This study was designed to evaluate the importance of local release of autonomic neuromediators when electrical stimuli are applied to the right atrium to measure the atrial fibrillation threshold (AFT). Experiments were performed in 16 open-chest dogs anesthetized with alpha-chloralose. The dogs were denervated by bilateral transection of the stellates and cervical vagi. The AFT was determined in 11 dogs by delivering either a train of stimuli (14 pulses, 4 ms, 100 Hz) or a single stimulus (10 ms) to the right atrium during its vulnerable period. In eight dogs, beta-adrenergic blockade with timolol (0.1 mg/kg) had no effect on the AFT determined with either method. Atropine (0.2 mg/kg), given after timolol, significantly increased the train-of-pulses AFT from 4.7 +/- 0.4 to 32.3 +/- 4.6 mA (P less than 0.001). The single-pulse AFT increased from 16.5 +/- 1.5 to 17.8 +/- 1.5 mA (P less than 0.05). Atropine had a similar effect on the AFT when it was given in the absence of timolol (n = 3). In five additional dogs, a monophasic action potential was recorded while a 10-mA train was delivered to the atrium during its absolute refractory period. There was marked shortening of the monophasic action potential duration (55 +/- 6 ms) in the first beat after the train. The shortening was totally abolished by atropine (0.2 mg/kg). The results suggest that a train of stimuli liberates local stores of acetylcholine, which cause a shortening of atrial repolarization time and a profound decrease in the current necessary to evoke fibrillation.
Two Varieties of the Onset of Atrial Fibrillation Studied by Monophasic Action Potential Recording
