Article Text

Download PDFPDF

Atrial fibrillation in mainland China: epidemiology and current management
  1. S Zhang
  1. Arrhythmia Center, Cardiovascular Institute and Fuwai Hospital, Chinese Academy of Medical Sciences, Beijing, China
  1. Dr S Zhang, Arrhythmia Center, Cardiovascular Institute and Fuwai Hospital, Chinese Academy of Medical Sciences. 167 Beilishi Rd, Beijing, China 100037; shuzhang{at}


With the economic development of China, research in atrial fibrillation has improved progressively, but due to a lack of international communication, many significant achievements have not been widely recognised. A brief overview of epidemiology, current therapy and research on atrial fibrillation in mainland China is provided. Chinese electrophysiologists would like to contribute more to international research in cardiac electrophysiology.

Statistics from

With the economic development of China, research in atrial fibrillation (AF) has improved progressively. However, due to a lack of international communication, many significant achievements have not been widely recognised.13 Following is a brief overview of epidemiology, current therapy and research on AF in mainland China.


China is the most populated country in the world and covers a vast area. Although it is difficult to carry out a well-designed survey of AF in such a large population spread across such a wide expanse, we have recently completed a cross-sectional survey of AF conducted in 10 various populations, located in different areas of China, including four groups of urban residents, five groups of rural farmland residents and one fishing community.4 These populations were selected to be representative of the geographic and economic diversity in China. A total of 19 368 participants (8636 men, 10 732 women) aged ⩾35 years were included. The age-adjusted prevalence of AF in Chinese adults aged ⩾35 years was 0.74% in men and 0.72% in women. The prevalence was higher among older adults of both sexes. The prevalence in subjects <60 years old was 0.43% in men and 0.44% in women, but the prevalence in subjects aged ⩾60 years was 1.83% in men and 1.92% in women. Out of all the AF patients, 30.9% were first diagnosed,33.0% as paroxysmal AF, 7.2% as persistent AF and 28.9% as permanent AF. In men, 81.3% were diagnosed as having non-valvular disease, and 18.8% were diagnosed as having valvular disease. In women, 68.8% presented with non-valvular disease and 31.3% with valvular disease. A multiple logistic regression analysis of variables including age (⩾60 years), myocardial infarction, left ventricular hypertrophy (demonstrated by the ECG), hypertension, diabetes mellitus, obesity, hyperlipidaemia (total cholesterol ⩾200 mg/dl), smoking and drinking showed that age (⩾60 years), history of myocardial infarction and electrocardiographic left ventricular hypertrophy were significantly associated with a higher risk of AF, and high cholesterol with a lower risk of AF in both men and women. Obesity and smoking were significantly related to the risk of AF only in women, and alcohol consumption only in men. AF was fairly common in older Chinese people. About a third of the cases were undiagnosed, and about three-quarters of the cases were non-valvular in origin.

Another epidemiological survey5 of AF in China was performed with 29 079 patients aged 30 to 85 years old. The crude rate of the prevalence of AF in this study was 0.77%, and 0.61% after being standardised. The prevalence was higher in men than in women (0.9% vs 0.7%, p = 0.013). Among all the AF cases, valvular, non-valvular, and lone AF was 12.9%, 65.2% and 21.9%, respectively.

A retrospective analysis6 of hospital records was conducted in 9297 patients (mean age 65.5 years, range 18 to 99 years) with a primary diagnosis of AF, discharged between January 1999 and December 2001 from 41 hospitals across China. During the study period, AF admissions increased as a proportion of the cardiovascular admissions, from 7.65% (in 1999) to 7.90% (in 2000) and then to 8.16% (in 2001). The distribution of AF increased with age. The main factors that might cause AF in those patients were age (58.1%), hypertension (40.3%), coronary heart disease (34.8%), heart failure (33.1%), rheumatic valvular disease (23.9%), idiopathic AF (7.4%), cardiomyopathy (5.4%) and diabetes (4.1%). Permanent AF accounted for almost half of the patients (49.5%), and paroxysmal and persistent AF accounted for 33.7% and 16.7%, respectively.


Drug therapy is still the primary treatment for AF in China. However, many patients in rural areas still do not receive any medicine because they do not seek medical attention. A retrospective analysis6 of the hospital records in cities during 2000 to 2002 showed that paroxysmal AF was treated mainly by rhythm control (56.4%). Rate control was also the main strategy in 82.8% of the patients with chronic AF.

Angiotensin-converting enzyme inhibitors have been used in AF recently. A prospective randomised study7 was performed to compare amiodarone vs amiodarone plus losartan and amiodarone plus perindopril for the prevention of AF recurrence in patients with lone paroxysmal AF. One-hundred and seventy-seven patients with lone paroxysmal AF were randomly assigned to three treatment groups: low-dose amiodarone alone, low-dose amiodarone plus losartan and low-dose amiodarone plus perindopril. After 24 months of follow-up, the combination of perindopril or losartan with low-dose amiodarone was more effective than low-dose amiodarone alone for the prevention of AF recurrence. Adding losartan or perindopril to amiodarone can inhibit left atrial enlargement in this group of patients.


The correlation between AF and strokes has been well recognised. A retrospective study6 showed that the prevalence of strokes in Chinese AF patients was 17.5%. In the non-valvular AF patients, risk factors associated with strokes were an advanced age, history of hypertension, coronary heart disease and the type of AF. In a case-control study8 from 18 hospitals in China during January 2000 to April 2002, a total of 4511 adult patients were diagnosed as having AF. There were 1086 patients with rheumatic valvular AF and 3425 patients with non-valvular AF. Among the valvular AF patients, 273 (26.88%) had strokes, and among the non-valvular AF patients, 827 (24.15%) had strokes. A multivariate analysis in the non-valvular AF patients showed that age ⩾75 (OR1.76; 95% CI 1.08 to 2.98), history of hypertension (OR 1.52; 95% CI 1.28 to 1.80), diabetes (OR 1.39; 95% CI 1.11 to 1.76), high systolic blood pressure (OR 1.71; 95% CI 1.21 to 2.28) and LA thrombi (OR 2.77; 95% CI 1.25 to 6.13) were independently associated with strokes. The epidemiological survey5 of AF showed that ischaemic stroke was the most frequent type observed in the AF cases, and the stroke rate among those cases with AF was significantly higher than that those without (12.1% vs 2.1%, p<0.01).

The efficiency and safety of anticoagulation therapy in patients with AF in China have been studied in recent years. The earlier anticoagulation studies were mainly observational or similar with western countries. A retrospective study6 by analysis of the hospital records from 2000 to 2002 showed that 64.5% of the patients received antithrombotic therapy, predominantly with antiplatelet agents. A retrospective multicentre study9 included 435 patients with AF who were hospitalised from 2000 to 2002 and given warfarin for prevention of thromboembolism. The average dose of warfarin was 2.77 (0.83) mg, and the median duration of anticoagulation therapy was 7 months (from 1 month to 3 years). In total, there were 31 confirmed bleeding events, with major haemorrhage occurring in five patients. The age of the patients in the haemorrhage group was not significantly higher than that in the control group (65.09 (9.99) vs 62.01 (12.19), p = 0.259). Chronic heart failure or hypertension increased the risk of bleeding during warfarin therapy. A multivariate analysis showed that an INR ⩾3.0 was an independent risk factor for haemorrhage (OR 3.74, 95% CI 1.28 to 8.98). There were 76 (17.47%) thromboembolic events, and the risk of a thromboembolism increased steeply with an INR below 1.5.

Another prospective multicentre randomised trial10 compared the effect of aspirin 150–160 mg once daily to an adjusted-dose of warfarin (international normalised ratio, 2.0–3.0). The primary end point was ischaemic stroke or death from any cause, and the secondary end-point was peripheral arterial embolism, transient ischaemic attack (TIA), acute myocardial infarction or serious bleeding. Of the 704 patients, 420 (59.7%) were male. The mean age was 63.3 (9.9) years. The median follow-up period was 19 months (2∼24 months). The mean dose of warfarin was 3.2 (0.7) mg. Compared with the aspirin group, the primary end point of death or ischaemic stroke was reduced by warfarin (2.7% vs 6.0%, p = 0.03) and the relative risk decreased by 56%. The thromboembolic events in the aspirin group were significantly higher than those in the warfarin group (10.6% vs 5.4%, p = 0.01). The other end points were also reduced but not statistically significant in the warfarin group as compared with that in the aspirin group, while the combined end point was statistically decreased by an adjusted dose of warfarin (8.4% vs 13.0%, p = 0.047). Warfarin treatment was associated with an increased bleeding rate compared with aspirin (6.9% vs 2.4%, p<0. 05) with the major bleeding rate of 1.5%. All the major bleeding events occurred with an INR above 3.0. In the warfarin subgroup (335 patients),11 among the 3482 INRs measured during the study, 2378 (68.3%) were in the target range. Of the 19 thromboembolic events that occurred during the warfarin therapy, 15 occurred with an INR of less than 2.0. The independent risk factors for thromboembolic events during the warfarin therapy were age (>75 years), history of a stroke, left ventricular ejection fraction (LVEF) <0.40 and INR >2.0. There were 23 (6.9%) bleeding events, including five cases (1.5%) with minor bleeding and 18 (5.4%) with major bleeding. The independent risk factors for haemorrhage during the warfarin treatment were age (>75 years), systolic blood pressure ⩾160 mm Hg, elevated serum creatinine level and INR >3.0. An INR of 2.0 to 3.0 was associated with the lowest combined rate of bleeding and thromboembolisms.

However, unlike western countries, China has a higher incidence of strokes (especially bleeding events) than myocardial infarctions. A suitable anticoagulation intensity for AF needs to be established in the Asian population in general. Non-Caucasians are at a greater risk for intracerebral haemorrhaging than Caucasians in the general population, and this also applies to patients with AF. Warfarin therapy was associated with a comparable risk of intracerebral haemorrhaging in non-Caucasians, especially Asians.12 Two studies in Japan suggested that a low-intensity warfarin anticoagulation treatment might be safer. One study showed that the low-intensity warfarin anticoagulation treatment (INR 1.5–2.1) was safer than the standard-intensity warfarin anticoagulation treatment (INR 2.2–3.5) for the secondary prevention of strokes in persons with non-valvular AF, especially in older patients.13 The other study14 concluded that major ischaemic or haemorrhagic events occur often in older non-valvular AF patients, in whom an INR value of between 1.6 and 2.6 seems optimal to prevent such events. Thus, a racial difference in optimal anticoagulation treatment may exist between western and eastern populations.

A recent study15 assessed different anticoagulation intensities for non-valvular AF focusing on the Chinese population (unpublished data). The study was conduced in 84 medical centres in China between 2002 and 2005. The enrolled patients were randomised into three groups: a standard-intensity warfarin anticoagulation group (INR 2.1–2.5), low-intensity warfarin anticoagulation group (INR 1.6–2.0) and aspirin group (200 mg/d). A primary endpoint event was defined as a thromboembolic event including an ischaemic stroke, TIA or peripheral artery thromboembolism. The data showed that in Chinese patients with NVAF, warfarin therapy (INR 1.6∼2.5) for the prevention of thromboembolic events was superior to aspirin. No difference was observed between the low-intensity anticoagulation treatment (INR 1.6∼2.0) and the standard intensity anticoagulation treatment (INR 2.0∼2.5). For the optimal intensity of warfarin therapy in Chinese patient population with NVAF, a large clinical trial will be needed in future.


A registration study16 showed that in 2000, there were 10 811 catheter ablation procedures for arrhythmias performed in 136 hospitals in China. The number increased to about 20 000 in 2006.17 Catheter ablation of AF has increased dramatically since the first case was performed in China in 1998.18 A registration study of catheter ablation in patients with AF in China up to 2005 included a total of 3196 cases from 40 hospitals, with 2193 males and 1003 females.19 The mean age was 54.77 (5.98) years. The proportions of paroxysmal, persistent and permanent AF were 85.67%, 11.51% and 2.82%, respectively. Further, 45.93% of the patients had one or more than one basic disease, and left atrium thrombus accounted for 0.9%. The diameter of the left atrium was 37.02 (3.98) mm, left ventricular end-diastolic dimension 46.81 (4.05) mm and left ventricular ejection fraction 0.59 (0.06). The energy used for the catheter ablation included radiofrequency (95.96%), ultrasound (3.00%) and cryoablation (1.22%). Factors impacting the success and recurrence rates included gender, age, basic disease, cardiac function and structure, experience of the physicians, type of AF, course of AF, procedures and energy used. After catheter ablation, the administration of antiarrhythmic drugs decreased markedly, and the anticoagulation therapy was strengthened. The overall rate of complications was 7.48%, and severe complications, such as a pericardial tamponade or pulmonary vein stenosis, accounted for 3.19%. In 2005, there were 1427 patients who underwent catheter ablation for AF, and the main procedures included focal ablation in 10 (0.70%), segmental PV ablation in 279 (19.55%), circumferential ablation of the PV ostia in 624 (43.73%), left atrial substrate modification in 53 (3.71%) and PV atrum modification in 461 (32.31%).

Circumferential pulmonary vein isolation (CPVI) guided by 3D mapping systems (CARTO or EnSite) is currently the major ablation type in China.20 The endpoint can be achieved either by closing gaps along circular lines or by segmental pulmonary vein isolation inside the circular lines after creation of initial circumferential pulmonary vein ablation (CPVA) lesions. A study21 was carried out to investigate whether the clinical outcome depended on the pulmonary vein isolation approach used during the first-time CPVA procedure. One hundred patients (69 males; age 56.7 (11.6) years) who underwent a first-time CPVA for the treatment of symptomatic AF were enrolled. Pulmonary vein isolation was randomly achieved either by CPVA alone (aggressive CPVA (A-CPVA) group, n = 50) or by a combination of CPVA with segmental PV ostia ablation (modified CPVA (M-CPVA) group, n = 50). The conclusion was that when pulmonary isolation was the endpoint of the CPVA, the efficacy of the A-CPVA approach was better than that of M-CPVA. Other ablation strategies for atrial fibrillation have also been performed in China, including a stepwise linear approach guided by non-contact mapping (EnSite Array).22


Chinese surgeons began surgical treatments for atrial fibrillation during the 1990s.23 New technologies and methods have been tried in the large medical centres in China recently.2426 However, the total number of cases undergoing surgical treatment for atrial fibrillation is still relatively small.


Chinese investigators have made great progress in AF basic research, especially in genetic studies, partly because China has the largest population in the world. The first familial AF causative gene, mutation (S140G) in the KCNQ1 gene on chromosome 11p15.5, was identified in a Chinese family.27 The KCNQ1 gene encodes the pore-forming subunit of the cardiac IKs (KCNQ1/KCNE1), KCNQ1/KCNE2 and KCNQ1/KCNE3 potassium channels. A functional analysis of the S140G mutation revealed a gain-of-function effect on the KCNQ1/KCNE1 and KCNQ1/KCNE2 currents. Thus, the S140G mutation is likely to initiate and maintain AF by reducing the action potential duration and effective refractory period in atrial myocytes. Subsequently, KCNE2 R27C,28 KCNJ2 V93I29 and KCNE3 R53H30 mutations were identified in other Chinese families with a history of AF, and the single nucleotide polymorphism (SNP) KCNE4 E145D was also found to be related to AF in the Chinese population.31 32

To determine the association between the C-reactive protein (CRP) and recurrence of AF after a successful electrical cardioversion, a meta-analysis33 including seven prospective observational studies with 420 patients (229 with and 191 without AF relapses) was performed, which suggested that increased CRP levels are associated with a greater risk of AF recurrence, although there was significant heterogeneity across the studies. The use of the CRP levels in predicting sinus rhythm maintenance appears promising but requires further study.


Chinese Society of Pacing and Electrophysiology (CSPE) is the main organisation in charge of EP training, education, research and communication. The CSPE was founded in 1994, and the mission was to improve the medical care of the patients and to assist in the economic construction through building a qualified specialty team with a high moral, providing continuing medical education, improving scientific research and promoting national and international intercourse in arrhythmia area. The CSPE has four branches: cardiac pacing team, cardiac electrophysiology team, medication team, and basic and non-invasive study team. The National Working Group on Atrial Fibrillation and National Working Group on Ventricular Arrhythmias were recently established. The CSPE hosts and sponsors various kinds of academic conferences. Scientific Sessions have been held biannually since 1994. The China Atrial Fibrillation Forum is held annually now. The official journal of the Chinese Journal of Cardiac Arrhythmias has been published since 1997. The CSPE also introduced a Chinese edition of the Journal of Cardiovascular Electrophysiology in 2006. The National Working Group on Atrial Fibrillation published the Current Knowledge and Management Recommendations of Atrial Fibrillation in 2001 and updated it in 2006.34 It also published the Current Knowledge and Recommendations of Pulmonary Vein and/or Superior Vena Cava Isolation by Radiofrequency Catheter Ablation in Patients with Atrial Fibrillation in 2004.35 The National Working Group on Atrial Fibrillation also has been instrumental in the registration of catheter ablations of AF in China.

Another electrophysiological specialty society in China is the Chinese Heart Rhythm Society belong to Chinese Society of Biomedical Engineering, and publishes the Chinese Journal of Cardiac Pacing and Electrophysiology. The two societies often coordinate in many large electrophysiology events in China.

Historic reasons and language barriers have hampered the communication between China and the rest of the world. A language gap between research publications in China and western countries still exists, although the Chinese research community is working hard to expand international communication and introduce new technologies both within China and internationally. Chinese electrophysiologists would like to contribute more to international research in cardiac electrophysiology.



  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.