Résumé : Aims: “CLOSE”-guided pulmonary vein isolation (PVI) is based on contiguous (≤6 mm) and optimized radiofrequency (RF) ablation lesions (ablation index [AI] ≥ 400 posteriorly and ≥ 550 anteriorly]. However, the optimal RF power to reach the desired AI is unknown. Therefore we evaluated the efficiency of an ablation strategy using higher power (40 W) during a first “CLOSE”-guided PVI. Methods: Eighty consecutive patients undergoing “CLOSE”-guided PVI for symptomatic paroxysmal atrial fibrillation were ablated with 40 W (group A). Results were compared with 105 consecutive patients enrolled in the “CLOSE to CURE”-study and were ablated using the same protocol with 35 W (group B). Results: In group A, ablation was associated with shorter ablation procedure time (91 vs 111 minutes; P <.001), shorter fluoroscopy time (5 vs 11 minutes; P <.001), shorter PVI time (48 vs 64 minutes; P <.001), shorter RF time (20 vs 28 minutes; P <.001), lower RF time per application (22 vs 29 seconds; P <.001), less RF applications (52 vs 58; P <.001), and less catheter dislocations (1 vs 2; P =.002). The impedance drop (12 vs 13 Ω; P =.192), first-pass isolation rate (99% vs 93%; P =.141) and acute reconnection rate (6% vs 4%; P >.733) were similar in both groups (groups A and B, respectively). No complications occurred. In group A, a gastroscopy—performed in five patients with esophageal temperature rise more than 42°C—did not reveal any esophageal lesion. Postprocedural recurrence of atrial tachyarrhythmia at 1 year was not significantly different between both groups. Conclusions: Using the “CLOSE”-protocol, increased power increases the efficiency of PVI without compromising patients' safety.