Effect of dispersion in nerve on compound action potential and impedance change: a modelling study.

Objective: Electrical impedance tomography (EIT) is capable of imaging fast compound electrical activity (compound action potentials, or CAPs) inside peripheral nerves. The ability of EIT to detect impedance changes (dZ) which arise from the opening of ion channels during the CAP is limited by the dispersion with distance from the site of onset, as fibres have differing conduction velocities. The effect is largest for autonomic nerves mainly formed of slower conducting unmyelinated fibres where signals cannot be recorded more than a few cm away from the stimulation. However, as CAPs are biphasic, monophasic dZ are expected to be detectable further than them; testing this hypothesis was the main objective of this study.
Approach: An anatomically accurate FEM model and simplified statistical models of 50-fibre Hodgkin-Huxley and C-nociceptor nerves were developed with normally distributed conduction velocities; the statistical models were extended to realistic nerves.
Main Results: Fifty-fibre models showed that dZ could persist further than biphasic CAPs, as these then cancelled. For realistic nerves consisting of Aα or Aβ fibres, significant dZ could be detected at 50 cm from the onset site with signal-to-noise ratios (SNR, mean  ±  s.d.) of 2.7  ±  0.2 and 1.8  ±  0.1 respectively; Aδ and rat sciatic nerve-at 20 cm (1.6  ±  0.03 and 1.6  ±  0.06), rat vagus-at 10 cm (1.6  ±  0.05); C fibres-at 1-2 cm (2.4  ±  0.02).
Significance: This study provides a basis for determining the distance over which EIT may be used to image fascicular activity in electroceuticals and suggests dZ will persist further than CAPs if biphasic.