DRWF Research: What does blood glucose have to do with brain health?
Recently published DRWF-funded research offers important clues.
Many people living with diabetes are understandably concerned about their long-term brain health.
Diabetes and brain health
We have known for some time that diabetes, particularly type 2 diabetes, is linked with a higher risk of memory problems and dementia. But why this happens has been much less clear. Is it high blood glucose itself? Insulin resistance? Changes in the brain’s structure? Or something else entirely?
A major new UK study, partly funded by DRWF, and published in Diabetes, Obesity and Metabolism, has taken an important step towards answering these questions.
Using cutting-edge genetic methods and data from hundreds of thousands of people in the UK Biobank, researchers – headed up by DRWF Research Fellow Dr Victoria Garfield at the University of Liverpool – have tried to “untangle” the complex relationship between glucose, insulin and brain health.
Their findings don’t provide all the answers – but they offer valuable new insight that could shape future research, prevention strategies and potentially even treatment.
Dr Victoria Garfield was the DRWF Professor David Matthews Non-Clinical Fellowship recipient in 2022, receiving an award of £128,663.95 over a two-year period.
The title of Dr Garfield’s proposed research programme is Understanding how diabetes and hyperglycaemia causes cognitive decline, dementia and abnormal brain structure. An integrated genetic epidemiological and deep phenotyping approach to disentangle pathways and interplay of risk factors.
In a summary of the research, Dr Garfield said: “Dementia is one of the most feared diabetes complications. Risks are elevated in people with diabetes, but glucose-lowering trials have been disappointing.
“However, Mendelian randomisation is a genetic tool which can help uncover true causal relationships, as genes are randomly distributed at birth and not influenced by external risk factors. Using this, I have shown that diabetes itself does not appear to cause dementia.
“I now want to use this tool to identify which diabetes-related factor is the true culprit, in large datasets which include measures of cognition, brain structural damage and dementia.
“There are four potential explanations: i) processes underlying diabetes, e.g. insulin resistance, ii) associated metabolic disturbances, e.g. amino acids, iii) associated risk factors, e.g. blood pressure (BP), iv) risk factors upstream of diabetes, e.g. obesity.
“By identifying true causal determinants, we will be in a better place to pinpoint interventions to be tested in clinical trials to reduce dementia risk in diabetes.”
Why this question matters
Dementia is one of the most feared health conditions of later life. Alzheimer’s disease is the most common form, while vascular dementia – caused by problems with blood flow in the brain – is also common. Rates of both are rising as the population ages.
At the same time, diabetes and prediabetes are increasingly widespread. We already know that people with type 2 diabetes have a higher risk of developing dementia. But diabetes is not one single thing – it involves several related problems, including:
- higher fasting blood glucose (blood glucose levels after not eating)
- higher post-meal (postprandial) blood glucose
- insulin resistance (when the body’s cells don’t respond properly to insulin)
- changes in blood vessels and Inflammation
Because all of these factors tend to happen at the same time, it has been difficult for scientists to work out which ones are causing harm to the brain, and which are just associated with it.
A clever way to study cause and effect: Mendelian randomisation
Rather than relying only on observational studies (which can be influenced by lifestyle, socioeconomic factors and other health conditions), this research team used a technique called Mendelian randomisation (MR).
In simple terms, this approach uses genetic differences between people as a kind of “natural experiment”.
Some people are born with genetic variants that predispose them to slightly higher fasting glucose, higher post-meal glucose or greater insulin resistance.
By studying these genetic patterns across very large populations, researchers can get closer to understanding causal relationships – that is, whether these factors truly increase dementia risk, rather than just being correlated with it.
The UK Biobank is a large national research database containing detailed health, genetic and lifestyle information from around half a million volunteers in the UK.
The researchers analysed data from up to 357,883 UK Biobank participants, mostly aged between 40 and 69, and focused primarily on people of White British ancestry to ensure reliable genetic comparisons. They looked at genetic markers linked to:
- fasting glucose
- fasting insulin
- 2-hour post-load glucose (a measure taken after a glucose drink, similar to what is used in some diabetes tests)
- insulin resistance
They then examined how these genetic tendencies related to:
- risk of all-cause dementia
- risk of Alzheimer’s disease specifically
- risk of vascular dementia
- brain structure measures from MRI (magnetic resonance imaging) scans, including total brain volume, hippocampal volume (a key memory area), and markers of small vessel damage in the brain (white matter hyperintensities).
The standout finding: post-meal glucose matters
The most striking result came from the analysis of 2-hour post-load glucose (2hPG) — that is, blood glucose levels after a glucose challenge.
In the UK Biobank sample, people with genetic variants that predisposed them to higher 2-hour post-meal glucose had a 69% higher risk of Alzheimer’s disease.
This suggests that repeated spikes in blood glucose after eating could play an important role in increasing dementia risk, even independently of other diabetes-related factors.
Interestingly, this association was not seen when the researchers looked at fasting glucose, fasting insulin or insulin resistance. This suggested that what happens to blood glucose after meals may be more relevant to Alzheimer’s risk than baseline (fasting) levels.
However – and this is important – when the team tried to replicate this finding in a separate international genetic dataset (a large Alzheimer’s genome-wide association study), the same strong association did not clearly reappear. That does not mean the finding is wrong, but it does mean it needs to be treated with caution and tested in further studies.
What about the brain scans?
One might expect that if higher post-meal glucose increases dementia risk, it would also be linked to visible changes in brain structure – for example, smaller hippocampal volume or more white matter damage.
Surprisingly, the study did not find clear genetic evidence that higher glucose, insulin or insulin resistance directly caused measurable changes in these MRI brain measures.
In other words:
- Higher 2-hour glucose was linked to higher Alzheimer’s risk.
- However, it was not clearly linked to changes in brain volume or visible brain damage on scans.
This suggested that if post-meal glucose was contributing to dementia, it could be doing so through effects on amyloid and tau proteins (markers of Alzheimer’s disease) – rather than simply shrinking parts of the brain that we can easily measure on MRI.
Strengths of the study – and why DRWF support matters
This research had several major strengths:
- It used one of the world’s largest and most detailed biomedical datasets — the UK Biobank.
- It applied two different Mendelian randomisation methods, increasing confidence in the results.
- It included both dementia diagnoses and brain imaging data.
- It carefully checked key assumptions of the genetic method to reduce the risk of misleading results.
Importantly, the study was part-funded by DRWF. This highlights DRWF’s commitment to supporting high-quality, impactful research that explores the long-term complications of diabetes and related metabolic conditions – not just blood glucose control in the short term.
What we still don’t know
The researchers were careful to acknowledge several important limitations:
- UK Biobank participants tend to be healthier and better educated than the general population, which could mean the findings underestimate some risks.
- Most participants were of White British ancestry, so the results may not fully apply to all ethnic groups.
- Dementia diagnoses were based largely on hospital and registry records, which may miss some cases.
- The genetic approach estimates lifelong tendencies toward higher glucose or insulin resistance – not the effects of changes in diet, lifestyle or treatment.
Most crucially, the key finding about 2-hour glucose and Alzheimer’s risk needs further replication in other populations before it can be considered definitive.
What does this mean for people with diabetes today?
This study does not mean that everyone with high post-meal glucose will develop Alzheimer’s – far from it. However, it does add to a growing body of evidence that:
- managing blood glucose levels after meals may be just as important – if not more so – than focusing only on fasting levels
- reducing glucose spikes through diet, physical activity and appropriate medication could have benefits that extend beyond heart health to brain health as well
- preventing or delaying type 2 diabetes, and managing prediabetes may also help reduce future dementia risk
What happens next?
The researchers conclude that genetically higher post-meal glucose likely contributes to increased Alzheimer’s risk. However, they call for further studies in more diverse populations to confirm and expand these findings.
Future research may focus on:
- identifying the precise biological mechanisms linking glucose spikes to Alzheimer’s disease
- exploring whether medications that specifically reduce post-meal glucose could help protect brain health
- studying similar questions in people with established diabetes, rather than just in genetic risk groups
As more people live longer with diabetes, understanding – and ultimately preventing – complications such as dementia becomes increasingly important. While we do not yet have all the answers, this research moves us closer to understanding how everyday blood glucose patterns might shape brain health decades later.
Dr Garfield, said: “We first need to replicate these results in other populations and ancestries to confirm the link and better understand the underlying biology. If validated, the study could pave the way for new approaches to reduce dementia risk in people with diabetes.”
Read the report in full in Diabetes, Obesity and Metabolism
Read more about the Professor David Matthews Non-Clinical Fellowship and DRWF Research
Read more about type 2 diabetes
This article originally appeared in the Spring 2026 edition of Diabetes Wellness News. To subscribe visit here
I would like to make a regular donation of
I would like to make a single donation of
There are lots of ways to raise money to support
people living with all forms of diabetes.
Bake, Swim, Cycle, Fly ... Do It For DRWF!
Fundraise with us
Recent News
