PERFORMANCE OF A NEW ATRIAL FIBRILLATION DETECTION ALGORITHM IN A SINGLE CHAMBER ICD

B. Schousek, G. Rovaris, M.L. Brown, P.A. Friedman

1. Azienda ospedaliera San Gerardo, Monza, MB, Italy; 2. CRHF Resaerch, Medtronic, plc, Mounds View, MN, USA; 3. CRHF Development, Medtronic, plc, Mounds View, MN, USA; 4 Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic, Rochester, MN, USA

Abstract

Introduction: Approximately 15% of ICD recipients with no history of AF experience new onset of AF within 12 months of implant. AF is usually not detected in single chamber ICDs, leading to morbidity including stroke and heart failure. We hypothesized that the AF detection algorithm used in Medtronic implantable cardiac monitors based on a Lorenz plot of ?RR intervals, could lead to effective AF detection in a single chamber ICD leading to earlier intervention and reduced complications.
Methods: Seventy-four Holter recordings containing ICD measured RR intervals from single chamber ICD patients were used. Total recording time was 1367 hours containing 165 hours of AF. AF duration sensitivity and specificity were evaluated over a range of detection duration settings.
Results: The ICD algorithm detects only episodes longer than 6 minutes. The table contains AF detection algorithm performance. AF duration sensitivity was as high as 91.0%. AF duration specificity was as high as 95.0%.
Conclusions: The new single chamber ICD permits detection of AF with high duration sensitivity and specificity. This may provide early AF detection with better outcomes for patients.


Univariate and multivariate analysis of risk factors
Detection Threshold Patient Average Gross Patient Average Gross
6 minutes 91.0% 92.5% 94.2% 99.7%
10 minutes 89.1% 91.1% 93.9% 99.8%
20 minutes 84.4% 86.6%93.8% 99.8%
30 minutes74.6% 85.6% 95.0% 99.8%