Long COVID brain fog has a biological name — dopamine nerve loss in 3 brain regions

Long COVID is not a vague syndrome. For the first time, brain-imaging researchers have identified specific neuron damage — specific cells, specific brain locations, specific symptoms — in patients who have been told for years there was nothing measurable to find.

The damage is in the dopamine system. PET scans taken at CAMH’s Brain Health Imaging Centre in Toronto measured the density of a protein called VMAT2 — the vesicular monoamine transporter 2 — in patients with long COVID and compared it against healthy controls. VMAT2 density indicates how many functioning dopamine nerve terminals a brain region has. Across three sections of the striatum, long COVID patients had significantly less of it. Each depleted zone corresponded to a different set of symptoms that clinicians had not previously been able to measure.

How the researchers tracked the damage

PET scanning is not a standard brain MRI. It tracks the movement of a radioactive tracer through the body and maps where specific proteins are concentrated. In this study, the tracer targeted VMAT2, the molecular machinery dopamine neurons use to package and release dopamine from their terminals.

VMAT2 was chosen precisely because dopamine levels themselves fluctuate second by second and cannot be reliably compared across patients. Terminal density — how many nerve endings are actively present — is a more stable marker of underlying neurological integrity. Lower VMAT2 in a brain region means fewer functional dopamine terminals, which means that region receives less dopamine signal on demand.

The study design required patients with confirmed long COVID to undergo PET scanning and complete detailed symptom questionnaires. Their striatal VMAT2 readings were then compared with those of control participants who had no history of the illness. The striatum was divided into functionally distinct zones, and the correlation between VMAT2 loss and reported symptoms was analyzed per zone.

Three regions, three deficits

The striatum is not a single structure. It operates through overlapping but functionally specialized zones, and the study found that VMAT2 loss was not uniform — it tracked precisely with the symptom geography that clinicians have struggled to account for.

In the ventral striatum, the zone most associated with reward anticipation and the drive to initiate behavior, VMAT2 levels correlated with reduced motivation. Long COVID patients who reported the greatest loss of drive — the kind of fatigue where the will to do things is absent rather than physical energy — had the steepest reductions here. This is a specific dopaminergic signal, not a general energy deficit.

In the dorsal putamen, a region involved in motor control and the smooth execution of movement, VMAT2 correlated with slower physical movement. Patients who reported noticeable motor sluggishness — distinct from muscular weakness — had lower density in this zone.

In the caudate putamen, which is involved in executive cognitive function and the organization of memory, VMAT2 loss correlated with memory difficulties. These were not reports of severe memory failure; they were the quieter deficits: losing words mid-sentence, struggling to hold a chain of thought, failing to remember recent decisions. The caudate putamen is the structure whose dopamine supply supports exactly that kind of working memory maintenance.

Dr. Jeffrey Meyer, the study’s senior author and a Canada Research Chair at the Brain Health Imaging Centre, described the findings as “compelling evidence that long COVID involves the loss of dopamine-releasing neurons.” The study, published in eBioMedicine, calls this “the strongest evidence so far” linking long COVID to dopamine system injury.

Why this changes the clinical picture

For most of the period since the pandemic, clinical response to long COVID neurology was limited by the absence of objective biomarkers. Symptom reporting alone was not sufficient to justify aggressive pharmacological intervention; without a measurable biological target, treatment protocols defaulted to rest, pacing, and general support.

VMAT2 density, measured by PET scan, is a biomarker. It can be quantified, compared across patients, tracked over time, and linked directly to dopamine pharmacology that is already well-mapped from Parkinson’s disease, depression, and attention research. Several existing drugs act on exactly the pathways this study identifies as depleted.

The researchers plan to test two classes of treatment: dopamine precursors, which provide neurons with more raw material to manufacture dopamine, and inhibitors of dopamine metabolism, which slow the breakdown of whatever remaining terminals can still produce. Both approaches are already in clinical use for other conditions; the challenge was not pharmacological novelty but the absence of a rationale for applying them to long COVID. That rationale now exists.

A clinical trial in collaboration with University Health Network is expected to begin within months. It will test whether dopamine-targeted treatment measurably improves the three symptom clusters this study identified.

What this study does not yet answer

The sample size was not reported in published detail, which limits the ability to estimate statistical power or determine whether results hold across the full range of long COVID severity and time since infection.

The study design was cross-sectional: patients were scanned once, not before and after infection. This means the data cannot establish whether VMAT2 levels were lower before COVID-19 infection in affected individuals, or confirm how far the loss progressed. A longitudinal cohort — scanned before infection and at intervals afterward — would be needed to trace the full progression.

The biological mechanism connecting SARS-CoV-2 infection to dopamine nerve terminal loss is also not fully mapped. Brain inflammation is the leading hypothesis and the researchers support it, but the molecular steps between viral exposure and VMAT2 depletion have not been described.

Finally, no direct comparison has been published between long COVID patients and ME/CFS patients — a condition sharing several of the same symptom clusters. Whether dopamine system depletion is specific to long COVID or common to post-viral syndromes more broadly is an open question.

Common questions about long COVID and the brain

What is VMAT2 and why does it measure dopamine damage?

VMAT2 — vesicular monoamine transporter 2 — is a protein on dopamine nerve terminals that packages dopamine before release. A PET scan measures its density across brain regions without any tissue sampling. Lower VMAT2 density means fewer functional dopamine-releasing terminals in that region — not that dopamine is low at any given moment, but that the structures releasing it are structurally reduced.

Can long COVID be treated with dopamine medications?

That is the hypothesis the upcoming clinical trial is designed to test. The CAMH researchers have not yet administered the drugs; the study established the biological rationale for doing so. The trial will use dopamine precursors and dopamine metabolism inhibitors, both already approved for other conditions, and will measure whether symptoms improve against the biomarker baseline.

Is this the same mechanism as Parkinson’s disease?

There is anatomical overlap — both involve dopamine nerve loss in the striatum — but the conditions are not equivalent. Parkinson’s is a progressive neurodegenerative disease anchored in the substantia nigra, with severe motor consequences and a well-characterized course. This study identifies VMAT2 depletion across a broader striatum profile in long COVID patients, with a different symptom pattern. Whether they share any underlying molecular mechanism is not established.

Are the effects permanent?

The cross-sectional study design cannot answer this. The scans represent patients at a single point in time; without baseline pre-infection scans or sequential follow-up scans, the data do not show whether VMAT2 density recovers, stabilizes, or continues to fall. The treatment trial will begin to address this indirectly: a response to dopamine-targeted therapy would indicate the system retains enough function to benefit from pharmacological support.

The clinical trial, when completed, will provide the first prospective data on whether dopamine augmentation restores function in long COVID patients. Results are expected in 2027.

Reference: Meyer et al., “Loss of vesicular monoamine transporter 2 in striatum of long COVID and relationship to neuropsychiatric symptoms,” eBioMedicine, 2026; 130: 106339. DOI: 10.1016/j.ebiom.2026.106339

Tags: neuroscience, brain fog, CAMH, dopamine, long COVID, striatum