The YPCDC is a longitudinal community-based cohort established within a multi-rural community project, which is part of the Korean Genetic Epidemiology Study funded by the Korean Centers for Disease Control and Prevention. It is a prospective population-based investigation of risk factors for AF in men and women ≥ 21 yr old. We enrolled 4,070 adult residents in Yangpyeong county (Fig. 1) who participated in this cohort from 2004 to 2008. Total populations of Yangpyeong county were 75,314 in 2005 and 82,802 in 2010 from Population and Housing Census data. We included all residents who were relatively asymptomatic healthy participants and underwent both electrocardiographic examination and serum blood test. Otherwise, we excluded 17 out of the 4,070 residents because of the absence of electrocardiography (ECG) data. Therefore, a total of 4,053 residents (mean age: 60.2±11.2 yr old; 1,582 men) participated in the present investigation. The majority of the subjects were farmers and housewives. A questionnaire about health status including past medical history and family history was filled out by the participants. Information on demographics, smoking, alcohol intake, medical history, and medications were collected using the questionnaire administered by well-trained interviewers. Among them, 943 participants answered questions about their medical history: hypertension, diabetes mellitus, hypercholesterolemia, ischemic heart disease, and stroke, as well as smoking and drinking status. All the participants underwent physical examinations, including blood pressure measurements, ECG, and blood tests after fasting for at least 8 hr (total cholesterol, serum fasting glucose, high density lipoprotein, triglycerides, serum uric acid, liver enzyme, blood urea nitrogen, serum creatinine, C-reactive protein, serum insulin, and adiponectin) which were analyzed by an ADVIA1650 Automatic Analyzer (Siemens, New York, NY, USA). Blood pressure was measured to the nearest 2 mmHg with a standard mercury sphygmo-manometer (Baumanometer) using the first and fifth Korotkoff sounds. Standardized measurements were performed with the participants seated after a five minutes rest, and were repeated three times at five minute intervals. For the analysis the mean of the second and third of the three BP measurements was calculated; however, if the difference was larger than 5 mmHg, up to five additional measurements were made, and the last two values were averaged. Body mass index was calculated as weight in kilograms divided by height in meters squared. Cardiovascular risk factors were defined as (7) 1) hypertension; use of antihypertensive medications, systolic blood pressure≥140 mmHg, or diastolic blood pressure≥90 mmHg; 2) diabetes mellitus; use of oral hypoglycemic agents or treatment by insulin injection at the current examination, fasting blood glucose≥126 mg/dL; 3) hypercholesterolemia; use of antihyperlipidemic agents or serum total cholesterol >220 mg/dL. Present illnesses were identified through medical interviews by physicians or survey interviewers. Clinical cardiovascular disease at baseline was defined by any of the following: a history of myocardial infarction, angina, stroke, and congestive heart failure.

A portable ECG machine was used to make 12 lead resting surface ECGs to detect AF, and all ECG records were reviewed by a physician.

AF was defined as disorganized atrial activity without discrete P waves, atrial activation ranging from 350 to 600 beats per minute, and irregularly irregular ventricular response.

The echocardiographic examination consisted of a standard 2-dimensional echocardiogram, including M-mode and Doppler echocardiography Sonos 2500 (Hewlett-Packard Co) during continuous ECG monitoring according to the American Society of Echocardiography guidelines; it was performed by a well-trained cardiologist and all records were reviewed and measured by the same physician. The exclusion criteria for echocardiography were poor acoustic window, dextrocardia, history of open heart surgery, and oblique M-mode interrogation angle >10° on the vertical axis of the left ventricle. Details have been given previously (8). The investigators who performed the echocardiographic measurements and the reviewer were independent observers blinded to participants' characteristics. This study was approved by the research ethics committee of Hanyang University Hospital, and written informed consent was obtained from each participant before the examination.

Continuous variables are presented as mean and standard deviation, categorical variables as observed number of participants and percentages. We used an independent t-test after performing Levene's test for equality of variances to compare all continuous variables. Categorical variables were tested with the chi-square and Fisher's exact tests. All parameters with a P value<0.1 resulting from the univariate comparisons were included in a multiple logistic regression model. Another logistic model was used to explore the association between AF and stroke, where the dependent variable was stroke (yes/no) and the independent variables based on previous studies (9) included age, diabetes, hypertension, dyslipidemia, smoking, and drinking. The criterion for inclusion of a variable in the regression model was P<0.05, and that for exclusion was P>0.10. Statistical analysis was performed with SPSS statistical software (SPSS Inc., release 15.0) and statistical significance was assumed at P<0.05. All tests were two-sided.

The institutional review board of Hanyang University College of Medicine reviewed and approved the protocol of this study (IRB approval number; 2010-R-31). Written informed consent was obtained from each participant before the examination.

Among the 4,053 survey participants, the crude overall prevalence of AF was 1.3% (n=54). It was higher in men (2.0%) than in women (0.9%), and there was a nonlinear increase with age in both men and women. The crude prevalence was highest between ages 61 and 70 in both genders (3.6% in men, 1.4% in women), and higher in those aged 41 to 70 than below 41 in both genders. The crude prevalence of AF in men was higher than that in women except for the 41-50 age group and those over 81. The crude overall AF prevalence was 0.6% in the 51 to 60 group, 2.3% in the 61 to 70 group, 1.9% in the 71 to 80 group, and 1.3% in those over 81 (Fig. 2). When age was adjusted on the basis of Korean age composition data (2005 National Population Census), the combined age-standardized prevalence of AF in Korea (≥40 yr) was 0.36% (0.45% for men and 0.32% for women).

Baseline characteristics and the results of the univariate analysis are presented in Tables 1 and 2. Of the demographic and blood test factors, age, male gender, body mass index, total serum cholesterol, serum uric acid, alanine transaminase, serum creatinine, and adiponectin level were related to AF (Table 1). Among the classical risk factors including smoking and alcohol status, congestive heart failure and ischemic heart disease were also related to AF (Table 2) as were the echocardiographic parameters, systolic and diastolic left ventricular internal dimensions, left atrial (LA) diameter, peak early rapid filling wave velocity (E), and LV ejection fraction (Table 3).

In a multivariate analysis, only ischemic heart disease (P= 0.002), LA diameter (P=0.006), and peak E (P=0.018) were significant risk factors for AF (Table 4).

Interestingly, in a multivariate analysis for stroke risk factors, neither age, gender, AF, diabetes, hypertension, hyperlipidemia, smoking, or drinking status were related to stroke events.