Avoiding a severe bout of flu in middle age may protect a person’s future brain health, according to a large population-based study. In the January 19 Neuron, researchers led by Mike Nalls at the National Institutes of Health, Bethesda, Maryland, reported that, among 450,000 people in Finland and the U.K., 12 common viral illnesses increased a person’s risk of developing any of six neurodegenerative diseases over 15 years. Pneumonia caused by the flu associated with five diseases, while viral encephalitis or meningitis posed the greatest risk for Alzheimer’s disease. The scientists suggest that vaccination against these viral illnesses might protect against neurodegenerative diseases.

  • Flu-related pneumonia associated with higher risk of five neurodegenerative diseases.
  • Viral encephalitis increased the risk of Alzheimer’s by 30-fold.
  • Risk abated over 15 years.

“This is a very interesting and important paper that strongly supports the link posited for over 33 years … between infections, herpes simplex virus 1 (HSV1) in particular, and AD,” wrote Ruth Itzhaki of the U.K.’s University of Manchester. Oliver Goldhardt, Technical University of Munich, Germany, was in agreement. “Because of the growing evidence, the link between viruses and NDDs, in my opinion, cannot be dismissed,” he wrote (comments below).

Previous research suggested that having cold sores or genital lesions, both caused by HSV, increases a person’s risk of developing AD (Feb 2021 news; Apr 2021 conference news). Likewise, someone who caught Epstein-Barr virus (EBV), which causes mononucleosis, is a whopping 32-fold more likely to develop MS (Bjornevik et al., 2022). While the impact of COVID on NDD risk is still unknown because data are scarce and the length of follow-up short, researchers have seen that severe COVID can worsen existing neurological problems and speed cognitive decline in the year after infection (Apr 2021 conference news; Mar 2022 news). However, nobody has systematically looked for links between NDDs and all common viral infections.

Now, co-first authors Kristin Levine and Hampton Leonard have analyzed medical records from 344,000 people from the Finnish nationwide biobank, FinnGen, and 106,000 in the U.K. Biobank. All participants were of European ancestry and older than 60 at baseline. About 405,500 were healthy, while 44,500 had been diagnosed with an NDD: 11,650 with AD, 2,750 with vascular dementia, 18,700 with all-cause dementia, 7,200 with Parkinson’s disease, 840 with amyotrophic lateral sclerosis, and 3,500 with multiple sclerosis.

Levine and Leonard compared NDD diagnosis to prior hospitalization with 32 common viral illnesses, including flu, pneumonia, viral warts, chickenpox/shingles, viral encephalitis, and meningitis. Viral exposure was based on hospital billing codes, so the researchers could not say whether it was the infection or something else that sent a person to the hospital. Any one of 12 of the illnesses correlated with a higher incidence of any of the six NDDs in both databanks. Notably, most of these viral illnesses are caused by neurotropic viruses—those that are able to get into the CNS and into neurons. These include influenza viruses, HSV, the herpes zoster virus that causes chickenpox/shingles, various enteroviruses that cause meningitis, and EBV.

Which viral illness posed the greatest NDD risk? People who had had viral encephalitis were 31 times more likely to develop AD and 40 times likelier to have dementia of any kind than were people not hospitalized for infection. Likewise, AD risk jumped a whopping 62-fold after meningitis. Nalls told Alzforum that he was surprised by the strength of these associations. People who were infected with EBV were four times more likely to get MS, an eightfold lower risk than previously reported. Nalls attributes the difference to the hospital billing codes used here, versus blood testing for the virus in prior studies. “What struck us is that our results produced a similar finding even though we mined a very different type of data,” he told Alzforum.

And it was not just infections of the brain. People hospitalized for a viral intestinal infection had three to five times the risk of developing AD or vascular dementia. Viral hepatitis tripled PD risk, while herpes zoster boosted the likelihood of developing vascular dementia and MS two- to sixfold.

Of all the infections studied, flu or flu-induced pneumonia were tied to the most NDDs. People from both cohorts who had been hospitalized for flu, with or without pneumonia, were two- to sevenfold more likely to have AD, all-cause dementia, vascular dementia, PD, or ALS.

Since infections are typically acute and neurodegenerative disease can advance slowly over years or decades, the authors wondered if the risk for NDDs wanes after infection. Levine and Leonard found that it was highest the first year, then fell over 15 years. For example, hazard ratios for all-cause dementia slid from 83 within a year of viral encephalitis to 24 over the next four years, then to five by year 15 (see image below). That’s riskier than having an ApoE4 allele or a TREM2 R47H mutation. “We were surprised to see any associations after that long,” said Leonard.

Risk versus Time. After a viral infection causing warts (blue), an unspecified illness (mustard), encephalitis (lavender), pneumonia (green), flu (crimson), or flu-induced pneumonia (light blue), hazard ratios for developing all-cause dementia are highest within the first year and drop over 15 years. [Courtesy of Levine et al., Neuron, 2023.]

What does this all mean for middle-aged folks? Hugo Lövheim of Umeå University, Sweden, noted that all these cases were in people who had severe symptoms. Therefore, they didn’t represent people who rode out the flu or shingles at home. Preventing severe viral illness by getting vaccinated against the flu, pneumonia, or shingles may partially protect someone from getting an NDD, the authors suggest. Indeed, a previous analysis concluded that people who got flu or pneumonia shots were less likely to get AD than the unvaccinated (July 2020 conference news).

As for what’s next, Levine said they are running similar unbiased analyses on links between NDDs and bacterial infections, specific types of intestinal infections, and sleep disorders. Nalls said they are also looking for enough data on COVID infection and NDD diagnosis to be able to analyze.

Lavinia Alberi Auber of the Swiss Integrative Center for Human Health, Fribourg, and others have proposed a worldwide consortium to study causative links between infectious agents and NDDs in a standardized way. “We need to study the human pathobiome as seriously as we do the human genome,” Alberi Auber wrote (comment below). “We are exposed to many pathogens every day, and what makes one individual more susceptible to developing a severe or chronic infection is the host-pathogen interaction.” The researchers are seeking funding for the consortium.—Chelsea Weidman Burke


  1. This is a very interesting and important paper, which strongly supports the major link posited for over 33 years (though all too often disregarded), between infections—HSV1 in particular—and AD. I was expecting a study of this sort following the important study by Bjornevik et al. (2022) on EBV and MS.

    Levine et al. mention the possible artefact that AD patients are more susceptible than controls to infection, because of likely damage to the blood-brain brain barrier. That might well be the case, but I should mention that the prevalence of HSV1 DNA in brains of patients and controls was found to be quite similar (Itzhaki et al., 1997), consistent with the mantra that “infected does not necessarily mean affected,” i.e., some controls are infected—but asymptomatically, a concept perhaps more readily understood now that it has been noted in some COVID-19 cases.

    The reason Levine et al. find a weaker link with a longer time interval between infection and diagnosis of brain disease is explainable if the concept that has been proposed for the HSV1-AD connection is correct: that the virus travels from the initial site of its latency—the trigeminal ganglia in the PNS—to the brain probably in late middle age, as the immune system declines. I suggested that because the prevalence of HSV1 DNA in brains of older people was much higher than in brains of young people (Jamieson et al., 1992). Consistently, intrathecal antibodies to HSV1 were present in older people, but absent in infants (Wozniak et al., 2005). Perhaps if infection occurs at an older age in some people, it travels more readily to the brain than in those infected when young, and possibly in some of the young, it never reaches the brain. (In the past, most people were infected in infancy, whereas now, primary HSV1 infection occurs from infancy onwards.)

    Are the results of Levine et al. explained by the viruses they investigated causing neuroinflammation and subsequent reactivation of latent HSV1 residing in brain? We have shown that in three-dimensional cell culture brain models, infection with varicella zoster virus (VZV) causes reactivation of quiescent HSV1 (Cairns et al., 2022), and we have since found that this occurs with several other infectious agents, and also, excitingly, with a completely different type of injury (Cairns, Itzhaki, Kaplan, to be published).

    In life, infections and other types of damage, such as axonal injury, stress, immuno-suppression, and burns, are known to cause reactivation of herpesviruses, with consequent inflammatory and direct viral damage; some cause reactivation in brain. The question is then: Does HSV1 reactivation in brain lead to AD/dementia? The answer is likely to be yes, because HSV1 infection of the three-dimensional brain model triggers the development of precisely the characteristic phenotypes of AD. In contrast, infection of the cells by VZV (Cairns et al., 2022), or by the other infectious agents that we investigated, does not lead to these phenotypes, so HSV1 might be unique in causing AD-like development. This seems consistent with the fact that no other virus has been implicated in AD/dementia, nor found at high prevalence in elderly brains. Strong support for HSV1 reactivation leading to AD/dementia comes from data showing that recurrent reactivation of HSV1 in brains of infected mice causes the development of cognitive defects and accumulation of AD hallmarks (De Chiara et al., 2019); also, epidemiological data show the high correlation between HSV1 reactivation (IgM-seropositivity) and risk of AD/dementia (Letenneur et al., 2008; Lövheim et al., 2015). 

    Lastly, it is good that Levine et al. mention bacterial involvement, even though only in relation to meningitis, as it seems likely that at least two or three other types of bacteria—Borrelia, C. pneumoniae, P. gingivalis—might be involved, as well as HSV1, in AD/dementia.


    . Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science. 2022 Jan 21;375(6578):296-301. Epub 2022 Jan 13 PubMed.

    . Herpes simplex virus type 1 in brain and risk of Alzheimer's disease. Lancet. 1997 Jan 25;349(9047):241-4. PubMed.

    . Herpes simplex virus type 1 DNA is present in specific regions of brain from aged people with and without senile dementia of the Alzheimer type. J Pathol. 1992 Aug;167(4):365-8. PubMed.

    . Productive herpes simplex virus in brain of elderly normal subjects and Alzheimer's disease patients. J Med Virol. 2005 Feb;75(2):300-6. PubMed.

    . Potential Involvement of Varicella Zoster Virus in Alzheimer's Disease via Reactivation of Quiescent Herpes Simplex Virus Type 1. J Alzheimers Dis. 2022;88(3):1189-1200. PubMed.

    . Recurrent herpes simplex virus-1 infection induces hallmarks of neurodegeneration and cognitive deficits in mice. PLoS Pathog. 2019 Mar;15(3):e1007617. Epub 2019 Mar 14 PubMed.

    . Seropositivity to herpes simplex virus antibodies and risk of Alzheimer's disease: a population-based cohort study. PLoS One. 2008;3(11):e3637. PubMed.

    . Reactivated herpes simplex infection increases the risk of Alzheimer's disease. Alzheimers Dement. 2015 Jun;11(6):593-9. Epub 2014 Jul 17 PubMed.

  2. Levine at al. have published an important work linking viral infections and neurodegenerative diseases (NDD), pointing to the possible contribution of various viruses to inflammation and neuronal damage in the brain.

    As a triggering factor for NDDs, one would expect increasing hazard ratios (HRs) with latency after viral infection. Instead, for example in the caseload data of pneumonia and Alzheimer's disease (AD), the HRs for AD, among others, are higher one year after infection than after longer periods. Because AD has a long preclinical stage that begins decades before diagnosis and likely predates the onset of pneumonia, there is also a possibility that NDDs, such as AD, are associated with higher susceptibility or vulnerability to (neurotrophic) viruses. As proposed by our group at TUM, one mechanism is the reduction of virostatic soluble amyloid in the brain, which is reduced to insoluble fibrils (Goldhardt et al., 2022).

    However, the authors note that HRs for AD after pneumonia (influenza) are higher than HRs for pneumonia after AD, suggesting a rather direct contributing effect of pneumonia to NDD progression. This interpretation should be considered with caution because this effect is also observed in etiologically different NDDs, and because the data, from hospital records, reflects only the more severe infection cases, as mentioned by the authors. There are additional considerations. First patients in whom NDD occurs before infection may have a shorter life expectancy, with a lower likelihood of developing pneumonia than have patients in whom pneumonia occurs first followed by an NDD. Conversely of course, people with pneumonia may have died and not been diagnosed with an NDD, even though it was present. Second, patients with pneumonia may be more likely to be diagnosed with NDD because cognitive impairment becomes apparent during treatment/hospitalization for the pneumonia. In this context, it would be interesting to review the data for the occurrence of delirium, which associates with cognitive deterioration, often requires diagnostic testing (observing bias), and increases likelihood for an NDD diagnosis.

    Because of the growing evidence, the link between viruses and NDDs, in my opinion, cannot be dismissed. To solve the chicken-and-egg problem (Is there an increased susceptibility to viruses in the presence of NDD, or do viruses have an influence on the development of NDDs?), it would be desirable to include vaccination registry data as well. The authors plan to add more time-sensitive analyses, and to control for one or the other limitation. I really look forward to the next update.


    . Herpes simplex virus alters Alzheimer's disease biomarkers - A hypothesis paper. Alzheimers Dement. 2023 May;19(5):2117-2134. Epub 2022 Nov 17 PubMed.

  3. The paper from Levine and colleagues is a retrospective observational study analyzing the association between 22 replicated infectious events occurring at defined point times (one, five, and 15 years) before the onset of a spectrum of dementia pathologies (AD, ALS, DEM, MS, PD, of such studies trying to address the intriguing hypothesis of an infectious etiology for neurodegenerative and VAS). The strongest associations were found for encephalitis and AD or all-causes dementia.

    These are not the first diseases (NDD) as overarching causal culprit using registry-data (Tzeng et al., 2018;  Lopatko et al., 2021; Sun et al., 2022). However, based on the large set of medical reports from two separate European biobanks analyzed, processed independently, and showing convergence of results for multiple infectious conditions, this manuscript certainly adds to the medical evidence driving interest at a global scale for such pathogenic mechanisms. Already, the report last year on the association between EBV and MS based on longitudinal records, and used here as a benchmark, was received to great acclaim (Butler and Walker, 2021).

    On the contrary, the past three decades of studies scraping for causal relationship between infectious agents and NDD using human and animal data were met with great skepticism, leaving those researchers rarely funded (Bjornevik et al., 2022). Then in 2020, COVID was declared a pandemic, and shortly after reports of post-viral syndromes with typical neurological traits, brain fog, memory and concentration deficit, smell and test decline, reopened the discussion and fueld interest for this highly debated research niche (Itzhaki et al., 2016). For such revived interest in the infectious origin of NDD to be fully transcended and actionable requires that global engagement be consolidated and validated using diversified and technically stringent approaches, comparable to the Human Genome Project, what I would call the “human pathobiome” project.

    For the present manuscript, the interrogation of historical data from the U.K. and Finnish Biobanks indicates that the majority of the relevant pathogens are respiratory or head-neck viruses. They likely prime neuroinflammatory processes, either by entry through the olfactory tract and/or by increased permeability of the blood-brain barrier as a consequence of peripheral inflammation.

    Previous studies using sterile mouse models of infections (PolyI:C and LPS) and direct nasal inoculation of herpes viruses and chlamydia pneumoniae, to name a few, provoked amyloid and/or tau pathology, microglia remodeling, and cognitive dysfunctions in rodents (Butler and Walker, 2021). Increasing reports about the presence of infectious agents in postmortem brains of sporadic AD patients support that invasion of microbial species into the brain are triggering, accompanying, and/or perpetuating the pathology (Xu et al., 2022; Bathini et al., 2023; Lee et al., 2008Sehl et al., 2020Little et al., 2014). With this in mind, correlative evidence from the present study matched to the past functional papers in rodents compels an investigation of postmortem brain autopsies from a subset of patients from the same biobanks, such that confirmatory positional evidence for pathogens in the brain can be established.

    The other notable observation, which was not captured in other studies, is the inverse correlation between the time of infection (one, five, and 15 years) and the risk for NDD onset, with hazard triplication closer to the time of infection onset. This can be explained by the depletion of the immune responses as we age, but also by increased permeability of the blood-brain barrier, allowing propagation to the brain not only of pathogens but also of inflammatory mediators, which can precipitate the pathology (Butler and Walker, 2021). This might also explain why the risk for PD, ALS, and MS conversion is lower one or five years before conversion among the NDD considered.

    Interestingly, the risk for AD, all-cause dementia, and vascular dementia (VAS) at the different time points of viral exposure are quite comparable, highlighting that, despite the different incidence of VAS in men than in women, as compared to AD, the infection-driven mechanism might be similar. Additional information about genetic immune variants already identified as risk factors for AD (CR1, CD33, MS4A, CLU, ABCA7, EPHA1, HLA-DRB5-HLA-DRB1) could help select those subject that might a highest risk for an infectious neurological condition.

    As the authors mention, it is clear that for such data to become meaningful prospectively, microbial surveillance and genetic fingerprinting in older adults would be needed to identify targets for intervention, not only as therapies for curbing the acute infectious symptoms, but also as pre-emptive measures for potential later neuropathological sequelae.


    . Anti-herpetic Medications and Reduced Risk of Dementia in Patients with Herpes Simplex Virus Infections-a Nationwide, Population-Based Cohort Study in Taiwan. Neurotherapeutics. 2018 Apr;15(2):417-429. PubMed.

    . Herpesvirus infections, antiviral treatment, and the risk of dementia-a registry-based cohort study in Sweden. Alzheimers Dement (N Y). 2021;7(1):e12119. Epub 2021 Feb 14 PubMed.

    . Hospital-treated infections in early- and mid-life and risk of Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis: A nationwide nested case-control study in Sweden. PLoS Med. 2022 Sep;19(9):e1004092. Epub 2022 Sep 15 PubMed.

    . The Role of Chronic Infection in Alzheimer's Disease: Instigators, Co-conspirators, or Bystanders?. Curr Clin Microbiol Rep. 2021 Dec;8(4):199-212. Epub 2021 Apr 24 PubMed.

    . Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science. 2022 Jan 21;375(6578):296-301. Epub 2022 Jan 13 PubMed.

    . Microbes and Alzheimer's Disease. J Alzheimers Dis. 2016;51(4):979-84. PubMed.

    . Long-term neurologic outcomes of COVID-19. Nat Med. 2022 Nov;28(11):2406-2415. Epub 2022 Sep 22 PubMed.

    . Systemic Inflammation Causes Microglial Dysfunction With a Vascular AD phenotype. Brain Behav Immun Health. 2023 Mar;28:100568. Epub 2022 Dec 21 PubMed.

    . Neuro-inflammation induced by lipopolysaccharide causes cognitive impairment through enhancement of beta-amyloid generation. J Neuroinflammation. 2008;5:37. PubMed.

    . An improved animal model for herpesvirus encephalitis in humans. PLoS Pathog. 2020 Mar;16(3):e1008445. Epub 2020 Mar 30 PubMed.

    . Detection of bacterial antigens and Alzheimer's disease-like pathology in the central nervous system of BALB/c mice following intranasal infection with a laboratory isolate of Chlamydia pneumoniae. Front Aging Neurosci. 2014;6:304. Epub 2014 Dec 5 PubMed.

  4. In this study, Levine et al. utilize the FinnGen and U.K. Biobank to examine associations between neurodegenerative diseases (NDDs) and viral infections. Using the recent discovery of Epstein-Bar virus’ association with multiple sclerosis as a positive control, they show significant risk associations between multiple virus infections and NDDs, especially Alzheimer’s disease (AD). These data build upon a growing wealth of findings from clinical retrospectives, gene-wide association, and molecular biology studies, implicating pathogens in NDD development. Of particular interest to Alzheimer’s disease is the inclusion of intestinal infections as a risk factor, but also the absence of varicella-zoster (VZV), which was shown by Schnier et al. to associate with a reduced incidence of AD after VZV vaccination.

    Unlike some previous observational cohort studies that utilize pathogen seropositivity to establish risk association between infection and disease, here the authors chose medically documented viral infection cases, narrowing the population pool to individuals with what most people would consider to be severe viral infections. Such aggressive infections requiring hospitalization could artificially generate a population with multiple damaged organs, including the central nervous system. The authors do address some lingering questions, such as the potential role for weakened immune systems from NDD, but their research opens the door to other exciting directions. For example, the observed pathogenic risk present from 15-year-old infections: Is this from pathogenic persistence, damaged immune regulation, or activation of NDD pathways?

    Our results examining the Ab protein in AD and its role as an antimicrobial peptide suggest that these infections observed by Levine et al. may activate a physiological neuro-immune pathway. We demonstrated that Ab binds to HSV1 surface glycoproteins, agglutinates the virus, and that these mechanisms provide a protective effect in mice challenged with HSV1 (Eimer et al., 2018). In conjunction with our earlier findings, these results provided the foundation for the antimicrobial protection hypothesis, which stipulates that Ab deposition is an early innate immune response to a pathogen, or perceived pathogen (as reviewed in Moir et al. 2018). Furthermore, this immune activation is not dependent on a particular pathogen but is a genetically conserved first line of defense, and activation (either initiation or exacerbation by pathogenic presence) may ultimately lead to increased risk for NDDs.

    This fascinating study by Levine et al. adds valuable insight into infection involvement in NDDs. It provides more ammunition for researchers to invest in examining not one, but many different pathogens, and how their interaction with our immune system impacts neurodegenerative diseases.


    . Alzheimer's Disease-Associated β-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron. 2018 Jul 11;99(1):56-63.e3. PubMed.

    . The antimicrobial protection hypothesis of Alzheimer's disease. Alzheimers Dement. 2018 Dec;14(12):1602-1614. Epub 2018 Oct 9 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Herpes Update—Virus Increases Dementia Risk in Sweden
  2. More Data on Herpes and Alzheimer’s Disease
  3. COVID-19 Worsens Neurological Problems, Delirium
  4. Mild COVID Infection Can Shrink Brain, Speed Cognitive Decline
  5. Could Common Vaccines Protect Against Alzheimer’s Disease?

Paper Citations

  1. . Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science. 2022 Jan 21;375(6578):296-301. Epub 2022 Jan 13 PubMed.

Further Reading

Primary Papers

  1. . Virus exposure and neurodegenerative disease risk across national biobanks. Neuron. 2023 Apr 5;111(7):1086-1093.e2. Epub 2023 Jan 19 PubMed.