Impact of periodontitis as representative of chronic inflammation on long-term clinical outcomes in patients with atrial fibrillation

Abstract

Objectives

Relationship between atrial fibrillation (AF) and inflammation was shown in previous studies. However, there was limited data about the association between the periodontitis and AF in the long-term follow-up. The aim of this study was to evaluate the impact of periodontitis on long-term clinical outcomes in patients with AF.

Methods

The Kosin University echocardiography, ECG and periodontitis database were reviewed from 2013 to 2015 to identify patients with AF. Those patients were divided into two groups according to the presence of periodontitis and clinical events including any arrhythmic attack, thromboembolic and bleeding and death were collected during a median of 18 months.

Results

Among 227 patients with AF, 47 (20.7%) patients had periodontitis. Major adverse cardiac events (MACE) were significantly higher in patients with periodontitis compared with those without periodontitis (p<0.001). Arrhythmias including AF, atrial tachycardia, atrial premature beat, ventricular tachycardia and ventricular premature beat also occurred in 44 (93.6%) patients, which was higher significantly higher incidence in patients with periodontitis than in those without periodontitis (p<0.001). In univariate analysis, age, CHA₂DS₂-VASc, left atrial volume index (LAVi) and periodontitis were significantly associated with arrhythmic events and MACE including bleeding events, thromboembolic events, arrhythmic events and mortality. In multivariate analysis, LAVi (p=0.005) and periodontitis (p<0.001) were independent risk factors for arrhythmic events and periodontitis (p<0.001) for MACE at the long-term follow-up.

Conclusions

The periodontitis as representative of chronic inflammation was an independent predictor of arrhythmic events and MACE in patients with AF.

Key questions.

What is already known about this subject?

• In recent studies, there have been increasing evidence linking local and systemic inflammation to broad spectrum of cardiovascular condition, such as coronary artery disease and atrial fibrillation (AF). In addition, emerging data supported the association between inflammation and AF.

What does this study add?

• AF patients with periodontitis, as representative of chronic inflammation, were more likely to develop arrhythmic events including AF, atrial tachycardia and atrial flutter AF and total any events than those without periodontitis.

How might this impact on clinical practice?

• The control of oral disease may improve inflammation and may be used as a possible therapeutic target to prevent embolic events and recurrence of arrhythmia in patients with AF.

Introduction

Atrial fibrillation (AF) is the most common clinically significant arrhythmia in clinical practice. AF is associated with increased morbidity and mortality that primarily occur as a result of complications, such as thromboembolic events and heart failure.¹

AF is often associated with other cardiovascular diseases (CVDs), including diabetes mellitus, hypertension, congestive heart failure, ischaemic heart disease, valvular heart diseases and other cardiomyopathies. However, in 10%–15% of the cases, AF occurs in the absence of any such comorbidities and is determined as lone AF.^{2 3} In recent studies, there has been increasing evidence linking local and systemic inflammation to broad spectrum of cardiovascular condition, such as coronary artery disease and AF. In addition, emerging data supported the association between inflammation and AF.^{4 5}

Various dental procedures, routine oral hygiene efforts and even gentle mastication may lead to the entry of bacteria or bacterial endotoxins from the oral cavity into the bloodstream; therefore, bacterial infection, antigens, endotoxins and pro-inflammatory cytokines triggering a systemic response may represent the link between periodontitis and coronary artery disease.⁶ Adult periodontitis is a chronic inflammatory disease resulting from infection by specific microbial species. Local inflammation in oral cavity causes an enhanced inflammatory response at distant sites.

Recently, the relations between periodontitis and CVDs have been recognised. Periodontitis plays a role in the pathogenesis of atherosclerosis, and relationship between AF and inflammation has been shown in previous studies.⁷

However, there have been limited data about the association between the periodontitis and AF in the long-term follow-up. The aim of this study was to evaluate the impact of periodontitis on long-term clinical outcomes in patients with AF.

Methods

Study populations

We retrospectively reviewed the medical records of 428 patients with non-valvular AF at Kosin University Gospel Hospital from January 2013 to November 2015.

We included non-valvular AF patients with/without periodontitis. Patients with a history of AF documented by a standard ECG or Holter-ECG were enrolled.

Patients with a recent history of acute infection or inflammatory disease except periodontitis were excluded. Exclusion criteria were as follows: age >80 years; a history of cardiomyopathy, or valvular or congenital heart disease; hepatic or renal disease (known chronic liver disease or aspartate aminotransferase [AST] >3 times than normal range, more advanced chronic kidney disease [CKD] or stage 3); an acute cardiovascular or cerebrovascular event within the preceding three months; any major trauma or surgery within the preceding three months; hyperthyroidism; uncontrolled hypertension; malignancy; connective tissue disease or ischaemic heart disease.

Finally, 227 consecutive patients with non-valvular AF (mean age 59.6±10.9 years, 68.7% male) were enrolled. And all patients were reviewed to evaluate major bleeding, thromboembolic events, arrhythmic events and death during follow-up according to the periodontitis.⁸ Trained and calibrated dentists examined the periodontal status of the patients. The WHO community periodontal index (CPI) was used to evaluate periodontitis and periodontitis was diagnosed when a CPI was greater than or equal to code 3.⁹

Data collection

After ECG and chest X-ray, cardiovascular status was evaluated for each patient using echocardiography, 24-hour Holter recordings and blood laboratory data from the initial visit, as determined by the attending physicians. From the database, the following information was collected: (1) patient data, including sex, age, height and weight; (2) cardiovascular risk factors, including hypertension and diabetes mellitus; (3) CVD status, including structural heart disease, congestive heart failure or a history of a disabling cerebral infarction or transient ischaemic attack (TIA); and (4) use of medication. We categorised the alcohol drinker was defined with moderate alcohol consumption (1–7 drinks/week).

Definitions of AF and atrial arrhythmia

In the present study, paroxysmal AF (PAF) at the initial visit was defined as sinus rhythm on ECG and previous diagnosis of PAF by referring physicians. Patients whose AF was estimated to continue for ≥7 days after the initial visit were considered to have persistent AF (PeAF) originally and were excluded from the analysis. Chronic AF was defined as an ongoing long-term episode. Asymptomatic AF was defined as AF documented on 12-lead ECG during a visit, in the absence of any new symptoms such as palpitations, tachycardia, fatigue, malaise, shortness of breath on exertion, dyspnoea, chest pain, syncope or pre-syncope related to AF or other illnesses. During the follow-up period, the onset of AF was defined as the first time in which all ECGs indicated AF after ≥3 consecutive ECGs at intervals of ≥1 week after the initial examination.¹⁰ When an ECG could not be obtained thrice during the period, the physicians made a clinical judgement regarding the onset time of AF progression. We calculated the CHADS₂ score (congestive heart failure, hypertension, age (≥75 years), diabetes mellitus; one point each, and history of stroke or TIA; two points). The CHA₂DS₂-VASc score was also determined, which also includes vascular disease (previous myocardial infarction, complex aortic plaque and peripheral artery disease), age 65–74.¹¹ HAS-BLED score was calculated (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile International Normalized Ratio [INR], elderly (>65), drugs/alcohol concomitantly—one point each). Atrial arrhythmia during follow-up was defined as atrial premature complex, atrial tachycardia and atrial flutter.

Clinical endpoints

The primary clinical endpoint was major adverse cardiac events (MACE) including major bleeding defined as that which was clinically overt and associated with any of the following: fatal outcome; involvement of a critical anatomic site (intracranial, spinal, ocular, pericardial, articular, retroperitoneal or intramuscular with compartment syndrome); fall in haemoglobin concentration >2 g/dL; transfusion of >2 U of whole blood; or packed red blood cells, thromboembolic events, which was the composite of stroke, TIA, systemic emboli and death during follow-up. And the secondary endpoint was arrhythmic events including AF, atrial tachycardia, atrial premature beat, ventricular tachycardia and ventricular premature beat.

Transthoracic echocardiography

All enrolled subjects underwent two-dimensional transthoracic echocardiography (TTE). All examinations were performed using a commercially available Vivid V.9 (GE Medical System, Vingmed, Horten, Norway) ultrasound system. All recorded echocardiograms were measured and interpreted with clinical information blinded using a computerised offline analysis station (Echopac V.6.3.4; GE Medical System).

Statistical analysis

For continuous data, statistical differences were evaluated using Student’s t-test or the Mann-Whitney U test, depending on the data distribution. Categorical variables are presented as frequencies (per cent) and were analysed using the χ² test. To determine whether any of the variables were independently related to periodontitis, a multivariate analysis of variables with a p value <0.05 in the univariate analysis was performed using linear logistic regression analysis. All statistical analyses were conducted using SPSS statistical software, V.19.0, and statistical significance was set at p<0.05 (two-sided).

Results

The baseline demographics for both groups are listed in table 1. This study consisted of 47 AF subjects with periodontitis and 180 AF subjects without periodontitis. Baseline characteristics were not statistically different between the periodontitis subjects and the non-periodontitis subjects, except for age (p<0.001), HAS-BLED scores (p<0.001) and PeAF (p=0.043). There was no patient with chronic AF in both groups—total 70 patients (30.8%) with PAF and 157 patients (69.2%) with PeAF.

Table 1.

Baseline clinical characteristics in patients with atrial fibrillation (AF) according to the presence of periodontitis

Variables	Periodontitis group (n=47)	Non-periodontitis group (n=180)	P values
Age (years)	71.1±8.7	56.7±9.4	<0.001
Gender (male, %)	27 (61.4)	129 (72.5)	0.197
PeAF (%)	11 (25.0)	59 (32.3)	0.043
DM (%)	12 (27.3)	26 (14.5)	0.071
HTN (%)	28 (65.1)	98 (54.7)	0.235
CHF (%)	5 (10.6)	19 (10.6)	1.000
CVA (%)	3 (1.7)	3 (6.8)	0.093
CHADS2	1.3±0.5.	1.4±0.4	0.789
CHA2DS2-VASc	1.8±1.5	1.8±0.7	0.845
HAS-BLED score	2.9±1.6	1.5±0.5	<0.001
COPD (%)	1 (2.3)	0 (0)	0.197
Alcohol (%)	6 (14.0)	48 (26.8)	0.112
Smoking (%)	5 (11.6)	30 (16.8)	0.491
Medication
Antiarrhythmics (%)
Amiodarone (%)	2 (4.0)	6 (3.5)	1.000
Propafenone (%)	6 (12.7)	25 (14.0)	0.450
Flecainide (%)	7 (1.5)	27 (1.5)	1.000
Digoxin (%)	2 (4.3)	11 (6.1)	1.000
Beta-blocker (%)	11 (23.2)	43 (24.0)	0.419
CCB (%)	13 (28.8)	44 (24.5)	0.365
ARB and ACEi (%)	12 (25.5)	42 (23.5)	0.407
Statins (%)	0 (0)	0 (0)	1.000
Aspirin (%)	4 (8.5)	64 (35.6)	<0.001
VKA	12 (25.5)	101 (56.1)	<0.001
Laboratory findings
WBC (103/μL)	8.7 ± 3.5	6.6 ± 2.8	< 0.001
NLR	7.8 ± 6.5	2.8 ± 2.2	0.001
Creatinine (mg/dL)	1.1 ± 1.0	0.9 ± 0.7	0.113
hsCRP (mg/dL)	1.6 ± 0.3	1.1 ± 0.4	0.048
Echo parameters
LVEF (%)	62.0 ± 9.7	63.8 ± 8.3	0.203
LVIDs (mm)	30.6 ± 5.9	32.0 ± 5.6	0.151
LVIDd (mm)	45.6 ± 9.7	46.3 ± 5.9	0.245
IVSD (mm)	11.7 ± 2.8	11.2 ± 1.8	0.127
LVPWD (mm)	10.4 ± 2.3	10.2 ± 1.5	0.441
LAVi (mL/m2)	44.2 ± 8.8	36.0 ± 13.1	< 0.001
E velocity (cm/s)	0.9 ± 0.4	0.8 ± 0.2	0.004
A velocity (cm/s)	0.8 ± 0.2	0.6 ± 0.2	< 0.001
E/A	1.0 ± 0.7	1.4 ± 0.7	0.016
E’	0.05 ± 0.2	0.1 ± 0.1	0.001
A′	0.1 ± 0.02	0.1 ± 0.02	0.090
E/E′	13.8 ± 6.7	9.8 ± 4.8	< 0.001

Values are mean±SD (range).

A, peak velocity of the late filling wave due to atrial contraction; A’, late diastolic mitral annulus velocity; ARB, angiotensin II receptor blocker; CCB, calcium channel blocker; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CRF, chronic renal failure; CVA, cerebrovascular accident; DM, diabetes mellitus; E, the peak mitral flow velocity of the early rapid filling wave; E’, early diastolic mitral annulus velocity; HTN, hypertension; hsCRP, high-sensitive C-reactive protein; IVSd, interventricular septal diameter; LAVi, left atrial volume index; LVEF, left ventricular ejection fraction; LVIDs, left ventricular systolic diameter; LVIDd, left ventricular diastolic diameter; NLR, neutrophil/lymphocyte ratio; PeAF, persistent atrial fibrillation; VKA, vitamin K antagonist; WBC, white blood cell count.

For the history of medications, there was no difference between the periodontitis subjects and the non-periodontitis subjects, except for aspirin (p<0.001) and vitamin K antagonist (VKA, p<0.001), that were used more often in the non-periodontitis group than in the periodontitis group.

In baseline laboratory findings, white blood cell counts (WBC, p<0.001), neutrophil/lymphocyte ratio (NLR, p=0.001) and high-sensitive C-reactive protein (hsCRP, p=0.048) levels were higher in the periodontitis group. And there was no difference of the echocardiographic findings between the periodontitis subjects and the non-periodontitis subjects, except for left atrial volume index (LAVi), E, A, E′, E/A and E/E′ velocity at baseline (table 1).

All patients were reviewed to evaluate arrhythmic events including AF, atrial tachycardia, atrial premature beat, ventricular tachycardia and ventricular premature beat, and MACE including major bleeding, thromboembolic events including stroke, TIA, systemic emboli and death during follow-up.

The clinical outcomes in patients with AF at 18-month follow-up are shown in table 2. MACE (40 patients (85.1%), p<0.001), total thromboembolic events (14 patients (29.8%), p<0.001) and arrhythmic events (44 patients (93.6%), p<0.001) were higher in patients with periodontitis compared with those without periodontitis during a mean follow-up of 18.1±10.2 months.

Table 2.

Clinical outcomes in patients with atrial fibrillation (AF) according to the periodontitis at 18-month follow-up

Variables	Periodontitis group (n=47)	Non-periodontitis group (n=180)	P values
Follow-up duration (months)	16.9±10.3	19.1±15.0	0.349
MACE (%)	40 (85.1)	44 (24.6)	<0.001
Total death (%)	1 (2.3)	1 (0.6)	0.356
Cardiac death (%)	1 (2.3)	1 (0.6)
Total thromboembolic events (%)	14 (29.8)	1 (0.6)	<0.001
CVA (new onset, %)	12 (25.8)	0 (0)
Peripheral thromboembolism (%)	2 (4.0)	1 (0.6)
Bleeding complications (%)	3 (6.8)	13 (7.3)	1.000
Arrhythmic events (%)	44 (93.6)	31 (17.4)	<0.001
SVT including AF, AFL or ATach (%)	44 (93.6)	30 (16.9)
VPC or VTach (%)	0 (0)	1 (0.5)
Treatment of AF
DC cardioversion for rhythm control (%)	7 (14.5)	18 (10.5)	0.042

Values are mean ±SD (range).

AFL, atrial flutter; ATach, atrial tachycardia; CVA, cerebrovascular accidents; MACE, major adverse cardiac events; SVT, supraventricular tachycardia; VPC, ventricular premature complex; VTach, ventricular tachycardia.

Total oral bacterial culture-positive (%) are 20 patients (42.6%) with periodontitis. ɑ-Streptococci (70%) are the most common oral pathogen for periodontitis. Oral fungal culture–positive (%) (Candida albicans (75%)) are four patients in table 3.

Table 3.

Periodontal pathogens in atrial fibrillation patients with periodontitis

Variables	Periodontitis group (n=47)
Total oral bacterial culture-positive (%)	20 (42.6)
Periodontal pathogens (%)
ɑ-Streptococci (%)	14 (70.0)
Klebsiella pneumoniae (%)	2 (10.0)
Escherichia coli (%)	2 (10.0)
Staphylococcus aureus (%)	1 (5.0)
Staphylococcus salivarius (%)	1 (5.0)
Oral fungal culture-positive (%)	4 (8.5)
Candida albicans (%)	3 (75.0)
Candida tropicalis (%)	1 (25.0)

In univariate analysis, periodontitis, CHA₂DS₂-VASc score, HAS-BLED score, LAVi and E/E′ velocity were significantly associated with MACE including bleeding events, thromboembolic events, arrhythmic events and mortality. In multivariate analysis, periodontitis (p<0.001) was independent risk factors for MACE in table 4A at the long-term follow-up.

Table 4.

Univariate and multivariate Cox analyses for major adverse cardiac events (A) and arrhythmic events (B) in patients with atrial fibrillation according to the periodontitis at 18-month follow-up.

Variable N (%)	Univariate analysis		Multivariate analysis
	OR (95% CI)	P values	OR (95% CI)	P values
(A) Major adverse cardiac events
Periodontitis	17.532 (7.330 to 41.937)	<0.001	17.776 (3.459 to 91.344)	0.001
CHA2DS2-VASc	1.280 (1.010 to 1.622)	0.041
HAS-BLED	2.119 (1.590 to 2.825)	<0.001
LAVi	1.030 (1.007 to 1.054)	0.012
E/E′	1.060 (1.060 to 1.118)	0.029
WBC counts	1.000 (0.998 to 1.000)	0.588
NLR	1.014 (0.980 to 1.051)	0.420
hsCRP	0.065 (0.001 to 7.037)	0.253
(B) Arrhythmic events
Periodontitis	40.688 (11.780 to 140.533)	<0.001	9.185 (1.241 to 67.964)	0.030
CHA2DS2-VASc	1.434 (1.089 to 1.888)	0.010
HAS-BLED	3.608 (2.242 to 5.805)	<0.001
LAVi	1.107 (1.064 to 1.151)	<0.001	1.055 (1.009 to 1.103)	0.019
E′	1.113 (1.041 to 1.191)	0.002
WBC count	1.000 (1.000 to 1.001)	0.011
NLR	1.102 (1.003 to 1.169)	0.006

E, the peak mitral flow velocity of the early rapid filling wave; E′, early diastolic mitral annulus velocity; hsCRP, high-sensitive C-reactive protein; LAVi, left atrial volume index; NLR, neutrophil/lymphocyte ratio; WBC, white blood cell count.

In univariate analysis, periodontitis, CHA₂DS₂-VASc score, HAS-BLED score, LAVi, E′ velocity, WBC counts and NLR were significantly associated with arrhythmic events including AF, atrial flutter, atrial tachycardia, ventricular premature beat, ventricular tachycardia and sick sinus syndrome. In multivariate analysis, periodontitis (p=0.030) and LAVi (p=0.019) were independent risk factors for arrhythmic events in table 4B at the long-term follow-up.

Flow diagram of our study is shown in figure 1. Kaplan-Meier curves show that the event-free survival from MACE including bleeding events, thromboembolic events, arrhythmic events and mortality (p<0.001; figure 2A) event-free survival from arrhythmic events (p<0.001; figure 2B) are lower in patients with periodontitis compared with those without periodontitis at 18-month follow-up.

Figure 1.

Flow diagram. AF, atrial fibrillation.

Figure 2.

Kaplan-Meier analysis for event-free survival in patients with atrial fibrillation according to the presence of periodontitis: (A) major adverse cardiac events including bleeding events, thromboembolic events, arrhythmic events and mortality; (B) arrhythmic events.

Discussion

In this study, we showed that the periodontitis in patients with AF is associated with higher incidence of arrhythmic events and MACE including major bleeding, thromboembolic events including stroke, TIA, systemic emboli and death in the long-term follow-up, suggesting more intensive medical therapy to control periodontitis as representative of chronic inflammation with close clinical follow-up will be required.

The link between periodontal disease and CVD is complex and many publications have provided support to the hypothesis of a causal association.¹² Periodontitis induces a peripheral inflammatory and immune response, reflected in elevated CRP and IgA class antibodies to periodontal pathogens. The prevalence of CVD seems to be the highest in those individuals in whom periodontitis coexists with elevated CRP levels. And elevated inflammation cytokines contribute to the pathogenesis of AF. The CRP levels were associated with the burden of AF. This may be due to genetic reasons and may also apply to chronic low-grade infections.⁷ And several studies have shown that there are many studies that have indicated that inflammation might play a significant role in the initiation, maintenance and perpetuation of AF.¹³ In our study, there were higher WBC, NLR and hsCRP levels in the periodontitis group compared with those in the non-periodontitis group, which is consistent with a previous study.¹⁴

Chronic inflammation and autonomic imbalance have been proposed as plausible pathophysiological mechanisms of AF.¹⁵ And structural and electrical alternations have been reported in the development and perpetuation of AF and local inflammation resulting from periodontitis, including inflammatory infiltration cardiomyocytes hypertrophy and disturbed atrial structure.⁷

Periodontal bacteria and bacterial components have easy access to the periodontal connective tissue and enter the blood circulation frequently. Previous study reported that bacteria such as Porphyromonas gingivalis and Prevotella intermedia stimulate a humoural response and antibodies generated in response to the bacterium may indeed contribute to tissue damage.¹⁶ Once the damage is confirmed in the myocardium, it may be responsible for the substrate of arrhythmia. In our study, total oral bacterial culture-positive (%) are 20 patients (42.6%) with periodontitis. ɑ-Streptococci (70%) are the most common oral pathogen for periodontitis. Oral fungal culture–positive (%) (C. albicans (75%)) are four patients. And there was no difference of clinical outcomes among the periodontal pathogens in patients with periodontitis in our study. Unfortunately, there are no data currently on ways to risk-stratify these patients according to periodontal pathogens in this study because of small number of patients.

In our study, we hypothesised that periodontitis may also be a risk factor for AF on the basis of the following facts. First of all, periodontitis may impair endothelial function and periodontitis can be associated with oxidative stress both locally and peripherally. Endothelial dysfunction and oxidative stress are both involved in the pathophysiology of AF. CRP downregulates endothelial nitric oxide (NO) synthase transcription in endothelial cells and destabilises endothelial NO synthase messenger RNA, resulting in decreased NO release.¹⁷ Previous study reported that many proofs have implied that AF may be an inflammatory disease. Inflammation caused by sterile pericarditis increased the CRP, decreased atrial effective refractory periods, delayed atrial conduction time and AF duration.¹⁸

Based on the predictors of bleeding, the HAS-BLED score was developed using a ‘real-world’ cohort of 3978 European subjects with AF from the EuroHeart Survey.¹⁹ In our study, there was also trend of higher HAS-BLED score in the periodontitis group compared with those in the non-periodontitis group. However, there was no difference of the incidences of bleeding events (p=0.239) including minor bleeding—gingival bleeding according to periodontitis at 18-month clinical follow-up in Kaplan-Meier analysis. And there was lower incidence of VKA medication in the periodontitis group (p<0.001), which reflect that the dentists did not want to use the anticoagulation medication during the treatment of periodontitis.

It is of interest that although there was no difference of the incidences of hospitalisation rate (p=0.721), total death (p=0.356) and bleeding complication (p=1.000) in both groups, the incidences of MACE (p<0.001), total thromboembolic events (p<0.001), arrhythmic events (p<0.001) and DC cardioversion for rhythm control (p=0.042) were higher in the periodontitis group compared with the non-periodontitis group.

In our study, various factors were associated with arrhythmic events in patients with AF, including periodontitis, CHA₂DS₂-VASc score, HAS-BLED score, LAVi and E′ velocity on TTE. In multivariate analysis, periodontitis and LAVi were independent predictors for arrhythmic events in patients with AF at 18-month clinical follow-up.

Previous reports that demonstrated that PeAF became more likely, which led to the adoption of the now oft-quoted adage ‘AF begets AF’.²⁰ In our study, there were higher incidences of PeAF (p=0.043) in patients with non-periodontitis. However, at 18-month follow-up, arrhythmic events (p<0.001) including AF, atrial tachycardia and atrial flutter were also higher in the periodontitis group, which is consistent with a previous study²¹ that both oxidants and inflammation play an important role in the initiation and maintenance of AF.

This is the first study to evaluate the impact of periodontitis on long-term prognosis in patients with AF and validate a risk stratification to assess the probability of arrhythmic events. In our study, the periodontitis in patients with AF is associated with higher incidence of MACE and arrhythmic events in the long-term follow-up, suggesting that, although the mechanism remains unclear, the control of oral disease may improve inflammation and may be used as a possible therapeutic target to prevent embolic events and recurrence of arrhythmia in patients with AF.

Further prospective studies are needed to determine if a true causal mechanism exists between periodontitis and AF, as well as to access whether the mechanisms is dependent on a specific subtype of AF.

Study limitations

There are some limitations to our study. First, this was a single-centre, retrospective study derived from a real-world practice with inherent limitations. Small numbers of patients with periodontitis were included in this study. Because a half of total patients included at baseline had no medical record or history of dental procedure, therefore, we excluded those patients. Hence, the results of our study should be considered as hypothesis generating, and future prospective studies are warranted to confirm our results. Second, asymptomatic episodes of AF may not have been recognised because AF recurrence was based on clinical symptoms and ambulatory monitoring for a short period. Third, the investigators may not screen AF patients with asymptomatic periodontitis or treated for periodontitis in other clinics. Fourth, patients with periodontitis could not be treated continuously. Therefore, there was limitation to generate the direct correlation of periodontitis treatment with clinical outcomes. However, chronic inflammation and autonomic imbalance have been proposed as plausible pathophysiological mechanisms of AF. And this study has presented a new vision for the periodontitis focusing on the arrhythmogenic effects beyond local chronic inflammation.

Conclusion

The periodontitis as representative of chronic inflammation was an independent predictor of arrhythmic events and MACE in patients with AF in the long-term follow-up, suggesting that the control of oral disease may improve inflammation and may be used as a possible therapeutic target to prevent embolic events and recurrence of arrhythmia in patients with AF.