Conduction block in myelinated axons induced by high-frequency (kHz) non-symmetric biphasic stimulation

Abstract

This study used the Frankenhaeuser–Huxley axonal model to analyze the effects of non-symmetric waveforms on conduction block of myelinated axons induced by high-frequency (10–300 kHz) biphasic electrical stimulation. The results predict a monotonic relationship between block threshold and stimulation frequency for symmetric waveform and a non-monotonic relationship for non-symmetric waveforms. The symmetric waveform causes conduction block by constantly activating both sodium and potassium channels at frequencies of 20–300 kHz, while the non-symmetric waveforms share the same blocking mechanism from 20 kHz up to the peak threshold frequency. At the frequencies above the peak threshold frequency the non-symmetric waveforms block axonal conduction by either hyperpolarizing the membrane (if the positive pulse is longer) or depolarizing the membrane (if the negative pulse is longer). This simulation study further increases our understanding of conduction block in myelinated axons induced by high-frequency biphasic electrical stimulation, and can guide future animal experiments as well as optimize stimulation parameters that might be used for electrically induced nerve block in clinical applications.

Non-monotonic kilohertz frequency neural block thresholds arise from amplitude- and frequency-dependent charge imbalance

“…There were no instances of anodal block in our computational models, and we also did not observe polarity effects in charge-balanced waveforms or waveforms with the lowest imbalances tested. Our data confirm the predictions from prior modeling studies 21,22 that reported non-monotonic block thresholds with frequency in charge-imbalanced asymmetric waveforms with a monopolar point source electrode in a homogeneous medium. However, we used more realistic preclinical computational models and validated our findings with in vivo experiments.…”

“…Meanwhile, changes in KHF amplitude needed for re-excitation occurred because virtual DC cathodes (or virtual DC anodes) at the distal contact strengthened (or weakened) the depolarization at the virtual cathodes of the KHF signal, which are the source of KHF re-excitation 42 . The observed polarity effects on block thresholds were consistent with in vivo DC block measurements from a previous study that used both monopolar and bipolar cuffs 43 and with prior modeling of monopolar electrodes 22 . There were no instances of anodal block in our computational models, and we also did not observe polarity effects in charge-balanced waveforms or waveforms with the lowest imbalances tested.…”

“…Phase differences for asymmetric waveforms were ± 2, ± 3, and ± 4 μs, enabling direct comparison between DC offset symmetric waveforms and charge-imbalanced asymmetric waveforms. The phase differences, in turn, were in a range similar to previous modeling work on asymmetric charge-imbalanced waveforms 22 , facilitating comparisons of the present symmetric and asymmetric work to previous studies. We evaluated all waveforms in vivo at 20, 40, 60, and 80 kHz.…”

“…Previous computational modeling studies 22 evaluated block thresholds of asymmetric rectangular waveforms, corresponding to hypothetical nerve block instruments that generate waveforms with unintended asymmetry. While such waveforms produced non-monotonic block thresholds, the individual contributions of asymmetry and charge imbalance were unclear.…”

Modeling the response of small myelinated axons in a compound nerve to kilohertz frequency signals

“…However, if we were to test higher kilohertz frequencies (100 kHz and above) that have been examined in other publications (e.g. [62,63]), then the quasi-static assumption may no longer be appropriate. Furthermore, there is a lack of available data on the permittivities and frequency-dependent conductivities of the tissues in compound peripheral nerves [64].…”