The influence of atrial and ventricular pacing on the incidence of atrial fibrillation: A meta-analysis

Background: The effect of atrial pacing on the incidence of atrial fibrillation (AF) is unknown. Furthermore, the threshold of ventricular pacing that is associated with a higher incidence of AF has yet to be determined. Thus, we set out to determine the optimal pacing modality in patients with sinus node dysfunction (SND) for the prevention of AF. Methods: Individual patient data from four contemporary pacemaker studies were gathered and analyzed. Since AF would inherently lead to a reduction in atrial pacing, percent atrial and ventricular pacing (%AP and %VP) were determined at the first follow-up visit and then used as a surrogate for all endpoints. Patients with >5 minutes of AF at the first visit were excluded. The primary endpoint was defined as 7 consecutive days of AF. Results: A total of 1,507 patients were included. During a mean follow-up of 14.3 ± 8.7 months, 77 patients developed AF (annual rate of 4.3%). The incidence of AF in the first (0-32%), second (32-66%), third (66-89%), and fourth (89-100%) quartiles of %AP was 1.3%, 5.3%, 5.8%, and 8.0%, respectively (P < 0.001). A multivariable analysis found that pacing above the first quartile was associated with a relative risk of 2.93 (95% confidence interval 1.16-7.39, P = 0.023). The grouping of %VP into first (0-2%), second (2-7%), third (7-84%), and fourth (84-100%) quartiles yielded an AF incidence of 2.4%, 3.4%, 6.6%, and 8.0%, respectively (P = 0.001). Conclusion: We demonstrated that in patients with SND both atrial and ventricular pacing are associated with a higher incidence of AF.