DEFIBRILLATION USING A HIGH-FREQUENCY SERIES OF MONOPHASIC RECTANGULAR PULSES - OBSERVATIONS AND MODEL PREDICTIONS

Introduction: Capacitor-discharge type waveforms are practical for def ibrillation devices but may not be optimum. Discharging a capacitor as a series of high-frequency (HF) pulses may allow effective waveform s haping by modulating the pulses. This approach could lead to improved defibrillation by allowing waveforms that would otherwise be unachieva ble with a capacitor-discharge approach. However, little is known abou t defibrillation with HF. Methods and Results: In open chest pentobarb ital anesthetized dogs, we measured defibrillation thresholds for cont inuous rectangular waveforms with 5-, 10-, and 20-msec durations and f or 10- and 20-msec long series of HF rectangular pulses. HF series had a 50% ''on-time'' duty cycle at 100 Hz to 20 kHz. At 1 kHz and above, defibrillation with HF required the same time-averaged current but ap proximately twice the peak current and energy as defibrillation with c ontinuous waveforms having the same envelope duration, At lower freque ncies, defibrillation peak current and energy approached values requir ed for the continuous waveforms. While waveforms were not actually fil tered, the heart responded as though the HF series were low-pass filte red, A filtered effective waveform model with a 3.7-msec time constant predicts these HF data and makes reasonable predictions for various c ontinuous waveform shapes. Conclusion: Defibrillation is possible usin g HF pulses up to 20 kHz and has a frequency response similar to a low -pass filter. A filtered effective waveform model predicts these HF re sults and may help explain how waveforms influence defibrillation effi cacy. While the unmodulated HF pulsing used in this study increased de fibrillation requirements, these findings support the concept that HF pulse modulation can be used to change the effective shape of a wavefo rm, which could permit more efficacious waveform shapes and a net redu ction of thresholds.