Internal Cardioverter Defibrillator Use after Implantation of Continuous Flow Left Ventricular Assist Device as Both Bridge to Transplant and Destination Therapy

Abstract

Introduction: Following pacemaker (PM) implant, both atrial fibrillation (AF) and complete atrioventricular block (cAVB), resulting in high burden right ventricular pacing, are independently associated with incident heart failure (HF); however, the combined effect of AF and cAVB in HF development is not well characterized. Methods: A MarketScan ® (MS) Commercial and Medicare Supplemental claims database was used to identify patients ≥18 years old undergoing de novo dual chamber PM implant between 4/2008-3/2014. Baseline cAVB and AF were determined using inpatient and outpatient billing codes in the year prior to PM implant. Patients with cAVB were identified by a diagnosis of 3 rd degree AVB or an AV node ablation, and were compared to those without any AVB (noAVB). Patients with a diagnosis of HF prior to implant were excluded, as were those without continuous MS enrollment for ≥1 year before and after implant. After stratification based on the presence of baseline cAVB and AF, patients were propensity score matched on age, gender, geographic region, implant year and 19 baseline comorbidities. The primary endpoint was incidence of new HF diagnosis following PM implant. Results: The four matched cohorts included: cAVB/AF (n = 2084); noAVB/AF (n = 2084); cAVB/no AF (n = 2075); and noAVB/no AF (n = 2065). Across all groups, mean age was 76 ± 11 years, 56% were male, and median duration of followup after PM implant was 2.28 [IQR 1.59, 3.27] years, without significant differences between groups. The incidence of new HF diagnosis following PM implant (per 100 pt-yrs of follow-up) was highest in the cAVB/AF group (21.5), followed by the noAVB/ AF group (16.7), cAVB/no AF (13.7) and was lowest in the noAVB/no AF group (11.4) (Figure). In multivariate Cox models, the hazard ratio (HR) for new HF diagnosis among those with cAVB compared to those with noAVB was 1.20 [95% confidence interval (CI) 1.06 -1.35, P = .004] and the HR associated with AF compared to no AF was 1.44 [95% CI 1.28 -1.62, P < .001]. There was no significant interaction between the presence of baseline AF and cAVB (P = .440). Conclusion: Both complete AV block and AF are associated with an increased incidence of HF following PM implant. Among patients with both diagnoses, the effect appears to be additive and the risk of developing HF was approximately 50% at 4 years following PM implant. These findings highlight the need for tools to predict which patients are most likely to develop HF following PM implant and strategies to mitigate that risk.