Abstract 2436: Automatic Determination of the Upper Limit of Vulnerability Using ICD Electrograms
Background: The upper limit of vulnerability (ULV) is the weakest shock that does not induce VF in the vulnerable period. It correlates with the defibrillation threshold and thus permits assessment of ICD defibrillation safety margins without inducing VF in most patients (pts). To determine the ULV, T-wave shocks must time at the most vulnerable intervals (corresponding to the strongest shock that induces VF), which are estimated using multiple ECG leads. To automate the ULV method, these intervals must be identified from an ICD electrogram (EGM). Our goals were to determine the range of most vulnerable intervals and to compare the accuracy of estimating them using timing points based on either ECG or EGM.
Methods: At ICD implant in 22 pts, we determined the ULV and most vulnerable intervals at paced cycle length 500 ms. The vulnerable zone was scanned by shocks at 20 ms intervals. Shock strength was varied in 3 J steps from an initial value of 12 J. We used an external ``ICD in a box” connected to a dual-coil ICD lead and left-pectoral Active Can Emulator. All 12 ECG leads and unfiltered EGMs were recorded on optical disk. EGMs were processed to reduce noise and differentiated to measure the maximum of the first derivative (dV/dtmax). Testing included 12.1±3.6 T wave shocks and 2.4±0.9 VF episodes per patient. The ECG timing point was the maximum voltage (Vmax) of the latest-peaking T wave in 12 leads. The EGM timing point was dV/dtmax of the T wave on the RV Coil - Can EGM. We computed timing points for each T-wave shock and compared the performance of optimal four-shock T-wave scans based on the ECG and EGM.
Results: The range of most vulnerable intervals was 20 – 80 ms (41± 17 ms). It was only 20 ms in 14 pts (64%). At least one of these intervals was identified by both the EGM and ECG scans in 20 pts (91%). With either method, the maximum error in estimated ULV was 3 J. The optimal four-shock scan extended from - 20 ms to + 40 ms relative to either timing point.
Conclusions: Timing T-wave shocks relative to EGM dV/dtmax identifies the most vulnerable intervals as accurately as the clinically-used, ECG-based method. An automated method can improve ease of determining the ULV without sacrificing accuracy.