Introduction: Automated implantable cardioverter defi brillator (AICD) interventions have the potential to be proarrhythmogenic. This is related to suboptimal programming and technical limitations of the device. We sought to categorize the stored events in Boston Scientifi c Latitude remote monitoring system to appropriate and inappropriate shocks and identify the proarrhythmic event. Methods: Currently patients with an AICD are monitored and stored remotely. We reviewed the Boston Scientifi c Latitude database for stored events and categorized them. Shocks delivered for deleterious arrhythmias (ventricular fi brillation) were considered appropriate. Shocks delivered for relatively benign arrhythmias (ex-sinus tachycardia treated with anti-tachycardic pacing) were considered inappropriate. Worsening of baseline arrhythmia secondary to implantable cardioverter-defi brillator treatment is considered proarrhythmic (ex-sustained ventricular tachycardia [VT] treated with shock resulting in ventricular fi brillation). Results: Of the 3049 stored events, 380 shock events were identifi ed. Among them, 132 events were induced during AICD implantation for testing purposes and thus excluded. One hundred and eighty were considered appropriate as the device shocked them out of sustained VT. Nine events were considered appropriate and proarrhythmic. Fifty-fi ve were considered inappropriate as the shock was delivered for supraventricular arrhythmia. Four events were considered inappropriate and proarrhythmic as the AICD shocked the clinically inappropriate rhythm resulting in worsening of arrhythmia. Conclusion: We noted 13 proarrhythmic events. Most of them are due to inadvertent recognition of supraventricular arrhythmia and delivering of therapies. Proarrhythmogenicity can be minimized by careful programming of the AICD.