Regular dental visits, periodontitis, tooth loss, and atherosclerosis: The Ohasama study

Abstract Objective We aimed to explore the association between regular dental visits and atherosclerosis and between periodontitis, number of remaining teeth, and atherosclerosis among community dwellers in Japan. Background Few studies have examined the association between regular dental visits, periodontitis, tooth loss, and atherosclerosis in community dwellers in Japan. Methods The participants of this cross‐sectional study included community dwellers aged ≥55 years and residing in Ohasama. Exposure variables were regular dental visits; periodontitis, defined as radiographic alveolar bone loss (BL); the Centers for Disease Control/American Academy of Periodontology (CDC/AAP) classification; and number of remaining teeth. The primary outcome was atherosclerosis, defined as maximum carotid intima‐media thickness ≥1.1 mm or confirmation of atheromatous plaque. Results Of 602 participants, 117 had atherosclerosis. In the multivariate model, compared to those with regular dental visits, the odds ratio (OR) (95% confidence intervals [CIs]) of atherosclerosis among those with the absence of regular dental visits was 2.16 (1.03–4.49). Regarding BL‐max, compared with those in the first quartile, ORs (95% CIs) of those in the second, third, and fourth quartiles were 1.15 (0.65–2.30), 0.65 (0.32–1.35), and 1.57 (0.81–3.01), respectively. Regarding CDC/AAP classification, compared to those with no or mild periodontitis, ORs (95% CIs) for those with moderate and severe periodontitis were 2.48 (0.61–10.1) and 4.26 (1.01–17.5), respectively. Regarding the number of remaining teeth, compared to those with ≥20 teeth, ORs (95%CIs) for those with 10–19 and 1–9 teeth were 1.77 (1.004–3.12) and 0.96 (0.52–1.80), respectively. Conclusion The absence of regular dental visits and presence of periodontitis are associated with atherosclerosis among community dwellers in Japan.


| INTRODUC TI ON
Cardiovascular disease (CVD), which is one of the major noncommunicable diseases (NCDs), and its clinical manifestations (e.g., atherosclerosis) remain a major public health issue, and efforts are ongoing to decrease their incidence rates and cause-specific mortality worldwide. 1 In addition, CVD and atherosclerosis are also related to dementia, which is increasing in incidence worldwide. [2][3][4][5] For healthy living in older stages of life, it is necessary to prevent the incidence and aggravation of CVD and the associated clinical disease, atherosclerosis.
Several epidemiological studies have demonstrated that periodontitis, which is also considered an NCD, is associated with atherosclerosis and CVDs such as myocardial infarction or stroke. [6][7][8] Ahn et al. 9 revealed an association between clinical attachment loss, which was measured by radiographic alveolar bone loss (BL) and peripheral arterial disease. Further, regarding subclinical cardiovascular disease, several studies have reported that carotid intima-media thickness (IMT) is related to periodontal disease. [10][11][12][13] In addition to periodontitis, several evidence-based studies have indicated that masticatory performance and the number of remaining teeth, which are associated with nutritional status and dietary intake, and oral health behavior, which can prevent periodontitis and tooth loss, are associated with atherosclerotic disease. [14][15][16][17][18] In this context, Sen et al. 19 elucidated the relationship between regular dental visits and the incidence of stroke. Regular dental visits are essential for the maintenance of oral health, such as periodontal status and the number of remaining teeth. Therefore, it is intuitive that regular dental visits are associated with atherosclerosis and CVD. However, few studies have examined the association between regular dental visits and atherosclerosis among community dwellers in Japan. Exploring the simple oral indicator associated with atherosclerosis leads to subsequent dental treatment and/or any health intervention, which may subsequently reduce the risk of CVD.
The primary aim of this cross-sectional study was to explore the association between regular dental visits and atherosclerosis among community-dwelling middle-aged and older adults in Japan. The secondary aim of this study was to explore the association between periodontitis, the number of remaining teeth, and atherosclerosis in the same population.

| Design and study participants
The present study was conducted as a part of the Ohasama study: a self-measurement of blood pressure at home and ambulatory blood pressure monitoring project since 1986. The project targeted community dwellers aged ≥55 years in Japan. The details of this project have been described elsewhere. 20

| Oral examination
Four well-trained dentists examined the periodontal status and the number of remaining teeth. In preparation for examining the oral cavity, all researchers were calibrated to target examinees using a periodontal probe for calibration, and the senior clinical researcher (T.O.) instructed the other researchers regarding oral cavity examination techniques. During the research period, constant meetings were held to ensure inter-and intra-examiner reproducibility.
Conducted by a questionnaire, regular dental visits were divided into regular (regular dental visits without any symptoms) and episodic (in cases of concern about oral status, inquiring about an oral problem, never going to the dental office for any symptoms) visits.
In this study, radiographic alveolar BL and the Centers for Disease Control/American Academy of Periodontology (CDC/ AAP) classification were used as indicators of periodontitis. 24 BL was measured for all remaining teeth on the mesial and distal Pharma Inc. and Takeda Pharmaceutical Co., Ltd.; The Health Care Science Institute Research Grant; Health Science Center Research Grant; Takeda Science Foundation moderate and severe periodontitis were 2.48 (0.61-10.1) and 4.26 (1.01-17.5), respectively. Regarding the number of remaining teeth, compared to those with ≥20 teeth, ORs (95%CIs) for those with 10-19 and 1-9 teeth were 1.77 (1.004-3.12) and 0.96 (0.52-1.80), respectively.

Conclusion:
The absence of regular dental visits and presence of periodontitis are associated with atherosclerosis among community dwellers in Japan.

K E Y W O R D S
alveolar bone loss, atherosclerosis, epidemiology, periodontitis, regular dental visit, tooth loss sides, except for the third molar, using fine panoramic dental X-ray equipment (Veraviewepocs X550, J Morita Mfg Corp) and digital image measurement software (Quick Grain Standard; Inotech). BL was defined as the ratio of the vertical distance from the cementenamel junction to the anatomical root apex to from the alveolar bone crest to the cement-enamel junction, based on the Schei ruler method. 25 Measurement for BL was conducted by a trained dentist.
BL-max, which was the highest score for each participant, was applied. Participants were divided into quartiles based on the BL-max score (e.g., the lowest BL-max score group was the first quartile).
Regarding the CDC/AAP classification, periodontitis was divided into three groups: normal (no or mild periodontitis), moderate periodontitis, and severe periodontitis. 9 The number of remaining teeth was defined as the total number of remaining teeth excluding residual roots and teeth with grade three mobility.

| Atherosclerosis
Atherosclerosis was evaluated by carotid ultrasound using a real- tively. An atheromatous plaque was defined as a discrete protruded lesion with an inflection point in the observation area. 26 IMT was defined as the highest IMT score in each area, except in areas with thickening caused by atheromatous plaque. Atherosclerosis was defined when max-IMT, which was the maximum value of IMT between each area, was ≥1.1 mm, or when an atheromatous plaque was confirmed. 27

| Covariates
The present study surveyed body mass index (BMI), current medical history (diabetes and dyslipidemia), antihypertensive medication use, systolic blood pressure (SBP), diastolic blood pressure (DBP), smoking, drinking alcohol, and educational attainment.
These covariates were selected based on previous epidemiological research focusing on regular dental visits, periodontitis, tooth loss, and atherosclerosis. 19,26 BMI (kg/m 2 ) was divided into underweight (<18.5), 18.5-<25.0 (normal), and ≥25.0 (overweight). 28 The definitions of diabetes and dyslipidemia were based on the diagnosis of these diseases, including information regarding current treatment.
Home blood pressure data were collected using a semiautomatic BP measuring device (HEM401C; Omron Healthcare Co, Ltd). The participants were required to measure their BP in the morning and the evening for four weeks. Detailed methods to collect home blood pressure data are reported elsewhere. 29 SBP (mm Hg) was categorized into <115, 115-124, 125-134, and ≥135, respectively, and DBP (mm Hg) was categorized into <75, 75-84, and ≥85, respectively.
These scores referred to the Japanese Society of Hypertension Guidelines for the Management of Hypertension. 30 Smoking and alcohol drinking statuses were divided into current, past, or no smoking/drinking. Educational attainment was defined as the number of years of education after elementary school, and the participants were grouped into those who received <10 years and ≥10 years of education after elementary school.

| Statistical analyses
The exposure variables were regular dental visits, BL-max (quartile), CDC/AAP classification (no or mild periodontitis, moderate periodontitis, severe periodontitis), and the number of remaining teeth (≥20 teeth, 10-19 teeth, and 1-9 teeth). The primary outcome was atherosclerosis (i.e., max-IMT ≥1.1 mm or confirmation of atheromatous plaque). In the analysis regarding the CDC/AAP classification, the participants with ≤3 teeth were excluded. The characteristics according to atherosclerosis status were examined using the Wilcoxon rank-sum test for continuous variables and Fisher's exact test for categorical variables. A logistic regression model was applied for statistical examination of the association between oral health variables and atherosclerosis, and the odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Three models were created for the analyses: Model 1 was a null model; Model 2 was adjusted for age and sex; and Model 3 was adjusted for age, sex, BMI, current medical history, current medications, SBP, DBP, smoking, alcohol consumption, and educational attainment. Sensitivity analysis was conducted for the statistical examination of the association between oral health and other atherosclerosis criteria (i.e., mean-IMT ≥0.9 mm or confirmation of atheromatous plaque). 31 Statistical analyses were conducted using JMP Pro Ver.15 (SAS Inc.). All analyses were performed as two-tailed tests, and p-values < .05 were considered statistically significant.  (Table S1). Table 2 shows the results of tests for an association between regular dental visits, BL-max, CDC/AAP classification, the number of remaining teeth, and atherosclerosis. Compared with participants who regularly visited the dentist, ORs (95% CIs) of atherosclerosis among those with episodic dental visits were significantly higher in Model 1 and Model 3. Regarding BL-max, compared with those in the first quartile, the ORs (95% CIs) of those in the second, third, and fourth quartiles were not significantly higher in any models.

| RE SULTS
Regarding CDC/AAP classification, compared to those with no or mild periodontitis, the ORs (95% CIs) of participants with atherosclerosis were significantly higher in those with severe periodontitis in Model 1 and Model 3. Regarding the number of remaining teeth, compared to those with ≥20 teeth, the OR (95% CIs) for atherosclerosis was significantly higher in those with 10-19 teeth in Model 1 and Model 3. The analysis of the relationship between each covariate and atherosclerosis using the multiple logistic regression model revealed that the OR increased significantly with age. The OR was significantly lower for females than males (Table S2).

| DISCUSS ION
The present cross-sectional study examined the association of oral health indicators with atherosclerosis in community-dwelling people aged ≥55 years. After adjusting for age, sex, BMI, current medical history, antihypertensive medication use, SBP, DBP, smoking, alcohol consumption, and educational attainment, regular dental visits were associated with a reduced presence of atherosclerosis, although the association was slightly attenuated in the sensitivity analysis (Table S3). Further, an additional multivariate logistic regression model with inverse probability of treatment weighting that was based on propensity score indicated that regular dental visits were still associated with atherosclerosis (average treatment effect = 0.121; p < .01, data not shown). BL-max was not associated with atherosclerosis. However, the CDC/AAP classification was associated with atherosclerosis in the primary analysis and showed a trend with atherosclerosis in the sensitivity analysis. The number of remaining teeth showed a significant association with atherosclerosis; however, this association disappeared in the sensitivity analysis.
To the best of our knowledge, this is the first study to demonstrate an association between regular dental visits and atheroscle- with lower overall health literacy and that those with irregular dental visits had little health knowledge. 36 If dentists can successfully persuade patients to change their visitation patterns from irregular to regular through approaches such as explaining the merits of maintaining oral health and the relationship between oral health and several health outcomes, the overall health literacy may improve, and the risk of atherosclerosis may be attenuated through improved oral health literacy. Regardless, future, longitudinal research is required to explore the relationship between regular dental visits and atherosclerosis, including information regarding oral health and overall health awareness.
Regarding the indicators of periodontitis, BL-max was not associated with atherosclerosis. In the present study, evaluations of BLmax focused on partial alveolar BL, whereas Ahn et al. allocated the alveolar BL as clinical attachment loss. 9 Hayashida et al. 37 reported that clinical attachment loss was less correlated with atherosclerosis than was periodontal pocket depth because a high score of clinical attachment loss was possibly attributable to non-inflammatory indicators, such as heavy toothbrushing pressure. Considering that inflammatory indicators such as proinflammatory cytokines mediate the relationship between periodontitis and atherosclerosis, [38][39][40] indicators that can evaluate the current inflammatory status of periodontal tissue may detect the association with atherosclerosis better than would alveolar BL. The CDC/AAP classification evaluated by both periodontal pocket depth and clinical attachment loss was associated with atherosclerosis; in the sensitivity analysis, a trend between the two was confirmed (p for trend < .01). These results were similar to those of previous studies that reported an association between periodontitis evaluated by the CDC/AAP classification and carotid IMT scores. 41 In the present study, as there were few participants with no or mild periodontitis, these participants were grouped into the same category. 9 Therefore, the association of periodontitis severity based on the CDC/AAP classification and atherosclerosis was possibly underestimated.
The number of remaining teeth was associated with atherosclerosis in only the 10-19 teeth group, which is not consistent with the findings of previous studies. 15 In this study, smoking, which has been previously observed to negatively affect atherosclerosis, was not associated with atherosclerosis. 43 The Ohasama study has been conducted since 1986 and has provided data for subsequent medical consultations and accompanying health interventions by physicians, public health nurses, and municipalities. Indeed, the current smoking rate of the participants in the present study (11.6%) was lower than the rate among those in their 60s from the national health and nutritional survey in Japan (19.4%). 44 One potential explanation for this discrepancy is that continuous public health attempts may have contributed to attenuating the association of smoking with atherosclerosis.
There are some limitations to this study. First, due to the crosssectional design, it is difficult to clarify whether regular dental visits precede the onset of atherosclerosis. Future research is expected to focus on the duration of regular dental visits, variation in the IMT score, or causal inference using prospective cohorts.
Second, this study did not consider information regarding health awareness or physical activity, which are possible confounders.
Although educational attainment is considered a surrogate indicator of health awareness and physical activity, 45 these factors need to be considered directly. Third, the sample size was relatively small, making conducting multifaceted analyses, such as stratified analysis with age, sex, and other variables challenging. Future research with larger samples will help to elucidate the relationship between regular dental visits, periodontitis, the number of remaining teeth, and atherosclerosis. Finally, the population of this study was community dwellers in a limited area of Japan. Further studies from various populations are needed to confirm the generalizability of the current findings.
In conclusion, this cross-sectional study in community-dwelling people aged ≥55 years indicates that regular dental visits and periodontitis are associated with a reduced presence of atherosclerosis.
Delineating information regarding the benefit of regular dental visits could help identify interventions that could reduce the prevalence of atherosclerosis.

ACK N OWLED G EM ENTS
We are grateful to the residents and staff members of Ohasama