Atrial fibrillation (AF) is growing into an epidemic affecting 1 in 4 adults. There is a need for research to elucidate the prevalence of ischemic cardiomyopathy in patients diagnosed with AF. The objective of this study is to demonstrate the utility of MRI in assessment of coronary artery disease. The specific objective is to demonstrate sensitivity/specificity comparable to that reported in meta-analyses of non-AF patients and adenosine (90% /80%) in an AF population using the time-efficient vasodilator regadenoson that requires only a single intravenous (IV).

Scientific Background/Rationale:

Atrial fibrillation (AF) is a large and growing healthcare problem worldwide. Over 7 million people in the U.S. and Europe currently suffer from atrial fibrillation, and this number is expected to double before 2050. The lifetime risk of AF is high: ~24% at age 40, and this risk remains fairly constant at older ages, with 22% lifetime risk at age 80 [1]. AF significantly increases the risk of stroke and mortality, and can greatly limit quality of life. Little research has been done on AF and ischemic cardiomyopathy, though it is a relatively common co-morbidity. CT found increased prevalence (41% vs 27%) of coronary artery disease (CAD) in patients with AF compared to patients with similar pre-test risk but no AF [2]. AF patients with a positive SPECT scan for CAD have a worse prognosis for cardiac events than patients with positive SPECT but without AF [3]. It was recently reported that in a study of 253 AF patients, that AF patients with positive SPECT studies had a very high number of false positives - only 15% of patients had significant CAD by angiography (compared to 67% in the control group) [4, 5]. Just over half of the patients were in sinus rhythm at SPECT, but since similar numbers of positives were seen in each half, the authors did not feel that imaging during AF was the cause of the poor specificity.

Use of MRI for assessment of CAD is a growing area that entails no radiation exposure to the patient. Advances in MRI have made it possible to accurately detect CAD, either as well or better than SPECT in unselected populations [6, 7]. However, adoption of MRI myocardial perfusion scans has been limited in part due to the challenges associated with the use of adenosine. Adenosine requires starting a second IV, and to use either a special expensive MRI-compatible infusion pump to deliver the drug, or long lengths of tubing to run to a pump outside the scanner room. Neither solution is ideal, and regadenoson would not require any such pumps or the starting of a second IV. Here the investigators propose to determine the sensitivity/specificity for dynamic contrast-enhanced myocardial perfusion MRI with the vasodilator regadenoson in a subpopulation of patients - those with atrial fibrillation.

The investigators have a great deal of experience with stress and rest myocardial perfusion MRI. In a preliminary study imaging three patients with AF that then went to X-ray angiography (cath), two of the three subjects had significant stenoses by cath, and one did not. This agreed with the regadenoson stress perfusion MRI findings. These MRI acquisitions were performed on a Siemens Verio MRI scanner. The Verio operates at twice the magnetic field strength (3 Tesla, or 3T) of most MRI scanners, which operate at 1.5T. The higher magnetic field offers images with significantly less noise (almost twice the signal-to-noise ratio).

Objective: To demonstrate sensitivity/specificity comparable to that reported in meta-analyses of non-AF patients and adenosine (90%/80% [6, 7]), in an atrial fibrillation population while using the time-efficient vasodilator regadenoson that requires only a single IV.

Study Design: This will be a prospective, open-label, comparative trial using MRI. Non-invasive MRI measurements of resting flow and flow at regadenoson stress will be obtained in each subject during a one hour MRI exam using our advanced MRI acquisition techniques. Sensitivity/specificity of regadenoson stress MRI will be determined using x-ray angiography as the standard. X-ray angiography will be done as standard-of-care, and is not an intervention of this clinical trial.

Study Procedure: Each subject will undergo a single MRI scanning session. Caffeine will be stopped 12 hours prior to the procedure. One IV will be started and subjects positioned in the scanner. Resting perfusion with Multihance Gd-BOPTA contrast agent will be performed first. The perfusion acquisition acquires 3-6 short axis slices each heartbeat and lasts for one minute. Then a standard regadenoson injection of 400ug/5cc will be given and MR imaging performed with a Gd-BOPTA contrast agent bolus 60-100 seconds later [9] [10].

Inclusion Criteria:

Patients with confirmed persistent or paroxysmal AF and suspected coronary artery disease who will undergo catheterization X-ray angiography

Exclusion Criteria:

1. Critically ill patients, patients on ventilators patients with hypotension, asthmatics, and other patients whose medical care or safety may be compromised from undergoing an MRI examination will be excluded.
2. Patients with claustrophobia will also be excluded since MRI is conducted in a closed environment.
3. Patients with contraindications to MRI (pacemaker, metal implants).
4. Pregnant subjects (or women who may become pregnant), minors, and prisoners will be excluded from this study.
5. Subjects are over 60 or have any suspicion of abnormal kidney function (a blood test to determine Glomerular filtration rate (GFR) will be performed prior to imaging. Subjects with GFR<30 will be excluded from the study. This is standard practice for clinical scans in Radiology due to the extremely small but not negligible relationship between gadolinium contrast agent and nephrogenic systemic fibrosis in patients with severely impaired renal function.

Subjects monitored for safety events during the entire magnetic resonance imaging session; up to 1 hour