Improving postural stability with proprioceptive rehabilitation in patients with diabetic a double-blind randomised controlled trial

Authors: Kavinda Malwanage, Esther Liyanage, Vajira Weerasinghe, Charles Antonypillai, Indumathie Nanayakkara

Abstract

Objectives

Postural stability (PS) refers to the ability to maintain balance during various movements. Proprioception deficits in patients with diabetic polyneuropathy (DPN) often lead to significant postural sway and postural unsteadiness during daily activities, increasing fall risk and diminished quality of life. Therefore, this study aimed to investigate the effectiveness of a 12-week proprioceptive rehabilitation programme (PRP) on PS in patients with DPN compared with conventional rehabilitation.

Methods

This double-blind randomised controlled trial involved 112 patients with DPN, who were randomly allocated to intervention (n=57) or control (n=55) groups. The intervention group received PRP three times per week for 12 weeks whereas the control group received a conventional balance and strength training programme (BSP) for the same duration. PS scores and anterior–posterior (AP) sway angle were measured under six conditions of the Clinical Test for Sensory Interaction on Balance test at baseline and post intervention.

Results

Following the 12-week PRP, PS scores significantly improved in the intervention group by 67.31%–187.81% across different platform types and visual conditions under the six test conditions. In the control group, the postural stability scores improved only in the eyes-opened condition by 36.45% and 50.05% on firm and foam surfaces. AP sway angle improved by 42.46%–56.84% across all test conditions in the intervention group, whereas in the control group there was no improvement.

Conclusion

This novel 12-week PRP had a positive impact on PS of patients with DPN compared with conventional BSP. Future randomised controlled trials may be carried out to examine how PRP affects diverse functional status, varying with difficulty levels.

Introduction

Diabetes mellitus (DM) often leads to serious long-term complications, with diabetic polyneuropathy (DPN) being a prominent microvascular complication. The underlying mechanisms may involve nerve damage and apoptosis caused by glucose toxicity,^1 2^ oxidative stress, inflammation or neurovascular ischaemia stemming from DM.

Proprioception refers to the ability to perceive joint position and orientation in space,^3^ as well as force generation, effort and awareness of the body in space.^4 5^ Proper integration and processing of movement patterns depend on accurate proprioceptive information. Conversely, proprioception deficits lead to inaccurate feedback mechanisms for motor control, resulting in ineffective and inefficient movements. As lower limb proprioceptive inputs are crucial for the adjustment and intramuscular coordination of postural movements, it is noteworthy that proprioception effectively controls the postural movements during activities of daily living (ADLs).^6^

Patients with DPN exhibit lower limb proprioceptive deficits compared with healthy controls, leading to impaired motor control, coordination and overall body awareness.^7 8^ These proprioception deficits in foot and ankle often act as the primary contributor to postural instability in DPN, compromising the essential feedback mechanism of the somatosensory system for maintaining postural stability (PS).^9^ Patients with DPN experience significant postural unsteadiness, evidenced by increased sway speed, velocity moment and medial–lateral (ML) and anterior–posterior (AP) displacements.1014 The concept of postural control encompasses achievement, maintenance or restoration of balance during various activities.^15 16^ It is crucial for ADLs, with somatosensory function contributing 60%–75% in controlling static posture on stable platforms.^17 18^

Majority of patients with DPN might not be aware of their postural instability, which substantially increases the likelihood of falls. Impaired proprioception, along with alteration of soft and hard tissues of the foot contribute for postural unsteadiness, with impaired proprioception being the key determinant.^19 20^ Despite the link between proprioceptive deficits and postural unsteadiness, evidence supporting rehabilitation of lower limb proprioceptive deficits to improve PS in DPN is sparse. Healthcare practitioners may often overlook unsteadiness, as patients often attribute unsteadiness to ageing rather than illness, leading to under-reporting. With no pharmacology treatments available, exercise-based interventions are highlighted as effective, accessible and cost-efficient for improving functional status as DPN was found to be best benefited from the exercises.^21^

Despite the benefits of balance and sensorimotor training in improving posture and reducing falls being established, challenges such as unavailability of sophisticated training equipment, costs, time constraints and mastering of training hinder their broader implementation. Notably, there is a dearth of research focusing on simple, cost-effective yet effective therapeutic interventions directly addressing postural instability. Therefore, this study aimed to assess the efficacy of a 12-week proprioceptive rehabilitation programme (PRP) to improve PS score and sway angle compared with conventional strength and balance training programme in patients with DPN.

Research design and methods

Study design and setting

Participants for this double-blind, randomised-controlled trial were selected from the Diabetic and Endocrinology Clinic, National Hospital, Kandy, Sri Lanka. Data collection was carried out at the Department of Physiology, Faculty of Medicine, University of Peradeniya. This study adheres to the Declaration of Helsinki.

Participants

Patients who were aged between 35 and 60 years with type 2 DM for more than 1 year and diagnosed with DPN were included. Patients with a history of acute cardiovascular diseases, strokes, autonomic dysfunctions, severe pain and paresthesia, diabetic foot ulcers, impaired vision and amputation were excluded. Participant enrolment, allocation, intervention and follow-up procedure are shown in figure 1.

Study protocol

Validated Sinhala and Tamil versions of the Michigan Neuropathy Screening Instrument (MNSI) were used to identify the presence of DPN. MNSI consisted of two a self-administered questionnaire assessing neuropathic symptoms and a physical examination conducted by a qualified healthcare professional. The examination included the evaluation of ankle reflexes using a reflex hammer, vibration sensation using a 128 Hz tuning fork, the presence of foot abnormalities and protective sensation with 10 g monofilament testing. Patients scoring over 4 in the self-administered questionnaire and over 2 in physical examination were considered neuropathic and were recruited to the study.^22^

After obtaining informed written consent, demographic data, that is, age, sex and duration of diabetes were recorded, and participants were randomly allocated either to the intervention (n=60) or the control (n=60) groups using computer-generated simple randomisation. The intervention group followed a novel 45-min PRP, three times a week for 12 weeks. The control group received a conventional balance and strength training programme (BSP) also for the same duration consisting of 45-min sessions three times a week for 12 weeks. Participants missing more than three consecutive weeks were excluded. Both participants and assessors were blinded to group allocation.

Outcomes of interest were PS score and postural sway under six different test conditions of the Clinical Test for Sensory Interaction on Balance (CTSIB) at the baseline and after trial completion. Sample size was calculated using GPower statistical software.^23^ Accordingly, 120 patients with DPN were recruited to detect a small effect size of PRP over conventional rehabilitation, with an alpha level of 0.05, and power of 0.80, assuming a 15% dropout rate. To minimise participant attrition, the study implemented proper education, weekly follow-up calls, and in-person progress assessments throughout the rehabilitation programmes.

Test and measurement

Assessment of postural stability

PS was assessed using CTSIB,^24^ the clinical version of the sensory organisation test, which altered somatosensory inputs (firm vs foam) and visual inputs (eyes-opened, eyes-closed and visual conflict dome) across six test conditions.

A medium-density viscoelastic foam was used to alter the support surface. A Chinese lantern, aligned with CTSIB guidelines,^25^ was affixed to the participant’s head used as a visual conflict dome to provide a challenging surrounding.

Test 1 served as a baseline reference of accurate somatosensory, visual and vestibular information. Test conditions 1–6 escalated the level of sensory conflicts and postural difficulty. Patient stood barefoot with arms parallel to the body on firm (floor) or foam (foam cushion) surface. Visual conditions were manipulated by instructing the participant to stand with closed/opened eyes and to wear the visual conflict dome. Tests 1–3 were performed on the floor whereas tests 4–6 were performed on foam cushion. Eyes were open for tests 1 and 4, and closed for tests 2 and 5, and a visual dome was used for tests 3 and 6. Each condition required maintaining posture for 30 s, with the time recorded until the point of failure in seconds using a stopwatch. The average time of three trials was calculated for each condition.

To obtain the maximum AP sway angle in degrees, a lateral-view videography was captured and analysed using Kinovea V.0.9.5 motion analysis software (Joan Charmant and Contributors, 2006), which has demonstrated excellent reliability for video-based sway analysis. AP sway angle was only measured as AP stability is more closely linked to the ankle and sole muscles, while ML stability is influenced by hip musculature. A third assessor, blind to treatment allocation, extracted all pre-test and post-test data to minimise inter-rater variability. During testing, the investigator stood nearby the participant to assist if needed. Trials were terminated if the participant opened their eyes during the eyes-closed condition, changed arm/foot position, stepped or received assistance to maintain balance or prevent from falling.

Composite and equilibrium scores were calculated based on the scores obtained from all six test conditions, with higher scores indicating better stability. Composite score was the sum of three 30-s trials across all conditions (total 540 s). The equilibrium score was the sum of mean values across all conditions (total 180 s). A composite score below 260 s indicated as the cut-off score for identifying postural instability and fall risk.^26^

Interventions

Proprioceptive rehabilitation programme

This novel 12-week PRP included these exercise static balance, dynamic balance, stability challenge, reaction time and fine motor skills. Exercises under each category were designed to address key elements; participant’s concentration, rationalisation, feedback, active exercises and repetitions^27 28^ which facilitate neuromuscular adaptation by recruiting proprioceptors to improve PS.

Progression of each exercise was achieved by changing movement parameters, that is, force, length, velocity and endurance; base of support (BOS); centre of gravity (COG) sway on BOS; exercise surface; upper limb tasks; obstacles in the walkway and visual cues (online supplemental file 5).

Conventional balance and strength training programme

The conventional BSP included standing weight shifts, static balance, dynamic balance, callisthenics, lower limb (LL) strengthening and walking balance,^25 29 30^ which were also progressed by reducing BOS and increasing COG sway over BOS.

Warm-up and cool down sessions were conducted before and after each session of both the exercise programmes (online supplemental file 5).

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Statistical analysis

Two-way, group (intervention vs control) × time (pre vs post) repeated analysis of variance (ANOVA) was conducted to analyse the statistically significant difference in PS measure between the interventions and control groups, with paired t-tests for pre-test and post-test comparisons. Baseline variations in proprioceptive abilities were controlled by analysing pre–post differences in postural stability. ANOVA with Bonferroni post hoc tests was used to assess the influence of the type of platform and visual status on PS measures. The statistical significance was denoted by p values, whereas clinical significance is represented by effect size. Cohen’s d (d) was computed to interpret the magnitude of the intervention effect. Statistical analyses were conducted using IBM SPSS statistical software (Chicago, Illinois, USA) V.11.0.

Results

Demographic characteristics

Of the 120 participants, 1 participant from the intervention group and 3 from the control group were excluded as they missed three consecutive sessions. After identifying two outliers each from the intervention and control groups, the analysis proceeded with a total sample of 112 participants.

Majority of the participants were women (66.96%), and age ranged from 39 to 60 years (mean±SD: 52.06±6.90). Mean body mass index of 25.35±2.79 indicated that the majority of participants were marginally overweight. Participants were suffering from diabetes for a mean duration of 11.04±7.38 years. There were no significant differences between the intervention and control groups at baseline (online supplemental file 1).

Effect of PRP on postural stability score

Both the intervention and control groups had significantly improved PS scores following two rehabilitation programmes under Test 1 (firm surface, eyes-opened), resulting in 67.31% (p<0.001) and 36.45% (p<0.001) improvement, respectively (95% CI: 0.56, 1.34). For Test 4 (form surface, eyes-opened), both the groups showed statistically significant improvement resulting in 103.48% (p<0.001) and 50.05% (p<0.001) improvement, respectively (95% CI: 0.89, 1.71). Only the intervention group showed significant improvements in PS at the end of 12 weeks. PS improved by 152.5% (p<0.001, 95% CI: 1.55, 2.46) for Test 2 (firm surface, eyes closed), by 96.48% (p<0.001, 95% CI: 1.34, 2.22) for Test 3 (firm surface, visual conflict dome), by 187.81% (p<0.001, 95% CI: 1.27, 2.13). Test 5 (foam surface, eyes closed) and by 137.93% (p<0.001, 95% CI: 1.22, 2.08) for Test 6 (foam surface, visual conflict dome) in the intervention group.

Composite and equilibrium scores significantly improved in both groups. Results revealed a threefold increase in PS score in the intervention group 108.14% (p<0.001) following 12-week PRP compared with the control group 35.31% (p<0.001) (95% CI: 1.34, 2.22). Pre-intervention and post-intervention mean scores for the six different conditions of CTSIB are summarised in figure 2 and online supplemental table 2.

Effect of PRP on anterior–posterior sway angle

Following 12-week PRP, only the intervention group demonstrated a significant reduction in AP sway angle across the six test conditions. Reductions were attributed to 43.49% (p<0.001, 95% CI: −0.71, 0.12), 46.11% (p<0.001, 95% CI: −1.43, 0.64), 56.14% (p<0.001, 95% CI: −0.34, –1.10), 56.84% (p<0.001, 95% CI: −0.34, –1.10), 44.69% (p<0.001, 95% CI: −0.43, –1.20) and 42.46% (p<0.001, 95% CI: −0.47, –1.24) for test conditions 1 through 6 respectively. No improvement was observed in the control group following conventional rehabilitation. Pre-intervention and post-intervention sway angle (in degrees) of CTSIB are summarised in figure 3 and online supplemental table 2.

Impact of platform type on postural stability

Platform type did not affect PS scores (F(1,110)=0.12, p=0.73). The intervention group showed PS improvements of 94.37% on firm and 132.67% on foam surfaces following 12-week PRP, compared with 22.5% and 25.9% in the control group (95% CI: −0.3, 0.74). (figure 4A).

In contrast, the reduction of AP sway angle depended on platform type (F(1,110)=4.08, p=0.04). In the intervention group, sway angle decreased by 49.12% on firm and 45.91% on foam surfaces, while the control group showed minimal reductions of 0.67% and 6.07%, respectively (95% CI: 0.01, 0.76).

Impact of visual status on postural stability

Visual status significantly affected PS scores (F(1,110)=38.11, p<0.001). The highest improvement occurred under eyes-closed conditions (167.37%), followed by visual conflict dome (113.44%) and eyes-opened conditions (80.26%) in the intervention group. The control group improved only under eyes-closed conditions (41.74%, 95% CI: 0.55, 1.33) (figure 4B).

Reduction in AP sway angle did not differ significantly by visual condition (F(1,110)=1.81, p=0.18), however significant time×group interaction (p<0.001) showed the intervention group achieved substantial reductions under eyes-opened condition (50.89%), eyes-closed condition (45.13%) and with visual conflict dome (46.75%). Control group showed non-significant trivial reduction (95% CI: 0.54, 1.01).

Online supplemental table 3 presents the pre–post comparison of PS and sway angle with improvement in percentages.

Discussion

In this study, it was observed that 12-week PRP significantly enhanced PS scores and reduced AP sway across different platforms and various visual conditions. The control group showed improvement only in eye-opened test conditions.

Role of 12-week PRP in improving postural stability

The improvement of PS observed following PRP may be attributed to various factors. First, restoration of lower limb proprioception, primarily in the ankle joint, is likely to provide sufficient information for maintaining PS, allowing patients with DPN to better respond to postural unsteadiness under different platforms and visual conditions. Second, PRP included all the fundamental physiological concepts to improve proprioception. Third, PRP progressively challenged sensory inputs through various postures, platforms, BOS and visual conditions. These strategies may aim to elicit automatic and reflective responses, ensuring the stabilisation of the body during demanding daily activities through enhanced proprioceptive function.

No improvement in AP sway in the control group indicates that conventional rehabilitation was unable to intensify proprioceptive feedback, making them inadequate to mitigate postural instability. Although it focused on strengthening lower limb muscles, it did not address proprioceptive demands required for optimal postural control.

In line with the present study, notable PS improvements have been observed following a 10-session^31^ and 6-week balance training programme,^32^ 12-week sensorimotor training programme,^33^ 12-week aerobic exercise programme,^34^ in patients with DPN. Persistence reduction of AP oscillations of centre of pressure (COP) was noted even 6-month post-sensorimotor rehabilitation,^33^ suggesting a potential carryover effect of PRP.

Regarding falls prevention, PRP improved the composite score from 198.86±72.61 s to 413.91±41.92 s (p<0.001), surpassing the 260 s cut-off (90% specificity and 44% sensitivity)^26^ for fall risk. In contrast, the control group improved from 172.73±55.65 s to 233.72±68.82 s (p<0.001), but failed to reach the cut-off.

Comparison with other sensorimotor rehabilitation programmes

In the context of sensorimotor rehabilitation for enhancing PS, a study on 12-week proprioceptive exercises reported no improvement in postural control, possibly due to lack of progression and load.^35^ In contrast, the present study’s systematic progression likely contributed to better outcomes. Similarly, 6-week and 12-week proprioceptive training in women with DM showed significant AP sway reduction.^36^

8 weeks of proprioceptive training combined with balance exercises (various surfaces, thera ball and trampoline) also showed a significant reduction in postural sway in patients with DPN.^37^ A study by Stolarczyk et al found the efficacy of a 12-week exercise programme on the Biodex Balance System with biofeedback in improving PS indices under eyes-opened and eyes-closed conditions^38^ in patients with DM.

A brief 3-week exercise programme on reactive movement and sensory strategies indicated an improvement only in ML sway under eyes-opened condition.^32^ Grewal et al found that 4-week sensor-based interactive exercise with real-time joint feedback effectively reduced centre of mass (COM) sway area in ML direction under both eyes-opened and eye-closed conditions in patients with DPN, which emphasises the importance of incorporating a longer training period of 8–12 weeks, as they did not observe significant improvements in postural control.^39^

Numerous studies have explored the impact of combining proprioceptive training with other forms of exercises. EI-Wishy and Elsayed reported a significant reduction in PS indices following 8-week proprioceptive training combined with conventional physiotherapy, compared with conventional rehabilitation alone in patients with DPN.^40^ Task-specific balance training incorporated with moderate-intensity aerobic training and strengthening exercise, and a 12-week moderate-to-high-intensity aerobic exercise programme effectively improved postural control in patients with DPN. Virtual reality sensorimotor training also showed a significant reduction in COM sway.^34^ A detailed comparison of similar studies is presented in tabular format in online supplemental file 4.

Despite these approaches improving PS, they limit the ability to identify which component predominantly contributes to the specific outcome. Determining the least influential component is crucial to streamline rehabilitation, reduce time required and lessen the patient’s physical burden. Further, some sensorimotor training programmes do not adhere to the principles of proprioceptive rehabilitation, and some rehabilitation protocols rely on sophisticated equipment, which may pose economic challenges for patients in developing countries. Limited access to equipment may further hinder patients from benefitting from these exercises, despite their effectiveness in enhancing PS.

Influence of type of platform and visual status on improvement of postural stability

Patients with DPN were found to exhibit increased postural instability under unstable conditions such as on foam surface,^9^ eyes-closed.^14^ Following PRP, greater improvement was achieved on foam surfaces compared with firm surfaces, as proprioceptors play a crucial role in stabilising posture in unstable conditions. This suggests that participants effectively used somatosensory information for PS under challenging conditions in which this effective reweighting of sensory information is crucial for achieving good postural control following 12-week PRP.

Under different visual conditions, PRP resulted in significant PS improvements, particularly with eyes-closed, where proprioceptive feedback compensated for the lack of visual input. Improvements were also observed with the visual conflict dome, followed by eyes-open conditions. PRP appears to facilitate somatosensory integration, enabling better PS under visually challenging conditions.

Therefore, PRP emerges as a cost-effective, straightforward solution for addressing postural instability in community-dwelling DPN patients. It can be implemented as a home-based or hospital-based rehabilitation programme, overcoming barriers like time, transport or financial constraints, while requiring no sophisticated equipment. This approach is particularly empowering for patients in low and middle-income countries, enabling them to proficiently manage postural unsteadiness during various physical activities, facilitating effective active participation in ADLs.

Clinical implication

The 12-week PRP is a simple, safe, cost-effective and adaptable rehabilitation protocol. With high patient satisfaction and compliance combined with minimal adverse effects, it can be implemented in both home-based and hospital-based settings. Physiotherapists, endocrinologists and other healthcare professionals may consider adopting this protocol. Policymakers may consider incorporating this protocol into DPN clinical practice guidelines.

Limitations

The study lacks post-intervention follow-up to determine potential carryover effects of PRP. Lack of objective tools for measuring PS and the sway index limits the external validity. Measuring PS as AP displacement in degrees, limits comparability with studies using COP or COM displacement. Future studies should explore these findings across varying DPN severities to determine if improvements are influenced by DPN severity.

Conclusions

This study investigated the effectiveness of a novel, 12-week PRP on PS score and AP sway angle in patients with DPN. Results revealed that 12-week PRP significantly improved PS and reduced sway angle across different platforms and various visual statuses compared with the conventional rehabilitation. Conventional strength and balance training improved PS only under eyes-closed conditions. Therefore, the 12-week PRP, developed in this study is a beneficial, cost-effective and safe intervention for patients with DPN. When incorporated with glycaemic control, PRP may serve as an effective strategy for preventing and managing long-term complications of DPN.

Supplementary material