This study was designed to compare the effects on atrial rhythm control between SGLT2 inhibitor and other oral hypoglycemic agents in patients with atrial fibrillation and diabetes mellitus. We are to compare efficacy and safety between these two groups.

This study is a multicenter, prospective open blinded-endpoint design, 1:1 randomized, and controlled study, with total of 716 patients enrolled from 4-5 centers. The primary outcome is recurrence rate of atrial fibrillation within a year, when recurrence of atrial fibrillation is defined as at least one significant atrial fibrillation documented on 24-hour Holter ECG. Secondary outcomes are atrial fibrillation free survival, atrial fibrillation burden at various time points, ablation rate within a year, atrial fibrillation recurrence rate after ablation within a year, volume of left atrium, NT-ProBNP, Symptom score and Quality of Life.

Inclusion Criteria:

• At least one episode of atrial fibrillation* that is documented during the prior year by any kind of ECG recording.
• Diabetes mellitus type II was diagnosed. (HBA1c >6.5%) and are using oral hypoglycemic agents only
• At least 20 years old and not older than 80 years old.
• Normal ECG parameters when measured in sinus rhythm (QRS width ≤120 ms, QTc interval <440 ms, and PQ interval ≤210 ms in a 12 lead ECG)
• Able and willing to give informed consent.

Exclusion Criteria:

• Any disease that limits life expectancy under 1 year
• Subject for another clinical trial within the past 2 months
• Under 20 years old or over 80 years old
• Pregnant women
• Lactating women
• Drug abuser
• Type 2 DM treated by recombinant insulin
• Diagnosis of Type 1 DM, MODY, or secondary DM
• HbA1c ≥12% or HbA1c <6.5% at diagnosis
• Previous treatment with any SGLT2 inhibitor
• Renal dysfunction (eGFR-CKD-EPI <30ml/min/1.73m^2)
• Chronic cystitis and/or recurrent genitourinary tract infections (3 or more in the last year)
• Unexplained hematuria at baseline study
• Systolic BP >180mmHg or diastolic BP >100mmHg at baseline study
• Systolic BP <95mmHg at baseline study
• Previous treatment with AF ablation
• Acute cardiovascular event [e.g., stroke, acute coronary syndrome (ACS), revascularization, decompensated HF, sustained ventricular tachycardia, return of spontaneous circulation (ROSC)] <8 weeks prior to baseline study
• Severe valvular disease or have prosthetic valve
• Treatment with chronic oral steroid (>30 consecutive days) at a dose equivalent to oral prednisolone ≥10mg/d, within the past 1 month
• History of any malignancy within 5 years
• Clinically profound hepatic dysfunction
• Clinically uncontrolled thyroid dysfunction
• Patients incapable of completing the trial due to any severe medical condition by clinical decision
• Patients with poor compliance (defined as 80 - 120%), except for reasonable situations judged by physician

The patients who diagnosed AF and T2DM in one year