Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Aug 9;16(33):43199–43211. doi: 10.1021/acsami.4c03715

An Implantable Self-Driven Diaphragm Pacing System Based on a Microvibration Triboelectric Nanogenerator for Phrenic Nerve Stimulation

Hao Zhong †,‡,§, Ke Zhang ∥, Mi Zhou †,‡,§, Cong Xing †,‡,§, Yang An ∥, Qi Zhang †,‡,§, Junrui Guo †,‡,§, Song Liu †,‡,§, Zhigang Qu ∥,*, Shiqing Feng †,‡,§,*, Guangzhi Ning †,‡,§,*
PMCID: PMC11346467  PMID: 39120580

Abstract

graphic file with name am4c03715_0006.jpg

Spinal cord injury poses considerable challenges, particularly in diaphragm paralysis. To address limitations in existing diaphragm pacing technologies, we report an implantable, self-driven diaphragm pacing system based on a microvibration triboelectric nanogenerator (MV-TENG). Leveraging the efficient MV-TENG, the system harvests micromechanical energy and converts this energy into pulses for phrenic nerve stimulation. In vitro tests confirm a stable MV-TENG output, while subcutaneous implantation of the device in rats results in a constant amplitude over 4 weeks with remarkable energy-harvesting efficacy. The system effectively induces diaphragmatic motor-evoked potentials, triggering contractions of the diaphragm. This proof-of-concept system has potential clinical applications in implantable phrenic nerve stimulation, presenting a novel strategy for advancing next-generation diaphragm pacing devices.

Keywords: spinal cord injury, diaphragm paralysis, microvibration triboelectric nanogenerator, diaphragm pacing, neural modulation

1. Introduction

Spinal cord injury (SCI) represents a profound central nervous system disorder, primarily resulting from various traumatic factors. SCI frequently causes devastating and enduring impairments in sensory, motor, and autonomic functions below the level of injury. A comprehensive epidemiological study estimated a global annual prevalence of 27.04 million (with a range of 24.98–30.15 million) cases.1−3 In China, the total number of patients with SCI is reported to exceed 759,302, with 66,374 new cases occurring annually.4 The degree of respiratory dysfunction following a traumatic SCI is influenced by the level of the spinal cord lesion. In tetraplegia, where the injury affects the cervical region, the reduction in vital capacity ranges from 20% to 60%, whereas in paraplegia, which occurs when the injury is located in the thoracic or lumbar region, the vital capacity is reduced to 80%–90%.5

Respiratory decline associated with high-level cervical injuries is characterized by a decrease in various lung function parameters such as forced expiratory reserve volume, forced vital capacity, inspiratory capacity, and total lung capacity, with an increase in residual volume.6,7 Respiratory failure can occur after acute SCI due to different factors, including partial or complete paralysis of the respiratory muscles, fatigue of the remaining intact muscles, flaccid paralysis of the intercostal muscles, or the occurrence of pleuropulmonary pathology.8 During inspiration, the diaphragm contracts and moves downward while the rib cage, affected by the paralyzed intercostal muscles, moves inward paradoxically, diminishing lung expansion.9 During the acute and subacute stages of SCI, individuals often encounter difficulties in maintaining spontaneous ventilation, which can have major consequences. The inability to sustain proper breathing can give rise to a range of complications, such as atelectasis, pneumonia, edema, pulmonary embolism, and aspiration,10 that pose serious risks to respiratory health.

Currently, various treatment approaches are employed to manage respiratory dysfunction in individuals with SCI.11,12 These include physical and functional assessment, including diaphragm fluoroscopy and ultrasonography, which may be used to assess diaphragm movement pattern and thickness,13 while ventilator support can be continued after those assessments. These techniques, including electrical epidural stimulation,14 intramuscular diaphragm pacing,15 phrenic nerve pacing,16 and mechanical ventilation techniques such as intermittent positive-pressure breathing, CPAP, and BiPAP, have proven effective in reshaping respiratory function.17,18

In recent decades, the U.S. Food and Drug Administration has granted approval for diaphragm pacing as a treatment for patients who have lost neurological control of respiration after cervical SCI.19 Currently, two modes of diaphragm pacing are used: external and internal.20,21 The external mode uses surgically implanted receivers, electrodes, and an external transmitter equipped with antennae worn directly over the implanted receivers. Electrical signals are sent down electrodes that are directly inserted into the diaphragm, causing the diaphragms to contract. The phrenic nerves initiate the contraction of the diaphragms, causing the inhalation of air. Once the pulses cease, the diaphragms relax, leading to exhalation. The repetition of this sequence of pulses generates a typical breathing rhythm.22,23 However, these devices still pose challenges in terms of their bulkiness and lack of convenience for daily treatment. This is mainly due to their limited battery capacity and the complexity of replacing them.24

Considering the relatively high infection and potential electrical risks of commercial diaphragm pacing devices, a fully implantable and battery-free novel device is urgently in demand.25,26 Wang and colleagues introduced the concept of the triboelectric nanogenerator (TENG) in 2012, based on Maxwell’s equations.27 The types of nanogenerators can be classified into piezoelectric, triboelectric, and pyroelectric nanogenerators. Biocompatible nanogenerators can be implanted inside subjects, enabling the extraction of micromechanical energy from muscle stretching and body movements, making this a valuable biomechanical energy–harvesting tool.28 The unique size of the TENG device allows it to operate efficiently on a small spatial scale, simplifying the implantation process. Moreover, TENG devices are characterized by their low cost, lightweight nature, and high efficiency.29 Presently, biocompatible nanogenerators have found widespread applications as implantable power supplies for close-loop nerve stimulation, harnessing energy from various physiological processes, including breathing, heartbeat, pulsing, body movements, gastrointestinal peristalsis, and muscle stretching.30−34 Specific applications of biocompatible nanogenerators include the replacement of pacemakers, restoration of fine touch sensation, mechanoneuromodulation of the autonomic pelvic nerve for underactive bladders, and weight control.35,36 Kim et al. reported on the development of capacitance-matched triboelectric implants driven by ultrasound under a safe 500 mW/cm2 intensity to realize a battery-free, miniatured, and wireless neurostimulator.37 Lee et al. proposed an on-demand bioresorbable neurostimulator that allowed for clinical operations to be manipulated using biosafe ultrasound sources.38 Thus, material design and well-defined triggering events can realize on-demand management.

Wireless transmission is also critical for monitoring physiological signals or manipulating devices in implantable nanogenerators.39 This is particularly crucial for long-term implantable medical devices. Simard et al. presented a novel approach to high-speed OQPSK and efficient power transfer for biomedical implants, optimizing the inductive link for enhanced data rates and minimizing crosstalk in multiple carrier systems.40 Furthermore, Hosur et al. introduced a hybrid magnetic-ultrasonic wireless interrogation platform, known as MagSonic, which uses a single magnetoelectric transducer for wireless power and data transfer in millimeter-scale biomedical implants.41 The advancements in wireless technology and energy harvesting present not just a theoretical possibility but a tangible future. The convergence of these technologies is poised to revolutionize the landscape of biomedical engineering. Karan et al. have introduced an innovative dual energy harvesting technology that effectively captures energy from magnetic fields and ultrasound, generating electricity within the human body safety limits.42 These developments highlight the significant progress achieved in the field of wireless energy harvesting for implantable medical devices.

In this study, we introduce a novel self-driven implantable self-driven diaphragm pacing (ISD-DP) system based on a microvibration TENG (MV-TENG). The system harnesses the pulse output generated by MV-TENG, which is converted from alternating current through a rectifying bridge and stored in a capacitor. This stored energy is then applied to a nerve cuff electrode that is securely fixed to the phrenic nerve of rats (Figure 1). Given that the phrenic motoneuron pools are located at the C3–C5 spinal cord level, we employed a C2 spinal hemisection (C2SH) model to disrupt the diaphragm movement controlled by the ipsilateral phrenic nerve. Our ISD-DP system exhibited marked efficacy in energy harvesting in vitro tests, consistently delivering a stable output throughout a two-week animal biological toxicity assessment. In addition, we recorded the diaphragm electromyographic (EMG) activity using implanted electrodes to quantitatively assess the impact of the ISD-DP system on diaphragm EMG. The results demonstrated that this system significantly improves diaphragm pacing in rats with C2 injuries. Therefore, we believe that the ISD-DP holds promising clinical application potential for in vivo phrenic nerve stimulation and presents a novel strategy for developing the next generation of diaphragm pacing devices.

Figure 1.

Figure 1

Open in a new tab

Design and working principle of the ISD-DP system. a) Schematic showing the structure of an implantable TENG and its application by the energy-scavenging process. The nerve cuff was placed on the phrenic nerve, and the intact phrenic nerve and diaphragm ensured normal pacing in the stimulation release. b) Schematic of its application via the energy-scavenging process: the MV-TENG worked as an energy harvester and could be placed in a subcutaneous pouch near the unaffected side of the chest. Created with BioRender.com.

2. Results and Discussion

2.1. Design and Characterization of MV-TENG

Building upon our prior studies,43−45 we explored various triboelectric materials to develop a robust, corrosion-resistant, mechanically flexible, impenetrable, and nontoxic device capable of harnessing friction-induced electrical effects. To maximize the electrification through friction, we selected polytetrafluoroethylene (PTFE) and polyamide 6 (PA6) films, which are recognized for their distinct frictional polarities. PTFE was chosen as the negatively charged dielectric material due to its marked electron-attracting properties. PA6 was used as the positively charged dielectric material because of its superior electron-capturing capability. Additionally, each of the two aluminum foils was divided into equal quarters. Four of these quarters were attached to the back of the PTFE, while the remaining four were attached to the back of the PA6 film, forming the conductive layer.

To augment the device sensitivity, we implemented a spacer structure (500 μm). By stacking the dielectric films comprising two distinct materials in proximity, surface charges were generated with opposing polarities when these materials came into contact. The subsequent separation between the films created an air gap, producing an inductive potential difference between the two electrodes. This innovative design allows for the improved capture of mechanical energy generated by rat movement, leading to higher energy output and more effective nerve stimulation.

Moreover, we elevated the frictional charge density by introducing silver nanowire structures and applying a silver nanowire solution during the PA6 film preparation. In addition to their exceptional electrical conductivity, nanosilver wires facilitate surface-enhanced plasma to significantly enhance the efficiency of the nanogenerator within the constraints of the same size. The silver nanowire was applied to the PA6 surface via a spray-on method to streamline the manufacturing process, making it more accessible for widespread use. Wang et al. included nanosilver wires with a high aspect ratio to enhance the output electrical performance of the nanogenerator, establishing additional conductive channels within the polymer matrix, which significantly increased the output voltage of doped silver nanowires.46 These silver nanowires have an approximate diameter of 60 nm, as depicted in the corresponding SEM images presented in Figure 2b. Hence, a specific content of AgNWs was selected to improve the output performance of the MV-TENG (Figure 5g).

Figure 2.

Figure 2

Open in a new tab

Design and characterization of microvibration triboelectric nanogenerator (MV-TENG). a) Structural schematic of MV-TENG. b) Scanning electron microscopy (SEM) images of nanostructures on positively charged dielectric films (PA6). c) Finite element simulation of the Inductive potential difference between PA6 dielectric film and PTFE dielectric film. d) Finite element simulation of the potential distribution between the two electrodes when the PA6 and PTFE dielectric films are separated in contact with different spacing. e) Finite element simulation of the output current between the two electrodes when the PA6 and PTFE dielectric films are separated in contact with different spacing. f) Relationship between output voltage and spacing during contact separation between PA6 and PTFE dielectric films. g) Output voltages of undoped and doped silver nanowires h) Output voltage of MV-TENG at different frequencies in an in vitro test. (i) Output current of MV-TENG in vitro test.

Figure 5.

Figure 5

Open in a new tab

Diaphragm pacing in cervical SCI model by the ISD-DP system. (A). Schematic of the diaphragmatic electromyogram. Representative examples of stimulation-induced diaphragmatic motor–evoked potentials in uninjured (B) and C2 hemisection animals during the acute (C), subchronic (D), and chronic (E) injury stages. All data shown as mean ± SD, n = 3. Significance between the groups is marked with asterisks (*) at p < 0.05; **p < 0.01 and #p < 0.0001.

A negative electrode was established by dividing a 100-μm aluminum foil layer into four equal parts and attaching them to the rear surface of the PTFE film. The core packaging layer comprised a 17.2-μm poly(dimethylsiloxane) PDMS film, enhancing the resistance of the device to corrosion and leakage. Ecoflex was used as the encapsulation layer, further elevating the biocompatibility. To accurately quantify the induced potential difference between the two electrodes, we developed a three-dimensional model for finite element simulation, as depicted in Figure 2c. A corresponding two-dimensional model was also developed, and finite element analysis was conducted to confirm the existence of the potential difference between the two dielectric materials. The outcomes unequivocally demonstrated that an increased spacing between these materials produced a higher output voltage and current (Figure 2d, e). The empirical relationship between output voltage and spacing is shown in Figure 2f. The basic working mechanism and photograph of MV-TENG are demonstrated in Figure S1 (see Supporting Information).

To assess the influence of various frequencies on MV-TENG power generation, we placed this under a 10-N pressure at different frequencies (0.5–2.5 Hz). The results demonstrated that the MV-TENG consistently generated the same voltage output across these diverse frequencies, as shown in Figure 2h. For in vitro testing, the performance of the MV-TENG’s current value was 0.83 ± 0.23 μA (Figure 2i).

The multilayer packaging structure significantly augmented the durability of the device. Specifically, the selected encapsulation material, exemplified by Ecoflex, has outstanding tensile properties and symmetrical mechanical responsiveness. This characteristic empowers the device to endure repeated torsion and extension, enabling unrestricted multidirectional movement with minimal force prerequisites. This exceptional flexibility and stretchability have negligible effects on the everyday activities of the subjects.

2.2. Long-Term Output Performance and Biological Safety of MV-TENG

Cervical SCIs, which interrupt the transmission of nervous signals from the respiratory centers in the medulla to the inspiratory motoneurons, often cause respiratory failure and dependence on artificial ventilation.47 The success of phrenic nerve pacing relies on the viability of phrenic motoneurons and axons, allowing for their activation through electrical stimulation methods.48,49 Stimulating the phrenic nerves with electrical impulses induces the contraction of the diaphragm, which decreases the intrathoracic pressure and facilitates the flow of air during inspiration. When the stimulation ceases, the diaphragm relaxes, enabling passive exhalation (Figure 3a). After a C2SH, phrenic motoneurons originating from the respiratory centers on the left side (ipsilateral) become deafferented, causing hemiparalysis of the diaphragm (as demonstrated in Supporting Video). The contralateral side remained intact, allowing partial retention of respiratory function. The lesion epicenter was stained with NeuN (red) and GFAP (green) to identify the area of injury. Representative immunofluorescence images of the injury epicenter are provided in Figure 3b. The injured area contains the ipsilateral ventrolateral funiculus, which contains the majority of bulbospinal pathways to phrenic motor neurons.50

Figure 3.

Figure 3

Open in a new tab

Long-term output performance and compatibility of MV-TENG. a) Mechanism of diaphragm modulation with phrenic nerve stimulation. b) Representative immunofluorescence staining image of a hemisection of the second cervical spinal segment demonstrating the completeness of the hemisection injury. The extent of both NeuN+ (red) and GFAP+ (green) staining are shown (7 days postinjury) c) The circuit schematic of the power conversion module (PCM). d) Charging voltage on various load capacitors for the MV-TENG with the PCM in parallel and series connections. e) Charging curve of the capacitor by using MV-TENG.

To explore the potential of electrical signals generated by MV-TENG in treating diaphragmatic paralysis caused by cervical SCIs, we implanted the MV-TENG-based diaphragm pacing device with an Ecoflex coating under the chest wall. To efficiently harvest biomechanical energy generated by intercostal muscle contraction, the MV-TENG was connected to a rectifier bridge with parallel and series connections.

To facilitate the widespread application of the MV-TENG network, a power conversion module (PCM) was designed to manage the energy harvested by the MV-TENG. The circuit schematic of the PCM is shown in Figure 3c. The PCM comprises an AC–DC rectifier circuit and a filter circuit. The specific components of the integrated circuit involve a full-wave rectifier bridge, a reed switch, a shunt diode (D), a series inductor (L–15 μH), and shunt capacitors (C1–10 pF, C2–10 pF, and C3–2.2 μF). The PCM controls the circuit through switches, storing and releasing electrical energy for diaphragmatic pacing. The LC unit in the charging circuit functions as a low-pass filter. The stability of the electrical signals is crucial for nerve stimulation in the diaphragmatic pacing system. Filters play a crucial role in eliminating noise, interference, or unwanted frequency components in electronic circuits, thereby enhancing signal quality and stability. The employed filter ensures the delivery of accurate, reliable, and appropriate signals to the stimulated neurons, promoting the overall effectiveness and safety of neurostimulation. The filter is also effective in removing interference signals, transforming irregular electrical output into a stable DC voltage. Additionally, the charging performance of the MV-TENG with PCM-loaded capacitors was compared under parallel and series connections, as shown in Figure 3d. The capacitor could be charged to 1 V within 15 s by the MV-TENG, and the pulse width was 17 ms (Figure 3e). Unlike the constant-voltage square wave pulses used in commercial devices, the voltage of the MV-TENG resembles an exponential waveform, which starts with rapid changes and then gradually slows down over time. Therefore, the pulse width of MV-TENG is slightly larger than those used in current commercial devices. For commercial phrenic nerve stimulators, square-wave pulses are the standard input signal. Some studies suggest that exponential waveforms may offer superior neuronal stimulation with reduced power consumption compared to square-wave pulses.51

Ensuring biological safety is an essential prerequisite to avoid undesirable biological responses. The use of Ecoflex as the encapsulation layer in the device yields substantial improvements in the flexibility and sealing of the MV-TENG. Ecoflex, known for its elasticity and flexibility, maintains these properties across a broad temperature range. Ecoflex not only demonstrates commendable biocompatibility but also exhibits exceptional resistance to chemicals and body fluids. Ecoflex is sufficiently resilient to withstand the corrosive effects of various solvents and chemicals and is therefore eminently suitable for packaging applications.

A pivotal aspect of our investigation centered on the stability of the device. Consequently, the sealed MV-TENG device was subcutaneously implanted within the chest wall of a rat for 4 weeks. The nanogenerator was implanted under the intact side of the subject and fixed subcutaneously and externally to the chest wall by using a subcutaneous pouch and sutures. To ascertain the long term performance of the system, we conducted assessments of output amplitude at four distinct time points: immediately after implantation and at 1, 2, and 4 weeks following implantation (0.99 ± 0.65, 1.35 ± 0.60, 1.31 ± 0.55, and 1.39 ± 0.71 V, respectively) (Figure 4c).

Figure 4.

Figure 4

Open in a new tab

Implantation, output performance, and biological toxicity assessment of the ISD-DP system. a) Photograph of the surgical procedures. The cuff electrode was secured around the phrenic nerve. b) Computed tomography images of MV-TENG in vivo after 4 weeks to ensure that no displacement has occurred. c) In vivo output performance immediately after implantation, and at 1, 2, and 4 weeks following that. The nanogenerator was implanted under the intact side of the rat and fixed subcutaneously and externally to the chest wall using a subcutaneous pouch and sutures without connecting to the nerve cuff electrodes on the phrenic nerve during this process. The output performance was stable throughout the in vivo experimentation. d) H&E stains of the liver, kidney, and skin around the incision site in the fourth week after implantation. e) Blood tests were performed 4 weeks after transplantation for white blood cells, lymphocytes (LYM), monocytes (MON), and neutrophile granulocyte percentage (NEU percentage). All data shown as mean ± SD, n = 4, ns means not statistically significant.

Notably, the waveform of the output signal exhibited inconsistencies due to the sporadic movements of the experimental subjects. Nonetheless, the peak voltage remained constant during the experimental duration, indicating the robust functionality of the MV-TENG. The output amplitude did not exhibit any significant decline during regular conditions throughout the testing period. Regarding the inconsistency between the frequency and peak-to-peak voltage output value, we speculate that this may be due to the excessive space of the pouch at the initial stage of implantation, which allowed the nanogenerator more room to move. This may have reduced the pressure on the surface and thus lowered the output efficiency. However, in the later stages, the encapsulation by tissue could have constricted the movement space of the nanogenerator, slightly improving the output efficiency. Another possible reason could be that in the later stages of SCI, the overall respiratory function may have improved with an increased respiratory movement range of the entire thoracic cavity. By the fourth week after implantation, microcomputed tomography scans confirmed successful implant integration and in vivo stability. No apparent displacement occurred, and the entire system remained intact (Figure 4b).

To comprehensively evaluate the biological toxicity of the MV-TENG device, histopathologic tests were undertaken on the kidney, liver, and skin surrounding the incision site. Hematoxylin and eosin (H&E) staining at 4 weeks after MV-TENG implantation revealed no instances of deformation or abnormal lymphatic cell invasion in the liver and kidney (Figure 4d). The skin tissue displayed benign proliferation and inflammatory changes, signifying no systemic side effects induced by the MV-TENG. Blood tests were performed 4 weeks after transplantation. The inflammation-related factors, including white blood cells, lymphocytes, monocytes, and neutrophil granulocyte percentage, did not show any abnormality, which further confirmed a lack of inflammation (Figure 4e). Collectively, these findings validate the nontoxicity and biological safety of the MV-TENG device, corroborated by its capacity for sustained and stable voltage output.

2.3. Evaluation of Diaphragm Pacing of the Phrenic Stimulation under the ISD-DP System

To verify the pacing efficacy exhibited by the ISD-DP system, a diaphragmatic electromyogram was employed as an objective assessment tool (Figure 5a). The cuff electrodes of the ISD-DP system were wrapped around the phrenic nerve on the injured side(the design of the cuff electrodes is presented in Figure S2). The inner diameter of the cuff electrode was 0.35 mm, with L2 measuring 1 mm and L1 measuring 2 mm. The receiving electrode was strategically inserted into the bilateral crus of the diaphragm.

The primary objective was to evaluate whether the ISD-DP system could effectively induce diaphragmatic motor–evoked potentials. To achieve this, we implanted the MV-TENG into the right thoracic wall of our model. This strategic placement facilitated the collection and storage of biomechanical energy generated by the periodic pressures exerted by the intercostal muscles. The final phase of our investigation entailed the application of the cuff electrodes to stimulate the left phrenic nerve, thereby reactivating the previously paralyzed diaphragm. Notably, under anesthesia, the animal can generate normal respiratory reflexes on the healthy side and can still produce active expiratory movements and passive inspiratory movements to provide sufficient deformation for the implanted nanogenerator. Therefore, this evaluation encompassed the energy-harvesting mechanism, storage capacity, and capacity to restore diaphragmatic functionality.

In uninjured animals, both bilateral diaphragms demonstrated analogous inspiratory rhythmic bursting activities, characterized by comparable discharge onsets and burst amplitudes (Figure 5b). The burst pattern of the right diaphragm, corresponding to the uninjured side of the experimental rats, did not exhibit significant deviations from that observed in uninjured animals, even following cervical SCI. However, a discernible disparity emerged with the left diaphragm, signifying the injured side. Notably, the left diaphragm typically displayed minimal burst amplitudes, reflecting the effect of the cervical SCI on this specific aspect of diaphragmatic function. This observation underscores the debilitating consequences of SCI, with the left diaphragm notably affected in terms of inspiratory rhythmic bursting activities.

Currently, commercial diaphragm pacemakers, exemplified by the Avery diaphragm pacemaker (https://averybiomedical.com/), generate a maximum voltage of 10 V, while the typical stimulus voltage ranges from 1 to 5 V. In this study, we used a 2.2 μF capacitor charged to 1 V to induce diaphragm contraction, similar to current commercial parameters. Our investigation encompassed three distinct time points: the acute phase (immediately after injury), the subacute phase (7 days after injury), and the chronic phase (14 days after injury). The aim was to evaluate the efficacy of the self-driven diaphragm pacing system in stimulating diaphragm contraction and activating diaphragm discharge across these temporal phases. Our findings demonstrated that the ISD-DP system was effective in stimulating diaphragm contraction and triggering diaphragm discharge in all three phases. In the acute phase, the system achieved a marked effectiveness of 70.27% (±13.12%, P = 0.0014), which increased even further in the subacute phase to 98.39% (±2.30%, P < 0.0001). Even in the chronic phase, the system performed with an efficacy rate of 90.44% (±12.58%, P = 0.0045). These results underline the consistent and reliable performance of the system in facilitating diaphragm function across various phases of SCI. Additionally, the phrenic nerve photographs demonstrate that the nerve structure remains intact following electrical stimulation, without any nerve damage (Figure S3a). Moreover, the phrenic nerve sections exhibited a normal pattern of β-III Tubulin with uniform intensity across the tissue, indicating the preservation of neuronal structure (Figure S3b). Our findings demonstrate that electrical stimulation is safe and does not result in tissue damage.

3. Discussion

Cervical SCIs frequently entail disruptions in the motor pathways that connect the respiratory center in the medulla with the inspiratory muscles. This disruption culminates in respiratory failure, necessitating artificial ventilatory support. Over 90% of cases of traumatic cervical SCIs are estimated to need intubation, with the majority of these cases requiring a tracheostomy.52 A univariate analysis by Marqués et al. proved that factors such as gender, age, ASIA grade, motor score, and injury severity score were related to the need for mechanical ventilation.53 Furthermore, long-term mechanical ventilation increases the risk of infections, challenges the weaning process, and impacts speech and swallowing functions, as well as having psychological implications.54

Various methods have been explored to enhance respiratory mechanics following acute cervical SCI. These approaches include breath stacking, cough-assist devices, and intermittent positive-pressure breathing. Among them, diaphragm pacing has exhibited promising outcomes. The currently used diaphragm pacing techniques can be divided into external diaphragm pacing (EDP) and IDP systems. The EDP technique was initially applied in the auxiliary ventilation treatment of diseases such as chronic obstructive pulmonary disease and chronic respiratory failure.55 This offers advantages such as simplicity of structure and noninvasiveness and overcomes the significant limitations of mechanical ventilation, including respiratory infection complications and the inability of patients to leave the intensive care unit. However, the EDP technique is limited by the need for long-term adjustment in positioning and fixation of the external electrodes on the skin for precise stimulation of the diaphragm and phrenic nerve to promote effective diaphragm contraction.56 IDP involves direct stimulation of the phrenic nerve to induce diaphragm contraction. This requires the surgical attachment of electrodes for phrenic nerve stimulation and the placement of antennae and receivers. The receiver converts this energy signal into an electrical current that is conducted to the phrenic nerves. Nerve stimulation causes the diaphragm muscles to contract, resulting in inhalation. This cycle repeats according to the predetermined respiratory rate set on the transmitter, triggering breaths at regular intervals.57 However, a notable limitation of IDP systems is their reliance on batteries, and battery failure is the primary cause of mechanical issues in IDP systems.58 Regular battery replacements are necessary, introducing inconvenience and the potential for interruptions in respiratory support.

Moreover, patients with diaphragm pacing systems may encounter activity restrictions to prevent damage to the implanted device, impacting their daily lives. Therefore, these battery-related limitations and the associated risks must be emphasized when considering using a diaphragm pacing system for respiratory support.59 Another equally important risk is the failure of radio frequency receivers in older systems caused by body fluid leakage through the encapsulation. Therefore, developing a closed-loop fully internal self-powered adaptive system is crucial. All presently used pacing systems follow an open-loop design, wherein predetermined levels set fixed values for inspired volumes and respiratory rates.

Herein, we present an ISD-DP system based on MV-TENG as a novel therapeutic strategy for cervical SCI–caused respiratory failure. The system delivers electrical signals that emulate the natural electrical impulses traveling through the phrenic nerve to the paralyzed diaphragm. The ISD-DP system, exhibiting no biological toxicity and antifluid immersion performance, has been developed to harvest biomechanical energy subcutaneously. This system comprises an MV-TENG, cuff electrode, and the PCM. The electric power generated by the MV-TENG is collected and stored in the PCM and used to initiate electrical stimulation. This approach represents a novel direction for developing and optimizing neurostimulation treatments. A distinguishing feature of this approach is its self-sufficiency, eliminating the need for battery replacements. The ISD-DP system was built based on a flexible MV-TENG attached to the subcutaneous pouch of the contralateral chest wall and could generate biphasic electric pulses during the respiration motion. The MV-TENG is connected to a nerve cuff electrode, where the bodily microvibrations and respiratory fluctuations generate an electrical signal that could directly stimulate the phrenic nerve, contributing to inhalation. Notably, the ISD-DP system demonstrated no biological toxicity, as evidenced by the absence of systemic toxicity or observable side effects in the histopathological analysis using H&E staining. Our pacing test demonstrated that the system could effectively promote diaphragm muscle contraction in the acute, subacute, and chronic phases after cervical SCI. Under spontaneous breathing conditions, the burst amplitudes of the left diaphragm were markedly diminished in the presence of a unilateral cervical injury at all postinjury time points examined. In addition to a decrease in burst amplitude, a delayed onset of inspiratory bursting was observed in the diaphragms of animals with C2SH. Significant muscle contraction signals were observed in our electromyography results. This proves that integrating our PCM module enables the efficient accumulation of small amounts of energy from multiple body movements and the release of this energy to deliver effective stimulation.

In comparison to traditional diaphragm stimulation devices, our approach offers several advantages. First, this precisely controls the activity of the target muscle, given that nerves typically innervate multiple muscle fibers. This allows for finer motor control, reducing unnecessary muscle activity. Additionally, nerve stimulation can be less uncomfortable or painful for some patients than direct muscle stimulation, as this is more localized. Furthermore, our approach helps prevent muscle fatigue as the stimulation can be distributed more effectively across multiple muscle fibers. Importantly, this typically consumes less electrical energy, extending the battery life of implanted medical devices. Furthermore, the system functions without requiring external wires or connections, offering patients comfort in their daily lives.

Our study has some limitations. We did not directly connect the nerve cuff electrode to the phrenic nerve to observe the direct long-term changes in respiratory function after implantation. In the preliminary research, we conducted similar experiments. However, because of limitations such as the size of animal models and a lack of interventional techniques (e.g., thoracoscopy), it was difficult for us to perform direct long-term observation of the effects of the ISD-DP system under our existing conditions. The relatively large size of the system was a consequence of our choice to use rats as the animal model for demonstrating the working system. This choice inherently imposes limitations on both the output capabilities and dimensions of the system. Nonetheless, despite these constraints, our system remains smaller than the current commercial diaphragm pacemakers; in this study, we have proposed a prototype device. Our study did not investigate the muscle responses during the release of stimulation, and we did not investigate the precise contraction, finer motor control of muscles under electrical stimulation, or tolerance to pain associated with the stimulation. Indeed, both EDP and IDP involve a particularly complex tuning process. Often, because of the difficulty in precisely positioning the electrodes, significant variability in efficacy can occur, and improper operation can easily cause diaphragm fatigue. Diaphragm fatigue is characterized by an initial increase and subsequent decrease in diaphragm movement amplitude, a decrease followed by a rebound in PaCO2, and patients reporting increased feelings of chest tightness. Diaphragm fatigue is closely related to the frequency (Hz), intensity (V), and number of stimulations per minute of the electrical pulses. Thus, the current single-output form requires further fine-tuning before implementation for diaphragm pacing, which will be addressed in future studies. Currently, our system only enables muscle contraction poststimulation, and because of the overall efficiency constraints of the nanogenerator system, this does not achieve the regular motion typical of powered diaphragmatic pacing systems.

Certain researchers have documented instances where diaphragm activation through phrenic nerve stimulation often causes insufficient production of inspired volume, and several potential reasons may account for this phenomenon. First, not all diaphragm activations may be successfully achieved because of the high threshold of some axons. Second, the regular respiratory pattern of the diaphragm is a highly complex process where subtle force modulation by different fibers enables the attainment of the maximum inspired volume. However, artificial pacing may cause a shift in the dominant myofiber population. Finally, adhering to the principles of normal breathing anatomy, collectively recruiting inspiratory intercostal and accessory muscles may be beneficial in facilitating this biomimetic process.

By further improving and perfecting the self-adapted diaphragm pacing system, we can place the TENG onto the diaphragm itself to complete the close-loop stimulation procedure by sensing the muscle diastolic pressure. Moreover, we can optimize the delivery of impulses based on the current sequence of impulse discharges. Each breath is initiated by administering a series of electrical pulses (usually at a frequency of ≤20 Hz) with a predetermined amplitude. The magnitude of inspired volumes can be primarily adjusted by altering the frequency. Most clinically approved PNP systems provide the capacity for a ramp increase in inspired volume by applying gradual increases in stimulus amplitude or frequency. Our prototype system has the potential to release electrical pulses causing tonic contraction of the diaphragm; therefore, in the future work, we will focus on the pulse discharge module60 in conjunction with the respiratory airflow monitoring system for refinement.

With ongoing advancements in materials science, we anticipate the development of devices with even higher energy densities in the future. This study provides preliminary validation of the developed device, although continuous pacing was not achieved, and introduces innovative ideas and unique insights for developing self-powered diaphragm pacemaker devices.

This study demonstrated the initial feasibility of a self-powered, implantable diaphragm pacemaker. Although several challenges remain to be overcome for the diaphragm pacing system to reach clinical applications, this prototype offers an unprecedented solution for patients needing long-term respiratory support, potentially revolutionizing the field of diaphragm pacing. In subsequent experiments, we plan to explore the optimal stimulation parameters (currents, voltages, and frequencies) and conduct further research in larger animals and even in humans. Our findings suggest that the self-driven diaphragm pacing system holds considerable promise as a neurostimulation strategy for repairing diaphragm paralysis following cervical SCI.

4. Conclusion

This study confirmed the pacing effect of the ISD-DP system based on MV-TENG, achieving effective diaphragm pacing. In our experiment, we were the first to propose the use of a biocompatible nanogenerator for diaphragm pacing after cervical spinal cord injury. We initially validated our approach from the perspectives of Biocompatibility, Output Performance, Anchoring or Fixation, Mechanical Stability, Long-term Monitoring, and Tissue Response. Our results confirmed the stability of the system under in vivo implantation conditions. Subsequently, in our immediate stimulation pacing experiment, we found that the collection of energy through the nanogenerator, coupled with the charging and discharging capacitor, could effectively stimulate the diaphragm. However, it is important to note that in our experiment, the nanogenerator-driven diaphragm pacemaker remains conceptually oriented. While this represents preliminary validation, continuous pacing was not realized. These findings offer fresh perspectives and unique insights that may catalyze further exploration in the application and development of self-powered pacemaker devices.

5. Experimental Section

5.1. Cervical SCI

Twelve adult Wistar rats weighing 180–210 g were used. All animal experiments were conducted following ethical guidelines and approved by the Animal Ethics Committee of Tianjin Medical University General Hospital (Protocol Number: IRB2023-DW-124). Before surgical procedures, all instruments were autoclaved to ensure sterility. The C2SH surgical procedures were performed under a microscope. Each rat was anesthetized using isoflurane, administered at a concentration of 4% during induction, and maintained at 1.5% throughout the procedure. The administration of isoflurane was conducted using a Veterinary Anesthesia Machine (RWD R640). Rats were positioned in a prone posture. Following the induction of complete anesthesia, the nape skin was shaved, and thorough disinfection was performed. An incision was carefully made on the neck, and the layers of muscle were separated step by step until the C2 vertebra, characterized by its prominent apophysis, was fully exposed. Subsequently, the spinous process and lamina of the C2 vertebra were meticulously removed using a rongeur. This procedure revealed the dorsal spinal cord. A lateral hemisection of the left side of the spinal cord was performed using iridectomy scissors. After a spinal cord hemisection injury, we first observed a lesion on the spinal cord, and the breathing pattern of the rat changed to pronounced abdominal breathing. Simultaneously, the frequency and amplitude of respiration both increased. Finally, the muscle and incision were sutured, and rats were placed on heating pads to maintain their body temperature during recovery. To prevent infections, antibiotic drugs were administered twice daily during the initial 3 days following the surgery.

5.2. MV-TENG Fabrication

The nanogenerator can be short-circuited because of fluid leakage in the body. To avoid this, a new multilayer encapsulation structure is proposed in this study. The device dimensions are approximately 20 × 18 × 4 mm. To fabricate the device, the surface of a glass substrate was cleaned with acetone and ethanol, and the glass substrate was then placed in an oven for drying. PA6 pellets (20%) were dissolved in formic acid (Aldrich, > 94.5%). The mixture was maintained at 70 °C with continuous stirring for 1 h, followed by natural cooling to 20 °C. Approximately 1 mL of the prepared solution was spin-coated onto the center of a glass substrate at 500 rpm for 10 s. Subsequently, the rotation speed was increased to 2000 rpm for an additional 20 s to achieve the desired thickness of the PA6 film. A solution of 5 mg/mL of silver nanowires was uniformly sprayed on the surface of the PA6 film. The scanning electron microscope image of the 60 nm silver nanowires is shown in Figure 2b. The 100-μm–thick aluminum foil was divided into four equal parts that were adhered onto the back of the PA6 film to form the positive electrode. PDMS and curing agent were mixed thoroughly at a weight ratio of 10:1 at 2000 rpm for 1 min. Subsequently, the blended solution was poured onto a dried glass substrate. Next, the glass substrate was placed on a rotary coater and rotated at 1500 rpm for 20 s before it was removed and placed in an oven at 80 °C. The 100-μm–thick aluminum foil was divided into four equal parts and adhered onto the back of the PTFE film to form the negative electrode. PDMS and Ecoflex are commonly used as packaging materials for nanogenerators because of their good flexibility and biocompatibility.35,61 We employed a layer of PDMS with a 1 mm thickness as the second layer, which was spin-coated over the entire device. This enhanced the corrosion resistance, ensuring exceptional flexibility and stability in its structural integrity. To improve the leak resistance and biocompatibility and mitigate potential issues such as corrosion, erosion, and adhesion in complex physiological environments, we prepared components A and B of ECOFLEX0030 in a 1:1 ratio. After thoroughly stirring for 1 min, the mixture was poured into a mold for the outermost packaging layer. The process formed a high-density, nonporous shell coating with degradation resistance. This multilayer encapsulation structure ensures the stability of the device within the body and its resilience to complex external environments in research experiments. We used a custom-made nerve cuff electrode equipped with two platinum contacts to ensure proper fit to the phrenic nerve. The electrode features a 1.5 mm internal diameter and is constructed with PTFE-insulated twisted stainless-steel leads.

5.3. Implantation of the ISD-DP System

We surgically implanted a compact MV-TENG device alongside a nerve cuff electrode for phrenic nerve stimulation in each experimental animal. Before implantation, thorough sterilization procedures were performed to ensure aseptic conditions. Following the creation of the spinal cord hemisection model, rats were placed in a supine position. A 2 cm incision was made beneath the left sternocleidomastoid muscle of the rat. Layer-by-layer dissection was performed to expose the muscles, and the brachial plexus nerves were separated between the internal jugular vein and the sternocleidomastoid muscle, the inner side of which is the descending phrenic nerve, and the phrenic nerve was isolated. A cuff electrode was then placed over the phrenic nerve and secured with surgical sutures (Figure 4a). The MV-TENG device was placed in a subcutaneous pouch within the thoracic region. This location was selected to secure the device and optimize its biomechanical energy–harvesting efficiency through deformation. The layers of muscle fascia were meticulously sutured using 4–0 nylon sutures. Finally, the skin incision was sealed using veterinary adhesive glue (vetbond, 3M, U.S.). Postimplantation, animals were allowed to recuperate and were continuously monitored while resting on a heated blanket pad. Animals were individually housed in separate cages and adhered to a 12-h light/dark cycle. Figure 3a shows an image of the cuff electrode in animal experiments.

5.4. In Vivo Biological Toxicity Assessments

The rats underwent the same anesthesia procedure as previously outlined. Subsequently, the MV-TENG devices were implanted through a dorsal incision. Four weeks after implantation, the rats were euthanized for histological analysis. Tissue samples, including liver, kidney, and the skin surrounding the incision site, were harvested. These collected tissue samples were fixed in 4% PFA for 24 h and subjected to a gradual alcohol dehydration process. Following dehydration, samples were embedded in paraffin and sectioned into 10-μm–thick-slices. H&E staining was applied to these sections for histological examination. Finally, stained tissue sections were scrutinized under a light microscope (IX73 Olympus, Japan) for further assessment.

5.5. Measurement of Nerve Electrophysiology

For diaphragmatic electromyogram measurements, rats were anesthetized with isoflurane, as previously described, and their body temperature was maintained at 37.0 °C using a heated pad. A 27-gauge needle electrode was inserted into the superficial layer of the diaphragm muscle (one lead per side). The reference electrode was positioned on the abdomen. The inhalation and exhalation of the rat caused the alternating expansion and contraction of the thorax. The MV-TENG harnesses the biomechanical energy from these respiratory movements as well as from muscle stretching during head turning. The electrical energy collected by the MV-TENG is stored in the PCM. When the voltage charges to 1 V, this starts releasing electrical stimuli upon the wireless trigger approach to the reed switch. The EMG activity was recorded at three different time points: immediately after the C2SH and on days 3 and 7 following the procedure. Sham surgeries were performed to establish a baseline for EMG activity, serving as a reference for comparison with the postinjury EMG recordings.

5.6. Immunostaining

The spinal cord and phrenic nerve were carefully fixed in 4% PFA for 24 h. Subsequently, the tissue underwent dehydration by being immersed in a 30% sucrose solution. After dehydration, the samples were embedded in OCT compound and frozen to facilitate sectioning. Tissues were precisely sectioned with a thickness of 10 μm using a CM 3050S Leica microtome. To prepare the sections for immunofluorescence staining, slices were permeabilized and blocked with a solution containing 5% bovine serum albumin (BSA) and 0.25% Triton X-100 for 1 h. Following the blocking step, sections were incubated overnight with primary antibodies: rabbit anti-GFAP (CST, 1:500), mouse anti-NeuN (CST, 1:500), and mouse anti-β-III tubulin (Abcam, 1:500). These primary antibody incubations took place in a solution of 5% BSA and 0.25% Triton X-100 in 0.1 M Tris-buffered saline with Tween (TBST) overnight. Sections were thoroughly washed with 0.1 M TBST on the following day, and the antigen–antibody complexes were visualized using secondary antibodies goat antirabbit Cy3 (Abcam) at a 1:500 dilution and goat antimouse Alexa Fluor 488 (Abcam) at a 1:500 dilution. After the secondary antibody incubation, slides were washed three times with TBST to remove any unbound antibodies or staining reagents. All stained slides were stored at −20 °C to preserve the fluorescence labeling for further analysis.

5.7. Statistical Analysis

Statistical findings are presented as the mean ± SD (standard deviation). Analysis of H&E staining and immunostaining images was conducted using Fiji and GraphPad Prism 8.0.2. Statistical evaluations were performed using SPSS (version 23.0). Unpaired t tests were applied to compare data between two groups, and a significance threshold of p < 0.05 was considered significant.

Acknowledgments

This work was supported by the following funding: Tianjin Key Medical Discipline (Specialty) Construct Project, National Natural Science Foundation of China (82272470, 82072439, 81930070), Tianjin Health Key Discipline Special Project (TJWJ2022XK011), Outstanding Youth Foundation of Tianjin Medical University General Hospital (22ZYYJQ01).

Data Availability Statement

All relevant data supporting the key findings of this study are available within the article and its Supporting Information files or from the corresponding author upon reasonable request.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.4c03715.

  • Working mechanism of microvibration triboelectric nanogenerator and photograph of MV-TENG, design of Cuff electrodes, and histological tests for the phrenic nerve (PDF)

  • After a C2SH, leading to hemiparalysis of the diaphragm (MP4)

Author Contributions

H.Z., K.Z., and M.Z. contributed equally.

The authors declare no competing financial interest.

Supplementary Material

am4c03715_si_001.pdf (264.1KB, pdf)
am4c03715_si_002.mp4 (786.1KB, mp4)

References

  1. Fan B.; Wei Z.; Feng S. Progression in Translational Research on Spinal Cord Injury Based on Microenvironment Imbalance. Bone Res. 2022, 10 (1), 35. 10.1038/s41413-022-00199-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Global, Regional, and National Burden of Traumatic Brain Injury and Spinal Cord Injury, 1990–2016: A Systematic Analysis for the Global Burden of Disease Study 2016. The Lancet. Neurology 2019, 18 (1), 56–87. 10.1016/S1474-4422(18)30415-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Lo J.; Chan L.; Flynn S. A Systematic Review of the Incidence, Prevalence, Costs, and Activity and Work Limitations of Amputation, Osteoarthritis, Rheumatoid Arthritis, Back Pain, Multiple Sclerosis, Spinal Cord Injury, Stroke, and Traumatic Brain Injury in the United States: A 2019 Update. Archives of Physical Medicine and Rehabilitation 2021, 102 (1), 115–131. 10.1016/j.apmr.2020.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Jiang B.; Sun D.; Sun H.; Ru X.; Liu H.; Ge S.; Fu J.; Wang W. Prevalence, Incidence, and External Causes of Traumatic Spinal Cord Injury in China: A Nationally Representative Cross-Sectional Survey. Frontiers In Neurology 2022, 12, 784647. 10.3389/fneur.2021.784647. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Alizadeh A.; Dyck S. M.; Karimi-Abdolrezaee S. Traumatic Spinal Cord Injury: An Overview of Pathophysiology, Models and Acute Injury Mechanisms. Frontiers In Neurology 2019, 10, 282. 10.3389/fneur.2019.00282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Reyes M. R. L.; Elmo M. J.; Menachem B.; Granda S. M. A Primary Care Provider’s Guide to Managing Respiratory Health in Subacute and Chronic Spinal Cord Injury. Topics In Spinal Cord Injury Rehabilitation 2020, 26 (2), 116–122. 10.46292/sci2602-116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Honda H.; Fukawa K.; Sawabe T. Influence of Adjuvant Arthritis on Main Urinary Metabolites of Prostaglandin F and E in Rats. Prostaglandins 1980, 19 (2), 259–269. 10.1016/0090-6980(80)90024-6. [DOI] [PubMed] [Google Scholar]
  8. Berlowitz D. J.; Wadsworth B.; Ross J. Respiratory Problems and Management in People with Spinal Cord Injury. Breathe (Sheff) 2016, 12 (4), 328–340. 10.1183/20734735.012616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Terson de Paleville D. G. L.; McKay W. B.; Folz R. J.; Ovechkin A. V. Respiratory Motor Control Disrupted by Spinal Cord Injury: Mechanisms, Evaluation, and Restoration. Transl Stroke Res. 2011, 2 (4), 463–473. 10.1007/s12975-011-0114-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bonay M.; Vinit S. New Perspectives for Investigating Respiratory Failure Induced by Cervical Spinal Cord Injury. Neural Regeneration Research 2014, 9 (22), 1949–1951. 10.4103/1673-5374.145367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Brown R.; DiMarco A. F.; Hoit J. D.; Garshick E.. Respiratory Dysfunction and Management in Spinal Cord Injury. Respir Care 2006, 51 ( (8), ).853. [PMC free article] [PubMed] [Google Scholar]
  12. O’Halloran K. D. Muscling in on Neurorehabilitative Strategies to Counter Respiratory Motor Dysfunction in Cervical Spinal Cord Injury. Journal of Physiology 2021, 599 (4), 1009–1010. 10.1113/JP281042. [DOI] [PubMed] [Google Scholar]
  13. Galeiras Vazquez R.; Rascado Sedes P.; Mourelo Farina M.; Montoto Marques A.; Ferreiro Velasco M. E. Respiratory Management in the Patient with Spinal Cord Injury. Biomed Res. Int. 2013, 2013, 168757. 10.1155/2013/168757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Malone I. G.; Nosacka R. L.; Nash M. A.; Otto K. J.; Dale E. A. Electrical Epidural Stimulation of the Cervical Spinal Cord: Implications for Spinal Respiratory Neuroplasticity after Spinal Cord Injury. Journal of Neurophysiology 2021, 126 (2), 607–626. 10.1152/jn.00625.2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Onders R. P.; Ponsky T. A.; Elmo M.; Lidsky K.; Barksdale E. First Reported Experience with Intramuscular Diaphragm Pacing in Replacing Positive Pressure Mechanical Ventilators in Children. J. Pediatr Surg 2011, 46 (1), 72–76. 10.1016/j.jpedsurg.2010.09.071. [DOI] [PubMed] [Google Scholar]
  16. Sitthinamsuwan B.; Nunta-aree S. Phrenic Nerve Stimulation for Diaphragmatic Pacing in a Patient with High Cervical Spinal Cord Injury. Journal of the Medical Association of Thailand-Chotmaihet Thangphaet 2009, 92 (12), 1691–1695. [PubMed] [Google Scholar]
  17. Bach J. R.; Burke L.; Chiou M. Noninvasive Respiratory Management of Spinal Cord Injury. Phys. Med. Rehabil Clin N Am. 2020, 31 (3), 397–413. 10.1016/j.pmr.2020.03.006. [DOI] [PubMed] [Google Scholar]
  18. Romitti F.; Busana M.; Palumbo M. M.; Bonifazi M.; Giosa L.; Vassalli F.; Gatta A.; Collino F.; Steinberg I.; Gattarello S.; Lazzari S.; Palermo P.; Nasr A.; Gersmann A.-K.; Richter A.; Herrmann P.; Moerer O.; Saager L.; Camporota L.; Marini J. J.; Quintel M.; Meissner K.; Gattinoni L. Mechanical Power Thresholds During Mechanical Ventilation: An Experimental Study. Physiol Rep 2022, 10 (6), e15225 10.14814/phy2.15225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Woo A. L.; Tchoe H. J.; Shin H. W.; Shin C. M.; Lim C. M. Assisted Breathing with a Diaphragm Pacing System: A Systematic Review. Yonsei Med. J. 2020, 61 (12), 1024–1033. 10.3349/ymj.2020.61.12.1024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cavka K.; Fuller D. D.; Tonuzi G.; Fox E. J. Diaphragm Pacing and a Model for Respiratory Rehabilitation after Spinal Cord Injury. J. Neurol Phys. Ther 2021, 45 (3), 235–242. 10.1097/NPT.0000000000000360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lohse A.; von Platen P.; Benner C.-F.; Deininger M. M.; Seemann T. G.; Ziles D.; Breuer T.; Leonhardt S.; Walter M. Evaluation of Electric Phrenic Nerve Stimulation Patterns for Mechanical Ventilation: A Pilot Study. Sci. Rep 2023, 13 (1), 11303. 10.1038/s41598-023-38316-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ricoy J.; Rodríguez-Núñez N.; Álvarez-Dobaño J. M.; Toubes M. E.; Riveiro V.; Valdés L. Diaphragmatic Dysfunction. Pulmonology 2019, 25 (4), 223–235. 10.1016/j.pulmoe.2018.10.008. [DOI] [PubMed] [Google Scholar]
  23. Wenzel C.; Spassov S. G.; Lengerer J.; Schmidt J.; Urban G.; Schumann S. Electromagnetic Stimulation of the Phrenic Nerve Preserves Diaphragm Muscle Strength During Mechanical Ventilation in a Rat Model. European Journal of Anaesthesiology 2022, 39 (1), 81–82. 10.1097/EJA.0000000000001590. [DOI] [PubMed] [Google Scholar]
  24. Lee K.-Z.; Liou L.-M.; Vinit S. Diaphragm Motor-Evoked Potential Induced by Cervical Magnetic Stimulation Following Cervical Spinal Cord Contusion in the Rat. Journal of Neurotrauma 2021, 38 (15), 2122–2140. 10.1089/neu.2021.0080. [DOI] [PubMed] [Google Scholar]
  25. Onders R. P.; Elmo M.; Kaplan C.; Schilz R.; Katirji B.; Tinkoff G. Long-Term Experience with Diaphragm Pacing for Traumatic Spinal Cord Injury: Early Implantation Should Be Considered. Surgery 2018, 164 (4), 705–711. 10.1016/j.surg.2018.06.050. [DOI] [PubMed] [Google Scholar]
  26. Kerwin A. J.; Yorkgitis B. K.; Ebler D. J.; Madbak F. G.; Hsu A. T.; Crandall M. L. Use of Diaphragm Pacing in the Management of Acute Cervical Spinal Cord Injury. Journal of Trauma and Acute Care Surgery 2018, 85 (5), 928–931. 10.1097/TA.0000000000002023. [DOI] [PubMed] [Google Scholar]
  27. Wang Y.; Yang Y.; Wang Z. L. Triboelectric Nanogenerators as Flexible Power Sources. npj Flexible Electronics 2017, 1 (1), 10. 10.1038/s41528-017-0007-8. [DOI] [Google Scholar]
  28. Xu C.; Zeng F.; Wu D.; Wang P.; Yin X.; Jia B. Nerve Stimulation by Triboelectric Nanogenerator Based on Nanofibrous Membrane for Spinal Cord Injury. Front Chem. 2022, 10, 941065. 10.3389/fchem.2022.941065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Shen J.; Li B.; Yang Y.; Yang Z.; Liu X.; Lim K.-C.; Chen J.; Ji L.; Lin Z.-H.; Cheng J. Application, Challenge and Perspective of Triboelectric Nanogenerator as Micro-Nano Energy and Self-Powered Biosystem. Biosens. Bioelectron. 2022, 216, 114595. 10.1016/j.bios.2022.114595. [DOI] [PubMed] [Google Scholar]
  30. Zhou M.; Huang M.; Zhong H.; Xing C.; An Y.; Zhu R.; Jia Z.; Qu H.; Zhu S.; Liu S. J. A. f. m. Contact Separation Triboelectric Nanogenerator Based Neural Interfacing for Effective Sciatic Nerve Restoration. Adv. Functional Materials 2022, 32 (22), 32. 10.1002/adfm.202200269. [DOI] [Google Scholar]
  31. Parandeh S.; Etemadi N.; Kharaziha M.; Chen G.; Nashalian A.; Xiao X.; Chen J. Advances in Triboelectric Nanogenerators for Self-Powered Regenerative Medicine. Adv. Functional Materials 2021, 31 (47), 2105169. 10.1002/adfm.202105169. [DOI] [Google Scholar]
  32. Ouyang H.; Liu Z.; Li N.; Shi B.; Zou Y.; Xie F.; Ma Y.; Li Z.; Li H.; Zheng Q.; Qu X.; Fan Y.; Wang Z. L.; Zhang H.; Li Z. Symbiotic Cardiac Pacemaker. Nat. Commun. 2019, 10 (1), 1821. 10.1038/s41467-019-09851-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kang M.; Shin H.; Cho Y.; Park J.; Nagwade P.; Lee S. Triboelectric Neurostimulator for Physiological Modulation of Leg Muscle. Nano Energy 2022, 103, 107861. 10.1016/j.nanoen.2022.107861. [DOI] [Google Scholar]
  34. Lee S.; Wang H.; Shi Q.; Dhakar L.; Wang J.; Thakor N. V.; Yen S.-C.; Lee C. Development of Battery-Free Neural Interface and Modulated Control of Tibialis Anterior Muscle Via Common Peroneal Nerve Based on Triboelectric Nanogenerators (Tengs). Nano Energy 2017, 33, 1–11. 10.1016/j.nanoen.2016.12.038. [DOI] [Google Scholar]
  35. Yao G.; Kang L.; Li J.; Long Y.; Wei H.; Ferreira C. A.; Jeffery J. J.; Lin Y.; Cai W.; Wang X. Effective Weight Control Via an Implanted Self-Powered Vagus Nerve Stimulation Device. Nat. Commun. 2018, 9 (1), 5349. 10.1038/s41467-018-07764-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lee S.; Wang H.; Xian Peh W. Y.; He T.; Yen S.-C.; Thakor N. V.; Lee C. Mechano-Neuromodulation of Autonomic Pelvic Nerve for Underactive Bladder: A Triboelectric Neurostimulator Integrated with Flexible Neural Clip Interface. Nano Energy 2019, 60, 449–456. 10.1016/j.nanoen.2019.03.082. [DOI] [Google Scholar]
  37. Kim Y.-J.; Lee J.; Hwang J.-H.; Chung Y.; Park B.-J.; Kim J.; Kim S.-H.; Mun J.; Yoon H.-J.; Park S.-M.; Kim S.-W. High-Performing and Capacitive-Matched Triboelectric Implants Driven by Ultrasound. Adv. Materials 2024, 36 (2), 2307194. 10.1002/adma.202307194. [DOI] [PubMed] [Google Scholar]
  38. Lee D.-M.; Kang M.; Hyun I.; Park B.-J.; Kim H. J.; Nam S. H.; Yoon H.-J.; Ryu H.; Park H.-m.; Choi B.-O.; Kim S.-W. An on-Demand Bioresorbable Neurostimulator. Nat. Commun. 2023, 14 (1), 7315. 10.1038/s41467-023-42791-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Karan S. K.; Maiti S.; Lee J. H.; Mishra Y. K.; Khatua B. B.; Kim J. K. Recent Advances in Self-Powered Tribo-/Piezoelectric Energy Harvesters: All-in-One Package for Future Smart Technologies. Adv. Functional Materials 2020, 30 (48), 2004446. 10.1002/adfm.202004446. [DOI] [Google Scholar]
  40. Simard G.; Sawan M.; Massicotte D. High-Speed Oqpsk and Efficient Power Transfer through Inductive Link for Biomedical Implants. IEEE Transactions on Biomedical Circuits and Systems 2010, 4 (3), 192–200. 10.1109/TBCAS.2009.2039212. [DOI] [PubMed] [Google Scholar]
  41. Hosur S.; Kashani Z.; Karan S. K.; Priya S.; Kiani M. Magsonic: Hybrid Magnetic-Ultrasonic Wireless Interrogation of Millimeter-Scale Biomedical Implants with Magnetoelectric Transducer. IEEE Transactions on Biomedical Circuits and Systems 2024, 18 (2), 383–395. 10.1109/TBCAS.2023.3334166. [DOI] [PubMed] [Google Scholar]
  42. Karan S.; Hosur S.; Kashani Z.; Leng H.; Vijay A.; Sriramdas R.; Wang K.; Poudel B.; Patterson A.; Kiani M.; Priya s. J. Magnetic Field and Ultrasound Induced Simultaneous Wireless Energy Harvesting. Energy Environ. Sci. 2024, 17, 2129. 10.1039/D3EE03889K. [DOI] [Google Scholar]
  43. Qu Z.; Wu L.; Yue B.; An Y.; Liu Z.; Zhao P.; Luo J.; Xie Y.; Liu Y.; Wang Q.; Wang Z.; Dai R.; Yin W. Eccentric Triboelectric Nanosensor for Monitoring Mechanical Movements. Nano Energy 2019, 62, 348–354. 10.1016/j.nanoen.2019.05.061. [DOI] [Google Scholar]
  44. Qu Z.; Wang X.; Huang M.; Chen C.; An Y.; Yin W.; Li X. An Eccentric-Structured Hybrid Triboelectric-Electromagnetic Nanogenerator for Low-Frequency Mechanical Energy Harvesting. Nano Energy 2023, 107, 108094. 10.1016/j.nanoen.2022.108094. [DOI] [Google Scholar]
  45. Qu Z.; Huang M.; Chen C.; An Y.; Liu H.; Zhang Q.; Wang X.; Liu Y.; Yin W.; Li X. Spherical Triboelectric Nanogenerator Based on Eccentric Structure for Omnidirectional Low Frequency Water Wave Energy Harvesting. Adv. Functional Materials 2022, 32 (29), 2202048. 10.1002/adfm.202202048. [DOI] [Google Scholar]
  46. Wang F.; Sun H.; Guo H.; Sui H.; Wu Q.; Liu X.; Huang D. High Performance Piezoelectric Nanogenerator with Silver Nanowires Embedded in Polymer Matrix for Mechanical Energy Harvesting. Ceram. Int. 2021, 47 (24), 35096–35104. 10.1016/j.ceramint.2021.09.052. [DOI] [Google Scholar]
  47. Fogarty M. J.; Sieck G. C. Spinal Cord Injury and Diaphragm Neuromotor Control. Expert Rev. Respir Med. 2020, 14 (5), 453–464. 10.1080/17476348.2020.1732822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tang J. E.; Saklayen S. L.; Savona S. J.; Essandoh M. K.; Augostini R. S. Remede̅ Systems: Transvenous Pacing of the Phrenic Nerve. Journal of Cardiothoracic and Vascular Anesthesia 2023, 37 (4), 627–631. 10.1053/j.jvca.2023.01.011. [DOI] [PubMed] [Google Scholar]
  49. Etienne H.; Dres M.; Piquet J.; Wingertsmann L.; Thibaudeau O.; Similowski T.; Gonzalez-Bermejo J.; Assouad J. Phrenic Nerve Stimulation in an Ovine Model with Temporary Removable Pacing Leads. J. Thorac Dis 2022, 14 (8), 2748–2756. 10.21037/jtd-21-1944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Boulenguez P.; Gestreau C.; Vinit S.; Stamegna J.-C.; Kastner A.; Gauthier P. Specific and Artifactual Labeling in the Rat Spinal Cord and Medulla after Injection of Monosynaptic Retrograde Tracers into the Diaphragm. Neurosci. Lett. 2007, 417 (2), 206–211. 10.1016/j.neulet.2007.02.047. [DOI] [PubMed] [Google Scholar]
  51. Lee S.; Wang H.; Wang J.; Shi Q.; Yen S.-C.; Thakor N. V.; Lee C. Battery-Free Neuromodulator for Peripheral Nerve Direct Stimulation. Nano Energy 2018, 50, 148–158. 10.1016/j.nanoen.2018.04.004. [DOI] [Google Scholar]
  52. Ropper A. E.; Neal M. T.; Theodore N. Acute Management of Traumatic Cervical Spinal Cord Injury. Pract Neurol 2015, 15 (4), 266–272. 10.1136/practneurol-2015-001094. [DOI] [PubMed] [Google Scholar]
  53. Montoto-Marqués A.; Trillo-Dono N.; Ferreiro-Velasco M. E.; Salvador -de la Barrera S.; Rodriguez-Sotillo A.; Mourelo-Fariña M.; Galeiras-Vázquez R.; Meijide-Failde R. Risks Factors of Mechanical Ventilation in Acute Traumatic Cervical Spinal Cord Injured Patients. Spinal Cord 2018, 56 (3), 206–211. 10.1038/s41393-017-0005-7. [DOI] [PubMed] [Google Scholar]
  54. Kumar A.; Khandelwal A.; Jamil S. Ventilatory Strategies in Traumatic Cervical Spinal Cord Injury: Controversies and Current Updates. Asian Spine J. 2023, 17 (4), 615–619. 10.31616/asj.2023.0094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Bao Q.; Chen L.; Chen X.; Li T.; Xie C.; Zou Z.; Huang C.; Zhi Y.; He Z. The Effects of External Diaphragmatic Pacing on Diaphragm Function and Weaning Outcomes of Critically Ill Patients with Mechanical Ventilation: A Prospective Randomized Study. Ann. Transl Med. 2022, 10 (20), 1100. 10.21037/atm-22-4145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhou Z.; Wang W.; Yang R.; Wang Y.; Zhu L.; Huang Y. Bio-Z-Based Feedback-Controlled External Diaphragm Pacing System. IEEE Transactions on Circuits and Systems II: Express Briefs 2023, 70 (8), 2779–2783. 10.1109/TCSII.2023.3255902. [DOI] [Google Scholar]
  57. Weese-Mayer D. E.; Berry-Kravis E. M.; Ceccherini I.; Keens T. G.; Loghmanee D. A.; Trang H. An Official Ats Clinical Policy Statement: Congenital Central Hypoventilation Syndrome: Genetic Basis, Diagnosis, and Management. Am. J. Respir Crit Care Med. 2010, 181 (6), 626–644. 10.1164/rccm.200807-1069ST. [DOI] [PubMed] [Google Scholar]
  58. DiMarco A. F. Phrenic Nerve Stimulation in Patients with Spinal Cord Injury. Respir Physiol Neurobiol 2009, 169 (2), 200–209. 10.1016/j.resp.2009.09.008. [DOI] [PubMed] [Google Scholar]
  59. Kerwin A. J.; Zuniga Y. D.; Yorkgitis B. K.; Mull J.; Hsu A. T.; Madbak F. G.; Ebler D. J.; Skarupa D. J.; Shiber J. R.; Crandall M. L. Diaphragm Pacing Improves Respiratory Mechanics in Acute Cervical Spinal Cord Injury. Journal of Trauma and Acute Care Surgery 2020, 89 (3), 423–428. 10.1097/TA.0000000000002809. [DOI] [PubMed] [Google Scholar]
  60. DiMarco A. F.; Kowalski K. E. High-Frequency Spinal Cord Stimulation of Inspiratory Muscles in Dogs: A New Method of Inspiratory Muscle Pacing. Journal of Applied Physiology (Bethesda, Md.: 1985) 2009, 107 (3), 662–669. 10.1152/japplphysiol.00252.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Liu Z.; Ma Y.; Ouyang H.; Shi B.; Li N.; Jiang D.; Xie F.; Qu D.; Zou Y.; Huang Y.; Li H.; Zhao C.; Tan P.; Yu M.; Fan Y.; Zhang H.; Wang Z. L.; Li Z. Transcatheter Self-Powered Ultrasensitive Endocardial Pressure Sensor. Adv. Functional Materials 2019, 29 (3), 1807560. 10.1002/adfm.201807560. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

am4c03715_si_001.pdf (264.1KB, pdf)
am4c03715_si_002.mp4 (786.1KB, mp4)

Data Availability Statement

All relevant data supporting the key findings of this study are available within the article and its Supporting Information files or from the corresponding author upon reasonable request.

Articles from ACS Applied Materials & Interfaces are provided here courtesy of American Chemical Society

RESOURCES