Options for the use of epidural stimulation in individuals with motor complete spinal cord injury

Overview

Spinal cord injury is associated not only with sensory and motor impairment below the level of the lesion, but also with other complications such as autonomic nervous system dysfunction, spasticity, or neuropathic pain. While intensive rehabilitation can alleviate neurological deficits in incomplete spinal cord lesions, the neurological picture in clinically complete lesions remains fundamentally unchanged. In recent years, the potential of epidural spinal cord stimulation has been investigated, showing promise as a method capable of partially restoring lost function even in complete spinal cord lesions. This review outlines the development of the method, from pain modulation to the restoration of voluntary movement using a digital bridge between the motor cortex and spinal cord stimulator. Over the past twenty years, significant progress has been made from slight improvement in mobility in incomplete lesions to the restoration of standing and walking in motor complete injuries. The work also includes a summary of the effects on autonomic functions, with impacts on the cardiovascular system, bladder control, and sexual functions. The limitations of these studies are primarily the heterogeneity of program settings, small patient cohorts, and the risks associated with stimulator implantation. Nevertheless, epidural spinal cord stimulation represents a significant advance in the treatment of spinal cord injury, with a positive impact on the quality of life for this population.

Keywords:

autonomic nervous system – spinal cord injury – Paraplegia – neuromodulation – epidural spinal cord stimulation

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Spinal cord injuries are among the most serious injuries that require demanding long-term medical and rehabilitation care and lead to severe permanent consequences. The incidence of spinal cord injuries in the Czech Republic ranges between two and a half and three cases per 100 000 inhabitants per year. Almost one third of them are classified as sensorimotor complete (AIS A) or sensorimotor incomplete (AIS B), i.e. with complete loss of mobility below the level of the lesion [1]. In these patients, there are very limited therapeutic options to affect sensorimotor and autonomic function.

One of the research directions is epidural spinal cord stimulation (eSCS). Already in 1967, this method was used to control chronic neuropathic pain [2]. Based on the results of preclinical studies, Dimitrijevic et al. used eSCS to induce rhythmic lower limb flexion-extension patterns (LLE) in six paraplegics, which they considered as evidence for the existence of a central movement pattern generator (CPG). The CPG is a group of interneurons that can generate rhythmic motor output without the presence of any external stimuli [3]. A breakthrough was the work of Herman et al [4], who used eSCS to improve motor function in a patient with a motor incomplete cervical spinal cord lesion (AIS C). Successful studies in people with incomplete spinal cord lesions led to a closer examination of the potential of this method in individuals with complete loss of movement [5].

However, the use of eSCS in patients with complete spinal cord lesions has some limitations. First, only certain specific spinal segments can be affected by eSCS. The selection of suitable patients is also hampered by the high demand for good cooperation, as well as the riskiness of the surgical procedure itself. On the other hand, the possibility of even minimal improvement and approximation to the pre-injury state is a great motivation for patients. This makes it all the more important to present the method and its results in a very rational way so that potential study participants do not have exaggerated expectations. The aim of this review article is to present the eSCS method and its use in patients with motor complete spinal cord lesions.

 

The principle of the method

The principle of eSCS is the transfer of electrical potentials from the electrode to the posterior horns of the spinal cord. Here, the signal is switched through the interneurons to the anterior spinal cord horns and from there transmitted through efferent pathways to the muscles or visceral organs. All this takes place below the level of the spinal cord injury. The muscle activity elicited is therefore primarily a response to spinal cord stimulation without any influence from higher centres. However, cases have been published where patients in the long term began to perceive impulses and be able to activate muscles even after the stimulator was switched off [6]. This is explained by the existence of so-called discomplex lesions, which have the clinical picture of a complete injury, but a certain part of the spinal pathways remains preserved and is only permanently strongly inhibited [3]. It is the long-term regular electrical stimulation that can lead to a kind of "awakening" of these strongly inhibited pathways. This could be one of the mechanisms of the effect of eSCS on the return of free motor activity [7]. In a recent experiment, it has been shown that a quarter to half of patients with clinically complete lesions may show residual activity that could be indicative of lesion incompleteness [8]. Thus, to maximize the effect of eSCS, it is necessary to choose a setting that also affects these residual spinal pathways at the lesion level. With an optimal setup, then not only the CPG but also the connections to the cortical movement centers can be activated. A greater effect on cortical centres has been demonstrated with subthreshold stimulation [9].

 

Implantation

For eSCS, flat electrodes made of polymer tape are usually used, in which metal contacts are placed in several rows. Each contact provides an individually programmable conducting surface, allowing a flexible combination of mono-, bi -⁠ or multipolar stimulation. Surgical access for placement of the electrode in the spinal canal is provided by standard hemilaminectomy. Cables exiting the electrode are tunneled to a remote subcutaneously placed pulse generator.

The key phase to ensure optimal eSCS function is the selection of the appropriate stimulation area and therefore electrode placement. This may vary according to the intended target of stimulation. The studies mentioned in this review article have primarily focused on influencing motor function of the ACC and therefore use electrode placement at the level of the spinal conus with the possibility of influencing the L1-S1 segments [10]. Due to individually different anatomical conditions, the spinal cone level needs to be verified peroperatively. For this purpose, perioperative EMG is used to verify the correct horizontal position of the electrode and also the optimal sagittal position of the electrode (Figure 1). Using remote access to the generator, training programs can be individually adjusted in patients immediately after surgery to achieve activation of specific muscle groups (Fig. 2).

 

Effect on motor function

The first paper describing the successful impact on motor function in a patient with a motor complete lesion was published by Professor Susan Harkema's team in 2011. Using a case study of a 23-year-old patient with a chronic thoracic sensory incomplete spinal cord lesion (AIS B), it was successfully demonstrated that eSCS can restore functional movements of the hip, standing and walking in individuals with complete motor loss. Prior to stimulator implantation, the participant underwent 26 months of intensive physiotherapy with 170 training units. He subsequently underwent implantation of a 16-contact epidural electrode (5-6-5 Specify, Medtronic, Minneapolis, MN, USA) placed on the lumbosacral extension of the spinal cord and a subcutaneously placed pulse generator (Restore Advanced, Medtronic, Minneapolis, MN, USA). After 80 training units with stimulation, the patient was able to stand without relief and without manual support from physiotherapists for more than 4 min. Standing was induced by stimulation of the caudal segments at a frequency of 15 Hz, whereas frequencies in the range of 30-40 Hz were used for movements of the hips. Surprisingly, after 7 months of physiotherapy and eSCS, the patient partially regained free control of the DKK movements during stimulation [11]. Three years later, this team published the results in three more patients with motor complete spinal cord lesions (one AIS B and two AIS A patients). All patients regained the ability to freely move the hips (e.g., ankle dorsiflexion or knee flexion in the supine position). However, in this experiment, there were side effects of stimulation, namely clonic muscle activity [12]. All three patients were also able to stand with only handrail support. It has been shown that patients achieved better standing results when the caudal contacts were individually stimulated at higher frequencies (range 25-60 Hz) [13]. In 2023, pooled results were published for 25 patients with chronic motor complete spinal cord lesions enrolled in the study between 2009 and 2020. All of these patients developed free mobility after implantation. Thirteen of these patients were followed up for more than 2 years after stimulator insertion, indicating the stability of the results obtained [14].

Similar results were achieved by a team from the Mayo Clinic (Rochester, MN, USA). Within 2 weeks after stimulator implantation, a 26-year-old paraplegic (AIS A) was able to stand independently and perform free movements of the hips similar to the walking stereotype [15]. In another paper, they presented this case along with another paraplegic with a similar outcome, where the ability to walk with a rollator was restored after stimulator implantation and subsequent rehabilitation training with only partial assistance from physiotherapists. Also, a frequency of 40 Hz was used to stimulate the gait pattern, whereas a frequency of 15 Hz was preferred for standing [16].

A study aiming to systematically evaluate the effect of different eSCS parameters on free movement and autonomic function was published by Darrow et al. They first enrolled two women aged 48 and 52 years with an interval of 5 and 10 years, respectively, after the onset of a complete spinal cord lesion (AIS A) in the thoracic region. The patients did not undergo any special neurorehabilitation before implantation. After insertion of the 16-contact electrode with pulse generator, free control of movement was restored due to stimulation [6]. Professor Darrow's team is currently investigating the use of eSCS to restore neurological function following motor complete cervical and thoracic spinal cord injury as part of the E-STAND project. The authors plan to enroll 100 participants in the study, who will be implanted with a pulse generator (St. Jude Medical Proclaim Elite 7, Abbott, Australia) with a 16-contact electrode. The aim of this large study is to determine whether eSCS could be useful in a wider range of spinal cord injury patients in the future [17]. In 2020, the same team published the results of follow-up of seven stimulated patients (six with spinal cord lesion extent AIS A and one AIS B). After prolonged motor function training during stimulation, which averaged 13.7 h/day for 255.3 days, four patients managed to maintain free control even when stimulation was switched off. However, the strength and accuracy of motor activation remained higher during active stimulation [18].

A shift in the neurological level of the lesion and partial return of motor function was also published by Kandhari et al. In their results, they reported very rapid recovery of motor function as early as 2 months after implantation in all ten patients with complete spinal cord lesions. Consistently, they observed an improvement in standing time over 10 min and also an improvement in sensation and a reduction in spasticity [19].

In contrast to American studies that use continuous eSCS, a team of researchers from Switzerland led by Professor Courtine chose so-called spatiotemporal stimulation, in which motor neurons are selectively stimulated according to the intended movement [20]. Using 3D kinematic analysis of gait and EMG recordings of muscle activity, they created a spatiotemporal map of motoneuron activation at different phases of the gait cycle in healthy subjects. This allowed the precise configuration of the implanted electrode to target posterior root stimulation with projection to spinal regions containing motor neurons involved in hip, knee and ankle movements. Based on the study conducted, this research team designed and constructed a custom stimulation electrode that will allow stimulation of a wider area of the spinal cord that also activates the muscles of the lower trunk. In addition, they have created software that can be used to quickly configure specific stimulation programmes. They tested this technology on three individuals with complete spinal cord lesion AIS A, who were able to recover the ability to stand, walk, ride a treadmill, swim and have trunk muscle activity, albeit only during active stimulation. A targeted stimulation program for each phase of the gait cycle was activated by ergonomic switches located in the handles of the rotator [21].

The latest research from this group is aimed at creating a digital bridge between the motor cortex and the spinal cord using electrodes that sense cortical activity. To monitor electrocorticographic signals from the sensorimotor cortex , the authors used Wimagine technology, which includes two fully implantable sensing electrodes with 64 contacts and a diameter of 50 mm [22]. The decoded signals were converted into stimulation pulses and transmitted in real time to a pulse generator that activates the stimulating electrode. The first test subject was a 38-year-old man 10 years after the onset of an incomplete spinal cord lesion in the C5/6 cervical segments, who had undergone eSCS 3 years earlier, followed by assisted ambulation in a rollator. Implantation of Wimagine technology and configuration with eSCS allowed free stimulation without the need for manual activation. The patient's motor and sensory function improved and he was able to walk on French canes in the field. The concept of a digital bridge between the brain and spinal cord thus heralds a new era in the treatment of motor deficits in neurological disorders [23].

 

Effect on autonomous functions

Since the first eSCS experiments in individuals with spinal cord lesions in 2011, changes in autonomic nervous system (ANS) function have also been observed. However, targeted investigations of the effect of eSCS on the ANS are still overshadowed by research on motor function, as the results are less consistent and less easily measurable. The effect of eSCS on cardiovascular function, the ability to empty the bladder or bowel, or sexual function is most often mentioned, and mostly in studies that primarily address the effect of eSCS on recovery of motor function. Only one study presented the effect of eSCS also on thermoregulation, where the stimulated proband regained the ability to sweat and tolerate environmental temperature fluctuations [11].

 

Effect on cardiovascular function

One of the first studies investigating the effect of eSCS on cardiovascular (CV) function was conducted by West et al. They presented the case of a chronic tetraplegic (AIS B) patient who was started on CV function testing 12 months after eSCS implantation. After optimization of the pacing electrode configuration, a significant increase in blood pressure, improvement in cerebral blood flow and disappearance of orthostatic hypotension symptoms occurred during the Head-Up Tilt test [24].

The effect of eSCS on blood pressure changes was also investigated by Aslan et al. They enrolled seven participants in the study. Three subjects with resting and orthostatic hypotension associated with low circulating catecholamine levels experienced an increase in blood pressure when stimulation was turned on. In contrast, in four subjects without CV instability, eSCS did not induce a significant increase in blood pressure [25].

The authors Bloom et al. focused on the assessment of CV stability and immunological profile in a woman with a chronic motor complete cervical lesion. After long-term stimulation, they demonstrated not only an increase in blood pressure and improved orthostatic tolerance during the Head-Up Tilt test, but also positive immunological changes in terms of down-regulation of the systemic inflammatory response [26].

Recently, Samejima et al. [27] published the results of the first study evaluating the effect of eSCS on the alleviation of autonomic dysreflexia (AD). In spinal cord injury above the 6th thoracic spinal cord segment, irritation below the level of the lesion results in the activation of disinhibited sympathetic neurons and the development of paroxysmal arterial hypertension. One of the stimuli is digital anorectal stimulation, which the authors used to induce AD during the examination. While anorectal stimulation resulted in elevation of systolic BP and bradycardia indicative of AD in all three participants, active eSCS during anorectal stimulation prevented the development of AD. The reason for the influence of sympathetic ganglia in the thoracic region even when the electrode was placed lumbosacrally is probably due to the blockade of visceral afferent inputs at the level of the lumbosacral spinal segments [28,29].

 

Effect on bladder and bowel emptying

The effect of eSCS on improving bladder control was addressed by Herrity et al. They included a tetraplegic patient (AIS B) in their study who was set parameters to influence the lower urinary tract during filling cystometry after completing a standard training program with eSCS. Subsequently, he was able to freely control micturition at a low stimulation rate of 30 Hz with low postmicturition residual. The efficacy of this eSCS configuration was successfully verified in four other participants [30].

The authors Walter et al. focused on monitoring changes in bladder and bowel emptying. In a tetraplegic patient (AIS B), there was an increase in external anal sphincter tone, pelvic floor muscles and detrusor pressure due to eSCS of the caudal segments. In addition, the time required to empty the bowel decreased by 55% during stimulation. Also, the neurogenic bowel dysfunction score (NDBS) decreased from 15 to 8 points [31].

The authors Di Marco et al. confirmed the positive effect of eSCS on bowel emptying. In five patients with cervical spinal cord lesions after eSCS implantation, they observed a significant reduction in the time required for emptying -⁠ from 118 min to only 18 ± 2 min. No patient developed fecal incontinence and four patients were able to eliminate the need for digital evacuation [32].

These positive changes were also registered in a study by Kandhari et al., who showed a reduction in the time required for defecation by an average of one-third and an increase in the frequency of defecation. In addition, patients reported improved perception of defecation in questionnaires and also reduced frequency of urinary and bowel incontinence [19].

 

Effect on sexual function

The results of eSCS on sexual function have been published only marginally so far. For example, Harkema et al. [11] described in the first spinal patient with eSCS, among others, an improvement in the perception of sexuality. Darrow et al. [6] described in one of two women with a complete thoracic spinal cord lesion a recovery of the ability to reach orgasm during or immediately after active stimulation. Surprisingly, Kandhari et al [19] in their study of ten subjects with complete thoracic lesions reported improved reflex and psychogenic erections using the International Index of Erectile Function (IIEF) questionnaire, although such changes are not expected given the spinal cord topography.

We have also started a study at Motol University Hospital to investigate the potential of eSCS to restore sensorimotor and autonomic functions in patients with chronic complete thoracic spinal cord lesions (NCT05690074). So far, we have published the results of two patients in whom eSCS was able to affect the ability to ejaculate. The first participant regained his ejaculatory reflex during one of the eSCS stimulation programs at home using penile vibrostimulation. This surprising finding was verified in clinical conditions and subsequently confirmed during eSCS in the same segment in the second participant. The stimulated L4 segment was previously verified as the segment where the interneurons forming the so-called spinal ejaculatory generator (SEG) are located. In our study, we were the first to target the SEG with eSCS and alleviate ejaculatory dysfunction [33].

As part of the above-mentioned E-STAND study, Prof. Darrow's research team evaluated the effect of eSCS on the sex life of three women after a sensorimotor complete spinal cord injury (AIS A) in the thoracic segments. In all patients, eSCS had a positive impact on quality of sexual life, sexual desire, arousal or ability to achieve orgasm. Sexual anxiety was reduced by 55% [29].

 

Limits of the method

Despite the groundbreaking results, this method has its clear limitations. The invasiveness of the surgery is undeniable, which poses a risk of perioperative and postoperative complications. The authors of Boakye et al. reported in their evaluation of 25 patients after eSCS implantation, two of them (8%) had infectious complications requiring removal of the stimulator and in one case the development of an ileus postoperatively. Among late complications, they presented one case of electrode malposition and one patient with femoral neck fracture during standing and walking training [14]. Authors Pino et al. evaluated the results of 14 subjects in the E-STAND study and reported no surgical complications, with only one patient experiencing stimulator failure requiring replacement after 4 months [34].

Another limitation is the software capabilities of commercially available stimulation systems. Due to the development of devices for the purpose of influencing pain conditions, manufacturers only allow a limited number of sequential programs, which limits the training of the gait stereotype. It is also difficult to isolate individual movements as triggering a single contact on the electrode leads from the posterior horns via interneurons to activate motor pathways for a wider group of muscles. Insufficient trunk stabilization is also a limiting factor, which poses a risk of overload and secondary damage to the lumbar spine and operated segments.

The impact on the autonomous system is also proving to be a significant limitation so far. This is due to the placement of the electrode with the primary goal of stimulating the L1-S1 segments, so that the sacral segments S2-S5, where the reflex erection centre or e.g. the sacral micturition centre is located, cannot be reached in most cases.

 

Conclusion

The long-term results of the use of eSCS are still under scrutiny. However, they already give some hope for further advances in the understanding of the pathophysiological processes after spinal cord injury and suggest a future trend in the treatment of spinal cord injury patients. Despite its invasiveness, the method offers a wide range of applications. In addition to restoring or improving motor skills, it offers the possibility of improving the perception of one's own body and influencing autonomic functions with a significant impact on quality of life.

 

Conflict of interest

The authors declare that they have no conflict of interest in relation to the subject of the study.