Currently, prosthetic heart valve testing is performed on animal models with no underlying cardiovascular pathologies. Unfortunately, unforeseen adverse events may occur when heart valves tested in animals in normal sinus rhythm are implanted in patients suffering from arrhythmias. For example, the Medtronic Parallel valve functioned well in pre-clinical testing, but a high rate of thromboembolic complications appeared when the valve was placed in patients with atrial fibrillation (AF). Given the increasing number of patients afflicted with AF, an animal model of the disorder is needed to more accurately predict a valve's function in the clinical setting. Among methods available for inducing AF, electrophysiological pacing is the most practiced, but the challenges associated with pacing have led to the development of alternative methods of inducing AF These methods include gene transfer and a pharmacologic approach with acetylcholine and catecholamines. Finally, although stem cells have been widely investigated in terms of their therapeutic benefits, the use of their well-reported pro-arrhythmic behavior shows great promise for the development of an AF model in sheep. Such a model would have the potential for detecting adverse outcomes with mechanical heart valves before implantation in the clinical setting.