Alert burden and clinical relevance of remote monitoring in heart failure patients with cardiac implantable electronic devices: Single-center experience in the reimbursement era

BACKGROUND: Heart failure (HF) affects 1%-2% of adults in developed countries and is associated with significant healthcare burden. Remote monitoring (RM) of cardiac implantable electronic devices may improve outcomes in HF patients, but itincreases the clinical management demands on device follow-up clinics. AIMS: The aim of our study was to assess the workload of a cardiac device follow-up clinic by analyzing RM alerts and associated clinical interventions. METHODS: A retrospective single-center study of consecutive HF patients with implantable cardioverter defibrillator or cardiac resynchronization therapy-defibrillator was conducted to analyze unscheduled RM alerts, categorizing them as clinically significant or insignificant and assessing resulting medical interventions. RESULTS: A cohort of 407 HF patients (median age 71 years, 15% female) with implantable cardioverter defibrillator or cardiac resynchronization therapy-defibrillator devices, predominantly implanted for primary prevention of sudden cardiac death and with predominantly ischemic etiology, was followed for a median of 18 months, with 10.6% mortality. A total of 19 788 alerts were generated (2.8 per patient-month), of which 1491 (7.5%) were clinically significant (0.21 per patient-month). The predominant clinically significant alert was a reduction in biventricular pacing below 85%. Clinically important arrhythmic alerts included 21 cases of first-detected atrial fibrillation, 107 episodes of sustained ventricular tachycardia and 37 episodes of ventricular fibrillation. The most frequent clinical intervention consisted of teleconsultation with pharmacotherapy modification. CONCLUSIONS: In this cohort, RM alerts identified clinically significant events, including newly detected atrial fibrillation and ventricular arrhythmias, and selected alerts were associated with subsequent clinical actions. The high alert burden and low proportion of actionable alerts underscore the need for optimized alert programming and structured monitoring workflows.