Periodontitis and Alzheimer's disease: a real association, unproven causation

What the major syntheses show

Dementia is cognitive decline that affects independence; Alzheimer's disease is its most common cause. A meta-analysis of 47 longitudinal studies (which follow people over time) estimated a risk of dementia 1.21 times higher (95% confidence interval: 1.07 to 1.38) in people with poor periodontal health, and a risk of cognitive decline 1.23 times higher; the authors rate the overall quality of evidence as low ^[1]. Another meta-analysis of 46 studies reached a similar figure for dementia (relative risk 1.22) ^[2].

A more recent meta-analysis, limited to six studies, found an odds ratio of 2.67 between periodontal disease and Alzheimer's disease, but with very high heterogeneity between studies (I² = 92%): the authors themselves urge caution ^[7]. The most robust order of magnitude therefore remains a modest increase in risk.

Recent cohorts, including the 2026 multicentre study

Korean multicentre study (2026). Using electronic health records from six institutions in South Korea, researchers followed 879,804 people aged 60 and over for ten years, 19,421 of whom had a diagnosis of chronic periodontitis ^[3]. Risk was higher in people with periodontitis for all-cause dementia (hazard ratio 1.51; CI 1.42–1.60), Alzheimer's disease (1.57; 1.45–1.70) and vascular dementia (1.60; 1.39–1.85). It was higher still in patients who had undergone periodontal surgery or extractions, which the authors interpret as a link with severity ^[3].

Limitations: the diagnosis relies on administrative codes rather than a standardised periodontal examination; factors such as education, smoking or lifestyle are imperfectly measured in this type of data; and early dementia can itself worsen oral hygiene (reverse causation).

Hisayama Study (Japan, 2026). In this community cohort of 1,396 people aged 60 and over, free of dementia at baseline and followed for ten years with a clinical periodontal examination, the most affected participants (gum bleeding, attachment loss, pocket depth) had a higher risk of dementia, and heavy bleeding was associated with a higher risk of Alzheimer's disease ^[8]. The association with vascular dementia was not significant. The authors point out that an observational study does not prove causation.

ARIC Study (United States). In this community cohort, severe periodontitis was associated with incident dementia (hazard ratio 1.22; CI 1.01–1.47), an association the authors describe as modest; total tooth loss was not significantly associated (1.21; 0.99–1.48) ^[9].

Studies that find no association

A cohort of 642 men in Northern Ireland, given a periodontal examination in 2001–2003 and a cognitive assessment about fifteen years later, showed no association between severe periodontitis and dementia or mild cognitive impairment (odds ratio 0.83; CI 0.45–1.50) ^[4]. On the other hand, each retained tooth was associated with a slightly lower risk (0.95 per tooth), and men who developed dementia had higher inflammatory markers at baseline ^[4].

Mendelian randomisation genetic studies, which try to approach causation, find no causal link between periodontitis and Alzheimer's disease; their genetic tools are, however, weak for periodontitis ^[10]. These negative results do not justify concluding that there is 'no link': they show that the nature of the association remains uncertain.

The mechanisms studied

In 2019, a study published in Science Advances detected gingipains, enzymes produced by Porphyromonas gingivalis, in the large majority of brains from Alzheimer's patients analysed, in relation to markers of the tau protein; DNA from the bacterium was found in the cerebrospinal fluid of 7 out of 10 patients ^[5]. In mice, oral infection with this bacterium led to brain colonisation and an increase in amyloid peptide; gingipain inhibitors reduced these effects ^[5].

Other avenues concern the passage of the bacterium to the brain, which might be facilitated by increased permeability of blood vessels and of the blood–brain barrier (the filter between blood and brain), and the lasting reprogramming of immune cells in the bone marrow by inflammation ^[11]. This work is mainly experimental.

The GAIN trial: a hypothesis tested, a negative result

The gingipain hypothesis has been tested directly. The phase 2/3 GAIN trial enrolled 643 patients with mild to moderate Alzheimer's disease, treated for 48 weeks with atuzaginstat (a gingipain inhibitor) or placebo. It did not meet its two primary endpoints (cognition and daily function) ^[6]. An analysis in a subgroup with P. gingivalis DNA in saliva suggested slower decline, but this type of result is not enough to draw conclusions. Raised liver enzymes led the FDA to halt development, and the programme was stopped in 2022 ^[6].

And periodontal treatment?

An American cohort observed fewer cases of dementia in older people who reported gum treatment than in those who had not received it ^[12]. But this is an observational study: people who seek treatment often differ in health, income or education. No randomised trial has shown that periodontal treatment reduces the risk of dementia or Alzheimer's disease.

Periodontitis is treated because it destroys bone and teeth, and keeping one's teeth makes it possible to chew and eat well. Treatment is described on the periodontitis treatment page; it is the same at any age, adapted to general health and current medication.

Should you worry, and what should you do in practice?

Reading the figure for what it is. A relative risk of 1.2 means that, in these studies, people with poor periodontal health developed dementia about 20% more often than others ^{[1] [2]}. This is a relative increase in a risk that depends first on age, family history and cardiovascular health; it is of the same order as that seen for many other common health factors, and part of it may be due to shared factors (smoking, diabetes, education) or to reverse causation: early dementia worsens oral hygiene ^[1]. Having periodontitis does not mean that you will develop Alzheimer's disease.

What is reasonable to do. First, treat periodontitis in its own right: stopping the loss of bone and teeth is a certain benefit, whatever the outcome of the scientific debate. Keeping one's teeth is also associated, in several cohorts, with a lower risk of cognitive decline ^[4]; again, this is an association. Next, act on the factors shared by the gums and the brain whose benefit is established: stopping smoking, controlling diabetes and blood pressure, physical activity, all of which are matters for your GP. Finally, do not buy tests or products 'against the Alzheimer's bacterium': no P. gingivalis test has demonstrated clinical value in this context, and no mouthwash or supplement has shown any effect on dementia risk (see blood tests and supplements).

If a relative has cognitive impairment. Brushing often becomes irregular, treatment sessions more difficult, and the mouth deteriorates quickly. At the practice in Levallois-Perret, Dr Hazout involves family or carers in the care, and the maintenance interval takes this risk into account. The aim is not to prevent dementia, but to keep a healthy mouth and functional teeth in a vulnerable person.

To place this link among the other relationships studied between periodontitis and general diseases, the periodontitis and general health page gives the overview, with the same standard of evidence.

Frequently asked questions