Can Postural Instability in Individuals with Distal Radius Fractures Be Alleviated by Concurrent Cognitive Tasks?

Abstract

Background

Although impaired postural control may be a risk factor for distal radius fractures (wrist fractures), which often are caused by falls, little attention has been given thus far to the various performance and neurophysiologic aspects involved. Although studies suggest that external focus and cognitive tasks can improve postural control, it remains unclear whether these benefits are observed in individuals with a history of distal radius fracture and to what extent.

Questions/purposes

(1) To compare patients with a history of distal radius fracture to age- and sex-matched controls in terms of postural stability while standing on stable and unstable support surfaces, using both postural sway and neurophysiological measures as endpoints; and (2) to determine whether internal- and external-focus strategies and cognitive tasks can improve postural stability in these patients.

Methods

Forty patients with distal radius fracture (33 females and seven males with a mean ± SD age of 56 ± 4 years) and 40 sex- and age-matched control participants participated in the study. We recruited patients with a history of fall-induced distal radius fractures occurring between 6 and 24 months before the start of our study. We excluded patients who had any of the following: fear of falling, taking any medication that may affect balance, neurologic disorders, dizziness, vestibular problems, Type II diabetes, musculoskeletal disorders or recent history of lower extremity fracture, any recent surgical interventions in the spine or lower limbs, and/or cognitive impairment. Of 120 patients who were being treated for distal radius fracture over the 18-month period, 91 (76%) agreed to participate and 40 eligible patients were finally enrolled. The control group included sex- and age-matched (within 2-year intervals) individuals who had never had a wrist fracture. This group was selected from attendants/relatives of the patients attending the neurology and physical medicine and rehabilitation outpatient departments, as well as other volunteers with no history of balance problems or wrist fractures. To address our primary research question, we compared the postural control of individuals with a history of distal radius fracture with the control group while quietly standing on different support surfaces (rigid and foam surfaces) using both postural sway measures obtained by a force plate as well as neurophysiological measures (electromyography [EMG] activity of tibialis anterior and medial gastrocnemius). To address our secondary research question, we compared the postural sway measures and EMG activity of the ankle muscles between different experimental conditions (baseline, internal focus (mentally focusing on their feet without looking), external focus (mentally focusing on rectangular papers, placed on the force plate or foam, one under each foot), difficult cognitive task (recalling maximum backward digits plus one) and easy cognitive task (recalling half of the maximum backward digits).

Results

Patients with distal radius fractures presented with greater postural sway (postural instability) and enhanced ankle muscle activity compared with their control counterparts, but only while standing on a foam surface (mean velocity: 5.4 ± 0.8 versus 4.80 ± 0.5 [mean difference = 0.59, 95% CI of difference, 0.44–0.73; p < 0.001]; EMG root mean square of the tibialis anterior: 52.2 ± 9.4 versus 39.30 ± 6 [mean difference = 12.9, 95% CI of difference, 11.4–14.5; p < 0.001]). Furthermore, a decrease in postural sway was observed while standing on both rigid and foam surfaces during the external focus, easy cognitive, and difficult cognitive conditions compared with the baseline (for example, mean velocity in the baseline condition compared with external focus, easy cognitive task and difficult cognitive task was: 4.9 ± 1.1 vs 4.7 ± 1 [mean difference = 0.14, 95% CI of difference, 0.11–0.17; p < 0.001], 4.6 ± 1 [mean difference = 0.25, 95% CI of difference, 0.21–0.29; p < 0.001], and 4.5 ± 1 [mean difference = 0.34, 95% CI of difference, 0.29–0.40; p < 0.001] in the wrist fracture group). The same result was obtained for muscle activity while standing on foam (EMG root mean square of tibialis anterior in the baseline condition compared with external focus, easy cognitive task and difficult cognitive task: 58.8 ± 7.2 versus 52.3 ± 6.6 [mean difference = 6.5, 95% CI of difference, 5.5–7.6; p < 0.001], 48.8 ± 7.1 [mean difference = 10.1, 95% CI of difference, 9–11.1; p < 0.001], 42.2 ± 5.3 [mean difference = 16.7 95% CI of difference, 15.1–18.2; p < 0.001] in the wrist fracture group).

Conclusions

The current results suggest that patients with a history of distal radius fractures have postural instability while standing on unstable support surfaces. This instability, which is associated with enhanced ankle muscle activity, conceivably signifying an inefficient cautious mode of postural control, is alleviated by external attention demands and concurrent cognitive tasks.

Clinical Relevance

The findings of this study may serve as a basis for designing informed patient-specific balance rehabilitation programs and strategies to improve stability and minimize falls in patients with distal radius fractures. The integrative methodology presented in this work can be extended to postural control and balance assessment for various orthopaedic/neurological conditions.

Introduction

Distal radius fractures are the most prevalent fall-induced fractures, accounting for up to 20% of all fractures treated in emergency departments [1, 4]. Falling forward on outstretched hands during walking is one of the main trigger mechanisms of distal radius fractures [1, 21]. The risk for such fractures is increased in people between 45 and 64 years old [21]. Distal radius fractures are also correlated with a two- to fourfold increase in consequent hip fractures [4, 20], although a small portion of this risk could be explained by osteoporosis, a well-known risk factor for fractures in general [30]. Impaired balance and postural control has also been implied as another important risk factor for distal radius fractures [14], although, it has not received sufficient attention thus far.

Postural control can be defined as the complex control mechanism that safely keeps the body´s center of mass within the base of support by integrating different sensory information (including somatosensory, vestibular, and visual inputs) with motor output to coordinate muscles action during the performance of an activity, the maintenance of a specific posture, or the reaction to an external perturbation [24]. In addition to sensory and motor information, cerebellar activity, emotional status, and cognitive processing have also been shown to affect postural control. Many conditions and disorders, including orthopaedic pathologies and injuries, may adversely affect postural control and, hence, predispose an individual to increased risk of falls and fall-related injuries [5, 22]. Two main modes of postural control have been proposed: a consciously controlled mode, which uses a stiffening strategy and has less efficiency and greater energetic cost, and an automatic control mode, which uses fast, unconscious, and reflexive processes with greater efficiency and less energetic cost [10, 35]. It has been shown that both external attention focus (for example, focusing one’s attention externally on markers located on a force plate while standing) and the performance of concurrent cognitive tasks typically withdraw the attention from postural control, which leads to increased regulation of posture by the automatic control mode, resulting in more effective motor function and improved postural stability [12, 27-29, 35].

A recent study reported that the dynamic postural stability of individuals with distal radius fractures is impaired compared with age- and sex-matched controls, as indicated by greater scores of dynamic motion analysis using the PROPRIO 5000 Balance Reactive System [14]. However, many important research questions pertaining to the postural control of patients with distal radius fractures remain unanswered. These include investigation at the neurophysiologic level using electromyography (EMG) activity, which could provide valuable information about the organization of motor control by the nervous system during different postural tasks and conditions, as well as the effect of different surfaces (stable versus unstable), which is well documented on healthy individuals but not yet explored for individuals with wrist fractures. Furthermore, it remains unclear whether individuals with a distal radius fracture would benefit in a similar manner to fracture-free individuals when their attention span has shifted from postural control via external focus and/or cognitive task performance. Finally, investigating postural control in patients with distal radius fractures is valuable for designing personalized patient-specific balance rehabilitation strategies to address various neuromuscular deficits and requirements.

We therefore sought (1) To compare patients with a history of distal radius fracture to age- and sex-matched controls in terms of postural stability while standing on stable and unstable support surfaces, using both postural sway and neurophysiological measures as endpoints; and (2) to determine whether internal and external focus strategies and cognitive tasks can improve postural stability in these patients.

Patients and Methods

Study Design and Setting

This study was designed as a matched-case control study comparing patients who had distal radius fractures with sex- and age-matched control participants. All participants provided written informed consent before participation, in accordance with the protocol approved by the university’s ethics committee.

Participants

The study participants were selected from a pool of 120 patients who were being treated for distal radius fractures over a period of 18 months. We recruited patients with a history of fall-induced distal radius fracture between 6 and 24 months before the start of our study. These cutoffs were set to decrease the impact of transient fracture effects, and to minimize the effects of changes in health status. A fall was considered “an event resulting in inadvertently coming to rest on the ground, and not as a result of overwhelming hazard or an intrinsic event” [31]. All patients who were potentially eligible were approached; 91 of 120 (76%) agreed to participate. A total of 40 individuals were enrolled, while the remainder were excluded (n = 51). This group included 33 females and seven males with a mean ± SD age of 56 ± 4 years; six were treated surgically. The control group was selected to match in sex and age (within 2-year intervals), and included individuals who had never had a wrist fracture. This group was selected from attendants/relatives of the patients attending the neurology and physical medicine and rehabilitation outpatient departments as well as individuals with no history of balance problems or wrist fractures. Participants (in both distal radius fracture and control groups) were excluded if they had any of the following: fear of falling (a score ≥ 23 on the Fall Efficacy Scale-international) [13]; taking any medication that may affect balance; neurologic disorders; dizziness; vestibular problems; Type II diabetes; musculoskeletal injuries or disorders such as low back pain, flat foot, or recent history of lower extremity fracture; any recent surgical interventions in the spine or lower limbs; and/or cognitive impairment (a score less than 24 on the Mini Mental State Examination). Participants also completed a questionnaire to provide data on demographics, history of neuromusculoskeletal conditions, medical comorbidities, current medications, and any information related to the bone health, including history of osteoporosis medications, calcium supplements, vitamin D, and dual x-ray absorptiometry.

Demographics, Description of Study Population

Forty patients with distal radius fracture (33 females, seven males) with a mean ± SD age of 56 ± 4 years and 40 age- and sex-matched controls (mean ± SD age, 55 ± 7 years) participated in this study. No difference was observed between the two groups regarding age, sex, weight, height, or body mass index (Table 1).

Table 1.

Demographic and fracture-related characteristics of the participants in each group with and without fracture

Description of Experiment, Treatment, or Surgery

A Kistler force plate (Kistler, Winterthur, Switzerland) was used to record the center of pressure sway as a quantitative indicator of postural control. The sampling frequency to collect data on the center of pressure was set at 100 Hz. Participants were asked to quietly stand barefoot in a bipedal straight position, with their arms hanging alongside their trunks and their feet close together because postural control is more challenging with a smaller base of support [33]. Postural performance was measured under five experimental conditions while the participants stood on a stable support surface (the rigid surface of the force plate) or unstable support surface (10.5-cm foam placed on the force plate). These included baseline, internal focus, external focus, as well as easy and difficult cognitive tasks. Participants were required to look straight forward during all conditions. The trials for each condition were performed twice, and each trial lasted 70 seconds with 60-second rest intervals between trials. Five-minute rest intervals were also implemented between the experimental conditions to minimize fatigue. The order of the conditions was determined randomly.

During the baseline condition, the participants were asked to stand quietly in the posture mentioned above. In the external focus condition, rectangular papers (30.5 × 17 cm), one under each foot, were placed on the force plate or foam, and the participants were instructed to concentrate mentally on these papers without looking at them. In the internal focus condition, the participants were asked to mentally focus on their feet without looking. To ensure that attention was properly allocated during both the external focus and internal focus conditions, we required the participants to score the percentage of attention they had allocated to the requested task. If the reported value was 50% or lower, the trial was immediately repeated.

A backward digit span test was performed as a cognitive task, as described in more detail elsewhere [19]. The participants were instructed to listen carefully to a string of random digits before we started recording center of pressure data, mentally rehearse the string in reverse order while the center of pressure data were recorded, and verbally report the string at the end of the trial. Two levels (easy and difficult) were considered for the cognitive task, based on the maximum backward digit span of each participant, as determined by the Wechsler Intelligence Scale. The difficult cognitive task consisted of recalling maximum backward digits plus one, while an easy cognitive task consisted of recalling half the digits of the difficult cognitive task. Three types of error (omission, wrong number, and order error) were recorded. If the error was greater than 1 during the easy cognitive task and greater than 2 during the difficult cognitive task, the trial was repeated.

EMG

During each experimental condition, the muscle activity of the tibialis anterior and medial gastrocnemius muscles was recorded using surface EMG (Myon EMG, Switzerland) at 1200 Hz. EMG electrodes were placed on the surface of the skin directly over the tibialis anterior and medial gastrocnemius muscles of both legs. These muscle groups were selected according to the results of previous studies [29, 15, 34].

Variables, Outcome Measures, Data Sources and Bias

To address our primary research purpose, we compared postural control of individuals with a history of distal radius fracture with control group while quietly standing on different support surfaces (rigid and foam surfaces) using a force plate. To evaluate postural control in different conditions, we calculated different postural sway measures including the mean velocity (cm/s), SD of velocity in both the AP and mediolateral directions (cm/s), as well as the path length (cm) based on the center of pressure sway data obtained from the force plate. These parameters have recently demonstrated high reliability and, thus, provide convenient quantitative measures of postural stability, where an increase in their values indicates postural instability [23, 11]. In addition, we combined neurophysiological measures (muscle activity of the tibialis anterior and medial gastrocnemius muscles [ìV] using EMG) with postural sway measures to comprehensively investigate postural control of these patients compared with age- and sex-matched control counterparts. Raw EMG data recorded from the tibialis anterior and medial gastrocnemius muscles of both legs during each trial were band pass filtered (10-499), and the root mean square value was calculated for the filtered EMG data. The mean root mean square of the EMG activity in both legs was consequently calculated for each muscle.

To address our secondary research purpose, we compared the above-mentioned postural sway measures and EMG activity of the tibialis anterior and medial gastrocnemius muscles between different experimental conditions (baseline, internal focus, external focus, easy cognitive task, and difficult cognitive task).

Clinically Important Effect Sizes

Currently, there are no published data on the minimal clinically important difference (MCID) for the endpoints used in the current study (different postural sway measures and root mean square of the EMG activity of the tibialis anterior and medial gastrocnemius muscles). We have, therefore, estimated practical/clinical meaningfulness using the partial eta squared effect size () from ANOVA throughout the results, where values ≥ 0.01, ≥ 0.06, and ≥ 0.14 indicated small, medium, and large effect sizes, respectively [7]. The signified the magnitude of standing support surface, condition and group effect, as well as their interaction on postural sway and neurophysiological measures, based on the ratio of the variation accounted for by the effect. Values varied between 0 and 1, with higher values indicating higher proportions of variance explained by the effect. For example, of 0.89 for a particular effect on the specific measure indicated that 89% of the variance in that measure explained by that particular effect [7].

Statistical Analysis

An a priori power analysis with Type 1 error probability of 0.05 and Type II error probability of 0.10 (statistical power was 0.90) was conducted for the SD of velocity in the mediolateral direction, which was obtained in a pilot study. The analysis indicated that 33 participants in each group would be sufficient to find a difference of 0.24 between the two groups. Considering a 20% dropout rate, we oversampled at 40 participants to ensure an adequate sample size. The SD of velocity in the mediolateral direction was selected because of its high reliability in single- and dual-task conditions, according to previous studies [17].

The Kolmogorov-Smirnov test demonstrated normal data distribution. The mean value of two trials of the same condition was calculated for each postural sway measure and the root mean square of the EMG activity of each muscle. To compare postural control (different postural sway measures including the mean velocity, SD of velocity in both the AP and mediolateral directions, and path length) and the ankle muscle activity (the root mean square of the EMG activity of the tibialis anterior and gastrocnemius muscles) between groups while standing on different support surfaces (rigid and foam surfaces) under different conditions, a 2 x 2 x 5 (group x support surface x condition) three-way repeated measure ANOVA with Bonferroni post-hoc test was performed. The significance level was set at 0.05. The effect size of both main effects and interaction effects was determined by calculation of , as previously explained. All statistical analyses were performed using SPSS version 13.0 (SPSS Inc, Chicago, IL, USA).

Results

Differences in Postural Stability Between Distal Radius Fracture Patients and Controls

Patients with a history of distal radius fractures demonstrated greater postural sway (postural instability) and increased ankle muscle activity compared with their control counterparts, but only while standing on a foam surface (Fig. 1, using the mean velocity as an example). We observed no differences in the postural sway or the ankle muscle activity while standing on a rigid surface as confirmed by group × standing surface interaction effect (Table 2). The mean ± SD of postural sway and neurophysiological measures in patients with distal radius fracture compared with the control group during standing on rigid and foam surfaces were as follows: mean velocity: 4.1 ± 0.8 versus 4.1 ± 0.4 (mean difference = 0.07, 95% CI of difference, -0.07 to 0.22; p = 0.54) (Fig. 2A); SD of velocity in the AP direction: 3.7 ± 0.7 versus 3.6 ± 0.7 (mean difference = 0.15, 95% CI of difference, -0.01 to 0.31; p = 0.07) (Fig. 2B); SD of velocity in the mediolateral direction: 3 ± 0.5 versus 2.9 ± 0.5 (mean difference = 0.16, 95% CI of difference, -0.0003 to 0.31; p = 0.06) (Fig. 2C); path length: 289 ± 54 versus 284 ± 27 (mean difference = 5, 95% CI of difference, -5.2 to 15; p = 0.54) (Fig. 2D); EMG root mean square of the tibialis anterior: 35 ± 5.2 versus 33 ± 5.8 (mean difference = 1.5, 95% CI of difference, -0.001 to 3; p = 0.05) (Fig. 3A) and EMG root mean square of the medial gastrocnemius: 33 ± 4.6 versus 32 ± 5.6 (mean difference = 1.3, 95% CI of difference, -0.23 to 2.8; p = 0.11) (Fig. 3B) during standing on rigid surface and mean velocity: 5.4 ± 0.8 versus 4.80 ± 0.5 (mean difference = 0.59, 95% CI of difference, 0.44–0.73; p < 0.001) (Fig. 2A); SD of velocity in the AP direction: 4.7 ± 0.7 versus 4.2 ± 0.7 (mean difference = 0.48, 95% CI of difference, 0.33–0.64; p < 0.001) (Fig. 2B); SD of velocity in the mediolateral direction: 4 ± 0.9 versus 3.5 ± 0.7 (mean difference = 0.44, 95% CI of difference, 0.29–0.60; p < 0.001) (Fig. 2C); path length: 377 ± 59 versus 336 ± 33 (mean difference = 41, 95% CI of difference, 31–51; p < 0.001) (Fig. 2D); EMG root mean square of the tibialis anterior: 52 ± 9.4 versus 39 ± 6 (mean difference = 13, 95% CI of difference, 11–15; p < 0.001) (Fig. 3A) and EMG root mean square of the medial gastrocnemius: 50 ± 9.2 versus 39 ± 7.1 (mean difference = 11, 95% CI of difference, 9.8–13; p < 0.001) (Fig. 3B) during standing on foam surface.

Fig. 1.

This figure shows the interaction effect of group × standing surface on the mean velocity of the center of pressure (COP). The post-hoc analysis revealed that the mean velocity of the COP was greater during the standing on a foam surface task than it was during the standing on a rigid surface task in both groups (wrist fracture group: 5.4 ± 0.8 versus 4.1 ± 0.8 [mean difference = 1.2; 95% CI of difference, 1.1–1.4; p < 0.001]; control group: 4.80 ± 0.5 versus 4.1 ± 0.4 [mean difference = 0.74; 95% CI of difference, 0.6–0.89; p < 0.001]). Moreover, during standing on a foam surface, the mean velocity of the COP was greater in the group of participants with a fracture than that in the group of participants without a fracture (5.4 ± 0.8 versus 4.80 ± 0.5 [mean difference = 0.59; 95% CI of difference, 0.44–0.73; p < 0 .001]). However, there was no difference between the groups for the standing on a rigid surface task.

Table 2.

Summary of ANOVA of postural sway measures: p values and effect sizes by variable

Fig. 2 A-D.

This figure shows the mean and SD of different postural sway measures (A) mean velocity, (B) SD of velocity in the AP direction, (C) SD of velocity in the ML direction and (D) path length during standing on rigid and foam surfaces in different conditions including baseline, internal focus, external focus, easy cognitive task and difficult cognitive task. ^*Indicates difference between the group with fracture compared with the group without fracture (p < 0.01). AP = anteroposterior; ML = mediolateral; IF = internal focus; EF = external focus, EC = easy cognitive task; DC = difficult cognitive task.

Fig. 3 A-B.

This figure shows the interaction effect of group × standing surface × condition on the EMG root mean square of (A) the tibialis anterior muscle and (B) the medial gastrocnemius muscle. The post-hoc analysis indicated that the EMG root mean square of the tibialis anterior and medial gastrocnemius muscles in participants in both groups was greater while standing on a foam surface than that while standing on a rigid surface in different conditions (baseline, internal focus, external focus, easy cognitive task and difficult cognitive task) (p < 0.05). Furthermore, during standing on a foam surface in different conditions (baseline, IF, EF, EC, and DC), the EMG root mean square of the tibialis anterior and medial gastrocnemius muscles was higher in patients with a fracture than that in participants without fracture (p < 0.05). However, there was no difference between the groups for the standing on a rigid surface task. During standing on a foam surface, EF and the EC and DC tasks resulted in a decrease in the EMG root mean square of the tibialis anterior muscle and medial gastrocnemius muscles compared with the baseline and IF conditions. The greatest decrease in the EMG root mean square of the tibialis anterior and medial gastrocnemius muscles was observed in the DC tasks (p < 0.05). TA = tibialis anterior muscle; GA = medial gastrocnemius muscle; EMG RMS = EMG root mean square; IF = internal focus; EF = external focus, EC = easy cognitive task; DC = difficult cognitive task.

The Effect of Internal and External Focus Strategies and Cognitive Tasks on Postural Stability

The results indicated that internal focus strategy did not improve postural stability, while external focus strategy and the performance of both easy and difficult cognitive tasks notably improved it (that is, they reduced postural sway) compared with the baseline condition. The greatest reduction in the postural sway measures was found in the difficult cognitive tasks. The mean ± SD of postural sway measures in the baseline condition compared with external focus, easy cognitive and difficult cognitive tasks were as follows, respectively: mean velocity: 4.9 ± 1.1 versus 4.7 ± 1 (mean difference = 0.14, 95% CI of difference, 0.11–0.17; p < 0.001), 4.6 ± 1 (mean difference = 0.25, 95% CI of difference, 0.21–0.29; p < 0.001), and 4.5 ± 1 (mean difference = 0.34, 95% CI of difference, 0.29–0.40; p < 0.001); SD of velocity in the AP direction: 4.5 ± 0.8 versus 4.10 ± 0.8 (mean difference = 0.43, 95% CI of difference, 0.37–0.50; p < 0.001), 4 ± 0.8 (mean difference = 0.48, 95% CI of difference, 0.40–0.55; p < 0.001), 3.9 ± 0.8 (mean difference = 0.65, 95% CI of difference, 0.56–0.74; p < 0.001); SD of velocity in the mediolateral direction: 3.8 ± 1.1 versus 3.6 ± 0.8 (mean difference = 0.22, 95% CI of difference, 0.10–0.34; p < 0.001), 3.4 ± 0.7 (mean difference = 0.43, 95% CI of difference, 0.23–0.64; p < 0.001), 3.2 ± 0.7 (mean difference = 0.63, 95% CI of difference, 0.41–0.85; p < 0.001); path length: 343 ± 71 versus 333 ± 72 (mean difference = 9.9, 95% CI of difference, 7.9–12; p < 0.001), 326 ± 71 (mean difference = 17, 95% CI of difference, 15–20; p < 0.001), and 319 ± 69 (mean difference = 24, 95% CI of difference, 20–28; p < 0.001) in the wrist fracture group and mean velocity: 4.6 ± 0.6 versus 4.4 ± 0.5 (mean difference = 0.19, 95% CI of difference, 0.08–0.31; p < 0.01), 4.3 ± 0.5 (mean difference = 0.28, 95% CI of difference, 0.13–0.43; p < 0.001), and 4.2 ± 0.5 (mean difference = 0.42, 95% CI of difference, 0.33–0.52; p < 0.001); SD of velocity in the AP direction: 4.2 ± 0.7 versus 3.8 ± 0.8 (mean difference = 0.37, 95% CI of difference, 0.16–0.58; p < 0.001), 3.7 ± 0.7 (mean difference = 0.43, 95% CI of difference, 0.23–0.63; p < 0.001), and 3.5 ± 0.7 (mean difference = 0.67, 95% CI of difference, 0.46–0.88; p < 0.001); SD of velocity in the mediolateral direction: 3.4 ± 0.6 versus 3.3 ± 0.6 (mean difference = 0.19, 95% CI of difference, 0.03–0.36; p = 0.01), 3.1 ± 0.6 (mean difference = 0.4, 95% CI of difference, 0.22–0.57; p < 0.001), and 2.8 ± 0.6 (mean difference = 0.63, 95% CI of difference, 0.44–0.83; p < 0.001); path length: 322 ± 43 versus 308 ± 35 (mean difference = 14, 95% CI of difference, 5.3–22; p < 0.01), 302 ± 37 (mean difference = 20, 95% CI of difference, 9.3–30; p < 0.001), and 292 ± 35 (mean difference = 30, 95% CI of difference, 23–36; p < 0.001) in the control group (Fig. 4 , using the mean velocity as an example).

Fig. 4.

This figure shows the main effect of condition on the mean velocity of the center of pressure (COP) is shown. The post-hoc analysis indicated there was a decrease in the mean velocity of the COP in the external focus (EF), easy cognitive task (EC) and difficult cognitive task (DC) compared with the baseline and internal focus (IF) conditions. The greatest decrease in the mean velocity of the COP was found in the DC tasks (p < 0.01 and p < 0.001). EF = external focus; EC = easy cognitive task; DC = difficult cognitive task; IF = internal focus.

In addition, during standing on the foam surface but not the rigid one, external focus and both easy cognitive and difficult cognitive tasks led to a decrease in the EMG root mean square of the tibialis anterior and medial gastrocnemius muscles compared with the baseline condition as indicated by the interaction effect of group × standing surface × condition (Table 3). The greatest decrease in the EMG root mean square of both the tibialis anterior (Fig. 3A) and medial gastrocnemius (Fig. 3B) muscles was observed during the difficult cognitive tasks. The mean ± SD of EMG root mean square of the tibialis anterior and medial gastrocnemius in the baseline condition compared with external focus, easy cognitive and difficult cognitive tasks during standing on foam surface were as follows, respectively: EMG root mean square of gastrocnemius: 59 ± 7.2 versus 52 ± 6.6 (mean difference = 6.5, 95% CI of difference, 5.5–7.6; p < 0.001), 49 ± 7.1 (mean difference = 10, 95% CI of difference, 9-11; p < 0.001), 42 ± 5.3 (mean difference = 17, 95% CI of difference, 15–18; p < 0.001) and EMG root mean square of gastrocnemius: 56 ± 7.9 versus 50 ± 6.6 (mean difference = 6.6, 95% CI of difference, 5.2–8.1; p < 0.001), 47 ± 7 (mean difference = 9.5, 95% CI of difference, 8.6–10; p < 0.001), 41 ± 5.6 (mean difference = 15, 95% CI of difference, 14–17; p < 0.001) in the wrist fracture group and EMG root mean square of tibialis anterior: 41 ± 6.4 versus 39 ± 6 (mean difference = 2.3, 95% CI of difference, 1.3–3.4; p < 0.001), 38 ± 5.4 (mean difference = 3.4, 95% CI of difference, 2.3–4.5; p < 0.001), 36 ± 4.6 (mean difference = 5, 95% CI of difference, 3.8–6.2; p < 0.001) (Fig. 3A) and EMG root mean square of gastrocnemius: 41 ± 7.6 versus 39 ± 6.8 (mean difference = 2, 95% CI of difference, 0.91–0.3; p < 0.001), 38 ± 6.5 (mean difference = 3.3, 95% CI of difference, 2.45–4.1; p < 0.001), 36 ± 6 (mean difference = 4.8, 95% CI of difference, 3.8–5.8; p < 0.001) in the control group (Fig. 3B).

Table 3.

Summary of the ANOVA of the EMG root mean square (RMS) of the tibialis anterior (TA) and medial gastrocnemius (MGA) muscles: p values and effect sizes by variable

Discussion

Despite the potential role of impaired postural control as a risk factor for distal radius fracture, which is a highly prevalent fall-induced fracture, little attention has been paid to various performance and neurophysiologic aspects. Although a recent study has reported the impaired dynamic postural stability of individuals with distal radius fracture compared with age- and sex-matched controls [14], it did not investigate the effect of different support surfaces. In addition, to the best knowledge of the authors, no studies thus far have quantified muscle activity, an important constituent of postural control, during standing on different support surfaces in these patients. The current study revealed that patients with distal radius fractures had greater postural sway measures (that is, impaired postural control) and increased EMG activity of the tibialis anterior and medial gastrocnemius muscles compared with the control participants, but only while standing on unstable support surface (foam surface), which is important because it implies that exposure to unstable/uneven support surfaces, which is quite common in real-life situations, may increase the risk of fall and fall-related injuries in these individuals. Postural control while standing on an unstable support surface was characterized by excessive attention involvement, potentially triggering inappropriate neuromuscular control. However, similar to their control counterparts, individuals with distal radius fractures were able to shift their attention from postural control to external focus and cognitive task performance while standing on both stable and unstable support surfaces, resulting in more automatic postural control.

The most important limitation of the current study was the incomplete evaluation of the muscles involved in postural control, such as the trunk muscles, which could provide further information regarding co-contraction or stiffening of the trunk. Two discrete control strategies (the ankle strategy and the hip strategy) are typically proposed for balance and postural control. The ankle strategy is proposed as the main strategy used during quiet standing and small perturbations, where the ankle plantar flexors/dorsiflexors act alone to control the posture. In more-perturbed situations and/or by increasing the difficulty of postural task, the use of hip strategy is typically boosted [2, 3]. The evaluation of postural control in the current study was conducted in quiet standing; and therefore, similar to previous studies [29, 15, 34], we focused on assessing the muscle activity of the ankle plantar and dorsiflexors (the gastrocnemius and tibialis anterior, respectively). However, the evaluation of other muscles, such as the trunk muscles, may provide further information regarding the postural control of participants with distal radius fracture history in different conditions and should be added to future work, as should the maximum voluntary contractions for the tibialis anterior and medial gastrocnemius muscles, which have not been measured here. Another noteworthy limitation is the use of surface EMG in terms of its inability of precise recording of individual motor units. However, despite its greater accuracy/selectivity, intramuscular EMG is an invasive method and is also limited by recording the activity of a small number of active motor units whose fibers are close to the position of the detection site. In addition, the identified intramuscular action potentials are not necessarily representative of the global muscle activity. Surface EMG, on the other hand, is a noninvasive technique, which provides a more global assessment of muscle activity. This study aimed to compare the global muscle activity of the tibialis anterior and medial gastrocnemius muscles during standing on different support surfaces under different attentional focus and cognitive conditions, and hence surface EMG was deemed as the most appropriate technique. This is in alignment with many recent studies which use surface EMG to evaluate the role of activity of different muscles in postural control [25, 32, 26, 34].

Another potential limitation of the current study is the limited age range of participants (45-64 years), which is most likely due to increased risk of this type of fracture between the ages 45 and 64 years [21], restricting our ability to generalize the results to people in other age groups. One may argue that the findings of the current study are partly observational, and that it is challenging to determine whether impaired postural control causes a fracture, or whether the fracture causes fear/anxiety, which in turn lead to poor postural control, or if these issues are unrelated. We have assessed the presence of fear of falling/anxiety from falling using Fall Efficacy Scale-international in both the distal radius fracture and control groups. The impaired postural group (the group with distal radius fracture history) showed no fear/anxiety (Table 1). We therefore support the scenario that the postural control impairment observed (measured) here has most likely contributed to the fracture, which may have been accompanied by fear/anxiety versus the second scenario where the fracture caused fear/anxiety leading to impaired postural control. It is recommended for future studies to explore this and other mechanisms that may contribute to this risk. Meanwhile, people with a fear of falling may reweight their attention differently during quiet standing while concurrently performing cognitive tasks. Therefore, it would be beneficial to compare the postural control of such individuals with that of those who have a fear of falling, while assessing EMG activity of the ankle and trunk muscles under different conditions of attentional focus and cognitive tasks.

The results of the present study confirmed that there were no differences in any postural sway measures between patients with distal radius fracture and age- and sex-matched controls while standing on a stable support surface. Greater postural sway measures were detected only when patients with a distal radius fracture stood on an unstable support surface (foam). It can therefore be hypothesized that small postural sway while standing on a stable support surface does not ensure postural stability or robustness in more difficult and complex conditions. Clinically, this implies that evaluating postural control is important for individuals who have sustained a distal radius fracture. An evaluation should be conducted for multiple postural tasks and various levels of difficulty, since simple tasks (such as quietly standing on a rigid surface) may be insufficient to identify postural instability adequately. Our results also showed that, in addition to greater postural instability, patients with a distal radius fracture had greater muscle activity than controls in both the tibialis anterior and medial gastrocnemius muscles while standing on a foam surface, but not on a rigid surface. This suggests that individuals with a distal radius fracture may apply an inappropriately enhanced conscious and cautious mode of postural control while standing on unstable support surfaces. Consequential enhanced musculoskeletal stiffening may restrict the flexibility of postural adaptation, hamper responses to sudden disturbances, and impose greater electrophysiological costs for maintaining balance [16], potentially leading to faster muscle fatigue and an increased fall risk [8]. We also found increased EMG activity in both the tibialis anterior and medial gastrocnemius muscles during standing on a foam surface, as opposed to standing on a rigid surface in both groups, indicating that an ankle stiffening strategy was used. Previous studies have also demonstrated that postural control while standing on unstable support surfaces involves increased activity of the tibialis anterior and medial gastrocnemius muscles [6, 18]. It has been suggested that the energetically inefficient ankle stiffening strategy may be used to consciously and precisely control posture during a potential postural threat (such as standing on an unstable support surface or at an elevated height), or to provide a less attention-demanding contraction mode, releasing attentional resources [9].

Our results also revealed a decrease in postural sway measures (that is, enhanced postural stability and robustness) during conditions of external focus and concurrent cognitive tasks compared with the baseline and internal focus conditions. This could be explained by increased automaticity of postural control, as evidenced by the results of EMG activity of the tibialis anterior and medial gastrocnemius muscles, which are discussed further below. Several previous studies have also demonstrated that external focus and concurrent cognitive tasks result in improved postural stability, as indicated by decreased postural sway velocity and amplitude [10, 12, 28, 29]. More importantly, in line with recent studies [27, 29], we confirmed that shifting attention away from postural control to cognitive tasks does indeed have a greater advantage to postural stability compared with external cues (Fig. 4). Such advantage could be explained by two possible mechanisms: maintaining attention for a longer time and providing a focus point more distant from the body [27, 29]. The effects of external focus and concurrent cognitive tasks on postural control that were observed in this study are valuable from a clinical perspective, in terms of devising personalized patient-specific posture and balance rehabilitation strategies and protocols for individuals who have sustained a distal radius fracture or have other orthopaedic conditions. This should be investigated in future studies.

The results of the current study also indicated that while participants stood on a foam surface, external focus and cognitive tasks led to a decrease in the EMG activity of the tibialis anterior and medial gastrocnemius muscles (reduced ankle stiffening) compared with the internal focus and baseline conditions. Shifting attention away from posture by using external focus or cognitive tasks seems to allow for regulation of posture by the more efficient automatic mode of postural control rather than the cautious mode [35, 29], leading to decreased EMG activity. Furthermore, performing cognitive tasks, especially difficult ones, resulted in a greater decrease in EMG activity compared with external focus. During standing on a rigid surface, no differences were detected between the different conditions (baseline, internal focus, external focus, easy cognitive task, and difficult cognitive task), which further confirms that the decrease in postural sway measures associated with external focus and cognitive tasks can be attributed to the enhanced automatization of postural control. This agrees with the findings of previous work, which did not identify any changes in the activity of the tibialis anterior and medial gastrocnemius muscles under the external focus and cognitive task conditions while healthy young adults stood on a rigid surface [29].

In conclusion, the current study reveals that the postural instability of patients with a distal radius fracture while standing on unstable support surface is associated with increased activation of ankle muscles, possibly as a compensatory strategy to ensure balance. This confirms the need for further research to investigate the postural control of these patients while performing various static and dynamic postural tasks on different support surfaces. Future research should include randomized controlled balance exercises (using both clinical and laboratory measures), along with electrophysiological activation of ankle and trunk muscles. An integrative approach, which combines the assessment of motor control strategies with quantitative kinematic analysis, EMG muscle activity evaluation, and functional balance and mobility tests (such as the timed up and go test, the Berg Balance Scale, the Functional Reach Test, or others) is highly recommended to bridge the current knowledge gap. From a clinical perspective, this study confirms the need to evaluate balance and postural control in individuals who have sustained a distal radius fracture. Our findings may serve as basis for designing informed patient-specific balance and posture rehabilitation programs to improve stability and balance in these patients by decreasing attentional focus to postural control, by performing a concurrent attention-demanding task, or by providing feedback and/or instructions as means of external attentional focus.