The investigators will conduct an observational trial, with outcome ascertainment at 3 months and 6 months after cardioversion. The primary objective is to demonstrate how cardioversion may aid in the regression of left atrial fibrosis in patients with persist atrial fibrillation as assessed by late gadolinium enhanced magnetic resonance imaging. The secondary objective is to assess the relationship between structural and functional changes with heart rhythm and heart rate.

Background: Atrial fibrillation (AF) is estimated to affect over 33 million people worldwide and is associated with significant co-morbidities such as embolic stroke, heart failure, dementia. Consequently, AF poses a significant burden to the healthcare system, in both direct and indirect costs. The management of AF is complex, especially for patients with persistent AF, which is defined as sustained AF for > 7 days. It is preferable to terminate the AF and restore normal sinus rhythm for relief of symptoms associated with AF or improvement of cardiac structure and function. The maintenance of AF, especially persist AF is associated with fibrosis of left atria. Furthermore, AF itself promotes fibrosis, which in turn leads to increased conduction heterogeneity within the atrial substrate resulting in further progression of AF. In animal models, alterations in myocytes after sustained AF resemble those of myocardial hibernation. Ultimately, these structural changes would lead to Calcium overload and metabolic stress, similar changes have been observed in humans. In humans, atrial dilatation and degenerative changes have been observed. Interstitial fibrosis is the prime cause of structural remodeling in left atrium. Whether reversal of sinus rhythm can reverse fibrosis partly? The introduction of Late Gadolinium enhancement magnetic resonance imaging (LGE-MRI) sequence now allows for non-invasive and dynamic assessment of the location and extent of atrial fibrosis.

To date, no studies evaluating the regression of left atrial fibrosis after cardioversion assessed by LGE-MRI have been performed. The investigators propose to use LGE-MRI to evaluate the effects of cardioversion on atrial fibrosis. It has been shown that the extent of fibrosis can predict the success of catheter ablation procedure. For these patients with more "scar", control the progression of fibrosis and simultaneously provide respite from AF recurrence would be an extremely desirable prescription.

Objectives: To determine the regression of left atrial fibrosis in patients with persist atrial fibrillation as assessed by late gadolinium enhanced magnetic resonance imaging.

Methods: This study will be an observational trial in patients with persist AF. Candidate inpatients or outpatients will be identified if MRI were performed after pharmacological or electrical cardioversion at least twice. The heart rhythm will be assessed by medical records and telephone survey.

Objective

Primary:

The primary objective of this study is to demonstrate how cardioversion may influence the regression of left atrial fibrosis in patients with persist atrial fibrillation as assessed by LGE-MRI, using longitudinal data from a retrospective study of patients diagnosed with atrial fibrillation over a six month follow up period.

Secondary:

To study the effects of sinus rhythm in global parameters of myocardial remodeling such as right and left atrial volumes and right and left ventricular volumes.

To assess whether characteristics of MRI after cardioversion can predict the rate of atrial fibrillation sinus rhythm maintenance.

Inclusion Criteria:

• Patients carry a diagnosis of Persistent Atrial Fibrillation
• Patients have given informed consent
• Pharmacological or electrical cardioversion has be performed successfully
• late gadolinium enhanced magnetic resonance imaging (LGE-MRI) were performed after cardioversion at least twice

Exclusion Criteria:

• Prior radiofrequency ablation treatment for atrial fibrillation
• Prior cardiac surgical procedures
• Individuals with cognitive impairments who are unable to give informed consent

Candidate inpatients or outpatients will be identified if pharmacological or electrical cardioversion has be performed and LGE-MRI were performed after cardioversion at least twice.