Improvements in bladder, bowel and sexual outcomes following task-specific locomotor training in human spinal cord injury

Abstract

Objective: Locomotor training (LT) as a therapeutic intervention following spinal cord injury (SCI) is an effective rehabilitation strategy for improving motor outcomes, but its impact on non-locomotor functions is unknown. Given recent results of our labs' pre-clinical animal SCI LT studies and existing overlap of lumbosacral spinal circuitries controlling pelvic-visceral and locomotor functions, we addressed whether LT can improve bladder, bowel and sexual function in humans at chronic SCI time-points (> two years post-injury).

Study design: Prospective cohort study; pilot trial with small sample size.

Methods: Eight SCI research participants who were undergoing 80 daily one-hour sessions of LT on a treadmill using body-weight support, or one-hour of LT and stand training on alternate days, as part of another research study conducted at the Kentucky Spinal Cord Injury Research Center, University of Louisville, were enrolled in this pilot trial. Urodynamic assessments were performed and International Data Set questionnaire forms completed for bladder, bowel and sexual functions at pre-and post-training time points. Four usual care (non-trained; regular at-home routine) research participants were also enrolled in this study and had the same assessments collected twice, at least 3 months apart.

Results: Filling cystometry documented significant increases in bladder capacity, voiding efficiency and detrusor contraction time as well as significant decreases in voiding pressure post-training relative to baseline. Questionnaires revealed a decrease in the frequency of nocturia and urinary incontinence for several research participants as well as a significant decrease in time required for defecation and a significant increase in sexual desire post-training. No significant differences were found for usual care research participants.

Conclusions: These results suggest that an appropriate level of sensory information provided to the spinal cord, generated through task-specific stepping and/or loading, can positively benefit the neural circuitries controlling urogenital and bowel functions.

Trial registration: ClinicalTrials.gov NCT03036527.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1. Methods flowchart.

Urodynamic and questionnaires (bladder, bowel and sexual function assessments) were conducted after clinical evaluation for eligibility. After approximately 80 locomotor training sessions or usual care (at home in usual routine between assessments for an equivalent time period for 80 training sessions), assessments were repeated. Each participant served as their own control. Individual and group data were analyzed at the conclusion of the study.

Fig 2. Detrusor pressure recording example.

Raw recording of detrusor pressures (Pdet) from research participant A57 for the initial fill cycle pre-training and post-training. Note that the urge sensation precipitated by the rise in the bladder pressure immediately preceded the onset of leak, which occurred at almost twice the fill volume post-training. On the pre-training fill/void cycle, the leak occurred at 05:54 (min:sec), while post-training the leak occurred at 10:33 (20 ml/min–constant fill rate).

Fig 3. Bladder capacity data summary.

Bladder filling ceased and capacity measured (leak + residual volumes) with the occurrence of either spontaneous urine leakage (n = 5; reflex void), autonomic dysreflexia (n = 2; Participants A60 and C42) or a voluntary void following a strong urge (AIS D Participant C43). (A) A comparison of pre- and post-training bladder capacity values in each of the eight participants. A binomial proportion test indicates that a significant majority of the research participants demonstrated an improvement in bladder capacity (vs. random occurrence, p

Fig 4. Blood pressure recording example during bladder filling.

Blood pressure recordings for research participant C42 during filling pre- and post-training. Filling was stopped at the onset of autonomic dysreflexia (increase of > 20 mmHg), as indicated by the asterisk (*). Note that the shift to a higher volume post-training is indicative of an increase in bladder capacity. Blood pressure returned to baseline by the time the bladder was emptied by catheterization for residual volume measurement.

Fig 5. Voiding efficiency data summary.

A comparison of pre- and post-training voiding efficiency (VE) values in the eight research participants. While the majority of participants demonstrated an improvement in VE, participant C42 did not have a leak/void and participant A60 just leaked a few drops. When considering the six research participants that had a measurable efficiency, a significant improvement in VE values occurred collectively post-training (p = .046; 39.6 ± 15.5% vs 63.9 ± 8.9%).

Fig 6. Leak point pressure data summary.

Intravesical bladder pressure values at the onset of leak/void in six of eight research participants during cystometry. (A) A comparison of pre- and post-training bladder leak pressure values in each of the six participants. A binomial proportion test indicates that a significant majority of the research participants demonstrated an improvement in overall bladder leak point pressure (vs. random occurrence, p 2O).

Fig 7. Bladder contraction data summary.

A comparison of pre- and post-training bladder contraction values (area under the curve and duration) for the eight research participants receiving LT. All research participants had an improvement in the detrusor contraction area (A) and 7/8 had improved duration (B). Collectively, the group values post-training (C) were significantly greater compared to pre-train values (area—p = .016; duration–p = .019). An example of the entire detrusor muscle contraction cycle is provided for participant A59 (D) to illustrate a greater capacity for bladder filling was maintained post-training and generated a larger voiding contraction for a longer duration.

Fig 8. Bladder compliance data summary.

Most research participants (85%) had low compliance (2O - ΔP/ΔV) at the pre-training baseline cystometric recording. However, post-training, 50% showed an improvement in bladder filling under lower pressures. A ranking and average of the compliance values at the pre-training time point indicate a median (horizontal line) and mean (small white square) value of 6.3 and 7.0 ± 4.8 cm/H2O, respectively, while post training, the median and mean values were 9.0 and 15.5 ± 16.8 cm/H2O, respectively (p>.05).

Fig 9. Participant estimated time for defecation.

Average time for defecation over “last four weeks” for seven of eight research participants (bowel data set used beginning with participant A57). Pre-/post-training group means were 57.9 ± 18.2 and 35.7 ± 26.2, respectively. A binomial proportion test indicates that a significant majority of the research participants demonstrated an improvement in bowel function (vs. random occurrence, p

Fig 10. Summary of sexual function data.

Sexual function domain scores for desire and satisfaction (total score from two 5 point questions). Group means are provided in the bottom histogram. Whereas scores for desire significantly increased post-training (*, p = .04), overall satisfaction remained unchanged. Note that scores above the two point minimum for satisfaction were given when research participants reported engagement in sexual activity.

Fig 11. Summary of cystometry data for usual care participants.

Summary of cystometry results for the four non-trained research participants at two different time points. No significant differences were found between any parameters. Shown are group means (± SD) for capacity, leak point pressure (LPP), maximum detrusor pressure (MDP) and contraction duration (CD).