Sources of off-target effects of vagus nerve stimulation using the helical clinical lead in domestic pigs

Objective. Clinical data suggest that efficacious vagus nerve stimulation (VNS) is limited by side effects such as cough and dyspnea that have stimulation thresholds lower than those for therapeutic outcomes. VNS side effects are putatively caused by activation of nearby muscles within the neck, via direct muscle activation or activation of nerve fibers innervating those muscles. Our goal was to determine the thresholds at which various VNS-evoked effects occur in the domestic pig—an animal model with vagus anatomy similar to human—using the bipolar helical lead deployed clinically. Approach. Intrafascicular electrodes were placed within the vagus nerve to record electroneurographic (ENG) responses, and needle electrodes were placed in the vagal-innervated neck muscles to record electromyographic (EMG) responses. Main results. Contraction of the cricoarytenoid muscle occurred at low amplitudes (~0.3 mA) and resulted from activation of motor nerve fibers in the cervical vagus trunk within the electrode cuff which bifurcate into the recurrent laryngeal branch of the vagus. At higher amplitudes (~1.4 mA), contraction of the cricoarytenoid and cricothyroid muscles was generated by current leakage outside the cuff to activate motor nerve fibers running within the nearby superior laryngeal branch of the vagus. Activation of these muscles generated artifacts in the ENG recordings that may be mistaken for compound action potentials representing slowly conducting Aδ-, B-, and C-fibers. Significance. Our data resolve conflicting reports of the stimulation amplitudes required for C-fiber activation in large animal studies (>10 mA) and human studies (<250 μA). After removing muscle-generated artifacts, ENG signals with post-stimulus latencies consistent with Aδ- and B-fibers occurred in only a small subset of animals, and these signals had similar thresholds to those that caused bradycardia. By identifying specific neuroanatomical pathways that cause off-target effects and characterizing the stimulation dose-response curves for on- and off-target effects, we hope to guide interpretation and optimization of clinical VNS

- Transection of the two somatic branches of the vagus were performed to test the hypothesis that vagus nerve stimulation causes action potentials that travel along the branches of the vagus into neck muscles, thus causing the apparent VNS-evoked neck muscle contractions. Transection of the recurrent laryngeal (RLT) always removed the long-latency EMG component and ENG signals with identical latencies (EMG artifacts). Transection of the superior laryngeal (SLT) always removed the short-latency EMG component and ENG signals with identical latencies (EMG artifacts). These experiments confirmed that VNS-evoked neck muscle contractions (and corresponding EMG signals) occur due to action potential signaling along the vagus nerve branches, as opposed to direct muscle activation. These experiments also suggest that neck muscle contractions can create EMG artifacts in ENGs that might be mistaken for nerve fiber signals.
Transection of the main vagus trunk after transection of the vagus somatic branches was performed to confirm that remaining ENG signals were caused by action potentials elicited at the stimulation electrode cuff that move down the vagus to where the LIFEs were located, as opposed to some unidentified artifact. Transection of the vagus trunk cranial ('Cr'anial Transection, CrT) to the stimulation electrode had no effect on any signals recorded. This is expected because action potentials generated under the cuff could still travel caudal to the cuff, which is where the LIFEs were located. Transection of the vagus trunk caudal ('Ca'udal Transection, CaT) to the stimulation electrode always removed all remaining signals in the ENG recordings. This suggests the ENG signals recorded at the LIFEs prior to transection of the main vagus trunk were indeed action potentials evoked by electrical stimulation at the stimulation electrode cuff, as opposed to some unidentified artifact. The stimulation artifact remains unchanged across conditions. RLT not performed.

Supplementary
Only animal with no apparent long-component EMG. Since we did not isolate the recurrent laryngeal branch for transection, there is a chance the recurrent was severed accidentally during surgical cutdown.
Short-component EMG is small but present, see zoom below comparing EMG "RLT" to "RLT+SLT".
Good example of how sometimes no EMG artifacts will happen in ENG signals despite large EMG components.
Two temporal zooms are presented. First, 0 to 15 seconds to match the rest of the animals. Second, 0 to 30 seconds to show the possible C-fiber, which was not observed in any of the other animals. Note that the possible C-fiber remains after both branch transections and is eliminated following transection of the main vagus trunk caudal to the stimulating electrode.
Only animal that SLT was performed before RLT. Respective effects on short-and longcomponent EMG signals is the same as RLT then SLT in other animals; RLT removes longcomponent EMG and SLT removes short-component EMG. Long-component EMG is small but present; see zoom comparing "SLT" to "SLT+RLT" below.
ENG fiber signals were not clear in the intact condition for this animal prior to transections.
ENG and EMG traces at every amplitude for intact (no transections and no vecuronium) and vecuronium, if applied, conditions. Arrows show identified thresholds for ENG fiber types and EMG components. EMG traces without vecuronium were always used to determine EMG component thresholds. ENG traces with vecuronium were used to determine ENG thresholds if vecuronium was applied, otherwise traces without vecuronium were used for determining Aα/Aβand Aγ-fiber thresholds since the latencies for these signals were always smaller than the fastest short-component EMG in any animal. Aδ/B-fiber signals could have the same latency as short-or long-component EMG signals for any given animal, so Aδ/B-fiber signals were not counted for animals without application of vecuronium. Red arrows in vecuronium traces indicate EMG signals and corresponding EMG-artifacts in ENGs that occurred despite neuromuscular junction block; incomplete block sometimes occurred at the end of stimulation dose response curves (random) as vecuronium was cleared or at higher amplitudes due to vecuronium being a competitive inhibitor.
Black arrows in ENG plots refer to Aα/Aβ, Aγ, and Aδ/Bin that orderfrom bottom to top, lowest amplitude to highest amplitude since the thresholds for each fiber were consistently Aα/Aβ < Aγ < Aδ/B. Likewise, black arrows in EMG plots refer to long-component and shortcomponentin that orderfrom bottom to top since the thresholds for each component were consistently long-component < short-component.
If available, a representative histological slice of the vagus nerve below the stimulation electrode is shown for each animal.
Two temporal zooms are presented. First, 0 to 15 seconds to match the rest of the animals. Second, 0 to 30 seconds to show the possible C-fiber, which was not observed in any of the other animals (hence not shown for any other animals).
Vecuronium was not used in this animal, though branch transections show that short and long EMG components can be completely removed leaving observable Aα/Aβ-, Aγ-, and C-fiber signals, which are then removed by transection of the main vagus trunk.
Some rational behind hesitation on C-fiber identification. No vecuronium in this animal, only branch transections; C-fiber signal could be caused by some unidentified motor pathway. Thresholds for all signals were generally higher in this animal than the rest of the cohort, which should mean any C-fiber should have a higher threshold than any other animal, yet this is the only animal with a C-fiber signal. No observation of Aδ/Bsignals, which should have lower thresholds than C-fibers.
From our perspective, the only explanation is that the ENG LIFEs were placed directly into fascicles containing C-fibers and were far away from fascicles containing Aδor B-fibers. Hence the hesitation. Smoothed version was used to calculate changes in all animals. The median of the 30 seconds preceding stimulation was used as the baseline value. The median of 5 data points surrounding the minimum value of the 30 seconds during stimulation was subtracted from the baseline value to calculate the change in heart rate due to stimulation.