Every few weeks, someone will say to me, half-joking and half-afraid, “My brain hasn’t been the same since COVID.” They are rarely dramatic people. They are professionals who forget simple words in meetings, students who reread the same page three times, parents who stand in the kitchen and cannot remember why they walked in.
For a long time, it was easy for systems to dismiss this as anxiety, burnout, or the collective exhaustion of a pandemic age. The science is now making that much harder to do.
Recently, researchers at Griffith University’s National Centre for Neuroimmunology and Emerging Diseases in Queensland used advanced MRI to look directly at the brains of people with long COVID, those who felt recovered after SARS-CoV-2 infection, and people who had never been infected.[1] Their work sits alongside a growing body of studies from the UK, Europe and the United States suggesting that SARS-CoV-2 can leave a measurable footprint in the brain – sometimes long after the cough and fever have gone.[2–5]
This is not the apocalyptic “everyone will get dementia” story some headlines imply. But it is also not nothing. It is a quiet, structural nudge to take the neurological consequences of SARS-CoV-2 far more seriously than we currently do.
What the Queensland study actually found
The Griffith group led by Thapaliya used multimodal MRI – essentially several different MRI techniques layered together – to study grey and white matter in people with long COVID, people who had recovered from SARS-CoV-2 infection, and never-infected controls.[1] Rather than relying on a single measure, they combined:
T1-weighted/T2-weighted (T1w/T2w) signal ratios, often used as a proxy for myelin and tissue microstructure.
Diffusion-weighted imaging, which looks at how water moves through brain tissue and can reveal subtle white-matter disruption.
Magnetic resonance spectroscopy, which estimates levels of key brain chemicals such as N-acetylaspartate (a neuronal marker), choline and glutamate–glutamine.[1]
Three things stand out:
Changes in both long COVID and “recovered” individuals110Please respect copyright.ＰＥＮＡＮＡ0zdskveqxh
T1w/T2w signal and diffusion metrics were altered not only in people with long COVID, but also in those who considered themselves fully recovered when compared with never-infected controls.[1] That is important: the brain changes are not limited to those who remain obviously unwell.The changes track with symptoms110Please respect copyright.ＰＥＮＡＮＡxVOGmSjDbC
Higher symptom burden – especially fatigue and cognitive complaints – correlated with greater abnormalities in these MRI measures.[1] In other words, people who felt worse were more likely to show measurable differences in brain structure and neurochemistry.Neurochemical imbalance, not just “scarring”110Please respect copyright.ＰＥＮＡＮＡqF0zOJG4sk
In a companion study, the same group examined brain neurochemicals in long COVID and myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).[2] They found altered levels of several metabolites, suggesting shared disturbances in brain energy metabolism and glial function, even though not every measure differed cleanly between the two syndromes.[2]
This is not a horror-movie brain infection. It is more subtle: a pattern of microstructural and neurochemical shifts that appear to link a past SARS-CoV-2 infection to ongoing cognitive and physical symptoms.
SARS-CoV-2 and the brain: how the wider literature fits in
If the Queensland study stood alone, we might be more cautious. It does not.
Large-scale imaging work from the UK Biobank has already shown that even relatively mild SARS-CoV-2 infection can be associated with small but measurable loss of grey matter in olfactory and limbic regions, and greater reduction in global brain volume than seen in matched controls who were never infected.[3] Participants also showed declines in cognitive performance, particularly in tasks testing attention and executive function.[3]
In parallel, animal work has demonstrated that a purely respiratory SARS-CoV-2 infection in mice – with no direct viral invasion of the brain – can still trigger:
1. Persistent microglial activation (brain immune cells in a pro-inflammatory state)110Please respect copyright.ＰＥＮＡＮＡ2tIR5roDyb
2. Loss of myelin-forming oligodendrocytes110Please respect copyright.ＰＥＮＡＮＡVviiUFRHBo
3. Impaired hippocampal neurogenesis (reduced birth of new neurons)[4]
These changes map uncomfortably well onto the “brain fog” many patients describe: slower thinking, memory glitches, and a sense that mental effort is suddenly expensive.[4]
Clinical epidemiology adds another layer. In a massive US veterans’ cohort, people with a documented history of SARS-CoV-2 infection were at higher risk of a range of neurological conditions over the following year – including stroke, cognitive impairment, movement disorders and mood disturbances – even when the initial illness was not severe enough to require hospitalisation.[5]
Review articles and commentaries from leading neurologists and psychiatrists now converge on a view that:
SARS-CoV-2 can disturb the brain via several overlapping pathways – neuroinflammation, microvascular injury, immune dysregulation, and possibly autoimmune phenomena – even if direct viral invasion of neurons is rare.[6–8]
Long COVID represents part of a broader family of post-acute infection syndromes that also includes sequelae after Epstein–Barr virus and other pathogens.[8,9]
The Queensland MRI findings, then, are not an isolated curiosity. They slot into an emerging picture: SARS-CoV-2 infection is capable of perturbing brain structure, chemistry and function in ways that are small on a population level but potentially devastating for a subset of individuals.
Reading the scans with humility
It is tempting, especially in policy circles, to treat a striking MRI image as the final word. “Look, there is the damage,” we say, as if a brighter or darker patch on the screen explains everything.
I think we should resist that temptation.
First, MRI is a proxy, not a biopsy. A change in T1w/T2w ratio suggests altered myelin or tissue microstructure, but it does not tell us exactly which cells are affected or whether the process is degenerative, inflammatory or compensatory.[1] Spectroscopy-derived metabolite levels are also indirect, averaged over a region rather than tied to specific cell types.
Second, sample sizes remain modest. The Griffith study, like many long COVID imaging projects, involves relatively small numbers of participants compared with the scale of the pandemic.[1] That is not a criticism of the investigators – this is resource-intensive work – but it does mean we must be careful not to over-generalise from early cohorts who may be healthier, more educated, or more motivated than the average person with persistent symptoms.
Third, associations are not destinies. The fact that some people with long COVID show measurable brain changes does not mean everyone who has had SARS-CoV-2 infection is on a one-way track to dementia. Long-term trajectories remain uncertain. Some studies suggest partial normalisation of white-matter metrics over time; others document persistent abnormalities alongside overall improvement in symptoms.[3,10]
Finally, we must acknowledge the flip side: normal imaging does not invalidate suffering. A normal MRI does not prove that everything is “in someone’s head” – at least not in the dismissive sense. Many disabling neurological or psychiatric syndromes have no clear radiological correlate on standard imaging. The absence of a visible lesion is not the absence of disease.
The value of the Queensland and related studies is not in providing a simple “yes/no” test for long COVID. It is in forcing the conversation away from vague scepticism (“surely they are just anxious”) and toward specific biological questions we can actually study.
Where the science needs to go next
If we take these findings seriously – and I would argue that we should – they raise uncomfortable, practical questions.
Who is at greatest risk?110Please respect copyright.ＰＥＮＡＮＡH9DflAHs8W
Current data hint that risk is higher with more severe acute illness, but it is not confined to those who were hospitalised.[3,5] We still do not have simple clinical tools to identify which patient with “just a mild case” will later struggle with concentration, mood or autonomic symptoms. Smoking status, cardiovascular risk, prior mental health, and social deprivation may all interact, but the models are still crude.[5,9]What is the dose–response relationship?110Please respect copyright.ＰＥＮＡＮＡFCiiAcD2y6
Many of us have now had SARS-CoV-2 infection more than once. We do not know whether repeated infections have additive, synergistic or plateauing effects on long-term brain health. Large longitudinal cohorts, ideally with pre-infection cognitive data (like UK Biobank), will be essential.[3]How do we separate injury from adaptation?110Please respect copyright.ＰＥＮＡＮＡHgkdTD1sUK
Some of the changes seen on MRI or PET may represent harmful processes (ongoing inflammation, demyelination), but others may be compensatory re-wiring of networks after an insult.[3,4,8,10] Knowing which is which matters: we do not want to “normalise” away an adaptation that is helping someone function.Where do treatment trials fit in?110Please respect copyright.ＰＥＮＡＮＡXky2cdb7Dd
At present, most long COVID care is symptomatic: pacing, sleep hygiene, cognitive rehabilitation, management of dysautonomia, treatment of co-existing mood disorders.[9,11] There are emerging trials of anti-inflammatories, antivirals and neuromodulation, but they are still early-phase and often small.[4,11] We urgently need methodologically rigorous trials that use objective cognitive and imaging outcomes rather than exclusively self-reported scales.How do we integrate brain health into SARS-CoV-2 policy?110Please respect copyright.ＰＥＮＡＮＡlcZw8viD66
Long COVID is now estimated to affect tens of millions globally.[9,11] When governments decide whether to invest in ventilation, masking in hospitals, or updated vaccination campaigns, they often focus on deaths and ICU admissions. Chronic brain and mental-health sequelae, which erode productivity and quality of life over years, are seldom factored into that calculus.[5,9,11] They should be.
There is a broader ethical point here. It is politically convenient to declare the pandemic “over”. It is less convenient to acknowledge that a proportion of those who “recovered” after SARS-CoV-2 infection have since left the workforce, dropped out of education, or reorganised their lives around a diminished cognitive bandwidth.
For clinicians, researchers and the rest of us
For clinicians, the practical message is straightforward but not easy: believe your patients, but keep your scepticism scientific rather than dismissive.
When someone who previously coped well with a demanding job reports new, persistent brain fog after SARS-CoV-2 infection, the default should not be “stress, next please.” It should be a careful history, consideration of other causes (sleep apnoea, anaemia, thyroid disease, depression, medication adverse effects), and an honest conversation about what we do and do not yet know about post-infectious brain changes.
Imaging is not mandatory for everyone, and a normal scan does not negate the illness. But in certain cases – unexplained focal deficits, progressive decline, atypical headaches, seizure-like events – MRI, and occasionally more advanced imaging, may be warranted, ideally in a research context that helps build the evidence base.
For researchers, the Queensland study is a welcome sign that smaller centres can make globally relevant contributions. The challenge now is standardisation and collaboration: harmonised protocols, shared data structures, and cross-cohort meta-analyses that can move us beyond a patchwork of single-site findings.[1–4,10]
For the rest of us, there is a more uncomfortable reflection. The story of SARS-CoV-2 and the brain is, in part, a story of how modern societies handle slow harms. We respond decisively to ambulance sirens and overflowing ICUs. We are less good at reckoning with millions of people who can still walk and talk, but no longer quite think, remember or plan as they once did.
A cautious conclusion
SARS-CoV-2 is not the only virus that can disturb the brain. It may not even be the worst. But it is the one that has swept through a naïve global population at extraordinary speed, leaving behind a trail of subtle, uneven and sometimes delayed neurological footprints.
The Queensland MRI work does not prove that everyone is destined for catastrophe. What it does is quietly insist that we stop treating long COVID as an abstract, psychosomatic curiosity and start treating it as what the data increasingly say it is: a real, biologically grounded condition in which the brain is one of several organs caught in the aftermath.[1–5,7–11]
We can continue to debate details: which pathways matter most, which interventions are worth funding, what level of absolute risk is acceptable. But we can no longer honestly say that there is “nothing to see” in the brain after SARS-CoV-2 infection. The scans – and, more importantly, the people attached to them – suggest otherwise.
References
Thapaliya K, Marshall-Gradisnik S, Inderyas M, Barnden L. Altered brain tissue microstructure and neurochemical profiles in long COVID and recovered COVID-19 individuals: a multimodal MRI study. Brain Behav Immun Health 2025;50:101142.
Thapaliya K, Marshall-Gradisnik S, Eaton-Fitch N, Eftekhari Z, Inderyas M, Barnden L. Imbalanced brain neurochemicals in long COVID and ME/CFS: a preliminary study using MRI. Am J Med 2025;138:567-574.e1.
Douaud G, Lee S, Alfaro-Almagro F, et al. SARS-CoV-2 is associated with changes in brain structure in UK Biobank. Nature 2022;604:697-707.
Fernández-Castañeda A, Lu P, Geraghty AC, et al. Mild respiratory SARS-CoV-2 infection can cause multi-lineage cellular dysregulation and myelin loss in the brain. Cell 2022;185:2452-2468.e16.
Xu E, Xie Y, Al-Aly Z. Long-term neurologic outcomes of COVID-19. Nat Med 2022;28:2406-2415.
Boldrini M, Canoll PD, Klein RS. How COVID-19 affects the brain. JAMA Psychiatry 2021;78:682-683.
Spudich S, Nath A. Nervous system consequences of COVID-19. Science 2022;375:267-269.
Monje M, Iwasaki A. The neurobiology of long COVID. Neuron 2022;110:3484-3496.
Davis HE, McCorkell L, Vogel JM, Topol EJ. Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 2023;21:133-146.
Bremner JD, Russo SJ, Gallagher R, Simon NM. Acute and long-term effects of COVID-19 on brain and mental health: a narrative review. Brain Behav Immun 2025;123:928-945.
Dacosta-Aguayo R, Alarcón-Ribas D, Vergara-Méndez LD, et al. Multimodal neuroimaging in long COVID and its correlates with cognitive symptoms and fatigue. Front Neurol 2024;15:1423300.
110Please respect copyright.ＰＥＮＡＮＡmSiITQYWrM
Disclaimer:110Please respect copyright.ＰＥＮＡＮＡotlkD2XdHS
The information in this article is for general information and educational purposes only. It is not a substitute for independent professional medical advice, diagnosis, or treatment. Always consult a suitably qualified healthcare professional with any questions or concerns about your health or a medical condition. Never ignore, delay, or disregard professional medical advice because of something you have read in this article.
and then Add to home screen.