Device interrogations
We reviewed all available device interrogations for the 44 patients included in this study, the dates of which ranged from 11/13/14 through 4/8/19. Devices included were single-chamber pacemakers, dual-chamber pacemakers, single-chamber ICDs, biventricular pacemakers and defibrillators, and subcutaneous ICDs. All interrogations, obtained both in the ambulatory and inpatient settings, were reviewed. For each CIED, the device type and implantation date were recorded. We calculated total number of transmissions and monitoring hours for all patients in the study. For VA analysis, the following were recorded: number of VT/VF events, number of non-sustained VT (NSVT) events, and ventricular pacing percentage. We defined VT as a wide-complex, regular tachycardia having a duration greater than thirty seconds and/or requiring either anti-tachycardia pacing (ATP) or ICD shock for termination. Thus, NSVT was defined as a wide-complex regular tachycardia lasting at least three beats but not meeting aforementioned VT criteria. For each NSVT, VT, and VF episode, the following details were recorded: date, time, day of week, duration, average atrial rate, average ventricular rate, and therapy requirement (i.e. ATP, ICD shock, or both).
In order to compare overall VA burden between groups, we calculated the total number of VA events divided by the product of the total number of patients per group experiencing any VA event and total number of monitoring hours for those patients. This resulted in events per patient-hour for both individual patients and for each group. We chose this calculation in order to account for the fact that patients monitored for longer periods of time may represent a survival bias.
In an attempt to better categorize ventricular pacing data, four quartiles for each group (0-25%, 26-50%, 51-75%, and 76-100%) were created and represented the pacing percentage reported since last device interrogation. Individual counts represented a device interrogation reporting a particular ventricular pacing percentage in the range of the specified quartile. Biventricular devices and subcutaneous ICDs were excluded from ventricular pacing analysis.