Author + information
- Received January 19, 1996
- Revision received July 18, 1996
- Accepted July 31, 1996
- Published online November 15, 1996.
- Jasbir S. Sra, MD, FACC∗,
- Cheryl Maglio, RN,
- Anwer Dhala, MD, FACC,
- Zalmen Blanck, MD, FACC,
- Michael Biehl, MD,
- Sanjay Deshpande, MD,
- Edward T. Keelan, MD,
- Mohammad R. Jazayeri, MD, FACC and
- Masood Akhtar, MD, FACC1
- ↵∗Address for correspondence: Dr. Jasbir Sra, 2901 West Kinnickinnic River Parkway, No. 470, Milwaukee, Wisconsin 53215-3660.
Objectives. This study assessed the feasibility of detecting atrial fibrillation (AF) and delivery of appropriately timed R wave shocks using an implantable atrial defibrillator.
Background. For atrial defibrillation therapy to be feasible in an implantable form, AF must be detected in a specific fashion, and the risk of ventricular proarrhythmia should be minimized.
Methods. Eleven patients with AF underwent testing with an implantable atrial defibrillator (METRIX 3000 Automatic Atrial Defibrillator, InControl, Inc.). Wideband electrograms (EGMs) were recorded from the right ventricular (RV) bipolar catheter and from the multipolar catheters located in the right atrium (RA) and coronary sinus (CS). Atrial fibrillation detection was performed using two serial algorithms—quiet interval analysis and baseline crossing analysis—that defect atrial activity on the RA-CS channel. Ventricular sensing using a minimal preceding synchronization interval of 500 ms as a criterion for synchronous shock delivery was performed from filtered RV and RV-CS EGMs.
Results. The AF detection algorithms were applied to 53 AF data segments and 18 normal sinus rhythm data segments. Atrial fibrillation was detected appropriately in 49 instances, and the specificity for detecting AF and normal sinus rhythm was 100%. Synchronization criterion efficacy was assessed by delivering shock markers and shocks. Of the 2,025 R waves processed, 557 (27.5%) were marked as suitable for shock delivery. In addition, 69 therapeutic and 11 test shocks were delivered during AF. All shock markers and shocks were delivered synchronously with the R wave, and the synchronization criterion was never violated.
Conclusions. Atrial fibrillation can be detected in a specific fashion using the RA-CS lead configuration and serial detection algorithms for atrial sensing. The delivery of properly timed shocks is feasible and should minimize the risk of ventricular proarrhythmia.
- Received January 19, 1996.
- Revision received July 18, 1996.
- Accepted July 31, 1996.