Abstract
Periodontitis is a chronic bacterial infection that elicits systemic inflammation. While retrospective studies have linked periodontal pathogens with Alzheimer's disease (AD) and dementia, few have combined cognitive assessments, pathogen exposure, and inflammatory markers. To investigate the longitudinal risk between periodontitis, cognitive impairment and dementia. We examined the relationship between periodontitis and onset of mild cognitive impairment (MCI) and dementia over 15.6 years (SD 1.6) in older men from Northern Ireland enrolled in the PRIME-COG cohort, using logistic regression. We also assessed associations between exposure to periodontal pathogens and blood inflammatory markers. Among 642 men, baseline periodontitis was not significantly associated with later onset of dementia and/or MCI (severe versus mild/none, OR 0.83, 95% CI 0.45–1.50, p = 0.923). However, having more teeth predicted lower risk (OR 0.95, 95% CI 0.91–0.99, p = 0.023). Dementia and/or MCI was associated with higher serum IL-6, IL-8, and IFN-γ at baseline, and IL-8 and TGF-β at follow-up. IgG levels to periodontal pathogens remained stable in men who developed dementia and/or MCI but declined in cognitively normal men. A positive correlation between IgG to periodontal pathogens and proinflammatory cytokines was observed in men who developed dementia and/or MCI. Clinical periodontitis was not associated with dementia or MCI onset, but tooth retention was protective. Elevated inflammatory markers in affected men suggest systemic inflammation may contribute to cognitive decline. Larger, more diverse cohort studies are needed to clarify the role of periodontal disease in dementia and AD risk.
Background
Periodontitis is a chronic bacterial infection that elicits systemic inflammation. While retrospective studies have linked periodontal pathogens with Alzheimer's disease (AD) and dementia, few have combined cognitive assessments, pathogen exposure, and inflammatory markers.
Objective
To investigate the longitudinal risk between periodontitis, cognitive impairment and dementia.
Methods
We examined the relationship between periodontitis and onset of mild cognitive impairment (MCI) and dementia over 15.6 years (SD 1.6) in older men from Northern Ireland enrolled in the PRIME-COG cohort, using logistic regression. We also assessed associations between exposure to periodontal pathogens and blood inflammatory markers.
Results
Among 642 men, baseline periodontitis was not significantly associated with later onset of dementia and/or MCI (severe versus mild/none, OR 0.83, 95% CI 0.45–1.50, p = 0.923). However, having more teeth predicted lower risk (OR 0.95, 95% CI 0.91–0.99, p = 0.023). Dementia and/or MCI was associated with higher serum IL-6, IL-8, and IFN-γ at baseline, and IL-8 and TGF-β at follow-up. IgG levels to periodontal pathogens remained stable in men who developed dementia and/or MCI but declined in cognitively normal men. A positive correlation between IgG to periodontal pathogens and proinflammatory cytokines was observed in men who developed dementia and/or MCI.
Conclusions
Clinical periodontitis was not associated with dementia or MCI onset, but tooth retention was protective. Elevated inflammatory markers in affected men suggest systemic inflammation may contribute to cognitive decline. Larger, more diverse cohort studies are needed to clarify the role of periodontal disease in dementia and AD risk.
Introduction
Periodontitis is a chronic inflammatory disease that affects the gums and surrounding tissues. This disease results from a bacterial infection and is associated with several conditions that have a vascular and/or inflammatory component including stroke, coronary heart disease, rheumatoid arthritis, and diabetic complications. 1 Severe periodontitis is the sixth most prevalent condition in the world; the global age standardized prevalence of severe periodontitis in 2010 was 11.2% rising steeply in those aged over 65 years. 2
Alzheimer's disease (AD) is the most common type of dementia, representing 60–70% of all cases. AD prevalence increases with age from 5% in the seventh decade to 50% by the tenth decade of life. 3 Sporadic late onset AD represents 98% of all AD cases and is likely due to a complex interaction of environmental, vascular, and genetic risk factors.4,5 In the context of environmental factors, the ‘Dementia prevention, intervention and care 2024 report’ by the Lancet Standing Commission identified 14 potentially modifiable risk factors for dementia prevention, which if addressed, could reduce dementia incidence by 45%, 6 highlighting potential interventions targeting environmental and lifestyle factors.
It has become increasingly clear that inflammation has a key role in the onset and/or progression of AD. Convincing evidence for an inflammatory influence comes from genetic studies of patients with late onset AD that have implicated genes encoding proteins within immune and inflammatory response pathways.7–14 It has also been hypothesized that the APOE ε4 allele, an established risk factor for AD, modulates neuroinflammation and significantly increases carriers susceptibility to environmental associated risk factors. 15 Our previous work has shown that individuals diagnosed with mild cognitive impairment (MCI) decline 5-fold faster when exposed to an acute systemic inflammatory event, such as a urinary tract infection. This accelerated decline in clinical symptoms is particularly evident in individuals with chronic low-grade inflammation, such as elevated serum levels of tumor necrosis factor alpha (TNF-α). 16
Investigations of the periodontitis prevalence in older adults and the inflammatory milieu has demonstrated associations with increased risk of reduced cognitive functioning17–21 and dementia 22 in cross-sectional studies. Further support from several retrospective population studies in 27,963 participants23–26 highlighted an increased risk of developing dementia in those with chronic periodontitis across a 1–10-year follow-up period. Other studies have shown that significant tooth loss predicts MCI, 24 while alveolar bone loss has been associated with reduced cognitive score (Mini-Mental State Examination (MMSE) < 25) after adjustments for age, sex and education. 27 Exposure to periodontal pathogens leads to activation of the immune system and elevated serum IgG and significantly higher periodontal pathogens levels in individuals that develop AD 10 years later.28,29 We demonstrated that individuals with mild to moderate AD decline significantly faster when co-existing periodontitis was present at baseline. 30 This increased cognitive decline was associated with a relative increase in the pro-inflammatory state during the follow-up period, although serum IgG titers to Porphyromonas gingivalis were similar. These contrasting findings support recommendations for well-designed longitudinal studies to address the knowledge gap related to periodontitis as a dementia risk factor, 31 further supported by a recent systematic review and meta-analysis that reported moderate quality of evidence and risk of bias amongst the identified studies. 32
The role of periodontitis as a dementia risk factor has been hypothesized by several mechanisms: (1) Pro-inflammatory cytokines produced by periodontal tissues during periodontitis enter the bloodstream leading to AD-related neurological changes.33,34 Systemic inflammation may activate the already primed microglial cells within the central nervous system and hasten neurodegenerative processes.35,36 (2) Periodontopathogens such as Porphyromonas gingivalis (Pg), Tannerella forsythia (Tf), Aggregatibacter actinomycetemcomitans (Aa), and Treponema denticola (Td) or their membrane components (lipopolysaccharides (LPS)) may reach the brain via the systemic circulation and accelerate neuroinflammation perhaps through an increasingly permeable blood-brain barrier (BBB) associated with increasing age or the circum-ventricular organs not protected by the BBB. 37 Td, Pg, and Tf have been isolated from the cerebral tissue of AD patients postmortem.38–40 (3) Periodontitis-induced cerebrovascular pathology, diabetic complications and cardiovascular disease could promote AD progression through a cascade of atherogenesis related systemic inflammation.41,42
The aim of present study is to investigate whether periodontal disease in 60–70-year-old men increases the risk of cognitive decline and the development of dementia after fifteen years follow-up and to investigate the role of inflammation in this process. Our a priori hypothesis was that diagnosis of periodontitis would be significantly associated with, and predict, incidence of both dementia and cognitive decline.
Study population
PRIME (Prospective Epidemiological Study of Myocardial Infarction) is a cohort study of cardiovascular disease in men from three centers in France and one in Northern Ireland (NI). During 1991–1994, 2748 men were recruited in NI which represented 5% of 50–60-year-old men in the greater Belfast area matching the local social class. 43 In 2001–2003, the surviving participants were invited for re-screening and the collection of data on social circumstances, demographic background, tobacco use and alcohol consumption, and personal measures including weight and height. Fasting blood samples were obtained and analyzed for lipids and inflammatory markers. In addition, a clinical periodontal examination was completed for 1400 (69.7%) of the 2010 men who attended for re-screening,44,45 and subgingival plaque sampling was carried out on a subsample of participants. Baseline and follow-up cognitive assessments included MMSE.
Between 2018 and 2020, the 1400 men who had a clinical periodontal examination were invited to a further follow-up visit for the present study (PRIME-COG), where lifestyle information was collected as previously, along with further clinical and cognitive assessments. Inclusion criteria were restricted to participants who had previously undergone periodontal examination and dental plaque sampling at the previous visit. Edentulous participants were excluded to reduce effects from reverse causality, and those participants formally diagnosed with previous cognitive impairment or who scored < 24/30 on MMSE at baseline visit. A total of 642 participants were included in the present analysis as they attended visits in 2001–2003 and again in 2018–2020. Approval for the project was obtained from the Office for Research Ethics Committees (Northern Ireland); reference number 06/NIR02/107. Written informed consent was obtained from all participants following a detailed explanation of the study aims. The study was compliant with the Declaration of Helsinki guidelines and STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement. 46
Periodontal examination
All periodontal examinations were completed by one of four dental hygienists who had completed a calibration assessment against a “gold standard” set by a senior clinical researcher prior to the study (GL). Regular monthly meetings took place to ensure inter- and intra-examiner consistency and reproducibility. Clinical periodontal measurements were made using a Michigan O periodontal probe (Hu-Friedy, Chicago, IL, USA) with William's markings. Measurements were made at the mesial, distal, buccal and palatal/lingual aspects of all teeth excluding third molars. Pocket probing depths were recorded as the distance from the gingival margin to the base of the clinical pocket with the probe tip parallel to the long axis of the tooth. Clinical attachment levels (CAL) were also recorded as the distance from the cement–enamel junction to the base of the clinical pocket. Measurements were made to the nearest millimeter and if doubt existed, the lower value was recorded. The number of teeth was also recorded, again excluding third molars. Periodontal disease was defined according to the Centre for Disease Control and the American Academy of Periodontology classification. 47 “Severe periodontitis” required two or more interproximal sites with CAL ≥ 6 mm, not on the same tooth, and one or more interproximal sites with PD ≥ 5 mm. “Moderate periodontitis” was defined as two or more interproximal sites with CAL ≥ 4 mm, not on the same tooth, or two or more interproximal sites with PD ≥ 5 mm, not on the same tooth.
Cognitive follow-up
In addition to repeating the baseline MMSE to assess change in cognition, a comprehensive assessment was carried out for the domains of attention/executive function, memory, language and visuospatial function using the Addenbrooke's Cognitive Examination Revised (ACE-R). Function was assessed using the Bristol Activities of Daily Living Scale (BADLS) 48 and mood/depression assessed using the Geriatric Depression Scale (GDS) short form. 49 Research nurses from the Northern Ireland Dementia Clinical Research Network were experienced in performing neuropsychological tests and venous blood collection. Serum was extracted, aliquoted and stored at −80°C for analysis. A comprehensive cognitive assessment included a consensus diagnosis of normal cognition for age, MCI, and/or dementia was made by three experienced clinical researchers (BMcG, PP, CH) who specialize in dementia/AD research using accepted clinical criteria.50–52
Determination of serum IgG levels to dental bacterial pathogens, and serum inflammatory markers
Samples collected at baseline (PRIME) and follow-up visit (PRIME-COG) were assessed for serum immunoglobulin G (IgG) levels to four common bacterial periodontal pathogens; Tf, Td, Aa, and Pg. The full details are included in the Supplemental Material.
Pro-inflammatory cytokines including interferon gamma (INF-ɣ), interleukin (IL)-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL12p70, IL-13, TNF-α, were determined using an MSD® multiplex assay (MSD Maryland, USA). CRP levels were determined using MSD® single plex, whereas transforming growth factor beta (TGF-β1, TGF-β2, TGF-β3) were determined using U-plex MSD® assays in line with recommended manufacturer's instructions.
Model covariates
The following covariates were included in the multivariate models. Baseline age, APOE ε4 allele status (2 versus 1 versus 0 alleles), following APOE genotyping using Taqman Real Time PCR Assays (Thermo Fisher Scientific). Cardiovascular disease was characterized by previous myocardial infarction or an intervention such as angioplasty with stenting or coronary artery bypass. Cerebrovascular disease was recorded for participants who previously had a stroke. Diabetes and hypertension were self-reported and high cholesterol confirmed from blood samples during the 2001–2003 study visit. Lifetime smoking was expressed in five categories: never smoked, smoked other than cigarettes, smoked < 15 cigarette pack years, smoked ≥ 15 but < 30 cigarette pack years and smoked ≥ 30 cigarette pack years. Alcohol intake was also expressed as five categories of consumption: none, 1–128, 129–265, 266–461 and ≥ 462 mL/ week. Education level was segregated into ‘Primary’, ‘Secondary’, ‘Technical’, and ‘Higher’. Socioeconomic status was categorized into ‘High’, ‘Middle’, and ‘Low’ using a composite measure of material conditions based on the type of living accommodation (rented or owned, mortgaged), number of cars/vans/motorcycles in the household and the number of baths and/or showers and toilets in the home. 53
Statistical analysis
Prior to statistical analysis, data was assessed for normality by way of visual inspection of Q-Q plots and histograms and performing Shapiro-Wilk tests. The data for blood triglycerides were not normally distributed, therefore, log transformed values were used for analysis. For descriptive purposes, median and interquartile ranges are presented. Comparison of characteristics between men with dementia and/or MCI and those without, were compared using chi-squared tests for categorical variables and t-tests for continuous variables.
The relationship between periodontitis and dementia incidence later in life was assessed using logistic regression adjusted for possible confounders. Dementia incidence among study participants was small (4.8%) and therefore models were constructed with incidence of dementia and/or MCI as the outcome variable and periodontitis as the main predictor variable. For the models, periodontitis was categorized into 3 groups; ‘severe’, ‘moderate’, and ‘mild or none’, and comparisons were made between ‘severe’ versus ‘mild or none’ and ‘moderate’ versus ‘mild or none’. In addition to the primary outcome, our secondary outcome consisted of cognitive decline as measured by a fall of ≥ 3 points from MMSE score at PRIME-COG rescreening. The first model was constructed in a minimal facet, with periodontitis as the predictor variable for the cognitive endpoints. Following this, potential confounders were added sequentially to construct a fully adjusted model. The confounders included in the analysis were age, level of education, socioeconomic status, number of APOE ε4 risk alleles, cardiovascular disease (coronary heart disease, cerebrovascular disease), diabetes, hypertension, blood cholesterol level, number of teeth, smoking status and alcohol consumption. Odds ratios (OR) and 95% confidence intervals were calculated from the regression coefficients. Model performance was assessed using the Akaike information criterion (AIC) to determine the best performing model.
For the analysis determining associations between serum IgG levels to common periodontal pathogens and pro-inflammatory cytokines with cognitive outcomes, CRP and TGF-β were not normally distributed, therefore, log transformed values were used. Comparisons between participants with dementia and/or MCI and those who were cognitive healthy were performed using the Kruskal-Wallis test, followed by the Mann-Whitney U test for multiple comparison. For all analyses, two-tailed p values < 0.05 were considered significant. Statistical analyses were performed in R Studio 54 SPSS 29.0 ((IBM, Armonk, NY) and Prism 10 (GraphPad software, San Diego, CA).
Results
At the PRIME rescreening visit between 2001 and 2003, of the 642 participants included in the present analysis, 127 (19.8%) had severe periodontitis and 148 (23.1%) had moderate, while 367 (57.2%) men had either mild, or no sign of periodontitis. During the follow-up period from the PRIME rescreening visit (PRIME-COG; mean 15.6 (SD 1.6) years), 31 (4.8%) were diagnosed with dementia, 112 (17.4%) with MCI, and 37 (5.8%) experienced a reduction of ≥ 3 points in their MMSE score, indicative of cognitive decline. Summary characteristics for participants with and without dementia and/or MCI at PRIME-COG follow-up are compared in Table 1. Participants with dementia and/or MCI were older (p < 0.001), had less teeth (p = 0.010), higher serum levels of IL-8 (p = 0.047) and TGF-β (p = 0.022), a higher waist to hip ratio (p = 0.019), attained worse scores in MMSE (p < 0.001), BADLS (p < 0.001) and GDS (p < 0.001), were more likely to carry a greater number of APOE ε4 risk alleles (p < 0.001), have hypertension (p = 0.002), be educated to a lower level (p < 0.001) and be within a lower social class (p = 0.002), compared to participants without dementia and/or MCI.
Periodontitis, dementia/MCI, and cognitive decline
Logistic regression models revealed no significant associations between periodontitis and incidence of dementia or MCI. In the sequential model design, moderate periodontitis did not display any further risk of dementia and/or MCI compared to mild or no periodontitis (fully adjusted model, OR 1.03, 95% CI 0.59, 1.77, p = 0.923). The OR for dementia and/or MCI in the presence of severe periodontitis was found to be lower compared to mild or no sign of periodontitis (fully adjusted model, OR 0.83, 95% CI 0.46, 1.45, p = 0.552) (Table 2), although this was not significant. In sensitivity analyses, we did not find any significant associations when testing for a trend across classification of periodontitis (‘severe’, ‘moderate’, ‘mild or no disease’), or when replacing periodontitis in the model with a differential classification of periodontal disease (mean clinical attachment loss measured in mm) (data not shown). Despite a lack of relationship between past periodontal disease, the fully adjusted logistic regression model identified number of teeth to display a significant inverse association with dementia and/or MCI (p = 0.023).
There was no significant relationship with periodontitis and the secondary endpoint, cognitive decline, with moderate periodontitis showing a negative association with cognitive decline compared to mild or no diagnosis of periodontitis (fully adjusted model, OR 0.48, 95% CI 0.13, 1.46, p = 0.234), while presence of severe periodontitis carried a positive association with cognitive decline (fully adjusted model, OR 1.34, 95% CI 0.48, 3.46, p = 0.562) (Supplemental Table 1), although neither were significant.
In the fully adjusted model, significant predictor variables associated with both dementia and/or MCI and cognitive decline included age (p = 0.002), all levels of education (primary, p < 0.001; secondary, p = 0.019; technical, p < 0.001), number of APOE ε4 risk alleles (1 versus 0, p = 0.018; 2 versus 0, p = 0.002) and hypertension (p = 0.018). Incident cardiovascular disease was significantly associated in model 3 (p = 0.046) (Table 2). In the models for cognitive decline, significantly associated variables included the number of APOE ε4 risk alleles (fully adjusted model, 1 versus 0, p = 0.012; 2 versus 0, p < 0.001), and primary versus higher education in models 2-4 (model 4, p = 0.030) (Supplemental Table 1).
Serum levels of IgG antibodies to four common dental bacterial pathogens
Combined data of all participants indicated an overall decrease in the serum levels of IgG antibodies to Td (p = 0.099), Tf (p = 0.035), Aa (p = 0.0007), and Pg (p = 0.615) over the 15-year follow-up period (Figure 1A). Stratification of participants by clinical group (i.e., dementia, MCI, cognitive normal/healthy) identified decreased serum IgG levels, over the 15–20-year follow-up period only in participants without cognitive decline (Figure 1D). In contrast, IgG levels to Td, Tf, and Aa remained constant over the study period in participants diagnosed with MCI (Figures 2 and 3) and dementia (Figure 3B), with the exception anti-Tf IgG, which decreased (p = 0.04). There was no significant difference in IgG antibodies levels for the four bacterial pathogens across the study groups (Supplemental Figure 1).
Serum levels of pro-inflammatory cytokines, CRP, and TGF-β
Comparison of serum levels of pro-inflammatory cytokines, CRP and TGF-β at baseline and follow-up visits were not significantly different between participants with dementia, MCI, or without cognitive impairment (Figure 2). Stratifying participants by clinical groups revealed that men with MCI had significantly higher serum IL-8 levels at baseline (p = 0.04; Figure 3A) and follow-up (p = 0.04; Figure 3D) and elevated levels of IL-6 at baseline (p < 0.001; Figure 3B). Men that developed dementia showed elevated levels of IFN-γ at baseline (p = 0.013; Figure 3C) and TGF-β1 at follow-up (p = 0.02; Figure 3B) compared to cognitive healthy men.
To control for age, we next compared serum levels of pro-inflammatory cytokines in older participants only (>81 years old at follow-up). Participants with MCI had significantly higher serum levels of IL-8 at baseline (p = 0.014; Figure 4A) and follow-up (p = 0.012; Figure 4D) and elevated levels of IL-6 at baseline (p = 0.043; Figure 3C). Older participants that developed dementia showed elevated levels of TGF-β1 at baseline (p = 0.007; Figure 3B) and follow-up (p = 0.049; Figure 3E) compared to cognitive healthy men. Older participants that developed dementia also showed significant higher IL-6 levels at baseline (p = 0.02; Figure 3C), compared to cognitive healthy men
We next assessed if higher levels of serum inflammatory markers were associated with serum IgG antibodies to four common bacterial periodontal pathogens. A correlation matrix between serum inflammatory markers and IgG antibodies levels identified variation between participants with MCI and/or dementia to those who were cognitive healthy (Figure 5). Participants that developed dementia over the study period showed a positive correlation between baseline serum levels of IL-12p70 (r = 0.523, r = 0.486), and IL-8 (r = 0.468, r = 0.389) and IgG serum levels to the periodontal pathogens Td and Tf, respectively (Supplemental Figure 2). These participants also showed a negative correlation with IL-12p70 (r = −0.437), IL-1β (r = −0.422), TGFβ1 (r = −0.391), and TGFβ2 (r = −0.424) and serum IgG levels of Pg (Figure 5 and Supplemental Figure 2). These associations between serum inflammatory markers and IgG levels of periodontal pathogens were persistent over the follow-up period (Figure 5). Participants that developed MCI during the follow-up period showed positive correlations between serum levels of IL-8 and IgG to periodontal pathogens (r = 0.88 for Td; r = 0.98 for Tf and r = 0.92 for Aa) and CRP (r = 0.96 for Td; r = 0.82 for Tf), which were not observed in participants that remained cognitive healthy (Figure 5 and Supplemental Figure 2).
Discussion
The main findings from the present study are that in older men residing in the Northern Ireland, a clinical diagnosis of periodontitis does not predict incident dementia, MCI or cognitive decline over a fifteen-year follow-up period. However, in men with a higher number of teeth at baseline, the risk of dementia and/or MCI was reduced, suggesting that maintaining adequate oral health and mitigating tooth loss may be a factor in reducing risk of cognitive decline in aging. Further, in men with dementia and/or MCI, blood levels of a number of inflammatory markers were higher compared to men with no cognitive impairment, highlighting the inflammatory component of perturbations in cognitive function. Serum levels of IL-6, IL-8, and IFN-γ were significantly elevated at baseline in men that developed MCI and/or dementia 15 years later, compared to men that remained cognitively healthy. The positive correlation between serum IgG levels to Td and Tf and inflammatory markers in participants that developed dementia and/or MCI may suggest an early and sustained immune response to common periodontal pathogens in men that develop MCI and/or dementia.
The lack of association between periodontitis and cognitive outcomes was somewhat unexpected, due to previous literature identifying significant longitudinal associations.25,55 However, it should be noted that this has not been consistently found, with other studies reporting a lack of significant longitudinal risk between oral health and dementia.56,57 Plausible reasons for the lack of significant associations in the present analysis could be due to study limitations: power, with low numbers of participants with dementia (4.8%), MCI (17.4%), and cognitive decline (5.8%). In addition, we did not carry out a periodontal examination at follow-up visits so we cannot rule out that disease improved for some men and was merely a transient occurrence. Oral status can change over the course of time, particularly when considering the ∼15 year follow up duration, therefore capturing tooth count at a single time point may bias results by not providing the entire picture for the relationship between longitudinal risk between periodontal disease and cognitive function. In the future, it would be advantageous to address such a limitation through repeat oral examinations and linkage to dental records, which would allow a fully comprehensive picture of both periodontal disease status, as well as overall oral health. In addition, our study population comprised Western European men only. A recent analysis reported that in individuals aged 60 to 69 years, AD prevalence in females was 1.9 times greater than that in males, 58 which could a suggest a lower incident in our male only participants. The reasoning for the choice of study population lies in the PRIME cohort, which recruited a male only study cohort, in France and Northern Ireland, with the Northern Ireland cohort captured in the present analysis. The location of the PRIME cohort in Northern Ireland facilitated the opportunity to apply a thorough examination included in the present analysis by way of comprehensive cognitive assessments alongside standardized dental examinations performed by dental experts. It would be pertinent to build on the present findings in study populations of greater variety in both gender and ethnicity, and we are presently exploring such avenues. As is common among observational studies, the possibility of residual confounding or failure to account for other relevant confounders may have had some influence on the associations between periodontal disease and cognitive outcomes. Attrition bias should also be considered, participants who were not mobile and/or cognitively deteriorated may have been unlikely to have participated (including men who were deceased) in the PRIME COG follow-up, which therefore may have weakened any potential associations.
Despite a lack of relationship between periodontitis, our study identified a significant inverse association between number of teeth and future development of dementia and/or MCI. These observations agree with other longitudinal studies. 30 Severe periodontitis is characterized by extensive bone loss, making teeth loose and prone to fall out. That being said, we did not find a difference of note between clinical attachment loss; a predominant clinical manifestation and determinant of periodontal disease, between men who developed cognitive impairment (mean 1.96 mm, SD 0.57) and those who did not (mean 2.24 mm, SD 0.87). While tooth loss due to periodontal destruction is a good indicator of previous susceptibility to periodontitis, 59 other causes of tooth loss cannot be ruled out. Another factor is that this metric again was captured at the same time point as number of teeth and was not followed up, which as previously noted, would be enlightening to include in future investigations. A recent systematic review reported that the available evidence elucidating the relationship between number of teeth and dementia risk may be due to reverse causality, as well as exhibiting heterogeneity and lacking robust methodology to draw firm conclusions. Considering this, it may be that further well-designed studies involving standardized periodontal and cognitive assessments, as well as controlling for reverse causality are warranted. Our finding suggests that maintaining good oral health, and specifically, retention of teeth, may be a factor in reducing risk of cognitive impairment.
Our current study identified significant associations between MCI and dementia with known risk factors for dementia, including increasing age and carriage of APOE ε4 risk allele/s, 60 lower education level and hypertension. 6 Our results therefore help further contribute to the evidence base and establishment of risk factors for declines in cognitive function and dementia.
T. denticola, T. forsythia, and A. actinomycetemcomitans and P. gingivalis, are widely accepted as highly virulent periodontal pathogens, with other species identified as accessory pathogens. 61 Multispecies microbial biofilms disrupt host tissue homeostasis and elicit a host immune-inflammatory response, resulting in expression of several cytokines and chemokines. 62 Exposure to periodontal pathogens also promotes a humoral immune response and serum IgG antibody levels to periodontal pathogens have been reported to be associated with AD. 29 In this study we found that serum IgG antibodies to periodontal pathogens remain constant during the study period in men that developed dementia and/or MCI, while a decrease was observed in cognitive healthy men. In men that developed dementia the serum IgG levels were associated with elevated levels of cytokines, and in particular IL12p70, IL-8, IL-6 and TGFβ. Men that developed MCI over the study period showed positive correlations between serum levels of IL-8 and IgG levels to periodontal pathogens. These observations highlight the potential role of systemic inflammation as a risk factor for dementia. The underlying mechanisms require further study, but may include a role for IL-8 mediated neutrophil function as reported previously. 63
The strengths of the present study include one of the longest follow-up investigations into the longitudinal association between periodontal disease and dementia. The study utilized the PRIME cohort, representing a well phenotyped sample of older men and homogenous sample: white Western-European men, of a similar age, who at original recruitment were representative of the general population of Northern Ireland. 43 The cognitive assessments were performed by experienced clinicians within our group who specialize in dementia/AD research, our analysis therefore benefitted from a consensus diagnosis of dementia, MCI and cognitive decline which was based on a comprehensive clinical cognitive assessment. The exposure variable was comprehensively assessed by a detailed full mouth periodontal examination, and the recommended case definitions of periodontitis for epidemiological research applied. 64 The periodontal examination was also performed by a calibrated examiner rather than relying on self-report or insurance data. Further, the models implemented were appropriately adjusted for important confounders well known to increase risk of cognitive decline including APOE ε4, sociodemographic characteristics and risk factors for cardiometabolic disease, highlighted by the significant relationships identified in the analysis.
In conclusion, in a cohort of older men residing in Northern Ireland, periodontitis did not influence longitudinal risk of dementia or MCI following a fifteen-year follow-up; however, a greater number of teeth was associated with reduced longitudinal risk of dementia and/or MCI. In addition, circulating blood markers of inflammation were higher, and IgG titers to common periodontal pathogens remained constant in men diagnosed with dementia or MCI, while IgG titers significantly reduced over time in cognitive normal men. Our findings therefore help contribute to the evidence base on the longitudinal associations between periodontitis and dementia. Periodontitis is treatable, and therefore if identified as a predictor of dementia, represents a modifiable risk factor. Intervention studies are warranted to provide more clear evidence for periodontitis initiating MCI and/or dementia.
Acknowledgements
The authors have no acknowledgments to report.
References
1 LindenGJ LyonsA ScannapiecoFA . Periodontal systemic associations: review of the evidence. J Clin Periodontol 2013; 40: S8–S19.10.1111/jcpe.1206423627336
2 KassebaumNJ BernabéE DahiyaM , et al. Global burden of severe periodontitis in 1990-2010: a systematic review and meta-regression. J Dent Res 2014; 93: 1045–1053.25261053 10.1177/0022034514552491PMC4293771
3 QuerfurthHW LaFerlaFM . Alzheimer's disease. N Engl J Med 2010; 362: 329–344.20107219 10.1056/NEJMra0909142
4 MayeuxR HonigLS TangMX , et al. Plasma A[beta]40 and A[beta]42 and Alzheimer's disease: relation to age, mortality, and risk. Neurology 2003; 61: 1185–1190.14610118 10.1212/01.wnl.0000091890.32140.8f
5 TangMX SternY MarderK , et al. The APOE-epsilon4 allele and the risk of Alzheimer disease among African Americans, whites, and Hispanics. JAMA 1998; 279: 751–755.9508150 10.1001/jama.279.10.751
6 LivingstonG HuntleyJ LiuKY , et al. Dementia prevention, intervention, and care: 2024 report of the standing commission. Lancet 2024; 404: 572–628.39096926 10.1016/S0140-6736(24)01296-0
7 BenitezBA JinSC GuerreiroR , et al. Missense variant in TREML2 protects against Alzheimer's disease. Neurobiol Aging 2014; 35: 1510.e1519–1526.10.1016/j.neurobiolaging.2013.12.010PMC396155724439484
8 HaroldD AbrahamR HollingworthP , et al. Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer's disease. Nat Genet 2009; 41: 1088–1093.19734902 10.1038/ng.440PMC2845877
9 HollingworthP HaroldD SimsR , et al. Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer's disease. Nat Genet 2011; 43: 429–435.21460840 10.1038/ng.803PMC3084173
10 JonssonT StefanssonH SteinbergS , et al. Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med 2013; 368: 107–116.23150908 10.1056/NEJMoa1211103PMC3677583
11 LambertJC Grenier-BoleyB HaroldD , et al. Genome-wide haplotype association study identifies the FRMD4A gene as a risk locus for Alzheimer's disease. Mol Psychiatry 2013; 18: 461–470.22430674 10.1038/mp.2012.14PMC3606943
12 NajAC JunG BeechamGW , et al. Common variants at MS4A4/MS4A6E, CD2AP, CD33 and EPHA1 are associated with late-onset Alzheimer's disease. Nat Genet 2011; 43: 436–441.21460841 10.1038/ng.801PMC3090745
13 WightmanDP JansenIE SavageJE , et al. A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nat Genet 2021; 53: 1276–1282.34493870 10.1038/s41588-021-00921-zPMC10243600
14 BellenguezC KüçükaliF JansenIE , et al. New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nat Genet 2022; 54: 412–436.35379992 10.1038/s41588-022-01024-zPMC9005347
15 LiuCC LiuCC KanekiyoT , et al. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol 2013; 9: 106–118.23296339 10.1038/nrneurol.2012.263PMC3726719
16 HolmesC CunninghamC ZotovaE , et al. Systemic inflammation and disease progression in Alzheimer disease. Neurology 2009; 73: 768–774.19738171 10.1212/WNL.0b013e3181b6bb95PMC2848584
17 Gil-MontoyaJA Sanchez-LaraI Carnero-PardoC , et al. Is periodontitis a risk factor for cognitive impairment and dementia? A case-control study. J Periodontol 2015; 86: 244–253.25345338 10.1902/jop.2014.140340
18 KamerAR MorseDE Holm-PedersenP , et al. Periodontal inflammation in relation to cognitive function in an older adult Danish population. J Alzheimers Dis 2012; 28: 613–624.22045483 10.3233/JAD-2011-102004
19 NascimentoPC CastroMML MagnoMB , et al. Association between periodontitis and cognitive impairment in adults: a systematic review. Front Neurol 2019; 10: 23.31105630 10.3389/fneur.2019.00323PMC6492457
20 GuoH ChangS PiX , et al. The effect of periodontitis on dementia and cognitive impairment: a meta-analysis. Int J Environ Res Public Health 2021; 18: 6823.34202071 10.3390/ijerph18136823PMC8297088
21 GaoC LarvinH BishopDT , et al. Oral diseases are associated with cognitive function in adults over 60 years old. Oral Dis 2024; 30: 3480–3488.37811600 10.1111/odi.14757
22 YooJE HuhY ParkS-H , et al. Association between dental diseases and oral hygiene care and the risk of dementia: a retrospective cohort study. J Am Med Direct Assoc 2023; 24: 1924–1930.e3.10.1016/j.jamda.2023.08.01137709259
23 ChenC-K WuY-T ChangY-C . Association between chronic periodontitis and the risk of Alzheimer’s disease: a retrospective, population-based, matched-cohort study. Alzheimers Res Ther 2017; 9: 56.28784164 10.1186/s13195-017-0282-6PMC5547465
24 ChenH-L WuD-R ChenJ-J , et al. Association between periodontitis treatment and dementia in Taiwanese adults. BMC Oral Health 2023; 23: 969.38057760 10.1186/s12903-023-03551-5PMC10701963
25 LeeYT LeeHC HuCJ , et al. Periodontitis as a modifiable risk factor for dementia: a nationwide population-based cohort study. J Am Geriatr Soc 2017; 65: 301–305.27685603 10.1111/jgs.14449
26 TzengNS ChungCH YehCB , et al. Are chronic periodontitis and gingivitis associated with dementia? A nationwide, retrospective, matched-cohort study in Taiwan. Neuroepidemiology 2016; 47: 82–93.27618156 10.1159/000449166
27 NilssonH Sanmartin BerglundJ RenvertS . Longitudinal evaluation of periodontitis and development of cognitive decline among older adults. J Clin Periodontol 2018; 45: 1142–1149.30076762 10.1111/jcpe.12992
28 Sparks SteinP SteffenMJ SmithC , et al. Serum antibodies to periodontal pathogens are a risk factor for Alzheimer's disease. Alzheimers Dement 2012; 8: 196–203.22546352 10.1016/j.jalz.2011.04.006PMC3712346
29 NobleJM ScarmeasN CelentiRS , et al. Serum IgG antibody levels to periodontal microbiota are associated with incident Alzheimer disease. PLoS One 2014; 9: e114959.10.1371/journal.pone.0114959PMC427077525522313
30 IdeM HarrisM StevensA , et al. Periodontitis and cognitive decline in Alzheimer's disease. PLoS One 2016; 11: e0151081.10.1371/journal.pone.0151081PMC478626626963387
31 AsherS StephenR MäntyläP , et al. Periodontal health, cognitive decline, and dementia: a systematic review and meta-analysis of longitudinal studies. J Am Geriatr Soc 2022; 70: 2695–2709.36073186 10.1111/jgs.17978PMC9826143
32 DibelloV CustoderoC CavalcantiR , et al. Impact of periodontal disease on cognitive disorders, dementia, and depression: a systematic review and meta-analysis. Geroscience 2024; 46: 5133–5169.38943006 10.1007/s11357-024-01243-8PMC11336026
33 KornmanKS . Interleukin 1 genetics, inflammatory mechanisms, and nutrigenetic opportunities to modulate diseases of aging. Am J Clin Nutr 2006; 83: 475s–483s.16470016 10.1093/ajcn/83.2.475S
34 WattsA CrimminsEM GatzM . Inflammation as a potential mediator for the association between periodontal disease and Alzheimer's disease. Neuropsychiatr Dis Treat 2008; 4: 865–876.19183779 10.2147/ndt.s3610PMC2626915
35 HolmesC CotterellD . Role of infection in the pathogenesis of Alzheimer's disease: implications for treatment. CNS Drugs 2009; 23: 993–1002.19958038 10.2165/11310910-000000000-00000
36 KamerAR CraigRG DasanayakeAP , et al. Inflammation and Alzheimer's disease: possible role of periodontal diseases. Alzheimers Dement 2008; 4: 242–250.18631974 10.1016/j.jalz.2007.08.004
37 LiR WangJ XiongW , et al. The oral-brain axis: can periodontal pathogens trigger the onset and progression of Alzheimer's disease? Front Microbiol 2024; 15: 1358179.38362505 10.3389/fmicb.2024.1358179PMC10868393
38 MiklossyJ . Alzheimer's disease–a spirochetosis? Neuroreport 1993; 4: 841–848.8369471
39 PooleS SinghraoSK KesavaluL , et al. Determining the presence of periodontopathic virulence factors in short-term postmortem Alzheimer's disease brain tissue. J Alzheimers Dis 2013; 36: 665–677.23666172 10.3233/JAD-121918
40 RiviereGR RiviereKH SmithKS . Molecular and immunological evidence of oral Treponema in the human brain and their association with Alzheimer's disease. Oral Microbiol Immunol 2002; 17: 113–118.11929559 10.1046/j.0902-0055.2001.00100.x
41 GhizoniJS TaveiraLA GarletGP , et al. Increased levels of Porphyromonas gingivalis are associated with ischemic and hemorrhagic cerebrovascular disease in humans: an in vivo study. J Appl Oral Sci 2012; 20: 104–112.22437687 10.1590/S1678-77572012000100019PMC3928781
42 XieJ Van HoeckeL VandenbrouckeRE . The impact of systemic inflammation on Alzheimer’s disease pathology. Front Immunol 2022; 12: 796867.35069578 10.3389/fimmu.2021.796867PMC8770958
43 YarnellJW . The PRIME study: classical risk factors do not explain the severalfold differences in risk of coronary heart disease between France and Northern Ireland. Prospective epidemiological study of myocardial infarction. QJM 1998; 91: 667–676.10024924 10.1093/qjmed/91.10.667
44 LindenGJ McCleanK YoungI , et al. Persistently raised C-reactive protein levels are associated with advanced periodontal disease. J Clin Periodontol 2008; 35: 741–747.18647204 10.1111/j.1600-051X.2008.01288.x
45 WinningL PattersonCC CullenKM , et al. The association between subgingival periodontal pathogens and systemic inflammation. J Clin Periodontol 2015; 42: 799–806.26309048 10.1111/jcpe.12450
46 von ElmE AltmanDG EggerM , et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370: 1453–1457.18064739 10.1016/S0140-6736(07)61602-X
47 PageRC EkePI . Case definitions for use in population-based surveillance of periodontitis. J Periodontol 2007; 78: 1387–1399.10.1902/jop.2007.06026417608611
48 BucksRS AshworthDL WilcockGK , et al. Assessment of activities of daily living in dementia: development of the Bristol activities of daily living scale. Age Ageing 1996; 25: 113–120.8670538 10.1093/ageing/25.2.113
49 SheikhJI YesavageJA BrooksJO3rd , et al. Proposed factor structure of the geriatric depression scale. Int Psychogeriatr 1991; 3: 23–28.1863703 10.1017/s1041610291000480
50 AlbertMS DeKoskyST DicksonD , et al. The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the national institute on aging-Alzheimer's association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 270–279.21514249 10.1016/j.jalz.2011.03.008PMC3312027
51 McKhannGM KnopmanDS ChertkowH , et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the national institute on aging-Alzheimer's association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 263–269.21514250 10.1016/j.jalz.2011.03.005PMC3312024
52 SachdevP KalariaR O'BrienJ , et al. Diagnostic criteria for vascular cognitive disorders: a VASCOG statement. Alzheimer Dis Assoc Disord 2014; 28: 206–218.24632990 10.1097/WAD.0000000000000034PMC4139434
53 WagnerA SimonC EvansA , et al. Physical activity patterns in 50-59 year men in France and northern Ireland. Associations with socio-economic status and health behaviour. Eur J Epidemiol 2003; 18: 321–329.12803372 10.1023/a:1023625110856
54R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2021. https://www.R-project.org/
55 KimDH JeongSN LeeJH . Severe periodontitis with tooth loss as a modifiable risk factor for the development of Alzheimer, vascular, and mixed dementia: national health insurance service-national health screening retrospective cohort 2002-2015. J Periodontal Implant Sci 2020; 50: 303–312.33124208 10.5051/jpis.2000600030PMC7606895
56 HanssonP Eriksson SörmanD BergdahlJ , et al. Dental status is unrelated to risk of dementia: a 20-year prospective study. J Am Geriatr Soc 2014; 62: 979–981.24828929 10.1111/jgs.12814
57 HolmerJ EriksdotterM HäbelH , et al. Periodontal conditions and incident dementia: a nationwide Swedish cohort study. J Periodontol 2022; 93: 1378–1386.35099831 10.1002/JPER.21-0518
58 CaoQ TanC-C XuW , et al. The prevalence of dementia: a systematic review and meta-analysis. J Alzheimers Dis 2020; 73: 1157–1166.31884487 10.3233/JAD-191092
59 CorbetEF . Oral diagnosis and treatment planning: part 3. Periodontal disease and assessment of risk. Br Dent J 2012; 213: 111–121.22878307 10.1038/sj.bdj.2012.666
60 van der FlierWM ScheltensP . Epidemiology and risk factors of dementia. J Neurol Neurosurg Psychiatry 2005; 76: v2–v7.10.1136/jnnp.2005.082867PMC176571516291918
61 SocranskySS HaffajeeAD CuginiMA , et al. Microbial complexes in subgingival plaque. J Clin Periodontol 1998; 25: 134–144.9495612 10.1111/j.1600-051x.1998.tb02419.x
62 HajishengallisG LamontRJ . Polymicrobial communities in periodontal disease: their quasi-organismal nature and dialogue with the host. Periodontol 2000 2021; 86: 210–230.10.1111/prd.12371PMC895775033690950
63 ZimnyA PłonczyńskaA JakubowskiW , et al. Porphyromonas gingivalis affects neutrophil pro-inflammatory activities. Front Cell Dev Biol 2025; 13: 1419651.39936030 10.3389/fcell.2025.1419651PMC11811088
64 HoltfreterB AlbandarJM DietrichT , et al. Standards for reporting chronic periodontitis prevalence and severity in epidemiologic studies: proposed standards from the joint EU/USA periodontal epidemiology working group. J Clin Periodontol 2015; 42: 407–412.25808877 10.1111/jcpe.12392PMC7441325