COVID and brain fog.
For millions, it’s the most disabling legacy of the pandemic.
Now, a new brain imaging study reveals a clear biological basis - and the parallels with other diseases are not encouraging.🧵
For millions, it’s the most disabling legacy of the pandemic.
Now, a new brain imaging study reveals a clear biological basis - and the parallels with other diseases are not encouraging.🧵
https://twitter.com/coviddatareport/status/1973236193321189432
Researchers used PET scans with a tracer ([11C]K-2) that binds to AMPA receptors - the tiny gates that neurons use to fire signals via glutamate, the brain’s main excitatory messenger.
These receptors are crucial for learning, memory, and attention.
These receptors are crucial for learning, memory, and attention.
In people with cognitive long COVID, AMPA receptors weren’t just slightly higher in one region.
They were globally increased across the entire brain!
It’s like the brain’s excitatory volume knob was turned way up everywhere.
They were globally increased across the entire brain!
It’s like the brain’s excitatory volume knob was turned way up everywhere.
That doesn’t mean sharper thinking.
Too many receptors - too much firing - neurons overwhelmed with noise.
Instead of clarity, the brain drowns in signals.
And that feels exactly like brain fog.
Too many receptors - too much firing - neurons overwhelmed with noise.
Instead of clarity, the brain drowns in signals.
And that feels exactly like brain fog.
The data confirmed it -
The higher the AMPA receptor signal, the worse people performed on tests of visual memory and word-finding.
The biology matches the symptom.
The higher the AMPA receptor signal, the worse people performed on tests of visual memory and word-finding.
The biology matches the symptom.
The PET signal also lined up with immune markers -
TNFSF12 up (a pro-inflammatory cytokine)
CCL2 down (a chemokine that normally regulates inflammation)
This suggests inflammation is literally pushing AMPA receptors onto neurons, driving hyper-excitation.
TNFSF12 up (a pro-inflammatory cytokine)
CCL2 down (a chemokine that normally regulates inflammation)
This suggests inflammation is literally pushing AMPA receptors onto neurons, driving hyper-excitation.
Put this together with earlier findings - reduced glucose metabolism (FDG-PET) and neuroinflammation - and a chain emerges -
Inflammation - astrocyte dysfunction - disrupted glutamate balance - more AMPA receptors - cognitive dysfunction.
Inflammation - astrocyte dysfunction - disrupted glutamate balance - more AMPA receptors - cognitive dysfunction.
Why does this matter?
Because it turns brain fog into something objectively measurable.
Not mood. Not imagination.
A molecular abnormality visible on a brain scan.
Because it turns brain fog into something objectively measurable.
Not mood. Not imagination.
A molecular abnormality visible on a brain scan.
Does this mean neurons are already dying?
We don’t know - because this study did not examine that.
It only shows receptor changes, not structural loss.
But in other diseases, chronic AMPA overdrive does lead to neuron death. That’s why the parallels matter.
We don’t know - because this study did not examine that.
It only shows receptor changes, not structural loss.
But in other diseases, chronic AMPA overdrive does lead to neuron death. That’s why the parallels matter.
Yes - increased AMPA receptors can lead to neuronal damage and death - if they persist long-term and are not balanced by inhibitory mechanisms.
Parallels -
Alzheimer’s, Huntington’s, epilepsy - glutamate overactivation - excitotoxic stress - gradual neuron loss.
MS, depression -
cytokines like TNF-α increase AMPAR trafficking - imbalance - cognitive problems.
EBV, ME/CFS -
post-infectious brain fog linked to glutamate dysregulation.
HIV brain disease -
viral proteins disrupt glutamate homeostasis, reshaping synapses.
Parallels -
Alzheimer’s, Huntington’s, epilepsy - glutamate overactivation - excitotoxic stress - gradual neuron loss.
MS, depression -
cytokines like TNF-α increase AMPAR trafficking - imbalance - cognitive problems.
EBV, ME/CFS -
post-infectious brain fog linked to glutamate dysregulation.
HIV brain disease -
viral proteins disrupt glutamate homeostasis, reshaping synapses.
What does that mean for long COVID?
It may be partly reversible - if inflammation settles and receptor levels normalize, cognition can improve (though often not fully).
Or it may be progressive - if excitotoxic stress continues, leading over months or years to neuron loss and brain atrophy.
We don’t yet know which path dominates.
It may be partly reversible - if inflammation settles and receptor levels normalize, cognition can improve (though often not fully).
Or it may be progressive - if excitotoxic stress continues, leading over months or years to neuron loss and brain atrophy.
We don’t yet know which path dominates.
Is there hope?
Yes - small.
Drugs like perampanel (an AMPA receptor antagonist) are already used in epilepsy.
But - this study did not test treatments.
Whether dampening AMPA activity helps in long COVID is only a hypothesis - not a proven therapy.
Yes - small.
Drugs like perampanel (an AMPA receptor antagonist) are already used in epilepsy.
But - this study did not test treatments.
Whether dampening AMPA activity helps in long COVID is only a hypothesis - not a proven therapy.
So, Long COVID brain fog is not vague or subjective.
It’s a measurable synaptic disorder - an overexcited brain struggling to process information.
It may resolve. Or it may progress.
Either way - it’s real, biological, and must not be dismissed.
It’s a measurable synaptic disorder - an overexcited brain struggling to process information.
It may resolve. Or it may progress.
Either way - it’s real, biological, and must not be dismissed.
This study is pioneering because it showed a molecular abnormality in the living human brain.
But it’s still a pilot with limitations - confirmation through larger cohorts, longitudinal follow-up, and therapeutic trials is still ahead of us.
But it’s still a pilot with limitations - confirmation through larger cohorts, longitudinal follow-up, and therapeutic trials is still ahead of us.
Caveats.
Only 30 patients, single-center (Japan).
Cross-sectional - a snapshot, not longitudinal.
PET shows receptor density, not neuron count or brain volume.
Confounders - reinfections, vacc, meds, sleep, metabolism.
Causality remains unclear.
Only 30 patients, single-center (Japan).
Cross-sectional - a snapshot, not longitudinal.
PET shows receptor density, not neuron count or brain volume.
Confounders - reinfections, vacc, meds, sleep, metabolism.
Causality remains unclear.
Fujimoto at al., Systemic increase of AMPA receptors associated with cognitive impairment of long COVID. Brain Communications, 2025. academic.oup.com/braincomms/art…
There is also a university press release (in Japanese) with details of the study. yokohama-cu.ac.jp/res-portal/new…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
