Abstract 2645: Longer Ventricular Fibrillation Detection Duration Does Not Increase Rate of Undersensing: Results from the PREPARE Study
Background: The PREPARE study was designed to evaluate ICD detection and therapy parameters strategically selected to safely reduce all-cause shocks in patients with primary prevention indications by treating only truly sustained fast VT/VF. Electrophysiologic properties have been shown to deteriorate during long-duration VF episodes. There was concern that electrogram deterioration might result in increased undersensing (US) and delayed detection of VF. In this analysis we compared the rate of US between episodes detected with shorter (number of intervals to detect (NID) =12/16) and longer (NID=30/40) detection settings to assess whether delayed detection times for VF increased the rate of US.
Methods: All induced VT/VF episodes from PREPARE patients programmed to NID 12/16 or 30/40 and sensitivity 1.2 mV were included in the analysis. US was determined automatically from the stored episode record. The numbers of US and the times from onset through initial detection and from detection to shock were computed. The rates of US over time were compared between the shorter and longer NIDs using a GEE analysis with time as the offset.
Results: Seven hundred fifty six VF inductions in 432 patients from the PREPARE study were included in the analysis. There were 319 VF inductions (180 pts) with NID 12/16 and 437 VF inductions (261 pts) with NID 30/40. The average time from VF onset to shock increased from 6.0 to 9.5 seconds with the longer VFNID with no evidence of an increase in the rate of US over time.
Conclusion: The longer NID delayed shock delivery by an average of 3.5 seconds but did not result in an increased rate of undersensing.