Modulating and Monitoring Autonomic Nerves for Glycemic Control

Abstract

Diabetic patients suffer from a long-term condition that results in high blood glucose levels (hyperglycemia). Many medications for diabetes lose their glycemic control effectiveness over time and patient compliance to these medications is a major challenge. Glycemic control is a vital continuous process and is innately regulated by the endocrine and autonomic nervous systems. There is an opportunity for developing an implantable and automated treatment for diabetic patients by accurately detecting and altering neural activity in autonomic nerves. Renal nerves provide neural control for glucose reabsorption in the kidneys, and the vagus nerve conveys important glucose regulation signals to and from the liver and pancreas. This dissertation investigated stimulation of renal nerves for glycemic control, assembled an implantation procedure for neural interface arrays designed for autonomic nerves, and recorded physiological action potential signals in the vagus nerve. In a first study, stimulation of renal nerves in anesthetized, normal rats at kilohertz frequency (33 kHz) showed a notable average increase in urine glucose excretion (+24.5%). In contrast, low frequency (5 Hz) stimulation of renal nerves showed a substantial decrease in urine glucose excretion (−40.4%). However, these responses may be associated with urine flow rate. In a second study, kilohertz frequency stimulation (50 kHz) of renal nerves in anesthetized, diabetic rats showed a significant average decrease (-168.4%) in blood glucose concentration rate, and an increase (+18.9%) in the overall average area under the curve for urine glucose concentration, with respect to values before stimulation. In a third study, an innovative procedure was assembled for the chronic implantation of novel intraneural MIcroneedle Nerve Arrays (MINAs) in rat vagus nerves. Two array attachment approaches (fibrin sealant and rose-bengal bonding) were investigated to secure non-wired MINAs in nerves. The fibrin sealant approach was unsuccessful in securing the MINA-nerve interface for 4- and 8-week implant durations. The rose-bengal coated MINAs were in close proximity to axons (≤ 50 μm) in 75% of 1-week and 14% of 6-week implants with no significant harm to the implanted nerves or the overall health of the rats. In a fourth study, physiological neural activity in the vagus nerve of anesthetized rats was recorded using Carbon Fiber Microelectrode Arrays (CFMAs). Neural activity was observed on 51% of inserted functional carbon fibers, and 1-2 neural clusters were sorted on each carbon fiber with activity. The mean peak-to-peak amplitudes of the sorted clusters were 15.1-91.7 µV with SNR of 2.0-7.0. Conducting signals were detected in the afferent direction (0.7-1.0 m/sec conduction velocities) and efferent direction (0.7-8.8 m/sec). These conduction velocities are within the conduction velocity range of unmyelinated and myelinated vagus fibers. Furthermore, changes in vagal nerve activity were monitored in breathing and blood glucose modulated conditions. This dissertation, to our knowledge, was the first to demonstrate glucose regulation benefits by stimulation of renal nerves, chronically implant intraneural arrays in rat vagus nerves, and record physiological action potential in vagus nerves using multi-channel intraneural electrodes. Future work is needed to evaluate the long-term glucose regulation benefits of stimulation of renal nerves, and assess the tissue reactivity and recording integrity of implanted intraneural electrodes in autonomic nerves. This work supports the potential development of an alternative implantable treatment modality for diabetic patients by modulating and monitoring neural activity in autonomic nerves.