Atrial fibrillation (AF) is the most common disturbance of cardiac rhythm and results in a reduced quality of life through poor response to exercise and a significantly increased risk of death. Although it is widely accepted that changes to the atrial muscle associated with heart disease make prolonged episodes of AF more likely to occur, the mechanisms underlying the origins of AF remain unclear. It is thought that sympathetic innervation plays an important role in the initiation of paroxysms of AF (Coumel, 1996). ATP-sensitive K+ channels (KATP channels) have been suggested to contribute to myocardial responses to β-adrenergic stress (Olson and Terzic, 2010). To investigate the role of KATP channel opening in the genesis of AF associated with β-adrenergic stress, we have examined the susceptibility to AF of excised perfused hearts from adult male Wistar rats (250-300 g). Hearts were excised according to the Animals (Scientific Procedures) Act (1986) and mounted on a Langendorff apparatus for retrograde perfusion with a Krebs-Henseleit solution via the aorta. Unipolar electrograms were recorded from the left atrial epicardial surface of perfused hearts using a 5×5 electrode array, allowing measurements of atrial effective refractory period (AERP), using a programmed extra-stimulation (S1-S2 protocol), and of conduction velocity and inhomogeneity in conduction through construction of activation maps. Paroxysms of AF were induced by 5 seconds-bursts of very rapid pacing (CL<10 ms) or through S1-S2 protocol and the incidence and duration of the arrhythmia noted. β-adrenergic stress was induced by perfusion of the hearts with isoprenaline (ISO). Perfusion of the hearts with ISO (10-9 M to 3×10-6 M) resulted in a concentration-dependent increase of heart rate during sinus rhythm and in the concentration-dependent shortening of AERP. While it was not possible to induce paroxysms of AF in control conditions, perfusion of the hearts with ISO rendered the left atrium susceptible to pacing-induced AF; the incidence and duration of which increased in a concentration-dependent manner. Time-matched control experiments indicated that hearts could be perfused with control solution for more than 4 hours without increase in susceptibility to AF. The shortening of AERP and incidence of AF induced by 10-6 M ISO were completely blocked in the presence of the KATP channel blocker, glibenclamide (10-5 M). A lower concentration of the KATP blocker (10-7 M) was without effect. Taken together, our results suggest that KATP channels can contribute to atrial arrhythmogenesis during β-adrenergic stress.