(30) Impact of Electrode Geometry on Off-Target Responses to Baroreceptor Activation Therapy

Introduction: Over 1.3 billion people worldwide have hypertension which can cause heart disease, stroke, kidney dysfunction, and premature death [1]. In 2019, the FDA approved baroreceptor activation therapy (BAT) to treat drug-resistant hypertension with an implanted device that electrically activates baroreceptors, which are stretch-sensitive neurons that detect changes in blood pressure (BP). BAT successfully reduced BP into a healthy range in 63% of patients [2], but it is often accompanied by intolerable side effects (neck discomfort, jaw pain) [3]. We implemented the first computational model of BAT, and we identified stimulation parameters and electrode geometries that increase activation thresholds for off-target pathways (i.e., reduce side effects).

Methods: We analyzed images and videos of a pig microdissection to parameterize a three-dimensional computational model of the carotid arteries and nearby ( < 2 cm) nerves including vagus, glossopharyngeal, superior laryngeal, hypoglossal, and spinal accessory (Figure 1). We used a mono-fascicular representation of each nerve, matching the total fascicular area from histological images. We evaluated responses for four monopolar geometries (Figure 2A) and three bipolar electrode configurations, each composed of two standard monopolar electrodes (Figure 2B ). The monopolar electrode was placed at the apex of the bifurcation of the common carotid and ascending pharyngeal arteries, and in bipolar configurations, we varied the location of a second electrode while maintaining the location of the first electrode. We simulated activation thresholds for 13 µm fibers along the centerline of each nerve in response to monophasic 350 µs stimulation.

Results: With the monopolar electrode, activation thresholds for off-target fibers were 0.4 to 2.7 mA. Modifying the geometry of the monopolar electrode had little effect on thresholds (Figure 2A). Activation thresholds for off-target fibers were 35 to 1255% higher with the opposite-side bipolar electrode compared to standard monopolar (Figure 2B). Electrode-to-nerve distance had an important effect on activation thresholds; for example, standard monopolar thresholds increased by 53% when the glossopharyngeal nerve moved by ~4 mm.

Conclusion: Bipolar electrodes increased the stimulation amplitude required to activate off-target fibers in our model of pig BAT, and thus are expected to reduce side effects relative to the monopolar electrode used clinically and in our in vivo experiments. Conversely, extending the insulation of the monopolar electrode did not affect off-target activation thresholds. Developing a fiber model to represent baroreceptors is a critical next step to evaluate relative changes in activation thresholds for target and off-target pathways.