Periodontitis and Dementia: A Spurious Causal Relationship?

To the Editor: We found the article by Lee et al. about periodontitis as a modifiable risk factor for dementia to be of great interest.1 As epidemiologists and geriatricians, we appreciate the potential impact of the findings because prevention of dementia is a critical public health priority in our aging populations. However, we would like to highlight some concerns about the study methods used by the authors to support their claim that periodontitis is a modifiable cause of dementia. Our first concern is about the exclusion criterion of dementia cases. The authors limited their exclusion of subjects with a diagnosis of dementia to those with a diagnosis of dementia within 1 year of the periodontitis index date. This design feature could lead to two significant problems: reverse causation and insufficient time for the underlying pathophysiological processes to occur. First, their exclusion criterion might not be sufficient to ensure that the onset of periodontitis preceded the onset of dementia. Current evidence suggests that a formal diagnosis of dementia is not made in 40–50% of persons living with dementia in high-income countries.2 When a diagnosis is eventually made, it can often be years after the actual onset of the disease.3 Moreover, the authors used ICD-9 codes to determine cases of dementia through administrative data. This method is particularly sensitive to selecting persons at more advanced stage of the disease.4 It is therefore quite plausible that a significant proportion of the study sample exhibited undiagnosed pre-clinical and clinical dementia prior to the periodontitis index date. These persons might have, because of their neurocognitive disorder, nutritional deficiencies or behavioural problems leading them to refuse or neglect oral care leading to poor oral health and periodontitis. This likely situation would result in reserve causation where dementia would, in fact, be preceding periodontitis. Moreover, the exclusion criterion does not account for the time needed for the underlying pathophysiological processes to occur. Two pathological processes could be at play in the potential causal relationship between periodontitis and dementia: periodontitis generates systemic inflammation and, in turn, systemic inflammation could increase the risk for development and progression of dementia by stimulating the production of beta-amyloid and tau protein in brain tissue, or promote thrombogenesis and lead to microvascular pathology.1, 5 The time needed for inflammation to lead to dementia is rarely discussed in the literature.6 In one study reporting the time needed for inflammation to lead to dementia, authors found that inflammation markers in periodontitis increase years before the onset of dementia and the median time between elevated serum immunoglobulin and diagnosis of dementia is almost 10 years.5 Therefore, the time allowed by the authors between periodontitis index date and diagnosis of dementia seems too short to support a causal relationship. Our second concern is about the potential confounding effect of education on the association between periodontitis and dementia. Stewart et al. have shown that education is a confounder in the relationship between almost all oral health measures and cognitive impairment.7 Lee et al. made an attempt to account for possible related sources of confounding in their analysis by adjusting for geographic region and urbanization level. Populations living in urbanized vs less urbanized areas might differ according to educational status. However, as Borell et al. have shown, adjusting for region or neighborhood is not sufficient to account for the relationship between education and periodontitis.8 The association between low education and developing periodontitis can remain as strong as a two-fold odds ratio even after adjusting for neighbourhood of residence.8 Therefore, the lack of adequately controlling for education in the analysis may have produced a spurious association between periodontitis and dementia. Mentioning the plausibility of a confounded association is critical in discussing the internal validity of the results. Sensitivity analyses to determine the potential magnitude of this bias are also strongly recommended.9 These three concerns generate considerable uncertainty regarding any putative causal relationship between periodontitis and dementia. While the authors' results are promising, a more thorough study allowing for a sufficient delay between periodontitis and dementia as well as adjustments for education and socioeconomic status would help to validate the claim of periodontitis prevention as a mean to prevent dementia. Conflicts of Interest: None. Funding: None. Author Contributions: Godard-Sebillotte has made substantial contributions to the conception, interpretation of the findings, drafted the article, and gave final approval of the version to be published. Sourial made substantial contributions to the interpretation of the findings, revised it critically for important intellectual content, and gave final approval of the version to be published. Kaufman made substantial contributions to the conception and interpretation of the findings, revised it critically for important intellectual content, and gave final approval of the version to be published. All the authors are accountable for the parts of the work he or she has done, authors are able to identify which co-authors are responsible for specific other parts of the work. In addition, authors have confidence in the integrity of the contributions of their co-authors.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.