is a well-established animal model of atrial fibrillation (AF). However, this model is limited by a high mortality rate and severe heart failure. The purpose of our study was to assess a new canine model of inducible AF. We performed acute, short-term, simultaneous atrioventricular pacing (SAVP) and RVP (in random order) in 14 dogs for 30 s. SAVP produced more echocardiographic pulmonary venous flow reversal, a greater increase in mean pulmonary capillary wedge pressure, and a significantly greater decrease in left atrial emptying function (Ϫ84.4 Ϯ 38.6% vs. Ϫ23.7 Ϯ 27.1%, P Ͻ 0.05) than RVP. Thirty dogs were randomized to three, longer-term, study groups: eight dogs in the control group (no pacing), eight dogs in the RVP group (2 wk at 240 beats/min followed by 3 wk at 220 beats/min), and fourteen dogs in the SAVP group (2 wk at 220 beats/min). SAVP induced less left ventricular dysfunction but more left atrial dysfunction than RVP. SAVP dogs had similar atrial effective refractory periods as RVP dogs but more heterogeneity in conduction and more AF inducibility (83% vs. 40%, P Ͻ 0.05) and maintenance (median 1,660 vs. 710 s, P Ͻ 0.05) than RVP dogs. SAVP induced more collagen turnover and was associated with a significantly greater increase in type III collagen in the atria compared with RVP dogs (6.9 Ϯ 1.5 vs. 4.8 Ϯ 1.6, respectively, P Ͻ 0.05 vs. 1.1 Ϯ 0.7 in unpaced control dogs). In conclusion, the SAVP model induced profound mechanical and substrate atrial remodeling and reproducible sustained AF. This new model is clinically relevant and may be useful for testing AF interventions. experimental atrial arrhythmia; extracellular matrix; remodeling ATRIAL FIBRILLATION (AF) is the most common sustained cardiac arrhythmia and frequently causes cardiovascular morbidity and mortality. Although AF can occur in patients with normal hearts (lone AF), the majority of patients with AF have coexisting structural heart disease, most commonly hypertension or heart failure (1). To better understand the relationship between the atrial substrate and AF vulnerability and to evaluate therapies for AF, multiple animal models of AF have been developed. They include sterile pericarditis, acute atrial ischemia, atrial volume overload, mitral regurgitation, pacemakerinduced atrial tachycardia, and ventricular pacing-induced heart failure (25).One of the most commonly used models employs rapid atrial pacing (RAP) at 400 beats/min with control of the ventricular response (atrioventricular node ablation) (24). This model helps to understand the mechanisms underlying the tendency of paroxysmal AF to become persistent but does not relate closely to the most frequent clinical conditions associated with AF, in which the left atrium (LA) dilates and is fibrosed, exhibiting conduction slowing (14, 21). The other commonly used model is rapid ventricular pacing (RVP) to induce heart failure (240 -280 beats/min for 3-8 wk), sometimes combined with RAP (RVP at 240 beats/min for 2 wk combined with additional RAP at 400 beats/min during the sec...