Reply to comment on: Short pulse width widens the therapeutic window of subthalamic neurostimulation

M. Reich,J. Volkmann

DOI: https://doi.org/10.1002/acn3.252

IF: 5.43

2015-09-17

Annals of Clinical and Translational Neurology

Abstract:We have read with interest the comment of McDermott et al.1 on our article “Short pulse width widens the therapeutic window of subthalamic neurostimulation”.2 While the authors provide an elegant and sophisticated argument trying to prove our data wrong, they disregard a simple rule of experimentation from the viewpoint of theoretical biophysics: Deep brain stimulation systems do not have a “charge knob” to control the output of the neurostimulator. As correctly pointed out by McDermott et al., charge is the product of pulse duration and current. Both variables are controlled independently by the programmer and may be used to improve the selectivity of neurostimulation within a mixed nervous tissue containing target elements of deep brain stimulation (DBS) as well as side-effect elements. Our article describes from a clinical and experimental point of view, how to use the chronaxie relationship in clinical programming in order to avoid typical side effects of neurostimulation such as dysarthria or impaired fine motor skills by inadvertent stimulation of the pyramidal tract. In this experiment, the impact of a single variable (pulse duration) on the outcome variable current was assessed based on a threshold response. Their revised graph was simply showing the theoretical relationship between charge and pulse duration (which is trivial, because charge is the product of current amplitude and pulse duration as pointed out above), but was not experimentally assessed. We have a long experience in teaching novel programmers in adjusting the parameters of neurostimulation, which turns out to be a difficult task for clinicians, and had to perform multiple troubleshooting sessions ourselves in patients with poorly programmed DBS systems. The theoretical sophisticism of McDermott et al. adds confusion rather than providing a much needed practical guideline for DBS programming. We emphasize based on our experimental data and our clinical experience that DBS systems should be programmed treating pulse duration as a “constant” fixed at a low value (30–60 μsec depending on the hardware used) while using current or voltage for “dosing” the therapeutic response irrespective of any theoretical charge-pulse duration relationship.