COVID injures the vessels your brain runs on.
Almost every adult now carries antibodies. The lasting mark lands on the small vessels and the blood-brain barrier, and the largest study measured the cognitive cost in points of IQ. Here is the mechanism, the toll, and the vaccine question answered without a side.
There is almost no unexposed group left.
By the end of 2023, antibodies from infection or vaccination were present in more than 19 in 20 adults.1 Each figure below is one person: red if they carry those antibodies, blue if they have none.
Based on CDC nationwide blood-donor antibody surveys, a convenience sample, not a random sample of adults. The never-infected group, the natural comparison for the harms below, is shrinking and blurred: most infections were never tested, so many people counted as never infected had been infected.
There is almost no one left to compare them to.
Measured, and almost certainly undercounted.
A study of 112,964 people measured cognitive loss that scaled with severity: about 3 IQ-equivalent points after a mild infection that cleared, 6 with lingering symptoms, 9 after hospital care.2
Those numbers are set against people counted as never infected, though many of them caught it without knowing, so the true gap is wider than any study can show.
The bar stays part empty on purpose: the top is the room for the infections you never counted. A rough illustration drawn from the severity pattern in the study above, not a personal prediction.
The same handful of complaints, over and over.
None of these proves anything on its own; they are common and have many causes. But several of them showing up together, all since one infection, is the recurring pattern.
A plumbing problem before it is a brain problem.
SARS-CoV-2 has a strong affinity for the endothelium, the thin lining of blood vessels, including the microscopic ones feeding the brain.3 Injure that lining and three failures follow: the wall leaks, flow drops, and the brain can no longer route blood to where it is needed.4
On brain imaging, post-COVID brain blood flow runs 10 to 40 percent lower in the regions hit hardest.4 The cognitive cost is larger after a severe infection.2
The seal breaks.
The tight junctions holding the vessel shut come loose, and the barrier that walls neurons off from the bloodstream fails. Inflammation seeps into the tissue, and the cells it reaches start to misfire.
The channel narrows.
The cells slow and bunch up as the vessel narrows, and the tissue on the far side gets less.
The timing breaks.
A healthy brain rushes extra blood wherever it is needed. When that timing slips, the shaded gaps open: the instant focus is needed and the blood is late or missing. That gap is the thought you lose, or the word that will not come.
Schematic, not to scale. As you scroll deeper, blood flow falls from full to roughly a third short, and the diagrams change with it.
What that shortfall does to your thinking
The brain keeps almost no fuel in reserve, so when the supply drops, thinking degrades in a fixed order, long before any tissue is at risk: the most expensive jobs fail first.
New memories go first. Laying one down is among the most energy-hungry things the brain does, and the hippocampus that does it sits in a thin part of the blood supply. This is the “wait, where was I going”.
Then working memory and focus. Holding something in mind and acting on it leans on the frontal lobe, which needs a steady, on-demand top-up of blood. In studies of low flow, frontal and temporal supply fall in step with these scores, while language and spatial sense are relatively spared.5
Then speed. The long wiring between regions, and the “watershed” zones fed by the thinnest ends of the supply, are starved first, so the hand-offs slow and thinking starts to feel like wading.6
Brain fog is mostly the timing failure. When a region gets to work, its vessels are meant to open and rush blood in; when that response is blunted, the supply never meets the spike in demand. This is graded and partly reversible once flow returns, because the cells are starved, not dead.
Two careful people can land on opposite answers.
The loudest voices treated this as obvious in both directions. It was not.
See the cited reasoning on both sides
Why a careful person got it
The clearest benefit was against severe disease and death, strongest in older and higher-risk people, where the downside of catching COVID unprotected was large.7
It also lowered the odds of long COVID by roughly a quarter to a third, including the cognitive and sleep symptoms this page is about.8 Heart inflammation was several times more likely from infection than from the shot.9 For them, the larger, better-measured risk was the virus.
Why a careful person held back
A young, healthy person faced a much smaller risk of serious illness and death from COVID to begin with, so the benefit they stood to gain was smaller, and the protection against catching it faded within months.7
Against that, the myocarditis signal was concentrated in young men after the second dose, roughly 10 to 20 cases per 100,000 in the highest-risk group.9 Most resolved quickly, but a rarer post-vaccination syndrome is still being studied.10 Waiting for more data, for the lowest-risk people, was a defensible call.
Both used real information. The honest answer depended on age, health and risk, which is why it was a personal call rather than a mandate. For most people it was not a cause of the blood-flow damage above. For the lowest-risk, never-infected person the math could genuinely fall the other way. Across the population, though, by lowering long-COVID risk and cutting how much virus people met,7 vaccination more often reduced that damage than added to it.
Did the vaccine’s spike protein cause the same damage as COVID?
Same protein, plausibly the same mechanism, but for most people a different dose, place and duration.
The spike protein on its own can injure vessel lining by suppressing ACE2 and starving the cell of energy.11 Vaccine-made spike can occasionally reach the blood: in the rare cases of post-vaccine myocarditis, free spike was measurable when it was absent in healthy vaccinated people.12
- Dose
- The virus copies itself without limit. A shot delivers a fixed, self-ending amount.
- Location
- The virus reaches vessels throughout the body. Vaccine particles stay largely in the arm and nearby lymph nodes.13
- Duration
- A translated dose clears in days to weeks; replicating virus can persist for weeks to months. A few studies report vaccine material lingering longer, contested.
Illustrative direction and rough size, not measured ratios; the infection bar runs off the chart because the virus has no dose ceiling. Crimson marks infection, blue marks the vaccine.
Infection also brings the whole virus and every other viral protein, not the spike alone. A bad infection and a bad reaction to a dose can land a person in the same low-perfusion place, so the same recovery steps apply to both.
COVID stacks onto machinery that was already there. Its lasting mark is on the small vessels and the blood-brain barrier,3 the same supply line behind much of this site. If your trouble is worst on standing, see when standing is the problem. If it is long-term memory and ageing, see the ageing brain.
The conditions it drives directly.
These are the conditions COVID owns most directly, through the vessels and the clotting it leaves behind. Open any to find the verdict and the evidence.
Stroke and TIA the acute case
Stroke is the moment brain blood flow fails. A clot blocks an artery and flow to that territory collapses: below a critical level neurons fall silent but stay alive, the salvageable penumbra, and lower still the tissue dies within minutes.14 That is why clot-busting and clot-removal treatment races the clock. A TIA is the same shortfall that resolves before any tissue dies, but it is a serious warning: a meaningful share of people go on to a full stroke within days.15 COVID itself raises short-term stroke risk, around three times in the first week and roughly double for months afterward, through clotting and vascular inflammation.16 About 90 percent of stroke risk traces to ten modifiable factors led by high blood pressure, so the absolute COVID-linked rise is small next to the controllable drivers.17
Concussion and TBI strong and direct
After a concussion, brain blood flow usually drops, and the drop often lingers after you feel back to normal, normalising more slowly in people who take longer to recover. The damage sits in the vasculature: the small vessels and neurovascular unit are physically strained, the blood-brain barrier and autoregulation are disrupted, and the vessels lose some ability to ramp flow up on demand.18 The chronic picture depends on severity and age, and repeated head impacts carry their own vascular signature, with the tau of CTE clustering around small blood vessels where impacts strain the brain most.19 No two concussions look alike, so perfusion imaging confirms the biology but cannot diagnose any single person.20
Multiple sclerosis a secondary feature
Multiple sclerosis is an immune attack on the brain’s myelin, but reduced blood flow comes with it. Perfusion scans repeatedly find lower flow not just in lesions but throughout normal-looking white and grey matter from the first attack, with some cohorts measuring whole-brain flow 15 to 20 percent below controls, and lower grey-matter perfusion tracking modestly with worse fatigue and slower thinking.2122 The direction of cause is unsettled. The one randomised trial that tried to reverse the hypoperfusion with a vessel-relaxing blocker did nothing for flow or symptoms, leaving open that the low flow may largely reflect tissue demanding less energy.23 Treat perfusion as a real, early but secondary feature, and ignore the debunked blocked-neck-veins (CCSVI) theory.24
Keep your brain better supplied.
Most research on cerebral blood flow never reaches the people it could help. We read it and send only what changes what you can do: a new way to raise your own blood flow, or a finding that moves the advice on this page.