This editorial refers to ‘The relationship between right ventricular pacing and atrial fibrillation burden and disease progression in patients with paroxysmal atrial fibrillation: the long-MinVPACE study’ by R.A. Veasey et al., on page 815. Atrial fibrillation (AF) is a frequent problem after pacemaker implantation. It is likely that paroxysmal AF had occurred in many patients before pacemaker insertion but was documented for the first time by stored atrial electrograms of dual-chamber devices. On the other hand, pacing from the right atrial appendage may promote the development of AF by mechanisms such as increased dispersion of atrial refractoriness, or scar tissue formation around the implanted atrial lead. The development of permanent AF after pacemaker implantation for sinus node disease even raises the question whether the device is still needed in this situation. 1 Randomized controlled trials have demonstrated that the incidence of AF is higher with asynchronous right ventricular pacing in the VVI(R) mode compared with dual-chamber pacing. 2 A post hoc analysis of the Mode Selection Trial (MOST) suggested that right ventricular pacing may be responsible for the development of AF, at least in patients with sinus node disease and narrow QRS. 3 In this analysis, a linear association between the percentage of right ventricular pacing and the risk of developing AF was observed especially for dual-chamber pacing. Reports on the development of heart failure with right ventricular pacing confirm and extend these concerns on the risks of unnecessary right ventricular pacing. Can this ‘proarrhythmic’ effect of pacing be prevented? It has been suggested that biventricular pacing might reduce the risk of developing AF with pacing. 4 Post hoc analyses 5 and small studies, 6 however, did not find a reduction in the incidence of AF with cardiac resynchronization therapy compared with right ventricular pacing or standard medical care. In contrast, the systematic prevention of unnecessary right ventricular pacing, facilitated by the use of dedicated pacing algorithms, seems to be associated with a reduction in the occurrence of AF. In the randomized Search AV Extension and managed ventricular Pacing for Promoting Atrioventricular Conduction (SAVE-PACe) trial, prevention of unnecessary right ventricular pacing (present during an average of 99% of the time in the conventional pacing group) was associated with a 40% relative risk reduction for the development of persistent AF. 3 Pacing used to prevent AF has been disappointing in most randomized trials. 7 – 9 In retrospect, this may have been caused by right ventricular pacing that may increase the risk of developing AF. Also biventricular pacing does not seem promising in the prevention of AF. 10 It raises the question: can pacing be more effective in the prevention of AFn if right ventricular pacing is prevented? Some analyses suggest that the reduction in right ventricular pacing may enhance the ability of atrial pacing to prevent AF. In the Pacing In Prevention of Atrial Fibrillation (PIPAF) trial, patients paced in the right ventricle for ,42% of the time had a significant reduction of the cumulative time in AF during atrial preventive pacing, while this effect was absent in patients paced in the ventricle for more than this. 11 In a single-centre experience in 153 patients, right ventricular pacing for ,40% of the time after 1 month was associated with a significantly lower mean AF burden during the following 5 years. 12 In fact, in these patients, AF burden was 6–11% compared with 22– 32% in patients who were paced in the right ventricle for .40% of the time. Other studies did not find this association. In a post hoc analysis in 331 patients from the Atrial Fibrillation Therapy (AFT) trial, AF burden was weakly associated with the percentage of right ventricular pacing with a 0.03% increase in AF burden for every 10% of right ventricular pacing. 13 However, the median percentage of right ventricular pacing in this trial was 97%, rendering any extrapolation questionable. In the same report, data from 79 patients of the Pacemaker Atrial Fibrillation Suppression (PAFS) trial showed no association between the percentage of right ventricular pacing and AF burden. However, the PAFS study consisted of four short (4 weeks each) cross-over study periods with different pacing modalities. Likely, these periods were too short, and there might have been significant carry-over effects. An