New onset atrial fibrillation after dual chamber pacemaker implantation: long term predictors

Abstract

Abstract Background Preserving atrioventricular synchrony has been accepted as a significant advantage of atrial and dual-chamber (DDD) pacing. However, little is known about the incidence of atrial fibrillation (AF) after DDD implantation and its prognostic predictors in long term. Purpose To determine the incidence of new AF episodes and to identify risk factors and prognostic predictors for new-onset AF and all-cause mortality after implantation of dual-chamber pacemakers. Population and methods 713 consecutive patients (P) who underwent colocation of DDD pacemaker, due to AV block (AVB) or sinus node disease (SND), with no prior history of AF, from 2011 to 2015. Through periodic PM interrogation, occurrence of AF (“automatic mode switch” episodes with documented AF), switch to ventricular pacing (VVIR), pacing site (apical or septal) and cumulative right ventricular (RV) pacing % were analysed. Results Follow-up data was available for 669 patients (93.8%) for a mean follow-up (FU) time of 47.8±22.7 months. Mean age was 72.9±10.8 years with 60.1% male. New occurrence of AF was observed in 345 P (51.6%) during the FU period; 45.7% of them were consequently anticoagulated (59.0% with NOACs). Median time to 1st AF episode since implantation was 21.6 months and in 50.9% of the cases it lasted ≥1h. In univariate analysis, 1st AF episode lasting more than 1 hour and existence of at least one episode longer than 24 hours were directly related to switch to VVIR (p<0.0005; p<0.0005; p<0.0005) as well as prescription of anticoagulation (p=0.001; p=0.011; p<0.0005). Compared to non-AF P, those with AF were older (74.0±9.9 vs. 71.8±11.7 years; p=0.008), had higher prevalence of SND (50.0% vs 40.20%; p=0.015), had superior % of RV pacing (65.9±39.3% vs. 58.3±44.3%; p=0.021) and more frequently had RV apical pacing (70.1% vs 57.3%; p=0.001). The prevalence of hypertension, diabetes mellitus and dyslipidemia were similar in the two groups. With multivariable Cox-regression, age (HR 1.02; p=0.017), SND (HR 1.49; p=0.010), admission for HF (HR 1.55; p=0.012) and % RV pacing (HR 1.01; p=0.003) were significantly associated with the incidence of FA. Predictors of all-cause mortality in Cox regression were the occurrence of AF in 1st of FU (HR 1.67; p=0.018) and % RV pacing (HR 1.01; p=0.043). Conclusions New onset AF is a frequent finding after DDD pacemaker implantation and is associated with all-cause mortality in long term. Age, admission for heart failure, sinus node disease and % of RV pacing were independent predictors for AF during follow-up. Funding Acknowledgement Type of funding source: None