Complete Liberation from Mechanical Ventilation Using Diaphragm Pacing in a Patient with Traumatic Spinal Cord Injury Despite Persistent Unilateral Diaphragmatic Paralysis

Shinya TOKUNAGA; Akinori MIYAKOSHI; Shinsuke SATO; Yoshihito HIRATA; Hiromasa ADACHI; Daisuke ARAI; Tsukasa SATO; Yoshifumi KAWANABE

doi:10.2176/jns-nmc.2025-0231

Abstract

We report a case of successful diaphragm pacing in a patient with severe traumatic spinal cord injury resulting in complete ventilator dependence. A 29-year-old man sustained a traumatic cervical spinal injury at the C3 level. On admission, he exhibited tetraplegia, complete sensory loss below the C6 dermatome, and acute respiratory failure. Emergency spinal decompression surgery was performed; however, neurological deficits showed no improvement. Despite intensive respiratory rehabilitation, the patient remained fully dependent on mechanical ventilation. Diaphragm pacing implantation was planned approximately 17 weeks post-injury. Intraoperative electrical stimulation revealed no responsiveness in the left diaphragm, whereas the right diaphragm demonstrated sufficient contractility. After estimating that adequate tidal volumes could be achieved through unilateral right diaphragmatic stimulation, bilateral diaphragmatic electrodes were laparoscopically implanted. Diaphragm pacing was initiated on postoperative day one, gradually increasing pacing duration daily while maintaining exertion levels below the Borg scale 4. By day 46 post-implantation, the patient achieved complete independence from mechanical ventilation despite persistent left diaphragmatic paralysis. Enhanced right diaphragmatic function was confirmed by increased diaphragmatic thickness and thickening fraction. Additionally, improved sputum clearance allowed withdrawal from cough-assist therapy and subsequent closure of tracheostomy. At 1-year follow-up, the patient maintained independent respiration without device-related complications. This case highlights the potential efficacy of early diaphragm pacing implantation in patients with high cervical spinal cord injury, demonstrating favorable respiratory outcomes even in the presence of unilateral diaphragmatic dysfunction.

Introduction

Traumatic spinal cord injury is a catastrophic event that may lead to severe permanent neurological deficits, such as tetraplegia and respiratory failure. The annual incidence rate of traumatic spinal cord injury in Japan is 49 per million inhabitants,¹⁾ which is higher than the global incidence rate of 23 per million.²⁾ The most frequently affected spinal level is the high cervical region (C1-C4: 45.8%).³⁾ The phrenic nerve originates from the C3-C5 segments of the spinal cord.⁴⁾ Therefore, high cervical spinal cord injury can result in severe respiratory failure, with approximately 2% of patients becoming permanently ventilator-dependent.⁵⁾ Ventilator-dependent SCI patients have a mortality rate 3.5 times higher than those who are not ventilator-dependent,⁶⁾ primarily because chronic mechanical ventilation leads to significant complications such as posterior lobe atelectasis, pneumonia, barotrauma, and tracheomalacia.⁷⁾ Thus, it is important to avoid long-term mechanical ventilation in spinal cord injury patients suffering from respiratory failure.

Phrenic nerve stimulation via implantation of a diaphragm pacing (DP) system has been developed to reduce dependence on mechanical ventilation in spinal cord injury patients (Fig. 1).⁸^,⁹⁾ Several studies have demonstrated the long-term efficacy of DP in reducing the need for mechanical ventilation in SCI patients.¹⁰^,¹¹⁾ However, insufficient diaphragm response during DP electrode implantation may lead to ineffective ventilation postoperatively.¹²⁾ Here, we report a patient with complete ventilator dependence resulting from a spinal cord injury at the C3 level. Despite the absence of a unilateral diaphragm response during DP electrode implantation, the patient successfully achieved complete ventilator weaning early in the postoperative period.

Figure 1

Schematic illustration of an external diaphragm pacing system (NeuRX, Synapse Biomedical, Oberlin, Ohio). This illustration was created by Shinya Tokunaga, adapted from a USCI Co. poster, with permission.

Case Report

A 29-year-old man was brought to our hospital by emergency transport following cervical trauma due to a fall at a building site. He had no significant past medical or family history. On arrival, he presented with tetraplegia and tachypnea and was intubated in the emergency room. He also exhibited loss of sensation below the C6 level. A computed tomography scan revealed fractures of the C3-C4 laminae, with displacement of the fractured C4 lamina into the spinal canal (Fig. 2A). Magnetic resonance imaging scan revealed severe injury of the spinal cord at the C3/4 level (Fig. 2B). He was diagnosed with traumatic cervical spinal cord injury classified as American Spinal Injury Association Impairment Scale grade A.

Figure 2

Imaging findings at initial presentation and after surgical intervention. (A) CT scan at initial presentation demonstrated fractures of C3-C4 laminae, with displacement of the fractured C4 lamina into the spinal canal. (B) MRI scan at initial presentation showed severe hyperintensity of the spinal cord at the C3/4 level. (C) Post-laminectomy CT imaging demonstrated adequate spinal canal decompression and successful removal of bone fragments. (D) Postoperative MRI showed severe edema of the spinal cord from the C3 to C5 level.

CT: computed tomography; MRI: magnetic resonance imaging

He underwent an emergent laminectomy from C2 to C7 on the day of admission. Postoperative computed tomography imaging demonstrated adequate decompression of the spinal canal and successful removal of bone fragments (Fig. 2C); however, spinal magnetic resonance imaging showed severe spinal cord edema (Fig. 2D). Despite surgical intervention, tetraplegia and complete loss of sensation below the C6 level persisted. Respiratory failure was severe due to the high cervical spinal cord injury, and he was unable to wean from mechanical ventilation. A tracheostomy was performed on day 9. Nevertheless, he experienced severe hypoventilation and difficulty with spontaneous breathing. Additionally, impaired secretion clearance resulted in copious sputum production. Consequently, he remained fully dependent on mechanical ventilation. His respiratory status was maintained with frequent suctioning, cough-assist therapy, and Continuous Positive Airway Pressure support (Positive End Expiratory Pressure) 10 cmH₂O, pressure support 5 cmH₂O.

He underwent respiratory rehabilitation for 3 months; however, ventilator weaning was unsuccessful. Therefore, we considered the use of a DP system (NeuRX, Synapse Biomedical, Oberlin, Ohio). As preoperative screening, phrenic nerve stimulation was performed to evaluate diaphragm movement. The left diaphragm was elevated and immobile, showing no response to stimulation. In contrast, the right diaphragm demonstrated good responsiveness; thus, we planned to implant a DP system (Fig. 3A). Laparoscopic diaphragmatic electrode implantation was performed on day 124 (Fig. 3B). The procedure was uneventful, although the left diaphragm did not respond even to direct stimulation. Considering that the patient was young and that only a short interval had passed since the injury, bilateral implantation of electrodes was performed. During intraoperative stimulation, a tidal volume of approximately 400 mL was confirmed, and the procedure was concluded.

Figure 3

Diaphragm response during phrenic nerve stimulation and diaphragmatic electrode implantation. (A) The right diaphragm responded positively to stimulation with elevation during stimulation; however, the left diaphragm showed no response. (B) Laparoscopic implantation of diaphragmatic electrodes.

We initiated external DP on postoperative day 1. When not pacing, he was supported with Continuous Positive Airway Pressure (Positive End Expiratory Pressure 10 cmH₂O, pressure support 3 cmH₂O). The duration of DP was gradually increased from 10 minutes per day, maintaining a Borg scale rating of 4 or below. On postoperative day 46 after DP implantation, he successfully achieved complete weaning from mechanical ventilation throughout the entire day (Fig. 4A). At the time of complete ventilator weaning, diaphragm function assessment showed improvements in diaphragmatic thickness and thickening fraction: calculated as the percentage increase in diaphragm thickness from end-expiration to end-inspiration (Fig. 4B).¹³⁾ From postoperative day 0 to day 78, the end-inspiratory diaphragm thickness improved from 2.4 mm to 3.2 mm on the right side and from 1.4 mm to 1.6 mm on the left side. However, the left diaphragm remained paralyzed. Subsequently, sputum production gradually decreased, and clearance capacity improved (Fig. 4C), allowing discontinuation of cough-assist therapy by day 106. Endotracheal suctioning was gradually tapered and eventually discontinued (Fig. 4C). He underwent closure of the tracheostomy at day 132 and was transferred to a local hospital. Approximately 1 year after diaphragm pacing initiation, no device-related complications occurred. He achieved a tidal volume of 590 mL, and he continues to breathe independently without mechanical ventilation support.

Figure 4

Respiratory outcomes after implanting diaphragmatic electrodes. (A) Daily duration of diaphragm pacing was gradually extended. (B) Diaphragm thickness and TFdi of the right diaphragm improved over time and were significantly higher at the time of complete ventilator weaning. (C) The frequency of endotracheal suctioning and use of cough-assist gradually decreased, ultimately becoming unnecessary by day 106.

TFdi: diaphragmatic thickening fraction

Discussion

We reported a case demonstrating the effectiveness of DP in a patient with severe high-level cervical spinal cord injury. The patient was initially completely dependent on mechanical ventilation; however, after implantation of a DP system, he successfully achieved complete ventilator independence. Additionally, sputum production gradually decreased, leading to improved secretion clearance, discontinuation of regular endotracheal suctioning, and eventual closure of the tracheostomy. Notably, the left diaphragm remained completely paralyzed throughout the clinical course.

Previous literature has demonstrated that DP can provide clinical benefit in 2 primary patient groups: patients with high cervical spinal cord injury (C1-C3 level) who have lost the ability for spontaneous breathing but have preserved diaphragm function at the lower motor neuron level, and patients diagnosed with congenital central hypoventilation syndrome (Ondine's syndrome).¹⁰^,¹¹^,¹⁴⁾ Additionally, DP has been tried in other patient populations, including patients with amyotrophic lateral sclerosis (ALS)¹⁵^-¹⁷⁾ and patients experiencing unilateral or bilateral diaphragmatic dysfunction unrelated to spinal cord injury or ALS.⁹⁾ In Japan, DP is covered by social insurance for 2 specific conditions: spinal cord injury and congenital central hypoventilation syndrome.⁷⁾ The first reported case of DP implantation for spinal cord injury in Japan was documented in 2022.¹⁸⁾ The DP system requires replacement of one D-cell battery every 96 hours and disposable connector socket replacement about once a day, but the cost burden on patients is low. To start using DP at the hospital, the electrode implantation surgery must be performed under the supervision of an experienced surgical mentor. In addition, all healthcare professionals involved in treatment are required to complete a training course on DP use.

DP is considered an advantageous alternative to mechanical ventilation due to several factors: improved clearance of sputum, enhanced spontaneous coughing ability, better speech and olfactory function, reduced risk of respiratory infections, and overall cost-effectiveness.¹⁴^,¹⁹⁾ Most patients with spinal cord injury experience respiratory benefits with DP, and between 25% and 61% achieve complete independence from mechanical ventilation over 24 hours.¹¹^,¹⁹⁾

Risk factors for unsuccessful pacing remain unclear; however, cases have been reported in which patients failed to gain respiratory benefit due to diaphragm atrophy and bilateral diaphragmatic paralysis.¹²⁾ Therefore, residual bilateral diaphragm function before electrode implantation is considered crucial. Intraoperative measurement of diaphragmatic residual function was assessed by evaluating tidal volume immediately following electrode implantation. In this case, a tidal volume of 400 mL was obtained, and this preserved tidal volume was one of the factors supporting the feasibility of successful diaphragmatic pacing.

However, the left diaphragm did not respond to direct intraoperative stimulation and remained completely paralyzed throughout the clinical course. The axial MR images also demonstrated slightly left-dominant spinal cord injury, suggesting that the right side was spared from complete paralysis. Although hemidiaphragmatic pacing has been reported, its effectiveness has typically been limited, achieving only about 3 hours of ventilation per day.²⁰⁾ Consequently, our case was expected to have limited success with pacing. However, the patient achieved complete independence from mechanical ventilation within 46 days and subsequently underwent successful closure of the tracheostomy.

Although the patient's young age may have contributed to successful pacing, the most critical factor among these was likely the early timing of implantation. In previous reports of multi-center studies, the average duration of mechanical ventilation prior to DP was 45.4 to 47.5 months.¹¹^,¹⁹⁾ In our case, DP was implanted at 5.4 months post-injury, which can therefore be regarded as a relatively early phase. Existing literature supports early DP implantation in SCI patients as a predictor of improved respiratory outcomes,¹¹^,¹⁹⁾ and our case may further support this finding. Mechanical ventilation is associated with diaphragmatic atrophy as early as 72 hours after initiation, and patients demonstrating improvements in diaphragm thickness tend to have better respiratory outcomes.²¹^,²²⁾ Ultrasound-based assessment of diaphragmatic function has recently gained attention and is increasingly being used as a noninvasive alternative to phrenic nerve stimulation for evaluating phrenic nerve paralysis.²³⁾ In our case, improvement in diaphragm thickness was observed. This condition further supports the beneficial effects of DP on respiratory outcomes in spinal cord injury patients, even when unilateral diaphragmatic paralysis is present.

Based on these findings from our case, it is suggested that early pacing implantation may facilitate successful ventilator weaning, even under unfavorable diaphragmatic conditions. Additionally, it may provide further respiratory benefits, such as improved sputum clearance and eventual tracheostomy decannulation.

Conclusions

We report a patient with spinal cord injury treated using DP. Despite persistent left diaphragmatic paralysis throughout the clinical course, complete independence from mechanical ventilation and subsequent tracheostomy closure were successfully achieved. Early implantation of DP in spinal cord injury patients may contribute to improved respiratory outcomes, even with unilateral diaphragmatic dysfunction.

Acknowledgments

The authors would like to thank Dr. Raymond P. Onders (Department of Surgery, University Hospitals Cleveland Medical Center) and Dr. Shunsaku Koriki (Department of Surgery, Shonan Fujisawa Tokushukai Hospital) for their technical support during the implantation of the diaphragm pacing stimulator.

Conflicts of Interest Disclosure

All authors have no conflict of interest.

Patient Informed Consent

Written informed consent for treatment and publication was obtained from the patient.

References

責任著者(Corresponding author)

J-STAGEへの登録はこちら（無料）