Shingles and brain health: an emerging link or healthy user bias?

Shingles, also known as herpes zoster, is an affliction caused by the varicella-zoster virus (VZV, the same virus responsible for chickenpox) and is marked by painful skin rashes and, frequently, postherpetic neuralgia, a form of prolonged nerve pain that may last months and even years in the worst cases. These effects alone are distressing and in many cases debilitating, and they alone provide plenty of justification to vaccinate against VZV. But recent evidence suggests that older adults may have an additional motivation for vaccination: the shingles vaccine may reduce risk for Alzheimer’s disease and other forms of dementia.

Shingles arises from the reactivation of VZV, which can remain dormant in nerve tissue for decades following an initial chickenpox infection. Since the virus impacts peripheral nerves, it can trigger neuroinflammation by stimulating immune responses in the nerve tissue. Neuroinflammation is widely recognized as an important factor in the development of dementia, leading to the hypothesis that viruses such as VZV could raise dementia risk – and thus, that vaccination against shingles might help to counteract this added risk.

The association between shingles and dementia

Researchers have attempted to establish a relationship between shingles and dementia through several observational studies, with varying success. For example, a 2023 meta-analysis including nine controlled trials and approximately 3 million patients found no association between shingles and dementia risk.¹ The pooled data indicate no statistically significant association between general shingles infection and dementia or Alzheimer’s disease risk (RR=0.99; 95% CI: 0.92–1.08 for dementia; RR=3.74; 95% CI: 0.22–62.70 for Alzheimer’s disease).

However, herpes zoster ophthalmicus (HZO), a specific manifestation of shingles involving the area around the eyes, was significantly associated with increased dementia risk (RR=6.26; 95% CI: 1.30–30.19; P=0.02), suggesting that HZO could pose a higher risk than shingles without ophthalmic involvement. (HZO affects 4-20% of people with shingles.²) This may be due to the unique ability of VZV to cause vasculopathies combined with the anatomical proximity to the cerebral arteries. In other words, the reactivated virus can travel to and infect the nearby arteries supplying the brain. Indeed, when controlling for many common risk factors for dementia (hypertension, diabetes, hyperlipidemia, stroke, and coronary heart disease, etc.), one study found an association between HZO and dementia (HR = 2.97; 95% CI: 1.90–4.67).³ While it seems HZO specifically may impact cognitive function, the current evidence is not sufficient to define a causal relationship between a majority of shingles cases and dementia risk.

What about the shingles vaccine?

Despite the lack of clear added risk of dementia among those with shingles, it nevertheless seems that shingles vaccines may confer some cognitive protection. In fact, a 2024 meta-analysis of four studies concluded that shingles vaccination indeed was associated with a modestly reduced risk of cognitive decline (pooled OR: 0.76; 95% CI: 0.60-0.96, P=0.02).⁴

Two new articles support this conclusion. In a paper from this past July, authors Taquet et al. show that receiving the recombinant shingles vaccine (compared to the less potent the live-virus version) correlated with a 17% longer dementia-free period over a six-year follow-up.⁵ The following month, investigators Xie et al. reported a 3.1% reduction in mild cognitive impairment over nine years among live-virus vaccine recipients compared to unvaccinated controls, which further points to a modest reduction in dementia incidence.⁶

The astute reader might currently be screaming, “but what about healthy user bias?!” The “healthy user bias” refers to the fact that those who are proactive about their health in one respect (for instance, by getting vaccinations) are more likely to engage in other healthy behaviors, as well, including others that might reduce dementia risk (such as exercising or getting enough sleep), creating a potential confounding effect. Some observational studies have attempted to subvert the healthy user bias with opportunistic study designs. For example, Taquet et al. take advantage of the fact that the live-virus vaccine had previously been the standard but was rapidly replaced by the more effective recombinant vaccine. This allowed them to compare apples to apples (or rather, vaccinated to vaccinated individuals) by comparing those who were vaccinated before versus after the transition. The authors also compare the outcomes in those vaccinated against shingles to those who received vaccinations for other viruses – specifically, influenza and tetanus & diphtheria & pertussis (Tdap) – and found that both shingles vaccines were associated with lower risk of dementia than the other vaccines, suggesting the effect is specific to shingles vaccination. 

More to the story

So it’s settled – right? Maybe shingles doesn’t cause dementia, but the shingles vaccine will protect us from it – right? 

Not so fast. While the newer studies utilize clever strategies to get around healthy user bias, they are still observational, so we must be cautious in interpreting causation. While it is true that Taquet et al. show modestly reduced risk with shingles vaccination compared to influenza and Tdap, this is just one study. A meta-analysis examining 17 studies including nearly 2 million participants shows a 35% lower dementia risk (HR=0.65, 95% CI: 0.60-0.71, P<0.001), regardless of vaccination type. Considering just the vaccines examined across multiple studies in this meta-analysis, shingles (HR=0.69, 95% CI: 0.67-0.72, P<0.001), Tdap (HR=0.69, 95% CI: 0.58-0.82, P<0.001), and influenza (HR=0.74, 95% CI: 0.63-0.87, P<0.001) were all significantly associated with decreased risk of dementia.⁷ This broad effect lends some support to the idea of nonspecific immunomodulation by vaccines. In other words, perhaps being vaccinated (regardless of the type of vaccine), confers an immunological benefit beyond the disease it prevents. 

Double clicking on that idea, note that vaccines are designed to protect against specific infections, but some research suggests they may have unexpected nonspecific effects that enhance the immune system’s ability to fight other diseases. For example, the BCG vaccine (for tuberculosis) has been shown to reduce overall mortality rates more than would be expected based on their protection against the targeted infections alone. This broader effect is thought to occur through two primary mechanisms: trained immunity and cross-reactive immunity. Trained immunity strengthens the innate immune system – the body’s first line of defense – by triggering cellular changes that enhance the ability to combat a wide array of pathogens. Cross-reactive immunity, on the other hand, primes adaptive immune cells, like T cells, to recognize and respond to pathogens that are unrelated to the original target. 

These combined effects help explain why vaccinated individuals may experience greater resistance to infections beyond those the vaccine was initially designed to protect against. A more robust immune system may prevent cognitive decline by reducing inflammation (a trained immune system can neutralize a threat more quickly, reducing the amount of time the inflammatory process is occurring). The hypothesis that this could prevent cognitive decline becomes more plausible when we consider diseases with neural tropism (i.e., a propensity to target the nervous system), like shingles. 

Notice I used qualifiers like “may” and “perhaps” in this explanation. It is of critical importance to point out that this is just a hypothesis and that much of the support for it comes from observational epidemiological studies. While there seems to be a reproducible trend of vaccines associating with reduced risk of dementia, we do not yet have sufficient data to support a specific mechanism of nonspecific immunity. In the end, it may indeed be healthy user bias.

Should you get the shingles vaccine?

For those over 50, shingles vaccination is already recommended and highly advisable for its proven benefits in shingles prevention, yet as of 2018, only 35% of individuals over age 60 get this vaccine.⁸ The recombinant shingles vaccine (Shingrix) is about 97% effective in preventing shingles for individuals aged 50–69 and 91% effective for those over 70. Given shingles’ high incidence and severe health implications in older adults, the primary reason to get vaccinated remains preventing this painful condition and its complications, particularly postherpetic neuralgia. While early epidemiological findings on dementia risk reduction are encouraging, it’s best to consider them as secondary perks rather than a primary motivation for shingles vaccination. The research on vaccination and cognitive outcomes remains preliminary and requires further investigation to confirm these potential effects definitively.

The shingles vaccine is recommended for individuals who have had chickenpox, as they carry latent VZV and therefore face a risk of shingles. If we are to believe that reactivation of VZV (especially in the case of HZO) is causing neuroinflammation or vasculopathies that lead to possible risk of dementia, people who have never contracted chickenpox would not have dormant VZV and thus would not face a similar shingles (or cognitive) risk. For these individuals, the dementia prevention potential observed in vaccinated individuals may not apply, as their risk of VZV reactivation – and its associated inflammation – is effectively zero. If we are to believe, however, that the vaccine is imparting some nonspecific immunity, those who have not had chickenpox may still benefit from the vaccine. 

The current form of the shingles vaccine, Shingrix, is a recombinant vaccine, meaning it contains only a portion of the virus. More specifically, Shingrix contains a protein from the varicella zoster virus. This is in contrast to an older version of a shingles vaccine, Zostavax, a live attenuated vaccine (i.e., a weakened version of VZV). By including only a protein from VZV, Shingrix is not infectious. Because the vaccine will inherently prompt a strong immune response, there are several likely side effects (injection site reaction, fever, nausea, muscle pain, fatigue) that resolve within days of receiving the vaccine. The recombinant vaccine, which more effectively prevents shingles, is associated with more injection site reactions than its live vaccine counterpart; however, serious adverse events are rare, showing no difference to placebo.⁹

The bottom line

While the shingles vaccine’s potential to reduce dementia risk remains an exciting prospect, I have yet to be convinced that it is not mostly attributable to healthy user bias or possibly nonspecific effects of vaccination. If shingles does increase dementia risk (and therefore the vaccine against shingles decreases dementia risk), it seems more likely it is related specifically to shingles with ophthalmic involvement. The vaccine’s well-documented effectiveness at preventing shingles should remain the top priority for older adults considering it (as anyone who has had, or known someone who has had, shingles will agree!). The possible added benefit of delaying dementia onset, while promising, will require more rigorous research to verify and fully understand. (For more on shingles and the shingles vaccine, see also my podcast interview with Dr. Michael Gershon.) In the meantime, consider the vaccine as a valuable tool to prevent the immediate, tangible effects of shingles, with the potential for a welcome bonus in cognitive health as research continues.

For a list of all previous weekly emails, click here. podcast | website | ama