Home‐based multi‐sensory and proximal strengthening program to improve balance in Charcot–Marie–Tooth disease Type 1A: A proof of concept study

People with Charcot–Marie–Tooth Disease (CMT) frequently report problems with balance, which lead to an increased risk of falls. Evidence is emerging of training interventions to improve balance for people with CMT, but to date all have relied on clinic‐based treatment and equipment. This proof‐of‐concept study explored whether a multi‐modal program of proprioceptive rehabilitation and strength training can be delivered at home, to improve balance performance in people with CMT Type 1A.

Discussion: The intervention was feasible to implement and safe, with some evidence of improvement in balance performance.This supports future studies to expand this intervention to larger trials of pragmatic, home-delivered programs through current community rehabilitation services and supported self-management pathways.

| INTRODUCTION
2][3][4][5][6] Two studies have reported falls events in CMT cohorts of 50% 1 and 86%. 4 Effective dynamic and static balance requires a complex interplay between motor control and the somatosensory, vestibular, and visual systems. 7Proprioceptive impairment and distal leg weakness have been correlated with static balance impairment for people with CMT Type 1A. 5,6Proximal disuse atrophy has also been observed, 8 with knee extensor strength being linked to forward reach distance. 95][16][17] These studies use clinic-based training methods, some with specialized equipment, and we suggest that training people at home will reduce burden of travel, promote self-management, and be better incorporated into daily lifestyles.
In this proof-of-concept study, we propose that a rehabilitation program of multi-sensory balance training and proximal strengthening exercises can be feasibly delivered in a person's own home.

| METHODS
A single-blinded randomized, two-arm design was used in this Phase 1 study.Ethical approval was granted (NRES Research Committee REC reference number 16/LO/0720) and informed consent was acquired from all participants.

| Study population
Adults were recruited from a specialist, neurological center.This study focused on people with CMT Type 1A to reduce group variability in this small cohort.They were included if they had a genetic diagnosis of CMT Type 1A, a history of falls and could walk for 50 m, with aids if needed.Participants were excluded if there were co-morbidities that effect balance and walking.

| Allocation
A single, one-to-one falls education session was delivered to all participants, who were then randomized to the control or intervention group, on completion of the baseline measurements (Figure S1).The control group were advised to continue with usual activities.The intervention group received a bespoke, 12-week exercise prescription of home-based multi-sensory balance rehabilitation and proximal strengthening exercises.An unblinded physiotherapist (L.E. L.) performed a risk assessment in the participant's home and delivered the program through three home visits and weekly phone calls.

| Multi-sensory balance training
Training was designed to address known contributors to balance impairment for people with CMT (Table S1).Postural stability was challenged daily using sensory and mechanical perturbations to train proactive and reactive balance responses.The program was individualized to baseline ability by altering sensory feedback and base of support and progressed in difficulty using these principles (Table S2).

| Strength training
Individualized, proximal lower limb and trunk, body weight-based, resistance exercises were performed on alternate days.Progression was achieved through increasing the difficulty of the anti-gravity position, degree of body weight, and time under tension.Exercise selection was a collaborative process and tailored to participant's preferences (Table S3).

| Self-management
The intervention was underpinned by an established self-management support framework and focused on augmenting engagement in the home-based training by promoting self-management techniques such as collaboration, problem solving, reflection and self-discovery. 18DZIEC ET AL.
Participants were supported to identify safe and effective ways to integrate the intervention into their daily routines and set personalized goals around engagement.

| Monitoring
The intervention group received 12 weekly phone calls and the control group 3 monthly calls.

| Feasibility and safety
Engagement with the program was monitored using weekly phone calls asking set questions to record program participation rate, challenges, and any adverse incidents (Table S4).This was supported by a smartphone app.Participants were also invited to individual, semistructured interviews after the study (Table S5).

| Demographics
At baseline, age and sex of participants was recorded with disease severity measured using the CMT Exam Score (CMTES). 19

| Balance and gait performance
Baseline measurements were conducted by blinded evaluators (M.M. D., C. M.) after enrolment and after the 12-week intervention or control period (Figure S1).
Functional balance and mobility were assessed using the Berg Balance Scale 20 the BEST Test, 21 the Functional Gait Assessment (FGA), 22 and the 10-m walk test (10MWT) at self-selected and fast pace. 23stural stability was measured using static posturography (details in Table S6) with the following variables recorded over 30-s standing trials: center of pressure path length (COP PL); center of pressure velocity (COP VEL); body sway path length (C7 PL); body sway velocity (C7 VEL). 24

| Muscle function
Lower limb strength was assessed with hand-held dynamometry using a "make-test" protocol (CITEC Handheld Dynometer model CT 3002, Netherlands). 25

| Patient-recorded outcome measures
Participants completed the Walk-12 scale measure perceived walking ability, 26 the Short Form 36 (SF36) 27 for health-related Quality of Life (QoL), International Physical Activity Questionnaire (IPAQ), 28 Hospital Anxiety & Depression Scale (HADS), 29 and the Falls Self efficacy scale 30 to explore fear of falling.

| Analysis
Feasibility was explored through participation in the program, as a proportion of the total prescribed sessions completed.Postintervention, individual interviews were transcribed, and qualitative data were analyzed using thematic analysis. 31e effect of the intervention on quantitative measures was examined using the Hedges G effect size calculation due to the small sample.Effect sizes were categorized as strong if 0.7 or over, moderate if 0.3-0.69,and small if 0.29 or less.

| Participant recruitment and engagement
Fifteen participants were recruited to this study, with two dropouts (Figure 1).There were no significant differences in demographics and disease severity (CMTES score) between the intervention and control groups (Table 1).

| Feasibility and safety
There was high engagement with 91% mean completion of strength exercises and 79% of balance exercises, and there were no adverse incidents.Interviews highlighted several themes relating to the intervention: Participants appreciated the individualized nature of the exercise prescription and acknowledged benefits of doing them at home.The physiotherapist was a key motivating and reassuring factor.Some participants described positive learning about their balance impairments, increasing their confidence in managing risk.Conversely, other participants described feeling more vulnerable and aware of their unsteadiness.
T A B L E 1 Participant baseline demographics.

| Muscle function
Changes in lower limb strength were inconsistent, with moderate effect sizes favoring the intervention group for hip extension and ankle dorsiflexion.Moderate effect sizes favored the control group for knee flexion and knee extension with a large effect for plantarflexion (Table 2).

| Patient-reported outcome measures
Perceived walking ability (Walk-12) and health related QoL (SF-36) demonstrating moderate effect sizes in favor of the intervention group (Table 2).

| DISCUSSION
In this small cohort of adults with CMT, a home-based balance intervention was feasible and safe, with some early signal for improvement in balance performance that is in keeping with clinic-based balance interventions. 14,16,17This individualized, flexible program was well received with high engagement.
Large effects in favor of the intervention were observed in multiple functional measures, indicating a consistent finding that is applicable to daily life balance challenges 32,33 and consistent with other balance studies. 14,16Moderate effect sizes indicated improvements in the posturography variables for the intervention group consistently in the eyes open, feet apart conditions.Balance training alone may not be sufficient to improve stability with more challenging conditions, for example, feet together and eyes closed, due to the degree of sensorimotor impairment, and we may need to look towards alternative strategies, for example, orthotics. 34,35ere were inconsistent effects on muscle strength that may be a function of the small sample, with previous studies of strength training including samples of 18-60. 13,36The sample sizes are insufficient at this stage to account for inter-subject variability at baseline and with training response.The large standard deviations highlight reliability issues with hand-held myometry, despite a good interrater reliability with isokinetic dynamometry. 37MRI could also be a valuable tool to explore changes within the muscle in response to training, as used in a study of distal muscle training in children and young adults with CMT1A. 36 Large effect sizes were observed in the FGA, the Berg Balance Test, and the Best Test in favor of balance training (Table2).Large effects T A B L E 2 Mean differences pre and post intervention or control, with Hedges G effect size statistic.