogenesis of dementia; however, the longitudinal associations with dementia and structural brain biomarkers remain unclear.
Until now the question needed something the field rarely has: tens of thousands of brain scans attached to years of follow-up.
Forty-five thousand scans
This prospective cohort study utilized data from the UK Biobank, including 45,306 participants who underwent 3.0 T multiparametric brain MRI scans, mean age 64 and 47.2% men.
Two measures were taken from the tissue: its volume and its signal intensity, both quantified automatically rather than by eye. Gray and white matter measures came from the same scans, and dementia was identified from hospital admission records and the death register.
Two measures, opposite directions
Age moved both, and moved them apart. Advancing age was correlated with increased choroid plexus volume and decreased signal intensity.
The risk figures follow the same split. Greater volume tracked with about 61% higher risk of all-cause dementia, while higher intensity was correlated with a reduced dementia risk - less than half.
A bigger, dimmer choroid plexus went with more dementia. A smaller, brighter one went with less. Whether those are two symptoms of one underlying change is not something the study resolves.
What sat in between
The link did not run in isolation. Reduced volumes of the hippocampus, amygdala, and nucleus accumbens, increased white matter hyperintensity volume and altered diffusion measures partly accounted for the association.
That is a coherent picture rather than an isolated correlation: the choroid plexus finding travels with the structural damage the field already recognises, which makes it more plausible as part of a process and less likely to be noise.
What a scan cannot show
The central problem is timing. Dementia begins decades before diagnosis, so a structure already altered at the moment of scanning may be reporting the disease rather than causing it.
The authors ran Mendelian randomization to probe that, which is the standard tool, and it rests on genetic assumptions that do not fully close the question.
And nothing here is clinical. There is no threshold, no guideline and no test to ask for. What the study contributes is a well-measured target for the research that would have to come next.
People also ask
What does the choroid plexus actually do?
It is tissue inside the brain's ventricles that produces cerebrospinal fluid, the liquid the brain floats in. It also acts as a barrier and helps regulate what passes between blood and brain. The authors describe it as playing a crucial role in cerebrospinal fluid secretion and the maintenance of brain homeostasis, which is why a change in it is potentially meaningful rather than incidental.
Why do volume and brightness point in opposite directions?
The study does not fully explain that, and it is the most interesting loose end. One reading is that they capture different things: volume may reflect enlargement or inflammation, while signal intensity may reflect tissue composition and function. Both changed with age in this data, volume up and intensity down, so they may be two views of the same deterioration.
Is this a cause or an early sign?
That is the central limitation and imaging cannot settle it. Dementia processes run for decades before diagnosis, so a structure that has already changed by the time of the scan may be reporting the disease rather than driving it. The study used Mendelian randomization to probe causality, but that method rests on genetic assumptions and does not close the question.
Could I be scanned for this?
There is nothing to ask for. These were research scans measured with automated software across tens of thousands of people, and the associations describe group-level risk. No clinical threshold exists, no guideline recommends it, and an individual measurement would not tell you anything actionable today.
How much does a 61% higher risk mean in practice?
It is a relative figure applied to a risk that is low for most people at 64. Relative increases of this size are common in imaging research and rarely translate into large absolute differences for an individual. The value here is in pointing research at a structure, not in re-estimating any one person's odds.