Mortality and device intervention in patients with a history of atrial fibrillation: a substudy of APPRAISE ATP

Background/Introduction Atrial fibrillation (AF) has previously been found to be associated with an increased occurrence of ventricular tachyarrhythmias. AF may also negatively impact all-cause mortality and cause inappropriate device interventions. APPRAISE ATP was a large, randomized trial assessing the efficacy of antitachycardia pacing (ATP) as a first-line strategy for the termination of fast ventricular tachycardias. Purpose To assess outcomes in patients with AF at baseline from the APPRAISE ATP clinical trial cohort and evaluate the occurrence of appropriate and inappropriate implantable cardioverter defibrillator (ICD) interventions, as well as the association between AF and all-cause death. Methods APPRAISE ATP (N=2595) was a prospective trial with high-rate programming (VT zone: 200-249 bpm, VF zone: 250+ bpm) and long delays to therapy (12s in the VT zone). All patients with ECG data at baseline (N=1933) were grouped by whether they had a history of AF at enrolment, had AF by ECG at implant, or had no history of AF. Kaplan-Meier methodology was then used to determine incidence of time to first all-cause shock, time to first appropriate shock, time to first inappropriate shock, time to death, or time to death or shock from any cause. Results Of the patients with ECG data at baseline, 234 had AF by ECG at implant (10 had AF noted in their medical history), 289 had a history of AF, and 1410 had no history of AF. At 60 months from randomization, the incidence of all-cause shock was 28.1% for patients with AF at implant, 23.8% for patients with a history of AF, and 13.8% for patients with no history of AF (Log-rank P=<.001; Figure). Similarly, first appropriate shock was 17.2%, 18.7%, and 11.4% (Log-rank P=.008) and first inappropriate shock was 14.7%, 8.5%, and 3.2% (Log-rank P=<.001), for each group respectively. Mortality was significantly associated with history of AF, with an incidence of 22.1% for patients with AF at implant, 24.8% for patients with a history of AF, and 17.0% for patients with no history of AF (Log-rank P=0.048; Figure). For the outcome of time to first shock or death, the incidence was 42.8%, 41.2%, and 26.3% (Log-rank P=<.001), respectively. Conclusions Our analysis of the APPRAISE ATP cohort found an association between AF at baseline and outcome during long-term follow up, in terms of treatable ventricular tachyarrhythmias and mortality. These findings suggest that AF at baseline characterizes patients at higher risk. Further analyses are planned to assess the outcome implications of device-detected AF and of new onset AF in this cohort.