Amerena et al and the sponsor of the report from the Registry on Cardiac Rhythm Disorders Assessing the Control of Atrial Fibrillation Asia Pacific (RecordAF-AP) deserve congratulations for their contribution in this issue of The American Journal of Cardiology . They present data on 2,721 patients with atrial fibrillation recruited from >100 hospital sites in 8 countries across the Asia-Pacific region. Atrial fibrillation is an important clinical entity because of its increasing incidence and prevalence, its association with embolic stroke, and the development of new antithrombotic medications, preventive approaches, and ablation and device therapies. Sixty-two percent of the patients in RecordAF-AP were on a rate-control strategy using primarily β blockers, and of the remainder (rhythm control), approximately half were receiving class III agents. The investigators conclude that upon the completion of the study, it will be possible to determine correlations between hard clinical end points and management strategies and to clarify whether there is a difference between rate and rhythm control in this region, given that a difference was not demonstrated in the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) study. The investigators appropriately describe their report as “insights into the management of atrial fibrillation,” implying that they are aware of the limitations of the study. They are contributing another piece to the tapestry of reports on atrial fibrillation worldwide. The risk factors (e.g., smoking) for and causes of atrial fibrillation may not be the same in all parts of the world. In Europe and the United States, rheumatic heart disease is a rare cause of atrial fibrillation, accounting for about 1% of cases, while in some countries in the Asia-Pacific region, it accounts for >10%. Also, because race-related differences have been reported, the investigators are justified in examining whether rhythm-control therapy would result in different outcomes in this region.

In my opinion, RecordAF-AP would benefit from a longer duration of follow-up. Also, in the absence of randomization, it will be difficult to ascertain whether rate or rhythm control will result in better outcomes. Clinical trials, registries, and administrative databases provide complementary information, and it is examination of all available types of studies that informs clinical decision making and the development of practice guidelines. Randomized, double-blind clinical trials such as AFFIRM provide the least biased information on efficacy and include detailed high-quality data, but they do not predict the effectiveness of an intervention or its cost-effectiveness in clinical practice. For example, the Randomized Aldactone Evaluation Study (RALES) showed a 30% decrease in mortality with spironolactone in patients with heart failure. However, the publication of the study was associated with an abrupt increase in hyperkalemia-associated morbidity and mortality. Prospective registries also contain detailed, high-quality clinical information, but when participation is voluntary and not inclusive of all cases in a given region, they may not be representative of the patients in the community at large. In contrast, administrative databases that include all patients in a given geographic area do not contain detailed clinical information and may include inaccuracies. Although the RecordAF-AP investigators were randomly selected to be representative of cardiologists who manage patients with atrial fibrillation on the basis of their expertise and the health care structure of each country, the actual data included in RecordAF-AP may not be representative of patients with atrial fibrillation in the Asia-Pacific region. Although the average age of the patients in RecordAF-AP (64 years) is similar to those in previous studies from the Asia-Pacific region, it is younger than that reported in a smaller previous study from Hong Kong and much younger than that observed in Western countries (73 years in the United Kingdom and the United States). The reported use of aspirin (51%) is higher than in previous studies from Hong Kong (23%) and China (38%), and smoking (13%) was much less prevalent than in the study from China (39%).

In summary, this well-conducted registry adds much-needed information to the mosaic of characteristics and management of atrial fibrillation. Compared to the United States and Europe, younger age, possibly related to more prevalent rheumatic heart disease and smoking, characterize atrial fibrillation in the Asia-Pacific region. The lack of a randomized design and possible limitations in generalizing the findings to all patients with atrial fibrillation in the Asia-Pacific region will limit the validity of the conclusions of the study as they pertain to morbid events.