par Vande Perre, Louis 
Président du jury Haut, Benoît
Promoteur Nonclercq, Antoine
Co-Promoteur Gorza, Simon-Pierre
Publication Non publié, 2025-03-26
Président du jury Haut, Benoît
Promoteur Nonclercq, Antoine
Co-Promoteur Gorza, Simon-Pierre
Publication Non publié, 2025-03-26
Thèse de doctorat
| Résumé : | Neurostimulation has been pivotal in advancing treatments for neurological disorders such as chronic pain, epilepsy, and neurodegenerative diseases, providing therapeutic options for patients unresponsive to conventional methods. Among the established techniques, electrical stimulation (ES) remains the most widely employed. However, ES presents limitations, including stimulation artifacts, low spatial selectivity, non-physiological activation of nerve fibers, and limited compatibility with Magnetic Resonance Imaging (MRI). In contrast, Infrared Neural Stimulation (INS), which uses transient near-infrared light to activate neurons is a promising alternative to overcome some of these challenges. Using light to stimulate nerves avoids stimulation artifacts in electrical recording and achieves spatial specificity due to high light absorption in tissue. It also eliminates the need for metal leads, making it compatible with MRI, and the order in which nerve fibers are recruited may differ between INS and ES. However, the exact mechanism of action of INS is not fully understood and several hypotheses, thermal or otherwise, have been proposed.This thesis aims to investigate the underlying mechanisms of INS in peripheral nerves, by studying the thermal requirements of nerve activation and identifying the recruitment of nerve fibers with INS. To address these questions, an experimental approach was adopted, combining customized optical stimulation setups and quantitative data analysis. The developed setup allows simultaneous measurements of surface temperature changes and the conduction velocities (CV) of the activated nerve fibers.The first part of this thesis investigates the thermal dynamics during INS in rat sciatic nerves. The use of a thermal camera allowed the monitoring of spatio-temporal temperature gradients at the surface of the nerve. As previous studies have shown the importance of the temporal temperature gradient in triggering an action potential, we were interested in the effect of heat accumulation and the resulting spatial temperature gradient on action potential generation. Our experiments show that the initiation of compound nerve action potentials (CNAPs) is primarily dependent on spatial temperature gradients rather than basal temperature increases. Using a 10 Hz, 4 mJ pulsed 1470 nm laser, a threshold spatial gradient of 20.0±5.4 °C/mm was identified, with CNAP amplitude progressively increasing with local heat accumulation. Further experiments combining a low-power continuous wave (CW) laser to induce heat accumulation with low repetition rate pulse stimulation successfully evoked CNAPs. However, when the basal temperature of the nerve was increased to test its effect on CNAP generation, it did not facilitate activation. These results highlight the critical role of spatial temperature gradients over absolute temperature in nerve activation. Heat transfer modeling supported the experimental results, suggesting that progressive heat accumulation enhances spatial gradients, leading to progressive fiber recruitment and the CNAP amplitude increase observed experimentally. These results challenge the hypothesis that temperature-sensitive channels may be the primary mechanism of INS. Nevertheless, our results are consistent with other hypotheses about the underlying mechanism of INS.The second part investigates fiber recruitment during INS using CV measurements to differentiate activated fiber types. Two animal models were used. In addition to the previously used rat sciatic nerve, the goat cervical vagus nerve was chosen because of its higher proportion of small-diameter fibers compared to other peripheral nerves and because of its greater length compared to the rat vagus nerve, allowing sufficient separation between recordings. In the ex vivo rat sciatic nerve at room temperature, CNAP obtained with INS showed an average CV of 8.10±2.82 m/s, which is lower than that of ES (9.81±3.18 m/s). It suggests preferential activation of smaller fibers. However, in the goat vagus nerve, INS did not evoke discernible CNAPs. This highlights the challenges of applying INS to larger nerves, where connective tissue layers are thicker and organization into fascicles is expected to reduce light penetration to the nerve fibers. This work represents the first use of CV measurements to evaluate fiber specificity in INS and provides fundamental insights for selective neuromodulation applications.Overall, this study presents an experimental approach that advances the understanding of the spatial temperature gradients required for effective INS. Future research could pursue exploring the underlying mechanism behind INS and, more specifically, the potential specificity of INS for small-diameter fibers. While we have successfully used CV measurements to gain insight into INS specificity, future experiments could compare the functional outcomes of INS to ES and establish its potential for selective neuromodulation therapies. |
