Table 3.

Characteristic of the studies included in the systematic review.

Author	Chen et al. [11]	Chen et al. [10]	Im et al. [14]	Chang et al. [9]	Sen et al. [15]
Year	2013	2016	2018	2020	2021
Country	Taiwan	Taiwan	Korea	Korea	United States of America
Study design	A nationwide, population-based, retrospective cohort (NHIRD)	A nationwide, population-based, retrospective cohort (NHIRD)	Case-control	A nationwide, population-based, retrospective cohort (NHIS-HEALS)	A large prospective cohort study (ARIC)
Follow-up duration	4.6 ± 1 years	4,075,682/3,405,292 person-years of follow-up	18 months	10.5 years	17 years
Subjects	16,955 adults who were age	787,490 subjects without previous history of AF/AFL	227 patients with non-valvular AF	161,286 adults who were age	5,958 subjects without previous history of AF
	60 or more without past history of cardiac arrhythmias	Exp: 393,745	Exp: 47	40-79 without past history of AF	M: 45.5%
	Exp: 3,391	Control: 393,745	Control: 180	M: 61%	Age: 59.5 ± 5.6 years
	Control: 13,564	M: 49%	M: 68%	Age: 52 ± 9 years	DM: 6.7%
	M: 55%	Age: 42 ± 17 years	Age: 60 ± 11 years	DM: 9%	HT: 34.5$
	Age: 68 ± 6	DM: 5%	DM: 17%	HT: 39%	HF: 0.6%
	DM: 22%	HT: 9%	HT: 56%	CKD: 8%	CAD: 3.3%
	HT: 44%	HF: 0.5%	HF: 11%
	HF: 5%	CAD: 3%	Stroke/TIA: 3%
	CAD: 21%	CKD: 1%
	Stroke/TIA: 13%
	CKD: 8%
Intervention/Exposure (Oral health indicators)
1) Periodontal disease	N/A	Presence of periodontitis Control: No periodontitis (1:1 age and sex matched) Note: Diagnosis of periodontitis was defined if one of the following criteria was present; 1) At least 1 dental visit recorded with periodontitis and received antibiotic therapy or periodontal treatment 2) Having more than 2 dental visits with periodontitis diagnosis in one year	Presence of periodontitis Control: No periodontitis Note: Periodontitis was evaluated using WHO CPI score and diagnosis of periodontitis was confirmed when CPI score ≥3.	Presence of periodontitis Control: No periodontitis Note: Diagnosis of periodontitis was defined if one of the ICD-10 codes was present; 1) Acute periodontitis, 2) Chronic periodontitis, 3) Periodontosis, 4) Other periodontal disease, and 5) Unspecified periodontal disease.	Presence of periodontitis (mild, moderate, severe) Control: Periodontal health Note: Periodontitis severity was diagnosed using the periodontal profile class (PPC)
2) Tooth loss				Number of missing teeth: 1–7, 8–14, 15–21, ≥22 Control: no missing teeth
3) Dental scaling/professional dental cleaning	Frequency of dental scaling at least 1 once a year for 3 consecutive years Control: No dental scaling (1:4 matched age, sex, and significant underlying disease, including HT, DM, CHF, CAD, CKD, ischemic stroke/TIA, and asthma/COPD)	Covariate of periodontitis: Frequency of dental scaling (none, 1–2, >2 times per year) (patients with periodontitis received more frequent dental scaling than the others without periodontitis)		Professional dental cleaning (Yes) Control: no professional dental cleaning
4) Toothbrushing				Frequency of tooth brushing: 2, ≥3 (times/day) Control: 0–1 times/day
5) Dental service utilization				Dental visit for any reasons in the last year Control: no dental visit in the last year	Dental care utilization: regular users (those who sought routine dental care) Control: episodic users (those who sought dental care only when in discomfort, something needed to be fixed, never, or did not receive regular dental care).
Methods for oral health assessment	Review patients' record	- Oral examination by trained and calibrated dentists to assess periodontitis - Review patients' record to assess dental scaling frequency		- Oral examination by dentist to assess periodontitis (diagnosed when >2 times of claims), and number of missing teeth - Self-reported questionnaire to assess dental symptoms, dental visit for any reasons, oral hygiene behavior	- Oral examination by a single examiner to assess periodontitis - Self-reported questionnaire to assess dental care utilization
Outcome	Primary: Occurrence of new-onset AF	Primary: Time from the inclusion to the first AF/AFL diagnosis during outpatient or inpatient visit	Primary: MACE Secondary: arrhythmic events, including AT, AF, VT, PAC, PVC.	Primary: Occurrence of new-onset AF and heart failure (HF)	Primary: Occurrence of new-onset AF
AF diagnosis	To assure AF diagnostic accuracy, the occurrence of AF was only defined in subjects with AF diagnosis at discharge or repeatedly confirmed more than twice in outpatient department, by documented ICD9-CM code: 427.31	Incident AF/AFL was defined in patients with at least one outpatient or inpatient diagnosis of AF/AFL, by documented ICD9-CM code: 427.31–2	Paroxysmal AF was defined when previous ECG showed sinus rhythm. Persistent AF was defined when perpetuating 7 days or more. Chronic AF was defined as an ongoing long-term period. During follow-up, diagnosis of AF recurrence or onset of AF progression was based on the first time that all ≥3 consecutive ECGs at interval of ≥1 week indicated AF, or based on the clinical judgement of the physicians if ECGs were not obtained thrice during the defined period.	To assure AF and HF diagnostic accuracy, the occurrence of AF and HF were defined in subjects with documented diagnostic ICD10 code: I48 at least 2 claims per year.	Incident AF was defined by any of the followings; 1) Documented standard 12-lead ECG during follow-up (tracings with automatically-reported AF were reviewed by a cardiologist) 2) Hospital discharge diagnostic codes (defined by ICD9-CM codes 427.31 or 427.32) 3) Death certificates (defined by ICD9-CM codes 427.31 or 427.32)
Potential covariate/confounder adjustment		- Age, Sex - Comorbidities at baseline (heart failure, hypertension, diabetes mellitus, vascular disease, hyperlipidemia ischemic heart disease, valvular heart disease, chronic obstructive pulmonary disease, sleep apnea, renal disease, hypothyroidism, hyperthyroidism) - Average number of outpatient visits		- Age, sex, socioeconomic status - Regular exercise - Alcohol consumption - Anthropometric measurements (body mass index, systolic- and diastolic blood pressure) - Comorbidities (Hypertension, diabetes, dyslipidemia, current smoking, renal disease, cancer history) - Laboratory findings (total cholesterol, fasting blood sugar, aspartate aminotransferase, gamma-glutamyl transferase, proteinuria)	- Age, sex, education level, race - Comorbidities (hypertension, diabetes, LDL, obesity, smoking, alcohol use, CAD, CHF)
Results	Significantly lower occurrence of new-onset AF in subjects who received dental scaling more than 1 time per year (HR = 0.340, 95% CI = 0.248–0.489; p < 0.001). Dental scaling was independently associated with a reduced risk of new-onset AF (HR = 0.671, 95% CI = 0.524–0.859; p = 0.002).	Significantly higher occurrence of new-onset AF in subjects with periodontitis (HR = 1.31, 95% CI = 1.25–1.36).	Significantly higher MACE and arrhythmic events were found in subjects with periodontitis (adjusted OR = 17.8, 95% CI = 3.46–91.3; p < 0.001). Periodontitis was an independent risk factor for MACE and arrhythmic events (p < 0.001).	Significantly lower occurrence of new-onset AF (HR = 0.90, 95% CI = 0.83–0.98) and HF (HR = 0.88, 95% CI = 0.82–0.94) in subjects with tooth brushings at least 3 times/day.	Significantly higher occurrence of new-onset AF in subjects with severe periodontitis (adjusted HR = 1.31, 95% CI = 1.06–1.62). Significantly lower occurrence of new-onset AF in regular dental care utilization users (adjusted HR = 0.88, 95% CI = 0.78–0.99).
Main findings and comments	Oral health promotion by dental scaling was associated with lower occurrence of new-onset AF.	Patients with periodontitis had an increased risk of new-onset AF/AFL. Dental scaling at least 2 times/year in this study was associated with increased AF/AFL risk because it indicated presence of periodontitis.	Periodontitis in patients with known valvular AF was associated with higher incidence of arrhythmic events and MACE	Lower risk of incident AF was associated with higher frequency of tooth brushings (≥3 times/day) Lower risk of incident HF was associated with higher frequency of tooth brushings (≥2 times/day), professional dental cleaning, and higher number of missing teeth (≥22 teeth).	Patients with severe periodontitis and episodic dental care users had higher risk of incident AF than those with periodontal health and mild/moderate periodontitis, and regular denture care users. The association between severe periodontitis and ischemic stroke mediated by AF.

AF, atrial fibrillation; AFL, atrial flutter; ARIC, Atherosclerosis Risk in Communities; AT, atrial tachycardia; CAD, coronary artery disease; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CPI, community periodontal index; DM, diabetes mellitus; ECG, electrocardiogram; Exp, exposed; HF, heart failure; HT, hypertension; HR, hazard ratio; ICD, International Classification of Diseases, M, male; MACE, major adverse cardiovascular events; NHIRD, National Health Insurance Research Database; NHIS-HEALS: National Health Insurance System-National Health Screening Cohort; PAC, premature atrial complex; PVC, premature ventricular complex; TIA, transient ischemic attack; VT, ventricular tachycardia; WHO, World Health Organization.