Non-Invasive Neuromodulation for Upper Extremity Function in Spinal Cord Injury: A Crossover Trial

Project Aims

The key objective of this pilot is to explore the feasibility of a comparative trial between two non-invasive neuromodulatory approaches for restoration of upper extremity function after SCI, and to work towards the identification of participant cohorts most suited for each intervention (i.e. responders). This objective will be achieved through the following:

1. Evaluation of clinical safety and effectiveness of each intervention, within participants, through safety monitoring and the use of standardized assessments of upper extremity strength, sensation and movement performance.
2. Evaluation of clinically meaningful improvements through participant reported outcomes related to activities of daily living, QoL and global impressions of change.
3. Neurophysiological assessments to assess the mechanisms of recovery following each interventional period, but also to aid in the identification of neurophysiological correlates of "responders" to each therapy.

In this randomized crossover study, participants will receive two months of non-invasive tSCS over the cervical spinal cord in combination with functional task practice (minimum of 12 sessions/month) preceded or followed by two months of PCMS and functional task practice, with a 4-week washout period between each intervention. With this design we will be able to examine the relative efficacy of each paradigm within a single participant and begin to identify neurophysiological and/or clinical characteristics that may help to guide the clinical identification of suitable participants for each intervention.

The minimum training dose for each interventional block is 12 sessions/month, for a total of 24 sessions of training during each treatment phase. Participants will be invited to complete up to 5 sessions per week if feasible. Participants will be required to complete a comparable number of intervention sessions (i.e., within 10%) in each of the two interventions to ensure that any effects are not attributed to differences in dosage. Each intervention session will include at least 60 minutes of upper extremity functional task practice (FTP). The FTP will be completed with typical rehabilitation activities (e.g., range of motion, strengthening activities, moving paperclips, clothespins, cups, etc.) as well as the ReJoyce rehabilitation workstation (Rehabtronics, Alberta), which incorporates computer games into the rehabilitation activities.

Transcutaneous Cervical Spinal Cord Stimulation (tSCS) Cervical transcutaneous spinal cord stimulation will be delivered following previously reported guidelines. Circular (1.25") electrodes will be placed over the cervical spinal cord between the C3 and C4 vertebral level, with rectangular return electrodes placed either over the clavicles or iliac crests. Biphasic stimulation will be delivered continuously at 30Hz, with a 1ms pulse width, and overlapping 10kHz carrier frequency (Gelenitis et al 2024; Moritz et al 2024). Stimulation will be delivered at a current intensity below motor threshold, that is below the intensity required to elicit a muscle twitch. Cervical tSCS will be delivered for at least 60 minutes during each training session, and FTP will occur during the active stimulation.

Paired Corticospinal Motoneuronal Stimulation (PCMS) Paired transcranial magnetic stimulation and nerve stimulation will be implemented following a previously-published protocol for individuals with SCI (Jo et al 2023). Two figure-eight coils will be used to deliver bilateral magnetic stimulation of the motor cortex and will be combined with synchronous electrical stimulation of the brachial plexus and ulnar nerves, bilaterally. The timing of stimuli will be determined to ensure that electrical volleys arrive at the corticospinal motoneuronal synapse at an appropriate interval to induce long term potentiation. During each session, 360 paired stimuli will be delivered at a frequency of 0.2 Hz. The stimulation paradigm should take approximately 30 minutes. Following this, participants will then complete 60 minutes of FTP.

Functional Task Practice (FTP) A typical rehabilitation program will be defined and delivered in consultation with physical and occupational therapists. The program will consist of functional tasks including repetitive activities of gross upper extremity movement, isolated finger movements, simple and complex pinch, and grasping activities. For each category, 4-10 activities with various difficulty levels will be identified. Each participant will perform at least 1-2 exercises within each category during each treatment session. Activities in each category will be chosen according to the participants' ability and adjusted based on the evolution of the ability to perform the functional task over time (graded training).

Assessment Measures

Assessments will occur at baseline then monthly for five months, corresponding to the beginning, midpoint and end of each treatment block. All participants will also have a three-month follow-up assessment to evaluate retention of any intervention effects (i.e. carryover neuroplasticity).

Outcomes Clinical measures will be completed by an occupational or physical therapist who is blinded to participants' group allocation. Assessments will include clinical measures of upper extremity strength, sensation and function including the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) and related subscores such as Upper Extremity Motor Score and Sensory Scores as well as the Graded Redefined Assessment of Strength Sensation and Performance (GRASSP). Patient reported outcomes will evaluate self-reported ability to perform activities of daily living (SCIM-III), overall quality of life (WHO-QoL BREF) and participant global impressions of change (PGIC) at the beginning and end of each intervention block (Figure 1).

Neurophysiological assessments will include transcranial magnetic stimulation (TMS) over the upper limb motor cortex to assess the excitability of the corticospinal tract, as well as stimulation to cervical dorsal roots to evaluate changes in cervical spinal cord excitability from the evoked multisegmental motor responses (cMMRs). TMS of the cortex has been used extensively in humans to assess the integrity of the descending corticospinal tract (Di Lazzaro et al 2008). Participants will be seated in their wheelchair. Single pulse TMS will be delivered at increasing stimulation intensity, with three stimuli (5s apart) at each intensity, until a plateau in evoked responses is reached, or we reach 100% maximal stimulator output or participant tolerability. Surface EMG will be recorded from multiple upper extremity muscles, bilaterally.

In a single visit, cMMRs will be evoked transcutaneously using single stimuli (1ms monophasic pulse). Muscle responses will be collected from multiple upper extremity muscles as per above. Stimulation will be delivered at an intensity below motor threshold for all muscles and will then be increased progressively until a plateau in evoked responses across the multiple muscles is obtained, or until participant tolerance is reached. The intensity increments will be chosen to ensure at least 9 measurements along the linear portion of the recruitment curve. Participants will receive 3 stimuli at each intensity, and stimuli will be delivered 5s apart (D'Amico et al 2013). Maximal compound muscle action potentials (Mmax) will be obtained at completion of the session by supramaximal stimulation of the brachial plexus (0.2ms pulse width).

Data Analysis Neurophysiology Measures The area of the reflex responses and MEPs obtained from TMS will be normalized to the Mmax area corresponding to the same muscle. The areas (%Mmax) will be plotted against the stimulation intensity (as a % maximum stimulator output (%MSO) or % of motor threshold (xMT)) and fit with a 3-parameter sigmoid function (D'Amico et al 2020, Klimstra \& Zehr 2008). The stimulation intensity required to elicit a response 50% of the maximum response (S50) and the maximum evoked responses (cMMRmax) will be calculated for each muscle by fitting the experimental data with a three-parameter sigmoid function (D'Amico et al 2013, Klimstra \& Zehr 2008). These measures will then be compared between the two interventions (tSCS and PCMS) within each participant.

Clinical and Patient-Reported Outcome Measures Scores on each patient-reported outcome measure and clinical assessment will follow the standard protocol for scoring each assessment. All scores will be compared between baselines, end of each intervention period, and 3-month follow-up.