Epidemiologic Features of the Korean Atrial Fibrillation Population in a Single Center

International Journal of Arrhythmia 2014;15(2): 17-24.

ORIGINAL ARTICLES

Epidemiologic Features of the
Korean Atrial Fibrillation
Population in a Single Center

Introduction

   Atrial fibrillation (AF) is the most common sustained cardiac rhythm disorder. An estimated 2.3million people in the United States alone are estimated to have AF; this number is projected to increase to 5.6 million by 2050.¹ AF is associated with significant morbidity and mortality rates in adults because of the corresponding stroke, heart failure, and cognitive dysfunction risks. As a result, the disease has vast socioeconomic implications.^2,3
   Some investigators have reported on distinct features in the Korean AF population.^4,5 However, these data are limited. While we know that the epidemiology of AF differs among races and is constantly changing,^6,7 the majority of our knowledge of the disease depends on data that relates only to Western populations, is out of date or both.^8,9 Much of the scant existing data on the epidemiologic features of AF in Asians is small-scale data from Japanese studies.^10,11 To gain better insight into the Korean AF population, we sought to evaluate, in a single-center study, both annual trends in AF diagnosis and management and related changes in stroke risk factor components.

Methods

   Our study protocol was approved, in accordance with the Declaration of Helsinki, by the Institutional Review Board of Seoul National University Hospital. Seoul National University Hospital is a teaching hospital with more than 1,700 beds; it provides highquality treatment to patients nationwide and is considered the representative medical center of Korea.
   A total of 15,593 patients in a single center who were diagnosed with AF (as documented by electrocardiography) between 2000 and 2012 were enrolled in our study. Patients under 18 years of age were excluded before enrollment. Medical records from the time of initial diagnosis were retrospectively reviewed, in consecutive order, and patients were classified according to their diagnosis of either valvular AF or non-valvular AF. Valvular AF was defined as any rheumatic mitral valve disease, regardless of severity or history of surgical repair or replacement of the mitral valve. Other vascular disease was defined as prior myocardial infarction, peripheral artery disease, or aortic disease.⁸ Patients were then grouped into 3 categories by their year of diagnosis: 2000-2003 was considered period 1, 2004-2007 was period 2, and 2008-2012 was period 3. Changes in risk factor composition and initial anticoagulation strategies were compared among patients in these categories.
   Continuous variables were presented as mean ± standard deviation. A 95% confidence interval (CI) was reported with sensitivity, specificity, and predictive values. The data were compared using Student t-test, while categorical variables were compared by Fisher’s exact test and the χ2 test. p-value ＜0.05 was considered statistically significant. Statistical analyses were performed using the SPSS Statistics 21.0 software package (IBM SPSS, New York, USA).

Results

Annual trends in valvular and non-valvular AF ratio

   Of the 15,593 total patients in the study, 13,998 (89.8%) were considered non-valvular AF patients and 1,595 (10.2%) were valvular AF patients. We observed that the percentage of newly diagnosed valvular AF decreased year-over-year, from 16.1% in period 1 to 9.9% and 6.0% in periods 2 and 3, respectively (Figure 1). In turn, the age of these patients at diagnosis increased year-over-year, as described in Table 1. Specifically, AF diagnosis in patients over 75 gradually increased from 14.0% in period 1 to 17.8% in period 2 and 24.4% in period 3. AF diagnosis in patients over 65 also increased, from 44.6% in period 1 to 53.9% in period 2 and 61.0% in period 3. This was the main cause of increased age at diagnosis, and was a consistent finding in separate analyses comparing patients by sex.

Stroke risk factor components in non-valvular AF patients

The average CHADS₂ and CHA₂DS₂-VASc system scores for our patients, as well as the percentages of their components each year, are demonstrated in Figure 2. The mean score significantly changed over time: the overall CHADS₂ score average was 0.89 and the overall CHA₂DS₂-VASc score average was 1.88. Among stroke risk factors, old age (over age 65) has increased in proportion, while congestive heart failure and diabetes (DM) have both decreased.
The CHADS₂ and CHA₂DS₂-VASc score distributions in our study were as follows: CHADS₂ score 0 (42.5%), 1 (35.0%), 2 (15.6%) and ≥3 (6.9%)/CHA₂DS₂-VASC score 0 (15.7%), 1 (28.6%), 2 (26.1%), 3 (17.3%), 4 (7.8%) and ≥5 (4.6%). The most common components contributing to CHADS₂ and CHA₂DS₂-VASC scores of 1 were hypertension (HTN) and old age, respectively. Likewise, a CHADS₂ score of 1 was influenced by HTN (40.0%), age ≥75 years (29.1%), DM (17.7%), and congestive heart failure (CHF, 13.2%). However, the component ratio has changed with time: patient age of over 75 years comprised 23.9% in period 1, but increased to 26.0% in period 2 and 34.8% in period 3. On the contrary, DM comprised 21.4% in period 1, but decreased to 17.8% in period 2 and 15.1% in period 3. HTN comprised 39.7% in period 1, slightly increased to 43% in period 2, and then decreased to 37.9% in period 3. A CHA₂DS₂-VASC score of 1 was influenced by patient age between 65-74 years (39.2%), female sex (29.2%), HTN (17.1%), DM (7.9%), CHF (5.4%), and vascular disease (1.2%). Among all contributing components, patient age of 65-74 years increased from 32.3% in period 1 to 38.1% in period 2 and 34.2% in period 3.

Annual prescription trends for non-valvular AF management

   The most frequent anticoagulation approach chosen for CHADS₂ score 1 patients was a single antiplatelet agent (44.2%), followed by no medication (28.4%), an anticoagulation agent (26.7%), and a combination of both antiplatelet and anticoagulation agents (1.6%). Similarly, the most frequent approach chosen for CHA₂DS₂-VASC score 1 patients was a single antiplatelet agent (44.3%) followed by no medication (28.4%), an anticoagulation agent (5.7%), and a combination of both antiplatelet and anticoagulation agents (1.6%). Anticoagulation was the second most prescribed approach in period 1, but only the third most prescribed in periods 2 and 3.
   We found that the initial management strategy chosen for patients at diagnosis also changed over the years. For rate control purposes, a total of 46.5% patients were prescribed beta-blockers (BBs) and 21.4% were prescribed calcium-channel blockers (CCBs). The incidence of BB prescription decreased from 52.6% in period 1 to 46.0% in period 2 and 42.8% in period 3. CCB prescription incidence also decreased, from 24.8% in period 1 to 22.2% in period 2 and 18.5% in period 3. For rhythm control purposes, prescription of either amiodarone or dronedarone slightly decreased over time, from 24.6% in period 1 to 24.7% in period 2 and 19.5% in period 3. Class Ic drug prescription has remained consistent at approximately 12% (12.6%, 11.6% and 11.5% in periods 1, 2 and 3, respectively).

Epidemiologic features of and treatment strategy for valvular AF patients

   There were a total of 1,595 valvular AF patients in our study population, with a mean age of 55.2 ±11.0 years-significantly less than that of the non-valvular AF patients in the study (65.2 ± 12.2 years, p＜0.001). The incidence of prior medical diagnoses (HTN, DM, vascular disease, previous stroke history or chronic kidney disease) was significantly lower in valvular AF patients (all, p＜0.001) than in non-valvular AF patients, with the exception of CHF (p=0.835). This was a consistent finding in separate analyses comparing study patients by sex and by year of diagnosis, respectively.
   While all of the valvular AF patients in our study who had undergone mitral valve surgery were receiving anticoagulation therapy, only 26.4% of all valvular AF patients were on an anticoagulation regimen. Approximately half (49.3%) of valvular AF patients were taking antiplatelet agents alone, and nearly one-quarter (22.4%) were not taking any medications stroke prevention purposes.

Discussion

   Although the data gleaned from our single-center, retrospective study may not represent the entirety of the Korean AF population, it is an unprecedentedly large data set. In addition, a complete enumeration survey was undertaken as part of our study so that the chronological changes and trends in this population could be documented. Although a broadened definition of valvular AF was used for the purposes of this study, the trend of decreasing valvular AF incidence we demonstrated is distinct. In addition, the trend of increasing age at diagnosis we reported aligns with data gathered in Western studies.¹² The increase in general life expectancy, and/or increased compliance with routine check-up appointments, may provide an explanation.
   As the age at AF diagnosis has increased, the composition of CHADS₂ and CHA₂DS₂-VASc system scores has also changed. We believe that age should be regarded as an important component of bleeding risk stratification schemes, such as HAS-BLED.¹³ Previous studies have reported that older age may increase the risk of anticoagulation-associated hemorrhage in AF patients.^14,15 Therefore, elderly patients (age over 65 or over 75 years) with absolute indications for anticoagulation should be strictly monitored throughout clinical follow-up.
   The use of therapy for rate control purposes has decreased over time, while therapy for rhythm control has remained consistent. Use of AF ablation has also markedly increased in recent years: in our study, we observed that most AF ablation cases (over 95%) were performed after 2008, and the incidence has doubled every year.
   The choice of medication prescribed for stroke prevention purposes has also changed over time; there is a clear trend toward lesser use of anticoagulation therapy for score 1 patients. Although the finding carries weak significance for CHADS₂ score 1 patients, physicians have tended to prescribe anticoagulation therapy more often for younger (age ＜ 75 years) patients than for the elderly (28.1% versus 25.6%, p=0.084). The aging of the patient population could continue to affect trends in stroke prevention prescriptions.
   Although the mean age of valvular AF patients is younger than that of non-valvular AF patients, and is also associated with fewer comorbidities, general guidelines recommend strict use of anticoagulation therapy for valvular AF patients. However, in our study, only 26.4% of valvular AF patients were receiving anticoagulation, while all patients who had undergone valve surgery were taking warfarin. This dichotomy may be explained by differing definitions of valvular AF among physicians. In addition, our study used a broadened definition of valvular AF, which included patients with any incidence of rheumatic mitral valve disease (stenosis or regurgitation), regardless of disease severity. Although definitions of valvular AF differ between investigators, studies, and official guidelines, few studies have reported on the epidemiology or prognostic implications of these definitions. A likely explanation is the low incidence of valvular AF in developed countries. Further studies are needed to determine absolute indications for anticoagulation therapy in patients with valvular AF.
   Understanding the epidemiology of valvular AF, as well as changes in the Korean AF population over time, is essential for mapping out future clinical management strategies. According to our data, most AF patients have been undertreated in terms of anticoagulation therapy. Physicians should be more aware of patients’ age to determine the optimal stroke prevention strategy.

References

Korean Heart Rhythm Society A-1604, Centreville Asterium Seoul, 372, Hangang-daero, Yongsan-gu, Seoul, Korea TEL: +82-2-318-5416 FAX: +82-2-318-5417 E-mail: khrs4@k-hrs.org, khrs6@k-hrs.org
Copyright © 2024 The Official Journal of Korean Heart Rhythm Society Editorial Board & MMK Co., Ltd.