Periodontitis, dental plaque, and atrial fibrillation in the Hamburg City Health Study.

BACKGROUND/AIM: Atrial fibrillation (AF) is a major health problem and causes heart failure and stroke. Pathophysiological mechanisms indicate a link with oral health including periodontitis (PD), but supporting data are scarce. The aim was to investigate the link between features of oral health and the prevalence of AF.
METHODS: This cross-sectional analysis of the Hamburg City Health Study included 5,634 participants with complete data on their PD and AF status. AF was assessed via self-reported questionnaire or medically diagnosed by standard 12-lead resting ECG. The oral health examination included full-mouth measurements of the dental plaque index (PI), the clinical attachment loss (CAL) at 6 sites per tooth, the bleeding on probing (BOP) and the decayed, missing and filled teeth (DMFT) index. Descriptive analyses for all variables stratified by the status of PD were performed. To test for an association between prevalent PD and prevalent AF, multivariable logistic regression models were used. Mediation analysis was used to test if interleukin-6 (IL-6) and/or C-reactive protein (CRP) mediated the association between PD and AF.
RESULTS: Atrial fibrillation (prevalence: 5.6%) and the severity of PD (prevalence: moderate: 57.7%, severe: 18.9%) increased with age in men and women. Prevalent severe PD, CAL ≥3 mm, PI, and BOP were all associated with prevalent AF in unadjusted regression analysis. However, no association except for PI (odds ratio (OR): 1.22, 95% confidence interval (CI): 1.1-1.35, p<0.001) could be observed after adjusting for age, sex, high-sensitivity C-reactive protein (hs-CRP), interleukin-6 (IL-6), body mass index, diabetes, smoking, and educational level. Participants brushing their teeth at least twice daily had a lower AF prevalence compared with those brushing only once daily. Hs-CRP, IL-6, and the odds of AF increased as a function of PD severity grades in unadjusted analysis. However, neither the DMFT index nor IL-6 or CRP was associated with AF after adjusting for age and sex. Mediation analyses could not provide support for the hypothesis that IL-6 or CRP acted as mediator of the association between prevalent PD and prevalent AF.
CONCLUSION: The study shows an association between prevalent AF and increased dental plaque levels indicated by a higher PI. In contrast, an association of prevalent PD with prevalent AF after adjustments for several confounders could not be demonstrated. Further studies are necessary to investigate the mechanisms underlying poor oral hygiene and AF as well as the influence of improved oral hygiene on AF onset.

Introduction

Atrial fibrillation (AF) is the most frequent cardiac arrhythmia and one of the biggest health problems with a prevalence of 1–2.5% in Western countries [1, 2]. AF comes along with major complications like stroke and heart failure [3, 4]. Different risk factors such as age, sex, and cardiovascular disease have been established [5]. In recent years, the influence of inflammation on the occurrence of AF has become more evident and previous studies have shown a correlation between systemic inflammation and AF [6, 7]. According to Jalife et al. [8], the structural remodeling of the left atrium plays an important role in the pathophysiology of AF. However, not all signal cascades that trigger structural remodeling have been identified. Elevated levels of inflammatory biomarkers, like C-reactive protein (CRP), interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α) [6, 9, 10] were observed in AF patients. Moreover, myocardial biopsies from AF patients showed inflammatory infiltrates in the atrial tissue [11].

A link between oral inflammation and several common diseases such as heart disease, stroke, and diabetes, has been observed [12-14]. Periodontitis (PD), which has a prevalence of 70% in German adults [15], is one of the most common oral disorders. The disease is primarily based on the dysbiosis in the accumulated biofilm [16]. Other co-factors are genetic predisposition (host response), smoking and various general diseases [17]. PD is characterized by a destruction of the supporting tissue of the teeth which causes gum bleeding recessions, and enlarged periodontal pockets resulting in tooth-loss if left untreated [18]. Several studies have confirmed increases of systemic inflammation markers in patients with PD [19, 20] and a reduction of serum inflammatory markers in PD patients after periodontal therapy was observed [21].

According to recently published results from the dental cohort of the Atherosclerosis Risk in Communities Study (ARIC), severe PD at baseline was associated with incident AF during follow-up and mediation analysis indicated that AF may mediate the association between PD and stroke [22]. Thus, the aim of the present study was to investigate whether an association between prevalent PD and prevalent AF could be demonstrated in the Hamburg City Health Study (HCHS), which is a large-scale population-based cohort with comprehensive oral and cardiovascular health data, in order to provide further evidence for a relationship between both diseases.

Materials and methods

Subjects, study design, and setting

The Hamburg City Health Study (HCHS) is an ongoing regional prospective cohort study launched in 2016 and performed by the University Medical Center Hamburg-Eppendorf [23]. The aim of the HCHS is to investigate major chronic diseases like stroke, AF, dementia, and their risk factors with an extensive baseline assessment including clinical examinations and self-reported questionnaires.

This study represents a cross-sectional analysis of data from the first 10,000 participants at baseline recruited between February 2016 and November 2018. The general inclusion criteria were: residence in Greater Hamburg, German language knowledge, aged between 45–74 years, and no abnormal medical history (current alcohol abuse or alcohol dependence, use of drugs or oral steroids, and headache due to severe head injury). Exclusion criteria were participants requiring endocarditis prophylaxis, missing periodontal examination and/or missing data for AF leaving 5,634 individuals for the analysis (see S1 Fig for a detailed description of the sample).

This investigation was performed in accordance with the Declaration of Helsinki with accepted ethical standards for research practice [24]. The study protocol was approved by the data protection officer of the University Medical Centre of the University of Hamburg-Eppendorf and the local data protection commissioner (reg.-no.: PV5131). All participants signed a written declaration of consent before registering and participating in the study, which was voluntary. The article was written in accordance with recommendations issued by the “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative” [25].

Assessment of dental variables

All included participants had received a full dental and oral examination (except the third molars). The examination was carried out by trained and certified study nurses, consisting of advanced dentistry students performing their DDS/DMD doctoral thesis work, especially trained dentist’s assistants, and other medical assistants with extensive experience in conducting the examination, which was performed according to a pre-specified SOP under the supervision of a certified dentist. The “study nurses” collected the raw data, such as number of teeth, pocket depths, number of bleeding points on probing etc., which were then used by two certified dentists to establish the diagnosis. In cases of disagreement, consensus was established by consulting a third dentist. Data accuracy was established by training and calibration of the staff, electronic data capture and transfer, longitudinal performance evaluation, and statistical monitoring. The Kappa statistics for the tooth count assessment ranged from 0.96 to 1.00, for untreated dental caries Kappa scores were 0.93 to 1.00. The overall Kappa statistics for identifying more complex traits, such as moderate and severe periodontitis according to the CDC/AAP case definition, ranged between 0.60 and 0.70, which was comparable with similar published results by Dye et al. for NHANES [26]. At the beginning, oral hygiene habits, antibiotic intake and a need for endocarditis prophylaxis were queried. Based on the measurements, a calculation of the decayed, missing and filled teeth (DMFT) index was performed and the plaque index (PI) of Silness-Löe (1964) [27] was assessed.

In order to assess the periodontal health, the probing depth of the gum pocket and the gingival (gum) recession were measured at 6 sites per tooth (mm) with a PCP-12 measuring probe. The clinical attachment loss (CAL) (mm) was calculated based on the measurement of probing depth and gingival recession. Additionally, the bleeding-on-probing (BOP) index (yes/no per tooth, expressed in % of bleeding sites) was measured. The diagnosis of PD and the classification of severity were performed based on the criteria of Eke and Page [28]:

none/mild periodontitis: ≥2 interproximal sites with attachment loss (AL) ≥3 mm, and ≥2 interproximal sites with probing depth ≥4 mm (not on same tooth) or one site with probing depth ≥5 mm

moderate periodontitis: ≥2 interproximal sites with AL ≥4 mm (not on same tooth), or ≥2 interproximal sites with probing depth ≥5 mm (not on same tooth)

severe periodontitis: ≥2 interproximal sites with AL ≥6 mm (not on same tooth) and ≥1 interproximal site with probing depth ≥5 mm

Assessment of cardiac variables

Demographic (age and sex) and anamnestic data were collected in advance by means of questionnaires. Based on the participant’s medical history and a 12-lead resting ECG, AF was diagnosed (self-reported and medically diagnosed AF). Diagnosis and data were interpreted by two experienced cardiologists and in cases of disagreement reviewed by a third cardiologist.

Assessment of biomarkers

For the biomarker analyses of high-sensitivity C-reactive protein (hs-CRP) and high-sensitivity interleukin-6 (hs-IL-6), venous blood samples were obtained from each participant. Hs-CRP was measured in the central lab of the UKE by routine measurements. The blood sample for IL-6 was frozen temporarily in tubes at—80°C for storage. After thawing, 0.2 ml of plasma was analyzed using the Human IL-6 Quantikine HS ELISA Kit (High Sensitivity) (sensitivity: 0.11 pg/mL assay range: 0.2–10 pg/mL).

Statistical analysis

Descriptive analyses were performed for all variables stratified by the status of PD. The frequency distributions of the categorical variables and medians including interquartile ranges (IQRs) for continuous variables were calculated. For bivariate analyses, p-values were calculated by the chi-square test for categorical variables and by a Kruskal-Wallis test for continuous variables. In order to test for associations between oral health measures and AF, multivariable logistic regression analyses were used. Three models were constructed: 1) bivariate unadjusted model, 2) adjusted for age and sex, 3) adjusted for several potentially relevant confounders (age, sex, body mass index, diabetes, smoking, educational level, and hs-CRP). Odds ratios (OR) with their 95% confidence intervals (CI) were reported. Values of p <0.05 were considered significant. Statistical analyses were performed with the RStudio Version 1.1.453 for Windows.

Statistical power considerations

Considering a sample size of n = 5,000, a prevalence of 25% for severe and of 30% for no or mild PD in older German people [15], a prevalence of AF of 6% [1], and assuming a true unadjusted odds ratio for exposed subjects relative to unexposed subjects of 1.5, the sample size is sufficient to reject the null hypothesis that this odds ratio equals 1 with a power of 81% (type I error probability: 0.05; statistical model: Continuity-corrected chi2 statistic).

Results

Baseline characteristics

The baseline characteristics of the cohort are summarized in Table 1 stratified by PD severity. Of participants with none/mild PD, 60.4% were women, 50.7% with moderate PD and 39.1% with severe PD. The severity of PD increased with age (median age in years [IQR]): none/mild PD 59 [52, 66]; moderate periodontitis 63 [55, 69]; severe periodontitis 66 [59, 71]). The prevalence of hypertension (72.5% in severe PD vs. 54.8% in none/mild PD) and of heart failure (2.7% in severe PD vs. 1.6% in none/mild PD) were higher in patients with severe PD compared with none/mild PD. AF was elevated in patients with severe PD (6.9%) in comparison to none/mild PD (4.3%). The proportion of patients affected by myocardial infarction or stroke was highest in the severe PD group. Median levels of inflammatory markers were higher in people with severe PD and the PI, number of teeth loss, and the BOP index increased according to PD severity (Table 1).

Table 1

Characteristics of the study population stratified by grades of periodontitis.

	Degree of Periodontitis
	None/mild	Moderate	Severe
	N = 1453	N = 3580	N = 1176
	Median [IQR] or N (%)
Socio-demographic characteristics
Sex = Female	878 (60.4)	1814 (50.7)	460 (39.1)
Age	59 [52, 66]	63 [55, 69]	66 [59, 71]
Cardiovascular risk factors
BMI	25.56 [23.01, 28.67]	26.02 [23.55, 29.01]	26.45 [24.11, 29.65]
Smoking
Current	235 (16.2)	608 (17.1)	293 (25.1)
Former	625 (43.2)	1581 (44.4)	550 (47)
Never	588 (40.6)	1369 (38.5)	326 (27.9)
Diabetes = Yes	85 (6.2)	242 (7.4)	122 (11.3)
Hypertension = Yes	768 (54.8)	2266 (66.3)	810 (72.5)
Systolic blood pressure	134 [123.5, 146.5]	137.5 [125.5, 150.5]	139.5 [127.5, 153]
Heart parameter
Heart rate	67.5 [61, 75]	68.5 [62, 76]	68.5 [61.5, 76.5]
LVEF	58.49 [54.58, 62.98]	58.57 [54.56, 62.67]	57.99 [53.45, 62.15]
Heart failure = Yes	23 (1.6)	76 (2.1)	31 (2.7)
Atrial fibrillation = Yes	57 (4.3)	181 (5.5)	75 (6.9)
Myocardial infarction = Yes	31 (2.1)	88 (2.5)	38 (3.3)
Stroke = Yes	27 (1.9)	98 (2.8)	45 (3.9)
hs-CRP	0.1 [0.06, 0.23]	0.11 [0.06, 0.25]	0.13 [0.07, 0.3]
hs-IL-6	1.47 [1.03, 2.08]	1.57 [1.16, 2.23]	1.8 [1.34, 2.69]
Dental parameter
Cleaning teeth
Never	3 (0.2)	1 (0)	1 (0.1)
Once per week	1 (0.1)	6 (0.2)	1 (0.1)
Once per day	172 (12.4)	425 (12.5)	160 (14.5)
2x per day	1214 (87.3)	2969 (87.3)	941 (85.3)
DMFT-Index	17 [14, 21]	19 [16, 23]	21 [17, 24.25]
Number of missing teeth	2 [0, 4]	2 [1, 5]	4 [1, 9]
BOP	2.08 [0, 7.14]	8.33 [2.17, 19.23]	21.05 [9.26, 41.67]
Plaque index	0 [0, 10.71]	8.93 [0, 27.78]	22 [5.77, 54.76]

Abbreviations: BMI, body mass index; LVEF, left ventricular ejection fraction; hs-CRP, high-sensitivity C-reactive protein; hs-IL-6, high-sensitivity interleukin-6; DMFT-Index, decayed, missing and filled teeth index; BOP, bleeding on probing

Subgroup results of periodontitis

Table 2 presents baseline characteristics stratified by age and sex. The prevalence of severe PD increased with age in both sexes (8.4% of the 45–54 years and 19.7% of the 65+ aged female participants suffered from severe PD). Male study participants were more often affected by PD than women: the severe PD group included 22.2% of the 55-64-year-old and 29.1% of the 65+ year-old male participants.

Table 2

Periodontal status, plaque index, BOP and AF related to age decades.

	Population
	Male			Female
	45–54	55–64	65+	45–54	55–64	65+
	N = 1046	N = 1577	N = 2269	N = 1227	N = 1745	N = 2136
	Median [IQR] or N (%)
Periodontitis							< 0.001
None/mild	199 (29.7)	183 (18)	193 (14.1)	306 (38.9)	311 (28.1)	261 (20.7)
Moderate	380 (56.6)	610 (59.9)	776 (56.8)	415 (52.7)	650 (58.7)	749 (59.5)
Severe	92 (13.7)	226 (22.2)	398 (29.1)	66 (8.4)	146 (13.2)	248 (19.7)
Plaque Index	8.7 [0, 26.92]	10.87 [0, 34]	17.39 [1.85, 46.86]	3.7 [0, 17.86]	4.35 [0, 1.43]	8.33 [0, 6.96]	< 0.001
BOP	7.14 [1.92, 17.35]	8.93 [2, 23.04]	9.26 [2.38, 22]	5.77 [1.79, 17.31]	7.41 [1.85, 9.64]	8.33 [1.98, 20.09]	< 0.001
AF = Yes	10 (1)	62 (4.3)	276 (13.2)	9 (0.8)	42 (2.7)	162 (8.5)	< 0.001

Abbreviations: BOP, bleeding on probing; AF, atrial fibrillation; IQR, interquartile range.

Subgroup results of atrial fibrillation

Fig 1 shows an increased prevalence of AF with increasing age in both sexes. The prevalence of AF in the group from 45–54 years was low. In men, 13.2% of the age group 65+ had AF, whereas AF was less prevalent in women of the same age group (8.5%). AF was slightly more prevalent in men aged 55–64 years who suffered from moderate/severe PD.

Fig 1

Prevalence of AF (atrial fibrillation) for females and males in different age decades (45–54, 55–64, 65+) classified according to the degrees of periodontitis none/mild and severe/moderate.

Interaction between AF and oral hygiene habits, plaque index and periodontitis

Participants aged ≥55 years who cleaned their teeth only once a week were more frequently affected by AF in comparison to patients with daily/twice a day cleaning frequencies (Fig 2).

Fig 2

Prevalence of AF (atrial fibrillation) (N, %) for ascending age decades (45–54, 55–64, 65+ years) and different oral hygiene habits.

Cleaning frequencies: Never, once per week, once per day, twice per day).

Multiple logistic regression models

Significant associations of prevalent severe PD, PI, and BOP with prevalent AF were observed in the bivariate unadjusted models. After adjusting for age, sex, hs-CRP, BMI, diabetes, smoking, and educational level, the relationships of AF with severe PD and BOP were no longer significant. In contrast, the PI was significantly associated with AF after these adjustments (p <0.001) (Table 3). As shown in Table 4, the DMFT index was not associated with AF after adjustment for age and sex.

Table 3

Logistic regression models (outcome: AF).

	Periodontal parameters						Periodontitis
	CAL ≥ 3 mm		Plaque index		BOP		Moderate		Severe
	OR (Cl)	p-value	OR (Cl)	p-value	OR (Cl)	p-value	OR (Cl)	p-value	OR (Cl)	p-value
Model 1	1.19 (1.07–1.33)	0.002	1.36 (1.25–1.48)	<0.001	1.11 (1.00–1.23)	0.046	1.31 (0.96–1.77)	0.085	1.66 (1.16–2.36)	0.005
Model 2	0.99 (0.88–1.12)	0.91	1.19 (1.09–1.30)	<0.001	1.07 (0.96–1.19)	0.24	0.93 (0.68–1.27)	0.64	0.92 (0.64–1.33)	0.67
Model 3	1.04 (0.91–1.18)	0.58	1.22 (1.10–1.35)	<0.001	1.07 (0.95–1.20)	0.28	0.93 (0.66–1.42)	0.70	0,94 (0.63–1.42)	0.78

Bold: p<0.05. Abbreviations: hs-CRP, high-sensitivity C-reactive protein; BMI, body mass index; CAL, clinical attachment loss; BOP, bleeding on probing; OR, odds ratio; CI, confidence interval.

Adjustment: Unadjusted (Model 1), adjusted for age, sex (Model 2) and for age, sex, log (hs-CRP), BMI, diabetes, smoking, and education (Model 3).

Table 4

Association between the DMFT index (exposure) and AF (outcome).

Predictors	OR (CI)	p-value
DMFT-index	1.00 (0.98–1.02)	0.854
Age	2.54 (2.25–2.88)	<0.001
Sex (women)	0.63 (0.52–0.76)	<0.001

Bold: p<0.05. Abbreviations: DMFT, decayed-missing-filled-teeth; OR, odds ratio; CI, confidence interval.

Subgroup analyses in participants with hypertension and participants lacking risk factors

To study the potential association between prevalent PD and AF in low and high risk groups, we performed multivariable regression analyses in participants with a history of hypertension (high risk group, n = 3,844) and a second group without a history of hypertension, stroke, diabetes mellitus, heart failure, myocardial infarction, current smoking, and/or BMI above 35 (low risk group, n = 1,319, 18 cases of AF, prevalence: 1.36%). As shown in S1 Table no association was observed based on the crude logistic regression model without any adjustments in the low risk group. The age- and sex adjusted multivariable logistic regression model for the low risk group, which is shown in S2 Table, shows that age was strongly associated with AF, whereas sex was not associated with AF. Similar results were observed in the high risk group except that both, age and sex, were associated with AF (S3 and S4 Tables).

Mediation analyses for CRP and IL-6

CRP, IL-6, and the odds of AF increased as a function of PD severity grades in unadjusted analysis (S5 Table). However, the estimates for the average causal mediation effects (ACME) observed for IL-6 and hs-CRP were not significant (S6 Table). The average direct effect (ADE) reached the level of significance for hs-CRP (p = 0.046) but was very small and negative.

Discussion

The cross-sectional data of the population-based cohort study shows associations between several measures of oral health, such as CAL, PI, severe PD, and BOP, and prevalent AF. All but PI lost statistical significance after adjustment for common confounders. The findings thus indicate that the observed associations of oral health parameters with prevalent AF are largely explained by sociodemographic, particularly age and sex, in this sample. This is also supported by the subgroup analyses on hypertensive participants and on participants lacking risk factors. The crude logistic regression model without any adjustments resulted in p-values clearly above 0.05, suggesting that an association between PD and AF was unlikely. However, it should be noted that due to the low prevalence of AF, the power to detect an association in the sub-group analyses may have been insufficient.

The findings are in agreement with results from a cross-sectional study on Danish octogenarians published by Holm-Pedersen et al. who observed an association between root caries and cardiac arrhythmias, but found no association between PD and AF [29]. Conversely, Im et al. observed a higher incidence of arrhythmias including AF in the PD group of a much smaller prospective study [30]. The median age of the non-PD group was below our median age and above in the PD group. Recently published results from the dental cohort of the ARIC study demonstrated that severe PD at baseline was associated with incident AF during follow-up in subjects without AF at baseline. In addition, mediation analysis indicated that AF may mediate the association between PD and stroke [22]. Mean follow-up for AF was over 17 years and Cox proportional hazards models adjusted for AF risk factors were used. The ARIC observations relate to a prospective analysis of AF incidences occurring during a long follow-up period in subjects without AF at baseline, whereas our observations relate to a cross-sectional analysis of the prevalences of PD and AF at baseline with no follow-up data yet available. Since PD and AF strongly correlate with age [15, 31], these divergent results may likely be explained by the differences of the study designs, differing impact of confounders, and/or different age ranges of the study samples.

We relied on a combination of self-reporting and ECG recordings to diagnose AF because paroxysmal and some persistent forms of AF are difficult to detect by a single ECG examination in an epidemiological study. Therefore, we included also the medical history to identify such cases. People are usually aware of their diagnosis and about ¼ of participants who reported a history of AF could actually be verified by ECG. In contrast, only 0.4% of those who reported no history of AF presented with AF according to the ECG. This shows that verifiable cases were highly enriched in the self-reporting group. Nevertheless, it cannot be excluded that including the self-reported cases may have led to some miss-classification. On the other hand, some cases of paroxysmal or persistent AF would have been missed, if we had decided to not consider the reports. Another limitation relates to the relatively low power of the subgroup analyses, which have therefore to be interpreted prudently.

Our data show a significant association between plaque accumulation and prevalent AF that is independent of age, sex, hs-CRP, BMI, smoking, diabetes, and educational status. This supports the hypothesis that dental plaque and the associated acute inflammatory response may have an impact on the development of AF, stronger than the BOP or PD as defined by clinical attachment loss. Various studies have already shown possible links between acute and/or chronic inflammation and AF (reviewed in ref. [32]). Higher systemic pro-inflammatory activity due to oral inflammation reflected by plaque burden may promote atrial remodeling and trigger AF. A potential contribution of Porphyronomas gingivalis, which is one of the key periodontitis-associated bacteria, to the risk of AF has been proposed [33]. P. gingivalis and its virulence factors as well as inflammatory mediators could be involved in cardiac remodeling. In our study, CRP, IL-6, and the odds of AF increased as a function of PD severity grades in non-adjusted analysis. Thus, our data do not exclude that the inflammatory markers play a role in the mechanism of AF. The statistical models show that neither the DMFT nor IL-6 or CRP is associated with AF after adjusting for age and sex, whereas aging and male sex are strongly associated with prevalent AF. It is known that the plasma concentrations of CRP and IL-6 are higher in healthy individuals aged over 65 years compared to younger people [34] and that healthy women have higher CRP levels than healthy men [35]. On the other hand, aging and sex are linked to a vast number of additional effects. Therefore, an investigation of the causal relationship between dental plaque accumulation and AF is beyond the scope of the study.

That oral hygiene may play a role in the prevention of AF is supported by the recent ARIC data [22] and by studies published by Chen et al. and Chang et al. who showed that the removal of biofilms by dental scaling or improving oral hygiene habits could reduce the risk of AF [36, 37].

In conclusion, this study shows an association between dental plaque accumulation and AF which was independent of the tested confounders. Further studies are necessary to investigate the relationship between oral inflammation and AF as well as the impact of improved oral hygiene on the prevention of AF.

Sampling chart.

9,087 ECGs, 9,895 medical histories, and 6,209 full-mouth periodontal examinations were evaluated for the study. Included were 5,634 participants with complete PD and AF classification data. The AF prevalence was 5.6% in the subgroup with complete data compared to 5.7% in the subgroup with AF data. Of 169 AF cases, which were identified by ECG, 129 cases (75%) had a positive medical history of AF.

(PDF)

Click here for additional data file.

Association between periodontitis and atrial fibrillation in low risk subjects: Crude logistic regression model.

The potential association between PD and AF was studied in a low risk group. Participants with a history of hypertension, stroke, diabetes mellitus, heart failure, myocardial infarction, current smoking, and/or BMI above 35 (N = 4,890) were excluded. This resulted in 1,319 subjects, which included 18 cases of AF (AF prevalence: 1.36%). Shown are the results of the crude logistic regression model without any adjustments.

(DOCX)

Click here for additional data file.

Association between periodontitis and atrial fibrillation in low risk subjects: Multivariable logistic regression model.

Age was strongly associated with AF, which corresponded with the results shown in the main paper for the whole sample of 5,634 participants. In contrast, sex was not associated with AF, which differed from the results obtained for the whole sample, in which women had lower odds of AF than men.

(DOCX)

Click here for additional data file.

Arterial hypertension and periodontitis severity grades.

In order to examine the association between PD and AF in a high risk group, subgroup analysis in participants with hypertension was performed. As shown in the Table, 3,844 participants had hypertension and the fraction of participants with severe PD was higher in hypertensive compared with normotensive participants.

(DOCX)

Click here for additional data file.

Association between periodontitis and atrial fibrillation in hypertensive subjects: Multivariable logistic regression model.

The Table shows the age- and sex adjusted multivariable logistic regression model for the high risk group. Age and sex were strongly associated with AF, which corresponded with the results shown in the main paper for the whole sample of 5,634 participants. However, no association between severe or moderate PD and AF could be observed.

(DOCX)

Click here for additional data file.

CRP and IL-6 plasma concentrations according to periodontitis grades.

CRP, IL-6, and the odds of AF increased as a function of PD severity grades in non-adjusted analysis.

(DOCX)

Click here for additional data file.

Mediation analysis for IL-6 and CRP.

Exposure: Periodontitis, outcome: Atrial fibrillation.

(DOCX)

Click here for additional data file.

26 Mar 2021

PONE-D-21-08970

Oral health and atrial fibrillation in the Hamburg City Health Study  ​

PLOS ONE

Dear Dr. Seedorf,

Thank you for submitting your manuscript to PLOS ONE. This is an interesting topics. Before sending out for external review, I would like to clarify some points. You may remove the patients with AF risk (i.e., HTN, DM, Stroke events, etc.) from the analyses. Then, you may see clear results without the statistical models.  Please analyze h-CRP and IL-6 as well. If the inflammatory markers do not play a role in AF, what would be potential mechanism? In addition, I do not think reference 25 would be a suitable to cite since these studies are  merely based upon insurance claims, not clinical data.

Please submit your revised manuscript by May 10 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Tomohiko Ai, M.D., Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments:

This is an interesting topics. Before sending out for external review, I would like to clarify some points. You may remove the patients with AF risk (i.e., HTN, DM, Stroke events, etc.) from the analyses. Then, you may see clear results without the statistical models. Please analyze h-CRP and IL-6 as well. If the inflammatory markers do not play a role in AF, what would be potential mechanism?

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables should be uploaded as separate "supporting information" files.

3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

[Note: HTML markup is below. Please do not edit.]

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 May 2021

Please, see the attached "Response to Reviewers" file

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

7 Jul 2021

PONE-D-21-08970R1

Oral health and atrial fibrillation in the Hamburg City Health Study  ​

PLOS ONE

Dear Dr. Seedorf,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process

Your paper was reviewed by three experts in the field. Although the topic seems to be interesting, several issues need to be clarified. Specifically, there are several confound factors need to be included to asses the PD status, and occurrence of AF needs to be validated. Please read the comments carefully, and address the issues accordingly. I would recommend the authors to consult statistical expert(s).

Please submit your revised manuscript by Aug 21 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Tomohiko Ai, M.D., Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: No

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I find it hard to justify the accuracy of the data collected without more information- Therefore I feel the article can be accepted if they provide more information about the data collected.

It would be useful if the authors shared data on how many patients were already diagnosed with AF, and how many they tested for AF who were diagnosed with unknown AF. Without this data it is hard to see a true representative of data- e.g the patient may have had good plaque control, was diagnosed with AF but since the diagnosis had then had not brushed teeth as much and resulted in poor plaque control. In a cross sectional study ideally the patient would not know they have AF, have poor plaque control and after investigation with ECG, has been diagnosed with unknown AF. If a patient was already diagnosed with AF in the medical was this information known to the researcher prior to the plaque index being carried out? This can introduce researcher bias. Also there is no cause or effect with this type of study as it is cross-sectional they cannot state oral health preventing AF based on their study. If there was an exclusion criteria for those who already had AF then can this be stated clearly in the methods

I have some other comments:

According to a recently published systematic review, only two small clinical studies had been looked at a potential association between PD and AF over the last fifty years [22].

Incorrect, this SR states two small clinical studies looking at acute infections, not PD, also here is a useful and recent citation

Sen S, Redd K, Trivedi T, Moss K, Alonso A, Soliman EZ, Magnani JW, Chen LY, Gottesman RF, Rosamond W, Beck J, Offenbacher S. Periodontal Disease, Atrial Fibrillation and Stroke. Am Heart J. 2021 May;235:36-43. doi: 10.1016/j.ahj.2021.01.009. Epub 2021 Jan 24. PMID: 33503409; PMCID: PMC8084947.

The examination was carried out by trained and certified study nurses.

Normally dentists are required to examine patients and diagnose PD, not nurses- if nurses then is diagnosis inaccurate? If nurses are trained then how are they trained to ensure accurate diagnosis are they performing the periodontal charts? more detail needed here.

Based on the measurements, a calculation of the decayed, missing and filled teeth (DMFT) index was performed and the plaque index (PI) of Silness-Löe (1964) was assessed

Would have liked to see more information in the methods about systematic and rigourous testing of scores- for example, were the plaque scores assessed by 2 different clinicians, if there was disagreement then would a third clinician be involved

Although (DMFT) index is a valid measurement tool it has disadvantages as people can lose teeth from trauma or for orthodontic treatment or tooth decay (eg not just from periodontitis).

The diagnosis of PD and the classification of severity were performed based on the criteria of Eke and Page [26]:

26. Chen SJ, Liu CJ, Chao TF, Wang KL, Chen TJ, Chou P, et al. Dental scaling and atrial fibrillation: a 241 nationwide cohort study. International journal of cardiology. 2013;168(3):2300-3. doi: 242 10.1016/j.ijcard.2013.01.192. PubMed PMID: 23453452.

Incorrect citation placed.

Assessment of cardiac variables

Authors have used a combined method of diagnosing for AF including self reporting and medical investigations with 24 hour ECG- 1) self reporting is considered inaccurate for diagnosing AF, diagnosing all patients medically would have been more accurate

Methods

would like to see a flowchart of patient journey and data which would make the results of the research more easily accessible and clear

Statistical analysis

adjusted for all confounders considered relevant (age, sex, body mass index, diabetes, smoking, and hs-CRP)

There are many more confounders considered relevant- please amend statement to adjusted for some confounders considered relevant

Figure 1 and figure 2 are not clearly labelled, figure 1 no label of unit measurement for y axis is it a percentage, figure 2 for me was not absolutely clear what the units of measurements were as the bar charts had numbers on the top but I do not know what they represented.

Reviewer #2: The authors present a cross-sectional study evaluating the relation between PD and AF. Patients were evaluated for PD using CAL, BOP, DMFTi and PI. AF was assessed using a 24-hour ECG monitoring or a single 12 lead ECG, according to risk, using the CHARGE-AF score. In addition, diagnosis was made using self-reported AF or medically diagnosed AF. Their results reflect what we already know about the epidemiologic patterns of AF, it increases with age and other risk factors such male gender and hypertension.

Overall, after adjustment for multiple confounders, there was no association between PD and AF; however, the authors found an association between markers of oral hygiene like plaque index and AF. They conclude that there might be an association between dental plaque accumulation and AF.

The article is interesting as it adds to the evidence of the link between PD and CVD; particularly the relationship between PD and AF still needs to be confirmed. The text is written in good English, methods and statistical analysis are clear, and results are presented appropriately.

Revisions

1. Could the authors explain the rational for using the CHARGE-AF score to decide if patients where having a 24-hrs ECG monitoring or not in the methods section.

2. Add the CHARGE-AF score for each group of PD as a part of the results.

3. The authors include the LVEF in different groups of PD as a part of their results. Do they have information of other echocardiographic parameters? It would be particularly interesting if they provide data on left atrium (volume, diameter), since it is a marker of FA risk.

4. This is a suggestion: Most of the data that shows an association between PD and cardiac structural and functional abnormalities detected with echocardiogram (mass, myocardial deformation), comes from studies in patients with diabetes mellitus, hypertension and chronic kidney disease. In all of these studies, there was a clear association between PD severity and left ventricular hypertrophy and abnormal ventricular function. (You can see: Periodontal Disease, Systemic Inflammation and the Risk of Cardiovascular Disease. Heart Lung Circ. 2018 Nov;27(11):1327-1334, for a description of some of these studies). In our laboratory, we have done some work in otherwise healthy subjects with different grades of PD using echocardiography, and we could not find an association between PD and LVH or abnormalities in LV function. Although the reasons can be diverse, there is a possibility that PD and the systemic inflammatory process that comes with it acts as a boost for CVD; perhaps, PD might not be enough to cause CV abnormalities, but when added to other types of stress to the CV system (Hypertension, DM) it can cause CVD.

Since AF is multifactorial, and abnormalities in cardiac structure and function can play a role, this can be important for your study. More than 50 of your study population have hypertension; perhaps, you can consider a last analysis including only those with HTN and divide them in those with none/mild PD and severe PD to look for an association between AF and PD, in unadjusted and adjusted analysis.

Reviewer #3: 1) Recent results from ARIC cohort study on PD-AF association not discussed

2) DMFT assessed, a measure primarily for Caries. Data not used for analysis

3) CRP and IL-6 are potential mediators, hence should not be adjusted for

4) Socioeconomic status is a potential confounder that is not assessed

5) Based on hypothesis oral care should be associated with less AF

6) Assessment of AF using questionnaire not a validated method

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Amaar Hassan

Reviewer #2: Yes: Edgar Francisco Carrizales-Sepúlveda

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

2 Sep 2021

6. Review Comments to the Author

Reviewer #1:

It would be useful if the authors shared data on how many patients were already diagnosed with AF, and how many they tested for AF who were diagnosed with unknown AF.

Rely: Of the 561 AF cases, 521 were already diagnosed with AF and 40 additional cases with previously unknown AF were newly identified in the study by the ECGs. The total number of ECGs was 9,087. We added a sampling flow chart to the supplementary material to better characterize the sample.

If a patient was already diagnosed with AF in the medical was this information known to the researcher prior to the plaque index being carried out?

Reply: No, this was not known.

There is no cause or effect with this type of study as it is cross-sectional they cannot state oral health preventing AF based on their study.

Reply: That is true. We removed the statement from the conclusions.

If there was an exclusion criteria for those who already had AF then can this be stated clearly in the methods

Reply: No, this was not an exclusion criterion. This is a cross-sectional study looking at prevalent AF and prevalent PD at baseline with PD as independent variable and AF as dependent variable.

According to a recently published systematic review, only two small clinical studies had been looked at a potential association between PD and AF over the last fifty years [22].

Incorrect, this SR states two small clinical studies looking at acute infections, not PD, also here is a useful and recent citation

Sen S, Redd K, Trivedi T, Moss K, Alonso A, Soliman EZ, Magnani JW, Chen LY, Gottesman RF, Rosamond W, Beck J, Offenbacher S. Periodontal Disease, Atrial Fibrillation and Stroke. Am Heart J. 2021 May;235:36-43. doi: 10.1016/j.ahj.2021.01.009. Epub 2021 Jan 24. PMID: 33503409; PMCID: PMC8084947.

Reply: Thank you. We discuss this excellent recently published work in the discussion of the revised manuscript.

The examination was carried out by trained and certified study nurses.

Normally dentists are required to examine patients and diagnose PD, not nurses- if nurses then is diagnosis inaccurate? If nurses are trained then how are they trained to ensure accurate diagnosis are they performing the periodontal charts? more detail needed here.

Reply: The term “study nurse”, which is used here, refers to extensively trained and calibrated staff, including advanced dentistry students performing their DDS/DMD doctoral thesis work, especially trained dentist’s assistants, and other medical assistants with extensive experience in conducting the examination, which was performed according to a pre-specified SOP under the supervision of a certified dentist. The “study nurses” did not “diagnose periodontitis”, as you say. They collected the raw data, such as number of teeth, pocket depths, number of bleeding points on probing etc., which were then used by two certified dentists to establish the diagnosis. Data accuracy was established by training and calibration of the staff, electronic data capture and transfer, longitudinal performance evaluation, and statistical monitoring. The Kappa statistics for the tooth count assessment ranged from 0.96 to 1.00, for untreated dental caries Kappa scores were 0.93 to 1.00. The overall Kappa statistics for identifying more complex traits, such as moderate and severe periodontitis according to the CDC/AAP case definition, ranged between 0.60 and 0.70, which was comparable with published results by Dye et al. BMC Oral Health (2019) 19:95 for NHANES. Thus, we are confident that the data are accurate. On the other hand, the high fraction of missing data, which was due to a shortage of sufficiently well trained staff, should be noted as a disadvantage. We provided more details in the methods section and added a sampling chart to the supplementary material (Fig. S1).

Based on the measurements, a calculation of the decayed, missing and filled teeth (DMFT) index was performed and the plaque index (PI) of Silness-Löe (1964) was assessed

Would have liked to see more information in the methods about systematic and rigourous testing of scores- for example, were the plaque scores assessed by 2 different clinicians, if there was disagreement then would a third clinician be involved

Reply: All scores that we used are in essence very simple and have been in use in dentistry already for a long time. We provided more details in the methods section.

Although (DMFT) index is a valid measurement tool it has disadvantages as people can lose teeth from trauma or for orthodontic treatment or tooth decay (eg not just from periodontitis).

Reply: We agree with you. However, we have not used the DMFT for any evaluations. Nevertheless, the DMFT may be interesting for some readers because it is a widely used indicator of overall dental health. The actual values show that our cohort in fact has a very good oral health.

The diagnosis of PD and the classification of severity were performed based on the criteria of Eke and Page [26]:

26. Chen SJ, Liu CJ, Chao TF, Wang KL, Chen TJ, Chou P, et al. Dental scaling and atrial fibrillation: a 241 nationwide cohort study. International journal of cardiology. 2013;168(3):2300-3. doi: 242 10.1016/j.ijcard.2013.01.192. PubMed PMID: 23453452.

Incorrect citation placed.

Reply: Thank you. We corrected this error.

Assessment of cardiac variables

Authors have used a combined method of diagnosing for AF including self reporting and medical investigations with 24 hour ECG- 1) self reporting is considered inaccurate for diagnosing AF, diagnosing all patients medically would have been more accurate

Reply: Paroxysmal and some persistent forms of AF are difficult to detect by a single ECG examination in an epidemiological study. Therefore, we relied also on the medical history to identify such cases. People are usually aware of their diagnosis and about ¼ of participants who reported a history of AF could actually be verified by ECG. In contrast, only 0.4% of those who reported no history of AF presented with AF according to the ECG. This shows that verifiable cases were highly enriched in the self-reporting group. Nevertheless you are right that self-reporting may lead to some miss-classification. On the other hand, we would have missed some cases of paroxysmal or persistent AF, if we had decided to not consider the reports. We added this consideration to the discussion section.

would like to see a flowchart of patient journey and data which would make the results of the research more easily accessible and clear

Reply: Very good suggestion. We added a flowchart to the supplementary material.

adjusted for all confounders considered relevant (age, sex, body mass index, diabetes, smoking, and hs-CRP)

There are many more confounders considered relevant- please amend statement to adjusted for some confounders considered relevant

Reply: Okay, done.

Figure 1 and figure 2 are not clearly labelled, figure 1 no label of unit measurement for y axis is it a percentage, figure 2 for me was not absolutely clear what the units of measurements were as the bar charts had numbers on the top but I do not know what they represented.

Reply: Please, see improved figures in the revised manuscript.

Reviewer #2:

Could the authors explain the rational for using the CHARGE-AF score to decide if patients where having a 24-hrs ECG monitoring or not in the methods section.

Reply: This was originally planned according to the study protocol in order to identify cases within the high risk group that did not show up in the 12-lead resting ECGs (i.e. paroxysmal, persistent AF). However, the devices were not well tolerated by the participants and only 330 participants received them. No additional AF cases could be identified by the 24-hrs ECGs. Thus, this strategy was cancelled early on in the study and CHARGE-AF scores were not determined systematically. We amended the text in methods section accordingly.

Add the CHARGE-AF score for each group of PD as a part of the results.

Reply: The scores are available only for a small fraction recruited at the very beginning. Thus, we refrain from reporting them.

The authors include the LVEF in different groups of PD as a part of their results. Do they have information of other echocardiographic parameters? It would be particularly interesting if they provide data on left atrium (volume, diameter), since it is a marker of FA risk.

Reply: These data are not yet fully quality controlled and could therefore not be used.

This is a suggestion: Most of the data that shows an association between PD and cardiac structural and functional abnormalities detected with echocardiogram (mass, myocardial deformation), comes from studies in patients with diabetes mellitus, hypertension and chronic kidney disease. In all of these studies, there was a clear association between PD severity and left ventricular hypertrophy and abnormal ventricular function. (You can see: Periodontal Disease, Systemic Inflammation and the Risk of Cardiovascular Disease. Heart Lung Circ. 2018 Nov;27(11):1327-1334, for a description of some of these studies). In our laboratory, we have done some work in otherwise healthy subjects with different grades of PD using echocardiography, and we could not find an association between PD and LVH or abnormalities in LV function. Although the reasons can be diverse, there is a possibility that PD and the systemic inflammatory process that comes with it acts as a boost for CVD; perhaps, PD might not be enough to cause CV abnormalities, but when added to other types of stress to the CV system (Hypertension, DM) it can cause CVD.

Reply: This is an interesting idea. We are constantly expanding the cohort to increase the case numbers and thus the study’s power. At present, the power is still not high enough for informative sub-group analyses of the kind you are suggesting.

Since AF is multifactorial, and abnormalities in cardiac structure and function can play a role, this can be important for your study. More than 50 of your study population have hypertension; perhaps, you can consider a last analysis including only those with HTN and divide them in those with none/mild PD and severe PD to look for an association between AF and PD, in unadjusted and adjusted analysis.

Reply: We did this. Please, see Results section and supplementary material.

Reviewer #3:

Recent results from ARIC cohort study on PD-AF association not discussed

Reply: Okay, fixed.

DMFT assessed, a measure primarily for Caries. Data not used for analysis

Reply: We have not used the DMFT for any evaluations because teeth loss can result from various issues, such as trauma, orthodontic treatment or tooth decay. Nevertheless, the DMFT may be interesting for some readers because it is a widely used indicator of overall dental health. The actual values show that our cohort in fact has a very good overall oral health.

CRP and IL-6 are potential mediators, hence should not be adjusted for

Reply: Very good point. We performed a mediation analysis for CRP and IL-6. Please see Results section and supplementary material.

Socioeconomic status is a potential confounder that is not assessed

Reply: Thank you, that is true. We included educational level in the revision as a proxy for the SES (It is difficult to obtain correct income information in surveys, which renders the SES unreliable).

Based on hypothesis oral care should be associated with less AF

Reply: The prevalence of AF was lower in participants brushing their teeth 2x and more compared with those brushing less than 2 times daily.

Assessment of AF using questionnaire not a validated method

Reply: Paroxysmal and some persistent forms of AF, which account for ~50% of all cases of AF, are difficult to detect by a single ECG examination in an epidemiological study. Therefore, we relied also on the medical history to identify such cases. People are usually aware of their diagnosis and about ¼ of participants who reported a history of AF could actually be verified by ECG in our study. In contrast, only 0.4% of those who reported no history of AF presented with AF according to the ECG. This shows that verifiable cases were highly enriched in the self-reporting group. Nevertheless you are right that self-reporting may have led to some miss-classification. On the other hand, we would have missed most cases of paroxysmal or persistent AF, if we had not included self-reported AF. We added this consideration to the discussion section.

Submitted filename: Reply_to_Reviewers_comments_rev1.docx

Click here for additional data file.

7 Oct 2021

PONE-D-21-08970R2Periodontitis, dental plaque, and atrial fibrillation in the Hamburg City Health StudyPLOS ONE

Dear Dr. Seedorf,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Your revised paper was reevaluated by the previous reviewers. Though the quality of paper has been improved, one reviewer made some suggestions to further improve the analysis. Please read the comments carefully and address the issues accordingly.

Please submit your revised manuscript by Nov 21 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Tomohiko Ai, M.D., Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: No further suggestions as the author has responded to all points and improved the overall validity of the article

Reviewer #2: The authors have made a great effort addressing the reviewers’ suggestions; the quality of the manuscript has improved.

Although the overall results of the study do not suggest an association between PD and AF, the study is still of relevance since it adds valuable information to the field, I think the paper is now ready to be considered for publication.

Reviewer #3: The reviewers comments have been addressed. I would suggest following changes:

1) Abstract conclusion starts with "no finding" first, and readers have to go sentences down to figure out what the positive findings of PI -->AF association

2) DMFT is an useful marker of oral health and dental caries. If the DMFT score is better in the group, please indicate that as a reason why it was not tested for association with AF. Alternately please include the analysis for readers to know that DMFT is not associated with AF

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Amaar Obaid Hassan

Reviewer #2: Yes: Edgar Francisco Carrizales-Sepulveda

Reviewer #3: Yes: Souvik Sen MD

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Oct 2021

Please see uploaded file "Respons to reviewers"

Submitted filename: Response to Reviewers_rev2.docx

Click here for additional data file.

25 Oct 2021

Periodontitis, dental plaque, and atrial fibrillation in the Hamburg City Health Study

PONE-D-21-08970R3

Dear Dr. Seedorf,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. I think this study will be an important reference regarding association between oral hygiene and AF. Congratulations!

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Tomohiko Ai, M.D., Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #3: Thank you for responding to all comments. All comments have been addressed including addition of DMFT data.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: Yes: Souvik Sen MD

2 Nov 2021

PONE-D-21-08970R3

Periodontitis, dental plaque, and atrial fibrillation in the Hamburg City Health Study

Dear Dr. Aarabi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Tomohiko Ai

Academic Editor

PLOS ONE