The Implantable Atrial Defibrillator: Are All the Problems Solved

Abstract

Atrial fibrillation (AF) is an extremely common arrhythmia seen in clinical practice, as shown by a number of epidemiologic studies. A significant number of patients either have recurrent attacks of atrial fibrillation despite pharmacological therapy, or are controlled, but complain of intolerable side effects. The atrial defibrillator (AID), a device capable of detecting and automatically terminating AF, may be an interesting non-pharmacological therapy of atrial fibrillation. Two devices are currently available for clinical evaluation: The Metrix system (In Control Redmond, USA) which is a stand-alone defibrillator, and the Medtronic 2030 which is a double-chamber defibrillator under evaluation solely in patients in whom a ventricular defibrillator is indicated, and who have paroxysmal atrial fibrillation. This presentation will focus on the Metrix system aimed at the trat-ment of patients whose major clinical problem is atrial fibrillation. This type of system has to deal with four issues: (1) the feasibility of atrial defibrillation, (2) the possible shock-related discomfort, (3) the safety of the device, and (4) the proper identification of patients who might benefit from the device. Several studies have shown that atrial defibrillation using low-energy shocks between two intracardiac catheters, in the coronary sinus and in the right atrium is feasible in patients with persistent spontaneous atrial fibrillation. As the minimum energy requirement in order to successfully terminate atrial fibrillation in 75% of patients averages 200-300 V (2–3 J), the conversion voltage needs to be above these values in order to have a satisfactory safety margin. Termination of atrial fibrillation with energy levels of less than 1 joule was found to be associated with little, if any discomfort. A good correlation was also found in our study between the level of discomfort and increasing voltage. The difference was statistically significant (p < 0.02) between 140 V and 220 V shocks and 220 V and 300 V shocks (p < 0.01). A marked inter-individual variation was noted. Therefore, this aspect should be tested before indicating the device in a given patient. Both in the XAD study and the Metrix study with the physician-activated device in more than 50 patients and more than 3000 shocks, no ventricular proarrhythmia was observed with synchronized shock. As shocks delivered following short (below 300 ms) RR intervals were associated with a low but definite risk of ventricular fibrillation, the shocks were delivered after an RR interval of 500 ms or longer.