The Scans Found Less Fuel Going to Your Brain. They Also Found the Recovery.

Body  ·  The Reading Room, Vol. 1  ·  9 min read  ·  What imaging at Weill Cornell measured, the recovery finding that gets left out, and what the scans did not measure

Based on the published research of Dr. Lisa Mosconi, PhD — Associate Professor of Neuroscience in Neurology and Radiology, Weill Cornell Medicine, and director of its Alzheimer’s Prevention Program and Women’s Brain Initiative. Faculty profile · lisamosconi.com · @dr_mosconi

You were mid-sentence. Not a difficult sentence — a sentence about the Q3 numbers, which you have said some version of forty times. And the word went. Not blurred, not slow. Gone, the way a name goes when you are introducing two people who are already shaking hands.

You covered. You always cover. You said the figure we discussed and moved on, and nobody in the room noticed, and you thought about it again at eleven that night.

The question underneath is rarely what can I take for this. It is usually narrower and worse: is something happening to my brain.

This is a reading of published imaging research on that question. We are not clinicians. We read the papers and translate them; the findings below belong to Dr. Lisa Mosconi and her colleagues at Weill Cornell Medicine, and the plain-English version is ours.

The short answer: yes, something measurable is happening — and it is a change in how the brain is fuelled, not a change in whether it works. Estrogen helps regulate how brain cells take up glucose, which is their primary fuel. As estrogen falls across the menopause transition, measured glucose use in parts of the brain falls with it. That much has been imaged directly. The part that rarely gets repeated is what the same research group found next: the brain does not simply keep declining. It compensates, and on two-year follow-up scans some measures had partly recovered — a recovery not seen in the age-matched men.

There is a name worth having for the window in between, because at some point you will need to say this out loud to a husband, a sister or a GP: the fuel gap — the stretch where the brain has stopped running well on the old supply and has not finished arranging the new one. It is not a diagnosis and it is not a term from the papers. It is a plain description of what the scans measured, and it is a great deal easier to say than “temporo-parietal glucose hypometabolism.”

The short version, with numbers

• 43 women, scanned in 2017: brain glucose use was about 19% lower in post-menopausal women than in pre-menopausal women of the same age range, in the regions most affected in Alzheimer’s disease. The widely repeated “more than 30%” figure is a caption on two example scans in the university’s press release, not a result from the paper.¹
• 161 women, scanned in 2021: the same direction, larger sample. Lower grey matter volume, lower white matter volume, and lower glucose metabolism in post-menopausal women compared with pre-menopausal women.²
• The brain was already compensating. The same post-menopausal group showed higher ATP relative to phosphocreatine in temporal regions and higher cerebral blood flow — the signatures of a system rerouting energy, not one shutting down.²
• Two years later, some of it had come back. Follow-up scans showed grey matter volume increases in the precuneus in the post-menopausal group — described by the authors as post-menopause-specific recovery, and absent in age-matched males.²
• None of these studies measured brain fog. They measured metabolism, volume and blood flow. Nobody has yet shown that the scan explains your Tuesday. That limitation is the honest centre of this piece.

Estrogen Is Not Only a Reproductive Hormone

This is the sentence that reorganises the whole question, and it is not controversial: estrogen receptors are distributed throughout the brain, including in the hippocampus and the temporal and parietal regions involved in memory and language. One of estrogen’s jobs there is metabolic. It participates in regulating how neurons take up and burn glucose.

Neurons are expensive. The brain is roughly two per cent of body weight and consumes something close to twenty per cent of the body’s energy at rest, almost all of it from glucose. A hormone that helps govern that supply line is not a peripheral detail. When it declines, the supply line is affected.

So the mechanism being proposed is not hormones make women emotional. It is narrower and far more testable: a fuel-regulation change, visible on a scanner.

The 2017 Study: The First Direct Look

Mosconi’s group imaged 43 healthy women aged 40 to 60 — 15 pre-menopausal, 14 peri-menopausal, 14 post-menopausal — using PET to measure glucose metabolism in the brain.¹

The peri- and post-menopausal women showed markedly lower glucose metabolism in several key brain regions than the pre-menopausal women. The paper puts the reduction at 19 per cent in the regions most affected in Alzheimer’s disease, comparing post-menopausal with pre-menopausal women. The intermediate comparisons were 13 per cent post-menopausal versus peri-menopausal, and 8 per cent peri-menopausal versus pre-menopausal.¹

A figure of “more than 30 per cent” circulates widely, and it is worth naming because you will meet it elsewhere. It comes from a caption beneath two illustrative scans in Weill Cornell’s own press release, describing the difference between those two particular images.¹ It is not the study’s group result, and we are not going to repeat it as one. The real number is smaller, and the smaller number is the one with 43 women behind it.

Two things to hold about that number. It is a real difference, and it comes from fourteen women per group. Mosconi said so herself at publication: “We really need to follow larger groups of women over long periods.”¹ A finding this size in a sample this small is a signal to go and check, not a settled fact.

The 2021 Study: Bigger, and With the Half Nobody Quotes

Four years later the group published a much larger imaging study in Scientific Reports: 161 women aged 40 to 65 — 30 pre-menopausal, 57 peri-menopausal, 74 post-menopausal — each compared against age-matched groups of men, which is the design detail that separates a menopause effect from simply getting older.²

The declines replicated. Post-menopausal women showed lower grey matter volume in temporal regions, lower white matter volume, and lower glucose metabolism in temporo-parietal areas, compared with the pre-menopausal group.²

Then the part that almost never travels with the headline.

In the same post-menopausal brains, the researchers measured higher ATP relative to phosphocreatine in temporal regions, and higher cerebral blood flow in frontal and temporal regions compared with the peri-menopausal group.² ATP is the cell’s usable energy currency. More blood arriving, and more energy in circulation, in tissue that is taking up less glucose, describes a system finding another way to run — not a system failing.

And on the two-year follow-up scans, the post-menopausal group showed grey matter volume increases in the precuneus. The authors describe this as post-menopause-specific recovery. It did not appear in the age-matched men.²

That is a different shape of story from the one that circulates. The circulating version is menopause shrinks your brain. The published version is closer to: the brain goes through a metabolically expensive transition, adapts during it, and on the measures followed, some of it comes back afterwards.

If you have been trying to work out which part of this is yours — the transition, the sleep you have not had in two years, or the load you have been carrying without counting it — the Quiet Audit is ten private minutes on exactly that question. No programme, no advice, nobody sees your answers. Start here.

What These Studies Did Not Measure

This is where most coverage of this research stops being useful, so it is worth being exact.

They did not measure brain fog. Not word-finding, not the reread email, not the name that goes at the handshake. They measured glucose metabolism, tissue volume, blood flow and amyloid deposition. The link from “temporo-parietal glucose uptake is lower” to “that is why the word went” is an inference. It is a reasonable one. It is not a demonstrated one.

The authors state their own limitations plainly, which is a mark in their favour. A causal link between the menopause transition and these brain measures “cannot be unequivocally established.” The cohort may be self-enriched for Alzheimer’s risk: 42 per cent were APOE-4 carriers against roughly 15 to 30 per cent in the general population. The sample was highly educated with a low proportion of minority participants, which the authors say limits generalisability. And some women classified as peri-menopausal may in fact have been early post-menopausal, and vice versa.²

The amyloid finding needs care. The 2021 paper did report higher amyloid uptake in post-menopausal than pre-menopausal women (27 per cent, p = 0.006), concentrated in APOE-4 carriers.² Amyloid is one biomarker associated with Alzheimer’s risk at the group level. It is not a diagnosis, it is not a prediction about any individual woman, and a cohort over-represented for the highest-risk genotype is exactly the cohort where you would expect to see it. Read carefully, this is a reason for researchers to keep looking. It is not a reason to lie awake.

Five Things That Look Like This and Are Not This

The section above protects the research. This one is for you, and it is the part most coverage of this topic leaves out, because naming alternatives means conceding that the article does not have the whole answer.

Five ordinary conditions produce word-finding trouble and slowed recall that is, from the inside, indistinguishable from what is described here. Four of them are a blood test.

• Thyroid. An underactive thyroid slows cognition alongside cold, weight change and fatigue, and it becomes more common in exactly this age band.
• B12. Deficiency produces memory and concentration symptoms, and it is more likely in anyone on metformin or long-term acid-reducing medication.
• Ferritin. Iron stores can sit low enough to affect concentration while a full blood count still reads as normal — which is why ferritin has to be asked for by name.
• Sleep apnoea. It rises sharply after menopause and shows up in women as daytime fog and morning headache rather than as snoring, so it goes unfound for years.
• Medication. Antihistamines, some blood-pressure drugs, sleep aids and anticholinergics all carry cognitive effects that are easy to file under the wrong cause.

None of these rules the transition in, and none of them rules anything else out. They stack: a woman in the transition who is also low on iron and waking six times a night is dealing with three things, and only one of them resolves on its own. The reason to have the list is not reassurance. It is that four of the five are cheap to check, and none of them gets checked unless somebody asks.

Who Else Has Measured This

One research group is never enough, and the gap above — that the scans did not measure the symptom — is exactly the gap other groups have been working on. Three independent lines of work are worth knowing about.

That last point is the one worth carrying out of this article. The imaging says the brain’s energy supply changes and then partly recovers. The cognitive testing, done by different people using different methods on a different cohort, says the performance dips and then comes back. Neither proves the other. Two independent methods agreeing is still the strongest thing on offer here.

A transition has a shape. Decline only has a direction.

What This Actually Changes

Not the treatment. Nothing here tells anyone what to take, and the trial evidence on whether hormone therapy improves cognition is a separate and much less encouraging question.

What it changes is the interpretation, and for a woman who has spent two years quietly building a case against herself, the interpretation is not a small thing.

The word that went in the meeting has been filed, somewhere, as evidence of decline — a first data point in a series you have been dreading. What the imaging supports is a different filing: a period of metabolic transition, with visible compensation happening inside it, and documented partial recovery on the other side. That reframe does not restore the word. It does change what the word means.

It also predicts something specific and useful. If the mechanism is fuel and adaptation, the fog should be uneven — worse when the system is already stressed, better when it is not. Which is exactly what most women describe, and exactly what nobody believes about themselves, because unevenness reads as unreliability rather than as a pattern.

What to Say at the Appointment

The sentence in the panel above is the useful part of it, and one sentence is usually not enough, because there is a standard reply waiting for it.

Neither of those is a demand, and neither asks for a named treatment. They are four ordinary tests and one question. Asking for them by name is the difference between a consultation that ends in reassurance and one that ends in information.

If what you want to take into that appointment is a clear account of what the last two years have actually contained, the Quiet Audit is ten private minutes on exactly that. No programme, no advice, nobody sees your answers. Start here.

The One Move

Track the pattern, not the incidents.

Incidents are what you currently collect, and they are the worst possible evidence because you only notice them when they are bad. Fourteen days, one line a day, three things: how you slept, roughly how heavy the day was, and whether the fog showed up. Nothing more elaborate.

At the end of two weeks you will have the thing you cannot get from a memory of a meeting: whether it tracks with sleep, whether it tracks with load, whether it tracks with cycle timing, or whether it tracks with none of them. Each of those answers points somewhere different, and all four are useful. The one thing this rules out is the version you have been carrying — that it is constant, that it is worsening, and that it is you.

Frequently Asked Questions

Why does menopause cause brain fog?
The leading mechanism is metabolic. Estrogen helps regulate how brain cells take up glucose, their main fuel; as estrogen falls, imaging shows measured glucose use falling in temporal and parietal regions.^1,2 The important caveat is that these studies imaged metabolism, not cognitive symptoms — the connection between the two is inferred rather than demonstrated.

Does perimenopause brain fog go away?
The imaging data are cautiously encouraging. On two-year follow-up scans, post-menopausal women showed grey matter volume recovery in the precuneus that age-matched men did not show, and some white matter and glucose measures partly stabilised or reversed.² This is a group-level finding about brain measures over two years, not a promise about any individual woman’s timeline.

Is menopause brain fog the same as early dementia?
They present differently. Hormonal fog typically fluctuates, affects word retrieval and short-term recall, and coexists with other transition symptoms. Progressive dementia tends to worsen steadily, affects orientation and familiar sequences, and is often noticed by others first. If the pattern looks like the second description, that is an appointment, not an article — and there is a specific question worth asking before you assume the worst.

Does the 30 per cent figure mean I have lost 30 per cent of my brain function?
No — and the figure itself needs correcting. The 2017 paper reports a 19 per cent reduction in glucose metabolism in the regions most affected in Alzheimer’s disease, comparing post-menopausal with pre-menopausal groups in a study of 43 women.¹ The “more than 30 per cent” you will have seen quoted comes from a caption on two example scans in the university press release, not from the study’s group results. Neither number refers to overall cognitive performance, and neither was measured in you. The same research group later found compensatory increases in blood flow and ATP in the same population.²

Related reading: whether a memory slip is normal, hormonal, or worth a check, what happens when a word you have used for thirty years goes missing, and why the 3 a.m. waking is a thermostat problem.

What to Carry Out of This

Fourteen days of tracking will tell you whether the fog follows your sleep or your load, and neither answer is bad news. The fuel gap has a far side. That is the whole finding, and it is the half that never travels with the headline.

The word will probably go again. It is worth knowing what it means when it does.

References
1. Mosconi L, et al. Perimenopause and emergence of an Alzheimer’s bioenergetic phenotype in brain and periphery. PLoS One. 2017;12(10):e0185926. journals.plos.org — institutional summary, and the source of the widely quoted “more than 30 per cent” image caption: news.weill.cornell.edu
2. Mosconi L, et al. Menopause impacts human brain structure, connectivity, energy metabolism, and amyloid-beta deposition. Scientific Reports. 2021;11:10867. nature.com — institutional summary: news.weill.cornell.edu
3. Maki PM. Faculty profile, Department of Brain and Cognitive Sciences and Department of Psychiatry, University of Illinois Chicago. psch.uic.edu — research programme: Menopause, Cognition and Mood, UIC College of Medicine
4. Brinton RD. Navigating uncharted territory in female brain aging. University of Arizona Health Sciences. healthsciences.arizona.edu — the “master regulator of the brain’s bioenergetic system” quotation is taken from this release.
5. Study of Women’s Health Across the Nation. SWAN Fact Sheet: Memory and Cognition During and After the Menopause Transition. Cognition papers led by Greendale GA, UCLA. swanstudy.org (PDF)

Miriam Alderton is Research Editor at Blue Leaf Journal. She reads the limitations section before the abstract, and every figure in this piece is linked to its source above. Published: August 27, 2026.