Online Bayesian optimization of vagus nerve stimulation.

Objective.In bioelectronic medicine, neuromodulation therapies induce neural signals to the brain or organs, modifying their function. Stimulation devices capable of triggering exogenous neural signals using electrical waveforms require a complex and multi-dimensional parameter space to control such waveforms. Determining the best combination of parameters (waveform optimization or dosing) for treating a particular patient's illness is therefore challenging. Comprehensive parameter searching for an optimal stimulation effect is often infeasible in a clinical setting due to the size of the parameter space. Restricting this space, however, may lead to suboptimal therapeutic results, reduced responder rates, and adverse effects.Approach. As an alternative to a full parameter search, we present a flexible machine learning, data acquisition, and processing framework for optimizing neural stimulation parameters, requiring as few steps as possible using Bayesian optimization. This optimization builds a model of the neural and physiological responses to stimulations, enabling it to optimize stimulation parameters and provide estimates of the accuracy of the response model. The vagus nerve (VN) innervates, among other thoracic and visceral organs, the heart, thus controlling heart rate (HR), making it an ideal candidate for demonstrating the effectiveness of our approach.Main results.The efficacy of our optimization approach was first evaluated on simulated neural responses, then applied to VN stimulation intraoperatively in porcine subjects. Optimization converged quickly on parameters achieving target HRs and optimizing neural B-fiber activations despite high intersubject variability.Significance.An optimized stimulation waveform was achieved in real time with far fewer stimulations than required by alternative optimization strategies, thus minimizing exposure to side effects. Uncertainty estimates helped avoiding stimulations outside a safe range. Our approach shows that a complex set of neural stimulation parameters can be optimized in real-time for a patient to achieve a personalized precision dosing.

Partial regression of peripapillary myelinated nerve fibers after non-arteritic anterior ischemic optic neuropathy.

A 71-year-old woman presented a non-arteritic anterior ischemic optic neuropathy in an optic nerve with previously registered superonasal peripapillary myelinated nerve fibers. Her past medical history was significant for controlled systemic hypertension, hyperlipidemia, and diabetes mellitus. The physiologic cup was absent in both optic discs. Non-arteritic anterior ischemic optic neuropathy mainly affected the temporal and inferior sectors of the peripapillary retinal nerve fiber layer, as could be demonstrated by retinal nerve fiber layer optical coherence tomography and optic disc optical coherence tomography angiography. Unlike other published reports, just a slight regression of the myelinated nerve fibers was observed after 1 year of follow-up. This occurred because ischemia mainly affected the temporal and inferior peripapillary sectors, whereas myelinated nerve fibers were superonasal to the optic disc.

Multiple averaged records to identify Aδ-fibers in sensory nerves.

BACKGROUND: Existing methods identify only ≈10 Aδ-fibers in human sensory nerves per recording. This study examines methods to increase the detection of Aδ-fibers. NEW METHOD: Two to 20 averages of 500 replicate responses to epidermal nerve stimulation are obtained. Pairs of different averages are constructed. Each pair is analyzed with algorithms applied to amplitude and frequency to detect replication of responses to stimulation as "simultaneous similarities in two averages" (SS2AVs) at ≥99.5th percentile of control. In a pair of averages the latencies of amplitude and frequency SS2AVs for the same response to stimulation may differ by ≤0.25 ms. Therefore, Aδ-fibers are identified by the 0.25 ms moving sum of SS2AV latencies of the pairs of averages. RESULTS: Increasing averages increases pairs of different averages and detection of Aδ-fibers: from 2 to 10 Aδ-fibers with two averages (one pair) to >50 Aδ-fibers with 12-20 averages (66-190 pairs). COMPARISON WITH EXISTING METHOD(S): Existing methods identify ≤10 Aδ-fibers in 10 averages/45 pairs with the medians of amplitude and frequency algorithms applied to all 45 pairs. This study identifies Aδ-fibers (i) by applying these algorithms at the 99.5th percentile of control, (ii) to each pair of averages and (iii) by the 0.25 ms sum of algorithm identified events (SS2AVs) in all pairs. These three changes significantly increase the detection of Aδ-fibers, e.g., in 10 averages/45pairs from 10 to 45. CONCLUSIONS: Three modifications of existing methods can increase the detection of Aδ-fibers to an amount suitable (>50 with ≥12 averages) for statistical comparison of different nerves.

Protocol for isolating and processing mouse sciatic nerve fibers for confocal immunohistochemistry.

Many motor and neurodegenerative diseases affect the peripheral nervous system (PNS). The myelinated axons in the sciatic nerves offer valuable insights into the pathology of these diseases. Here, we present a protocol for isolating and processing mouse sciatic nerves for confocal immunohistochemistry. We describe steps for mouse perfusion, removing and fixing the sciatic nerve, transferring nerves onto slides, staining, and imaging. This protocol can assist in characterizing pathologies of myelinated fibers resulting from diseases affecting the PNS. For complete details on the use and execution of this protocol, please refer to Chang et al. (2023).1.

Changes in myelinated nerve fibers induced by pulsed electrical stimulation: A microstructural perspective on the causes of electrical stimulation side effects.

Electrical brain stimulation technology is widely used in the clinic to treat brain neurological disorders. However, during treatment, patients may experience side effects such as pain, poor limb coordination, and skin rash. Previous reports have focused on the brilliant chapter on electrical brain stimulation technology and have not paid attention to patients' suffering caused by side effects during treatment. In this study, electrodes were arranged on the medulla oblongata. Pulsed electric fields of different frequencies were used to perform electrical stimulation to study the impact of electric fields on myelinated nerve fibers and reveal the possible microstructural origin of side effects. Transmission electron microscopy was used to analyze and quantify the changes in microstructure. The results illustrated that myelinated nerve fibers underwent atrophy under pulsed electric fields, with the mildest degree of atrophy under high-frequency (400 Hz) electric fields. Myelin sheaths experienced plate separation under pulsed electric fields, and a distinct laminar structure appeared. The microstructure changes may be related to the side effects of clinical electrical stimulation. This study can provide pathological possibilities for exploring the causes of the side effects of electrical stimulation and supply guidance for selecting electrical parameters for clinical electrical stimulation therapy from a distinctive perspective.

Microglial Refinement of A-Fiber Projections in the Postnatal Spinal Cord Dorsal Horn Is Required for Normal Maturation of Dynamic Touch.

Sensory systems are shaped in postnatal life by the refinement of synaptic connectivity. In the dorsal horn of the spinal cord, somatosensory circuits undergo postnatal activity-dependent reorganization, including the refinement of primary afferent A-fiber terminals from superficial to deeper spinal dorsal horn laminae which is accompanied by decreases in cutaneous sensitivity. Here, we show in the mouse that microglia, the resident immune cells in the CNS, phagocytose A-fiber terminals in superficial laminae in the first weeks of life. Genetic perturbation of microglial engulfment during the initial postnatal period in either sex prevents the normal process of A-fiber refinement and elimination, resulting in an altered sensitivity of dorsal horn cells to dynamic tactile cutaneous stimulation, and behavioral hypersensitivity to dynamic touch. Thus, functional microglia are necessary for the normal postnatal development of dorsal horn sensory circuits. In the absence of microglial engulfment, superfluous A-fiber projections remain in the dorsal horn, and the balance of sensory connectivity is disrupted, leading to lifelong hypersensitivity to dynamic touch.

Free water alterations in different inflammatory subgroups in schizophrenia.

BACKGROUND: Accumulating evidence suggests that inflammatory dysregulation both in blood and the brain is implicated in the pathogenesis of schizophrenia. Alterations in peripheral cytokines are not evident in all patients and there may be discrete altered inflammatory subgroups in schizophrenia. Recent studies using a novel and in vivo free-water imaging to detect inflammatory processes, have shown increased free water in white matter in schizophrenia. However, no studies to date have investigated the free water alterations in different inflammatory subgroups in schizophrenia. METHODS: Forty-four patients with schizophrenia and 49 controls were recruited. The serum levels of interleukin-1 beta (IL-1ß), IL-6, IL-10, and IL-12p70 were measured and used for cluster analysis with K-means and hierarchical algorithms. Diffusion tensor imaging (DTI) images were collected for all participants and voxel-wise free water and fractional anisotropy of tissue (FA-t) were compared between groups with Randomise running in FSL. Partial correlation analysis was employed to explore the association of the peripheral cytokine levels with free water. RESULTS: We identified two statistically quantifiable discrete subgroups of patients based on the cluster analysis of cytokine measures. The peripheral levels of IL-1ß (P < 0.001), IL-10 (P = 0.041), and IL-12p70 (P < 0.001) showed significant differences between the two different inflammatory subgroups. In the inflammatory subgroup with a predominantly higher IL-1ß level, increased free water values in white matter were found mainly in the left posterior limb of the internal capsule, posterior corona radiata, and partly in the left sagittal stratum. These affected areas did not overlap with the regions that showed significant free water differences between patients and healthy controls. In the inflammatory subgroup with lower IL-1ß levels, peripheral IL-1ß was significantly associated with free water values in white matter while no such association was detected in the patient group. CONCLUSIONS: Localized free water differences were demonstrated between the two identified inflammatory subgroups in our data, and free water appears to be a feasible in vivo neuroimaging biomarker guiding the target of inflammatory intervention and development of new therapeutic strategies in an individualized manner in schizophrenia.

Modulating functionally-distinct vagus nerve fibers using microelectrodes and kilohertz frequency electrical stimulation.

Modulation of functionally distinct nerve fibers with bioelectronic devices provides a therapeutic opportunity for various diseases. In this study, we began by developing a computational model including four major subtypes of myelinated fibers and one unmyelinated fiber. Second, we used an intrafascicular electrode to perform kHz-frequency electric stimulation to preferentially modulate a population of fibers. Our model suggests that fiber physical properties and electrode-to-fascicle distance severely impacts stimulus-response relationships. Large diameter fibers (Aα- and Aß-) were only minimally influenced by the fascicle size and electrode location, while smaller diameter fibers (Aδ-, B- and C-) indicated a stronger dependency.Clinical Relevance- Our findings support the possibility of selectively modulating functionally-distinct nerve fibers using electrical stimulation in a small, localized region. Our model provides an effective tool to design next-generation implantable devices and therapeutic stimulation strategies toward minimizing off-target effects.

Filtering property of myelinated internode can change neural information representability and might trigger a compensatory action during demyelination.

In this paper, for the first time, we showed that an Internode Segment (INS) of a myelinated axon acts as a lowpass filter, and its filter characteristics depend on the number of myelin turns. Consequently, we showed how the representability of a neural signal could be altered with myelin loss in pathological conditions involving demyelinating diseases. Contrary to the traditionally held viewpoint that myelin geometry of an INS is optimised for maximising Conduction Velocity (CV) of Action Potential (AP), our theory provides an alternative viewpoint that myelin geometry of an INS is optimised for maximizing representability of the stimuli a fibre is meant to carry. Subsequently, we show that this new viewpoint could explain hitherto unexplained experimentally observed phenomena such as, shortening of INS length during demyelination and remyelination, and non-uniform distribution of INS in the central nervous system fibres and associated changes in diameter of nodes of ranvier along an axon. Finally, our theory indicates that a compensatory action could take place during demyelination up to a certain number of loss of myelin turns to preserve the neural signal representability by simultaneous linear scaling of the length of an INS and the inner radius of the fibre.

Progressive Retinal Medullated Nerve Fiber Layer in a Young Boy.

This case report documents progression of myelinated retinal nerve fibers in a young boy from age 6 months to 2 years.

Modeling the excitation of nerve axons under transcutaneous stimulation.

Computational models enable a safe and convenient way to study the excitation of nerve fibers under external stimulation. Contemporary models calculate the electric field distribution from transcutaneous stimulation and the resulting neuronal response separately. This study uses finite element methods to develop a multi-scale model that couples electric fields within macroscopic tissue layers and microscopic nerve fibers in a single-stage computational framework. The model included a triaxial myelinated nerve fiber bundle embedded within a volume conductor of tissue layers to represent the median nerve innervating the forearm muscles. The model captured the excitability of nerve fibers under transcutaneous stimulation and their nerve-tissue interactions to a transient external stimulus. The determinants of the strength-duration curve, rheobase, and chronaxie for the proposed model had close correlations with in-vivo experimentation on human participants. Additionally, the excitability indices for the triaxial myelinated nerve fiber implemented using the finite element method agreed well with experimental data from the literature. The validity of the proposed model encourages its use for applications involving transcutaneous stimulation. Capable of capturing field distribution across realistic morphologies, the model can serve as a testbed to improve stimulation protocols and electrode designs with subject-level specificity.

Macular Imaging Characteristics in Children with Myelinated Retinal Nerve Fiber and High Myopia Syndrome.

Objectives: To investigate the macular imaging features in patients with unilateral myelinated retinal nerve fiber (MRNF) and high myopia syndrome. Materials and Methods: Six patients with unilateral MRNF and high myopia syndrome and 13 myopic controls were enrolled in this study. Spectral domain (SD) optical coherence tomography (OCT), SD enhanced depth imaging OCT, and OCT angiography (OCTA) imaging results of MRNF-affected eyes were compared with the fellow eyes and myopic controls. Results: All patients had abnormal foveal reflex and/or ectopia. No significant difference in retinal thickness parameters were noted between the groups. In OCT scans, posterior vitreous detachment (PVD) was observed in 4 out of the 6 MRNF-affected eyes. Regarding OCTA parameters, only a significant increase in acircularity index was noted in myelinated eyes (p=0.01). Conclusion: All patients demonstrated normal foveal contours, macular structure, and OCTA features except for a higher acircularity index. The incidence of PVD was notably increased in the myelinated eyes.

Micro-pharmacokinetics of lidocaine and bupivacaine transfer across a myelinated nerve fiber.

BACKGROUND: The aim of the present study was to predict the time to onset and duration of action of two local anesthetics (lidocaine and bupivacaine) based on experimental dimensions of a typical nerve and experimental octanol/water partition coefficients. METHODS: We began our compilation of experimental data with a numerical solution of the Smoluchowski equation for the transfer of lidocaine and bupivacaine across the axon membrane in the region of the node of Ranvier (axolemma) and across the Schwann cell. The difference between the aqueous and lipid environments of the neuron was simulated by including the coordinate-dependent chemical potential. In the second step, the permeation rates calculated using the diffusion equation were used to solve a system of four ordinary differential equations. This approach allowed us to simulate the cellular environment for a longer time and to compare our model with pharmacokinetic properties (time to onset and duration of action) of local anesthetics from the literature. The behavior of local anesthetics under physiological conditions and in case of local acidosis was also simulated. RESULTS: We demonstrated that local anesthetics cross the axolemma in a time span of less than 1 µs. The time to onset of action, controlled by diffusion from the epineurium to an axon with a typical distance of 500 µm, was 167 s and 186 s for lidocaine and bupivacaine, respectively. The calculated half-life, which is a measure of the duration of action, was 41 min and 328 min for lidocaine and bupivacaine, respectively. CONCLUSIONS: Duration of action is controlled by the storage capacity of lipophilic compartments around the axon, which is higher for bupivacaine but lower in local acidosis. For the latter case, the literature, including textbooks, provides a misinterpretation, namely that protonated species cannot penetrate the membrane.

Cerebral cortical regions always connect with each other via the shortest paths.

In human society, the choice of transportation mode between two cities is largely influenced by the distance between the regions. Similarly, when neurons communicate with each other within the cerebral cortex, do they establish their connections based on their physical distance? In this study, we employed a data-driven approach to explore the relationships between fiber length and corresponding geodesic distance between the fiber's two endpoints on brain surface. Diffusion-MRI-derived fiber streamlines were used to represent extra-cortical axonal connections between neurons or cortical regions, while geodesic paths between cortical points were employed to simulate intra-cortical connections. The results demonstrated that the geodesic distance between two cortical regions connected by a fiber streamline was greater than the fiber length most of the time, indicating that cortical regions tend to choose the shortest path for connection; whether it be an intra-cortical or extra-cortical route, especially when intra-cortical routes within cortical regions are longer than potential extrinsic fiber routes, there is an increased probability to establish fiber routes to connect the both regions. These findings were validated in a group of human brains and may provide insights into the underlying mechanisms of neuronal growth, connection, and wiring.

Saltatory conduction and intrinsic electrophysiological properties at the nodes of ranvier of Aα/ß-afferent fibers and Aα-efferent fibers in rat sciatic nerves.

Large-diameter myelinated fibers in sciatic nerves are composed of both Aα/ß-afferent fibers and Aα-efferent fibers to convey sensory and motor impulses, respectively, via saltatory conduction for rapid leg responses. Saltatory conduction and electrophysiological properties at the nodes of Ranvier (NRs) of these sciatic nerve fibers have not been directly studied. We used ex vivo sciatic nerve preparations from rats and applied patch-clamp recordings at the NRs of both Aα/ß-afferent fibers and Aα-efferent fibers in the sciatic nerves to characterize their saltatory conduction and intrinsic electrophysiological properties. The velocity and frequency of saltatory conduction in both types of fibers were similar. Resting membrane potentials (RMPs), input resistance, action potential (AP) threshold, and AP rheobase were also not significantly different at the NRs of the two types of fibers in the sciatic nerves. In comparison with Aα/ß-afferent fibers, Aα-efferent fibers in the sciatic nerves show higher amplitude and broader width of APs at their NRs. At the NRs of both types of fibers, depolarizing voltages evoked transient inward currents followed by non-inactivating outward currents, and the inward currents and non-inactivating outward currents at the NRs were not significantly different between the two types of fibers. Using AP-clamp, inward currents during AP upstroke were found to be insignificant difference, but amplitudes of non-inactivating outward currents during AP repolarization were significantly lower at the NRs of Aα-efferent fibers than at the NRs of Aα/ß-afferent fibers in the sciatic nerves. Collectively, saltatory conduction, ionic currents, and intrinsic electrophysiological properties at the NRs of Aα/ß-afferent fibers and Aα-efferent fibers in the sciatic nerves are generally similar, but some differences were also observed.

Validated computational models predict vagus nerve stimulation thresholds in preclinical animals and humans.

Objective.We demonstrated how automated simulations to characterize electrical nerve thresholds, a recently published open-source software for modeling stimulation of peripheral nerves, can be applied to simulate accurately nerve responses to electrical stimulation.Approach.We simulated vagus nerve stimulation (VNS) for humans, pigs, and rats. We informed our models using histology from sample-specific or representative nerves, device design features (i.e. cuff, waveform), published material and tissue conductivities, and realistic fiber models.Main results.Despite large differences in nerve size, cuff geometry, and stimulation waveform, the models predicted accurate activation thresholds across species and myelinated fiber types. However, our C fiber model thresholds overestimated thresholds across pulse widths, suggesting that improved models of unmyelinated nerve fibers are needed. Our models of human VNS yielded accurate thresholds to activate laryngeal motor fibers and captured the inter-individual variability for both acute and chronic implants. For B fibers, our small-diameter fiber model underestimated threshold and saturation for pulse widths >0.25 ms. Our models of pig VNS consistently captured the range ofin vivothresholds across all measured nerve and physiological responses (i.e. heart rate, Aδ/B fibers, Aγfibers, electromyography, and Aαfibers). In rats, our smallest diameter myelinated fibers accurately predicted fast fiber thresholds across short and intermediate pulse widths; slow unmyelinated fiber thresholds overestimated thresholds across shorter pulse widths, but there was overlap for pulse widths >0.3 ms.Significance.We elevated standards for models of peripheral nerve stimulation in populations of models across species, which enabled us to model accurately nerve responses, demonstrate that individual-specific differences in nerve morphology produce variability in neural and physiological responses, and predict mechanisms of VNS therapeutic and side effects.

Inflammation in multiple sclerosis: consequences for remyelination and disease progression.

Despite the large number of immunomodulatory or immunosuppressive treatments available to treat relapsing-remitting multiple sclerosis (MS), treatment of the progressive phase of the disease has not yet been achieved. This lack of successful treatment approaches is caused by our poor understanding of the mechanisms driving disease progression. Emerging concepts suggest that a combination of persisting focal and diffuse inflammation within the CNS and a gradual failure of compensatory mechanisms, including remyelination, result in disease progression. Therefore, promotion of remyelination presents a promising intervention approach. However, despite our increasing knowledge regarding the cellular and molecular mechanisms regulating remyelination in animal models, therapeutic increases in remyelination remain an unmet need in MS, which suggests that mechanisms of remyelination and remyelination failure differ fundamentally between humans and demyelinating animal models. New and emerging technologies now allow us to investigate the cellular and molecular mechanisms underlying remyelination failure in human tissue samples in an unprecedented way. The aim of this Review is to summarize our current knowledge regarding mechanisms of remyelination and remyelination failure in MS and in animal models of the disease, identify open questions, challenge existing concepts, and discuss strategies to overcome the translational roadblock in the field of remyelination-promoting therapies.

Morphological study of the phrenic nerve to determine a reference value for the myelinated fiber density in elderly individuals.

Phrenic nerves (PNs) play an important role in respiration; however, very few morphological studies have assessed them. This study aimed to provide control reference values, including the density of large and small myelinated PN fibers, for future pathological studies. We assessed a total of nine nerves from eight cases among consecutive autopsy cases registered to the Brain Bank for Aging Research between 2018 and 2019 (five men and three women, mean age 77.0 ± 7.0 years). The nerves were sampled distally, and their structures were analyzed using semi-thin sections stained with toluidine blue. The mean and standard deviation of the density of each myelinated fiber of the PN was 6908 ± 1132 fibers/mm2 (total myelinated fiber), 4095 ± 586 fibers/mm2 (large diameter myelinated fiber; diameter ≥7 µm), and 2813 ± 629 fibers/mm2 (small diameter myelinated fiber; diameter <7 µm). There was no correlation between myelinated fiber density and age. This study provides the density measurement of the human PN myelinated fiber, and these findings can be used as reference values for the PN in elderly individuals.

CIDP/autoimmune nodopathies with nephropathy: a case series study.

OBJECTIVE: The co-morbidity of chronic inflammatory demyelinating polyradiculoneuropathy (CIDP)/autoimmune nodopathies with nephropathy has been gradually known in recent years. This study was intended to explore the clinical, serological and neuropathological features of seven patients with CIDP/autoimmune nodopathies and nephropathy. METHODS: Among 83 CIDP patients, seven were identified with nephropathy. Their clinical, electrophysiological and laboratory examination data were collected. The nodal/paranodal antibodies were tested. The sural biopsies were performed in all the patients, and renal biopsies were operated in 6 patients. RESULTS: Six patients had chronic onsets and one had an acute onset. Four patients exhibited peripheral neuropathy preceding nephropathy while two showed concurrent onset of neuropathy and nephropathy, and one started with nephropathy. All the patients showed demyelination in electrophysiological examination. Nerve biopsies showed mild to moderate mixed neuropathies including demyelinating and axonal changes in all patients. Renal biopsies showed membranous nephropathy in all 6 patients. Immunotherapy was effective in all patients, with two patients showing good response to corticosteroid treatment alone. Four of the patients were positive to anti-CNTN1 antibody. Compared with anti-CNTN1 antibody-negative patients, antibody-positive patients had a higher proportion of ataxia (3/4 vs. 1/3), autonomic dysfunction (3/4 vs. 1/3), less frequent antecedent infections (1/4 vs. 2/3), higher cerebrospinal fluid proteins (3.2 g/L vs. 1.69 g/L), more frequent conduction block on electrophysiological examination (3/4 vs. 1/3), higher myelinated nerve fiber density, and positive CNTN1 expression in the glomeruli of kidney tissues. CONCLUSION: Anti-CNTN1 antibody was the most frequent antibody in this group of patients with CIDP/autoimmune nodopathies and nephropathy. Our study suggested that there might be some clinical and pathological differences between the antibody positive and negative patients.