News · Brain & Mental Health
Shingles that reached the eye tracked 37% more dementia than shingles elsewhere
Shingles has been tied to dementia, but not every case is the same. Across 71 US health systems, cases reaching the eye carried more dementia within three years, and prior vaccination tracked less of it in both groups.
- Shingles that reached the eye was followed by more dementia than shingles elsewhere.
- 2.45% against 1.88% within three years, among people who had all had shingles.
- The gap showed up only in people over 70, not in those aged 50 to 69.
- Vaccinated people had less dementia even when the shot failed to prevent shingles.
- Records from 71 US health systems, matched carefully but not randomly assigned.
Shingles and dementia have been circling each other in the literature for years. The awkward part of that link has always been the comparison: people who get shingles are older, more run down and more likely to be ill in other ways than people who do not, and all of those things lead to dementia on their own.
A study in the journal Cornea sidesteps the problem by comparing shingles against shingles. Every person in it had the infection. What differed was where the virus came back.
When it reached the eye, dementia followed more often: 2.45% within three years, against 1.88% when it appeared anywhere else.
Why the eye is a different case
Shingles (herpes zoster) is an infection that causes a painful rash, caused by the varicella-zoster virus left behind after chickenpox and reactivating along a single nerve decades later.
Which nerve it picks decides what happens. Usually it is a band of skin on the trunk. Sometimes it is the nerve serving the forehead and eye, and then the rash arrives on the face and the inflammation reaches the eye itself, which is an emergency for sight.
That branch of the nerve runs close to the brain, and it is the anatomical reason anyone would expect the two versions of the same infection to have different consequences beyond the eye.
The comparison that makes this worth reading
Comparing people who had shingles with people who did not is the standard approach and the weak one, because getting shingles is itself a marker of a failing immune system.
Here, both groups came from the same pool of shingles patients aged 50 and over, recorded between January 01, 2005 and January 01, 2024, and were matched against each other. Age, immune status and general frailty predict having shingles at all, and both groups have already cleared that bar. What is left is the difference in where it landed.
That is not a randomized trial, and it is a good deal closer to one than the usual design in this field.
What the records held
The data came from deidentified electronic health records from the TriNetX US Collaborative Network, covering 71 health care organizations and nineteen years of cases.
In absolute terms the gap is six extra dementia diagnoses per thousand people over three years, on a base of about nineteen. The figure of 37% sounds larger than the experience it describes, which is what happens when a percentage change is applied to an uncommon outcome.
More interesting is where the gap lived. Age-stratified analyses showed no significant difference in dementia risk for the eye group against the rest, in patients aged 50 to 69. It appeared in the seventies and eighties and not before.
A brain with reserve appears to absorb the insult. A brain already close to its limits does not, or at least does not within three years, which is all this study watched.
The vaccination result
The second finding is the one with something to do.
Among people who developed shingles despite having been vaccinated, dementia was less common over the next three years than among people who developed shingles having never been vaccinated. That held whether or not the eye was involved.
The authors put it carefully, as supporting current shingles vaccination recommendations and suggesting potential cognitive benefits even when zoster is not fully prevented. The phrase doing the work is not fully prevented: these were all vaccine failures, in the narrow sense that everyone in the comparison got shingles anyway, and the vaccinated ones still came out ahead.
If real, the mechanism would presumably be a milder infection rather than no infection. A vaccinated person whose shingles breaks through tends to have a shorter, less severe episode, and less of whatever it is about the episode that harms the brain.
Why the vaccination finding is the shakier half
People who take up vaccination are not a random subset of people. They see doctors more, are less frail, and are better off, and every one of those predicts a lower chance of a dementia diagnosis regardless of any vaccine.
That bias runs in exactly the direction of this result, and no amount of matching on recorded characteristics removes it, because the things that matter most are not in the record.
The eye comparison is far less exposed to that problem, since nobody chooses which nerve the virus travels down. It is the more trustworthy of the two findings, and it is also the less useful one, because there is nothing a person can do about it.
What follows
For anyone who has had shingles on the face, this is not a reason for alarm and it is a reason to mention it. Three years of slightly elevated risk in people over seventy is the sort of thing a doctor should know is in the history, and eye involvement should already have brought urgent care at the time.
For the vaccine, the case does not rest on this. Shingles vaccination is recommended because shingles is painful, sometimes sight-threatening, and can leave nerve pain that lasts for years. A possible cognitive dividend is a bonus that this study cannot confirm and a randomized trial has not yet tested.
What the paper adds is a clean comparison in a field short of them, and a specific, checkable claim: that where the virus reactivates changes what happens next.
People also ask
What did the study find?
Shingles affecting the eye was associated with higher 3-year incident dementia risk compared with uncomplicated nonophthalmic zoster (2.45% vs 1.88%; adjusted HR 1.37, 95% CI 1.24-1.48). Prior zoster vaccination was associated with lower dementia risk both among patients with eye involvement (aHR 0.77) and among those without (aHR 0.75).
What is shingles in the eye?
Shingles is caused by the chickenpox virus reactivating along a nerve. When the nerve involved supplies the eye and forehead, the rash and inflammation reach the eye itself, which can threaten sight and needs urgent ophthalmic care.
Why compare eye shingles against ordinary shingles rather than against no shingles?
Because both groups had shingles, so the comparison strips out most of what makes someone prone to it in the first place: age, immune status, general frailty. What differs is where the virus reactivated, which is closer to a natural experiment than comparing infected with uninfected.
How big is the difference in real terms?
About six extra dementia diagnoses per thousand people over three years, on top of roughly nineteen. Meaningful across a population and small for any one person.
Why did it only appear over 70?
Age-stratified analyses showed no significant difference for those aged 50 to 69, with elevated risk in the 70 to 79 and 80 to 89 groups. One reading is that the extra insult only matters to a brain already under strain. Another is that three years is too short to see anything in younger people.
What about the vaccination finding?
Among people who developed shingles anyway, those who had been vaccinated beforehand went on to less dementia than those who had not, in both the eye and non-eye groups. The authors read this as supporting current shingles vaccination recommendations and suggesting potential cognitive benefits even when zoster is not fully prevented.
Can this prove vaccination protects the brain?
No. People who get vaccinated differ from those who do not in ways records cannot fully capture, and that difference alone can produce a result like this. It is a reason to take the question seriously, not an established benefit. This is general information rather than medical advice.