Management of neurogenic lower urinary tract (LUT) dysfunction in people with spinal cord injury (SCI) has centered on pharmacotherapy, emptying techniques (e.g. intermittent catheterization), and surgical interventions. In other populations who experience LUT dysfunction, current practice emphasizes activity-based exercises targeting the pelvic floor and abdominal muscles (i.e. the ‘core’) as the first line of treatment. However, exercise-based approaches to address urinary incontinence (UI) in SCI may have been overlooked because of assumptions that motor function would be absent in the core muscles below the level of injury. However, recent findings from our laboratory present the possibility that the pelvic floor muscles (PFM), along with the other muscles of the core, demonstrate preserved function following SCI that can be elicited by exoskeleton-assisted walking. The overall goal of this project is to determine the feasibility and compare the possible effects of an overground vs. treadmill exoskeleton-assisted walking program on LUT function in people with motor-complete SCI.

This is a parallel-group randomized pilot trial of individuals with motor-complete SCI. Inclusion criteria are: American Spinal Injury Association Impairment Scale (AIS) A or B classification, at least 12 months post-injury, injury level above T10, presence of UI (defined as >1 leak per day), and no change to bladder medications in the past month. Subjects should have control of the triceps muscle in order to use hand-held walking aids; strap-on hand grips may be used if necessary. The exclusion criteria are: current urinary tract infection or other bladder irritations (e.g., bladder stones), injection of botulinum toxin A into the detrusor or external urethral sphincter less than 6 months ago, progressive spinal condition (e.g. spinal tumor and syrinx), lower motor neuron signs (e.g. cauda equina syndrome) or other neurological injury besides SCI, and presence of other conditions that are contra-indications for exoskeleton-based training.

Training Intervention: Participants were randomized using computer-generated sequences with allocation concealment to receive either Lokomat or the Ekso walking program. Group allocation was not revealed until after the completion of the baseline (pre-training assessments). Both training interventions consist of a 12-week exercise program involving of 45-minute sessions held 3 times/week in our laboratory. 

The EksoTM (Ekso Bionics, Richmond, CA) is a computer-controlled lower limb robotic suit used to practice standing and walking overground. Linear motors control flexion/extension at the hip and knee joint while a neutral ankle joint with a stiff spring supports dorsiflexion during swing. Users actively control walking through subtle forward-lateral weight-shifting motions to position their center of mass over the support limb in order to initiate the swing phase of the contralateral limb. A 2-wheeled walker or forearm crutches are required to support balance. Subjects are progressed from ‘FirstStep’ mode, where a physiotherapist manually triggers the initiation of each step to ‘ProStep’ mode, where the steps are automatically triggered when weight-shifting targets are achieved.

The Lokomat® (Hocoma, Switzerland) is a computer-controlled gait training system comprised of an exoskeletal driven gait orthosis, a dynamic body weight support system, and a treadmill. The device consists of a pair of adjustable exoskeletal arms that are strapped to the subject’s thighs and lower legs. Computer-controlled linear motors embedded within the exoskeleton control flexion/extension at the hip and knee; foot lifters are used to maintain the ankle in a neutral position.

Outcome Measures: Outcomes were assessed at pre-training (within 2 weeks of the start of training) and post-training (within 2 weeks following the last training session).

As a pilot trial, feasibility was our primary outcome, which was evaluated by various measures: recruitment rate (the number of participants approached, screened for eligibility, and randomly assigned, as well as reasons for exclusions), compliance (the number of participants who received their allocated intervention as well as their attendance rate to scheduled sessions, with reasons for drop-outs/missed sessions recorded), and adverse events.

Urodynamics were performed in accordance with the ICS “Good Urodynamic Practices”, by a nurse blinded to treatment allocation. Surface EMG perianal patches were applied to monitor PFM activity during urodynamics. Improvement in LUT function was defined by improvements in one or more of the following parameters: volume at first contraction (mL), compliance (mL/cmH2O), maximum detrusor pressure (cmH2O), and volume before leak/cystometric capacity (mL).

To assess LUT symptoms and quality of life (QoL), we used the Qualiveen-30 questionnaire, which has been shown to be a valid and reliable measure in individuals with LUT dysfunction due to SCI. We examined the four domains of this questionnaire by reporting the mean scores of limitations, constraints, fears, and feeling.

To-date, 12 participants were approached, 10 agreed to be screened for eligibility, and 6 have been enrolled. One participant dropped out after baseline testing due to initiation of botulinum toxin A injections, so 5 were randomly assigned. All participants received their allocated intervention and attended 100% of their scheduled sessions. One participant dropped out following 4 training sessions due to lower limb fracture unrelated to the study procedures. One participant experienced minor skin abrasion on both shins initially when using the Lokomat, but these minor wounds closed within a week and additional padding during subsequent training prevented further injury. Post-training results from one subject were omitted due to subject-reported illness including recurring bladder infection and flu symptoms. 

Urodynamics parameters showed no improvements in LUT function between pre- and post-training for the three participants with pre- and post-training urodynamic results (Table 1).

In contrast, self-reported LUT symptoms and QoL indicated that participants in the Ekso group tended to report stability or improvements across all four domains of the Qualiveen questionnaire while the two participants who completed Lokomat training reported stability or worsening of symptoms, except for one who had an improvement in the feeling domain (Fig. 1).

In this ongoing study, our data so far support the feasibility of testing exoskeleton-assisted walking programs in people with SCI, with an acceptable enrollment rate of 60% and drop-out rate of 20%, in line with other rehabilitation intervention studies of this length. Although the urodynamics data to-date did not provide an indication of objective changes in LUT function following training, the results from the Qualiveen questionnaire provide some promising benefits of Ekso-assisted training on self-reported bladder symptoms and QoL.

These preliminary results offer some interesting possibilities for the opportunity to integrate rehabilitation training programs in the management of LUT symptoms in people with SCI.