Atrial fibrillation (AF) is the most common cardiac rhythm disturbance in adults, with prevalence expected to rise significantly the coming decades. The occurrence of AF is associated with significantly increased mortality as well as morbidity of which cerebrovascular accidents is the most important. Unfortunately treatment options remain limited. Anti-arrhythmic drugs are widely used but have limited efficacy and the potential for toxicity and adverse events are recognized. Recent year's catheter ablation of AF continues to gain acceptance for symptomatic treatment, but recurrence rate are high with need for continuous medication. Thus there is a need to better understand what causes development and triggers episodes of AF as well to introduce new treatment options. Cardiometabolic factors such as obesity, inactivity and sleep apnea (SA) have therefore gained interest. Many patients with AF have chronic sleep apnea, and in the present study the investigators want to explore the interaction between SA and AF. The hypothesis of the present study is that SA may trigger AF and that treatment of SA will reduce the overall burden of AF as well as reduce the recurrence of AF after pulmonary vein ablation. To test the hypothesis the investigators will implant a Reveal device that continuously records the hearts rhythm of 100 patients with paroxysmal AF and concomitant SA. Initially the influence of SA on onset of AF will be examined, and the patients will then be randomized to treatment of SA or not and the influence on total AF burden recorded both before and after ablation.
Inclusion Criteria: * Age 18-75 years * Male or female * Patients with paroxysmal AF scheduled for first or second catheter ablation * Moderate-to-severe SA defined as an AHI ≥15/h (OSA and/or CSA) * Signed informed consent Exclusion Criteria: * Unstable patients * Patients with left ventricular ejection fraction (LV-EF) \< 45% * Unstable coronary artery disease, myocardial infarction or PCI within 3 months prior to the study * Bypass surgery within 6 months prior to the study * Patients with TIA or stroke within the previous 3 months * BMI \> 40kg/m2 * Drowsy drivers and/or sleepy patients with ESS (Epworth Sleepiness Score) \> 15 * Patients with interstitial lung diseases, severe obstructive lung defects, and thoracic myopathies or severe obstructive lung defects with FEV1 \< 50% of predicted * Oxygen saturation \< 90% at rest during the day * Poor compliance * Patients with single chamber pacemaker (or ICD) * Current use of PAP therapy * Patients using amiodarone