In this randomized controlled study we analyse and compare the acute and chronic effects of visual and acoustic cues on gait performance in Parkinson’s Disease (PD). We enrolled 46 patients with idiopathic PD who were assigned to 3 different modalities of gait training: (1) use of acoustic cues, (2) use of visual cues, or (3) overground training without cues. All patients were tested with kinematic analysis of gait at baseline (T0), at the end of the 4-week rehabilitation programme (T1), and 3 months later (T2). Regarding the acute effect, acoustic cues increased stride length and stride duration, while visual cues reduced the number of strides and normalized the stride/stance distribution but also reduced gait speed. As regards the chronic effect of cues, we recorded an improvement in some gait parameters in all 3 groups of patients: all 3 types of training improved gait speed; visual cues also normalized the stance/swing ratio, acoustic cues reduced the number of strides and increased stride length, and overground training improved stride length. The changes were not retained at T2 in any of the experimental groups. Our findings support and characterize the usefulness of cueing strategies in the rehabilitation of gait in PD.
Parkinson’s disease (PD) is a degenerative neurologic disorder characterized by motor and nonmotor symptoms. Gait disorders are a hallmark of idiopathic PD and several studies have highlighted a typical parkinsonian walking pattern characterized by reduced speed, increased duration of the stance phase, shorter stride length, and increased number of strides [
Several studies show that the use of external cues is effective in improving gait parameters [
Rehabilitation of gait is progressively becoming a mainstay in the management of advanced phases of PD. Several approaches have been proposed in recent years, including individual or group rehabilitation in the outpatient setting and home-based therapy [
The aim of the present study was the comparison and the characterization of the acute and chronic effects of visual and acoustic cues, used individually, in gait rehabilitation of PD. The study was conducted on PD patients hospitalized for neurorehabilitation at our Unit and was designed as a randomized controlled study for parallel groups, where patients were assigned randomly to one of the following groups for gait training: (1) use of acoustic cues (rhythmical sounds), (2) use of visual cues (stripes of contrasting colour), or (3) overground training without any cues. The objective of the study was to quantify the changes induced by the 3 different approaches applied for 4 weeks in an intensive rehabilitative programme on (i) gait parameters, measured by means of the kinematic analysis of gait, and (ii) the clinical picture, measured by means of the Unified Parkinson’s Disease Rating Scale (UPDRS) and the Functional Independence Measure (FIM).
The subjects were enrolled among consecutive PD patients hospitalized in the Neuro-Rehabilitation Unit of the C. Mondino National Neurological Institute of Pavia, Italy. Hospitalization for neurorehabilitation is a routine procedure at our Institute, as we know from our long-time experience and from data from the literature that inpatient-delivered rehabilitation, with strictly supervised physical therapy, is associated with a greater benefit in patients affected by PD with moderate-severe degrees of motor impairment [
Forty-six patients (24 males, 22 females; age
Inclusion criteria were Hoehn and Yahr stage between II and IV, MMSE > 23, and no changes in antiparkinsonian drug treatment in the previous 6 months. Exclusion criteria were positive history for neoplasms, cardiovascular disease, respiratory disease, clinically significant muscular-skeletal disease, other neurological conditions, uncorrected visual or auditory disturbances, or hospitalization in the previous 3 months.
Patients were divided into 3 groups who were randomly assigned to three different treatment approaches for gait training (with a 1 : 1 : 2 ratio): walking in the presence of rhythmical sounds (Acoustic Group,
Patients in all the 3 groups underwent 5 daily rehabilitation sessions per week for 4 consecutive weeks. These sessions consisted in 40 min treatment with passive muscle stretching, exercises for rigidity and joint mobility, specific motor exercise for hypokinesia, weight shifting, and balance training for posture and movement strategies to prevent falls. In addition, patients underwent 5 daily sessions per week for 4 weeks dedicated to gait training as described below. Each session lasted 20 minutes.
In the Acoustic Group, cues consisted in a rhythmical digital sound (“beep”) emitted by a digital metronome, with a frequency ranging between 60 and 120 Hz. The beep cadence was personalized and optimized for each patient during the first rehabilitative session by the physical therapist.
In the Visual Group, cues consisted in coloured stripes placed on the floor perpendicularly to the walking direction. The interstripe distance was personalized and optimized by the physical therapist during the first rehabilitative session. The physical therapist tested each subject with different distances between the stripes, starting from a minimum distance of 25 cm to a maximum of 60 cm. The therapist asked the patient to walk over the stripes trying to step over the next stripe and avoiding trampling on them.
In the Control Group, gait training was performed overground, without the use of any cue.
All patients were examined by a neurologist with expertise in Movement Disorder at the beginning of hospitalization (T0), at the end of the neurorehabilitation period (+4 weeks, T1), and 3 months after discharge from the hospital (T2). At each time point, the patients were tested with the Unified Parkinson’s Disease Rating Scale, motor part (UPDRS-III) [
For the evaluation of the
All patients enrolled in the study were tested in the morning, always in the ON condition.
Antiparkinsonian drugs schedule was kept steady for the entire study duration.
Kinematic analysis of gait was performed with a 6-camera optoelectronic system (ELITE, BTS Engineering, Milan, Italy) by an experienced laboratory technician with a sampling rate of 100 Hz. Twenty-one spherical reflective markers (15 mm in diameter) were applied along the body according to the Davis protocol [
We collected the following variables: number of strides needed to walk 7 meters, speed of gait, stride duration and stride length, percentage duration of swing and stance phases.
The local Ethics Committee approved the study protocol and all the participants gave their written informed consent before enrolment.
We considered as our primary outcome measure the chronic effect of gait rehabilitation with cues on the number of strides at the end of the 4-week rehabilitation period. We knew from our clinical experience that patients with PD employed an average of 6-7 strides to walk the 7-meter walkway of our laboratory. Based on our practice and on data from the literature we considered as clinically meaningful a difference between groups after rehabilitation greater than one stride, which corresponds to a difference of at least 20% between groups [
Therefore, we calculated the sample size with the following parameters: confidence interval (two sided) 95%; power 80%; difference between groups 20% (with a standard deviation between 20 and 25% for each group). The suggested sample size was of 42 patients. We planned to enlarge the study group of a further 10% considering possible drop-outs, so we decided to enroll 46 patients, to be distributed into the 3 different arms.
The Statistical Package for the Social Sciences (SPSS) for Windows, version 21.0, was used for the calculation.
For each variable we evaluated “skewness” and “kurtosis” to assess normality. Moreover the data were plotted using a “
Regarding the
Regarding the
Demographic and clinical characteristics of the 3 groups are shown in Table
Baseline parameters.
Acoustic cues | Visual cues | Controls | |
---|---|---|---|
Number of subjects | 11 | 11 | 24 |
Age (years, m ± sd.) | 78.1 ± 6.1 | 73.2 ± 6.9 | 72.1 ± 7.3 |
Sex (F/M) | 4/7 | 6/5 | 12/12 |
Disease duration |
10.0 ± 3.1 | 9.0 ± 2.4 | 10.5 ± 5.2 |
Patients with freezing (%) | 21.2% | 20.6% | 22.1% |
UPDRS-III | 32.1 ± 9.8 | 29.1 ± 7.9 | 32.8 ± 10.8 |
FIM score | 102.0 ± 10.2 | 105.8 ± 11.5 | 101.9 ± 19.2 |
Number of strides |
7.2 ± 3.3 | 6.8 ± 2.5 | 7.0 ± 4.1 |
Stride duration (ms) | 1250.5 ± 317.2 | 1362.9 ± 216.6 | 1336.7 ± 247.9 |
Stride length (cm) | 83.5 ± 25.7 | 84.8 ± 19.2 | 86.3 ± 20.5 |
Stance (% of stride) | 73.8 ± 7.5 | 71.3 ± 3.5 | 69.5 ± 6.0 |
Swing (% of stride) | 26.2 ± 7.5 | 28.7 ± 3.5 | 30.5 ± 6.0 |
Speed (m/s) | 0.63 ± 0.22 | 0.62 ± 0.1 | 0.64 ± 0.2 |
Use of acoustic cues induced a significant increase in stride duration and in stride length (Table
Acute effects of acoustic cueing: comparison of gait with and without cue conditioning. Data are expressed as mean ± sd. The right column reports the
Walking without |
Walking with cue |
|
|
---|---|---|---|
Number of strides | 7.2 ± 3.3 | 7.3 ± 2.5 | NS |
Stride duration (ms) | 1250.5 ± 317.2 | 1374.8 ± 381.0 | <0.05 |
Stride length (cm) | 83.5 ± 25.7 | 102.1 ± 31.6 | <0.05 |
Stance (% of stride) | 73.8 ± 7.5 | 75.5 ± 4.6 | NS |
Swing (% of stride) | 26.2 ± 7.5 | 24.5 ± 4.6 | NS |
Speed (m/s) | 0.63 ± 0.22 | 0.69 ± 0.32 | NS |
Acute effects of visual cueing: comparison of gait with and without cue conditioning. Data are expressed as mean ± sd. The right column reports the
Walking without |
Walking with cue |
|
|
---|---|---|---|
Number of strides | 6.8 ± 2.5 | 4.5 ± 1.3 | <0.05 |
Stride duration (ms) | 1362.9 ± 216.6 | 1456.7 ± 270.1 | NS |
Stride length (cm) | 84.8 ± 19.2 | 89.3 ± 12.0 | NS |
Stance (% of stride) | 71.3 ± 3.5 | 65.5 ± 2.2 | <0.05 |
Swing (% of stride) | 28.7 ± 3.5 | 34.5 ± 2.2 | <0.05 |
Speed (m/s) | 0.62 ± 0.1 | 0.55 ± 0.1 | <0.05 |
At the end of the 4-week rehabilitation programme, in the Acoustic Group we observed a significant decrease in the number of strides, an improvement in stride length, and an increase in the speed of gait (Table
Effect of acoustic cues on gait parameters: kinematic analysis of gait was performed in uncued conditions at baseline (T0), at the end of the 4-week rehabilitation period (T1), and at a 3-month follow-up (T2). Data are expressed as mean ± sd.
T0 | T1 | T2 |
|
|
|
---|---|---|---|---|---|
Number of strides | 7.2 ± 3.3 | 6.2 ± 1.7 | 7.0 ± 4.3 | <0.05 | NS |
Stride duration (ms) | 1250.5 ± 317.2 | 1246 ± 263.4 | 1292.5 ± 214.2 | NS | NS |
Stride length (cm) | 83.5 ± 25.7 | 106.7 ± 10.7 | 91.5 ± 11.7 | <0.05 | NS |
Stance (% of stride) | 73.8 ± 7.5 | 70.2 ± 3.1 | 74.5 ± 7.0 | NS | NS |
Swing (% of stride) | 25.5 ± 6.9 | 28.5 ± 4.3 | 24.9 ± 8.9 | NS | NS |
Speed (m/s) | 0.63 ± 0.22 | 0.77 ± 0.3 | 0.68 ± 0.32 | <0.05 | NS |
In the Visual Group we found a significant reduction in the number of strides, an increase in the speed of gait, and an increase in the duration of the swing phase with a corresponding reduction in the stance phase. At T1 the reduction in the number of strides was associated with an increase in stride length, which however did not reach a statistical significance (Table
Effect of visual cues on gait parameters: kinematic analysis of gait was performed in uncued conditions at baseline (T0), at the end of the 4-week rehabilitation period (T1), and at a 3-month follow-up (T2). Data are expressed as mean ± sd.
T0 | T1 | T2 |
|
|
|
---|---|---|---|---|---|
Number of strides | 6.8 ± 2.5 | 5.2 ± 1.0 | 7.1 ± 3.2 | <0.05 | NS |
Stride duration (ms) | 1362.9 ± 216.6 | 1332.9 ± 263.1 | 1384.1 ± 196.1 | NS | NS |
Stride length (cm) | 84.8 ± 19.2 | 94.0 ± 29.5 | 84.1 ± 17.0 | NS | NS |
Stance (% of stride) | 71.3 ± 3.5 | 62.6 ± 4.0 | 70.4 ± 4.5 | <0.05 | NS |
Swing (% of stride) | 27.6 ± 3.5 | 36.6 ± 3.5 | 29.1 ± 4.6 | <0.05 | NS |
Speed (m/s) | 0.62 ± 0.1 | 0.71 ± 0.2 | 0.65 ± 0.6 | <0.05 | NS |
In the Control Group we detected an increase in stride length and in gait velocity (Table
Effect of gait training without cues on gait parameters: kinematic analysis of gait was performed at baseline (T0), at the end of the 4-week rehabilitation period (T1), and at a 3-month follow-up (T2). Data are expressed as mean ± sd.
T0 | T1 | T2 |
|
|
|
---|---|---|---|---|---|
Number of strides | 7.0 ± 4.1 | 6.8 ± 3.5 | 7.4 ± 2.1 | NS | NS |
Stride duration (ms) | 1336.7 ± 247.9 | 1351.8 ± 267.7 | 1301.7 ± 254.1 | NS | NS |
Stride length (cm) | 86.3 ± 20.5 | 103.9 ± 20.7 | 93.3 ± 25.6 | <0.05 | NS |
Stance (% of stride) | 69.5 ± 6.0 | 68.8 ± 6.8 | 67.3 ± 5.1 | NS | NS |
Swing (% of stride) | 30.2 ± 6.0 | 31.1 ± 6.7 | 31.5 ± 4.4 | NS | NS |
Speed (m/s) | 0.64 ± 0.2 | 0.74 ± 0.3 | 0.66 ± 0.7 | <0.05 | NS |
When comparing the 3 groups (Figure
Effect of the different modalities of gait training on gait parameters recorded by means of the kinematic analysis. Baseline values are normalized to 100% and changes represented as % variation from baseline. ▲ Acoustic Group versus Controls
Interestingly, at T1 in the Visual Group we observed a significant increase in the time spent during the swing phase (with a corresponding decrease in the stance phase). This redistribution normalized the gait pattern of the patients, bringing the swing/stance ratio within the normal variability range in this treatment group (Figure
Distribution of stance and swing phases in the 3 treatment groups at T0 and T1. The first column on the left shows the normal percent distribution of the 2 phases of gait. The shaded horizontal bar represents the normal variability of gait pattern in healthy subjects (±4%). Note that parkinsonian gait is characterized by a reduction in the swing phase and that visual cues normalized the distribution of these 2 phases at T1. Visual Group: T0 versus T1
UPDRS-III significantly decreased at T1 in all the 3 groups under evaluation, whereas at T2 the improvement in UPDRS-III was no longer detectable.
FIM significantly improved at T1 in all groups of patients, but the gain was not preserved at T2. No statistically significant differences were found between groups at any time point in neither scale (Table
Scores at UPDRS-III and FIM at baseline and at follow-ups.
T0 | T1 | T2 |
|
|
||
---|---|---|---|---|---|---|
UPDRS-III | Acoustic cues | 32.1 ± 9.8 | 24.1 ± 9.3 | 31.6 ± 8.7 | <0.05 | NS |
Visual cues | 29.1 ± 7.9 | 22.0 ± 4.6 | 28.8 ± 8.3 | <0.05 | NS | |
Controls | 32.8 ± 10.8 | 27.8 ± 6.3 | 30.4 ± 8.5 | <0.05 | NS | |
|
||||||
FIM | Acoustic cues | 102.0 ± 10.2 | 111.7 ± 9.8 | 103.1 ± 11.3 | <0.05 | NS |
Visual cues | 105.8 ± 11.5 | 111.5 ± 11.2 | 104.3 ± 10.6 | <0.05 | NS | |
Controls | 101.9 ± 19.2 | 107.7 ± 14.7 | 102.2 ± 15.4 | <0.05 | NS |
In the last years rehabilitation has assumed a growing importance as part of a multidisciplinary approach to PD. One of the most affected motor tasks in PD is gait, due to a deficit of internal rhythmic signals, which interferes in motor performance [
Data from the literature show that external stimuli (acoustic, visual, and somatosensory) are able to modulate the motor pattern in PD, helping the patients to start and maintain a rhythmic motor task [
Most of the randomized controlled trials aimed at evaluating the effect of auditory and visual cues on gait in PD have focused on the immediate effect on gait of the cues [
To the best of our knowledge, no randomized controlled trial has analyzed comparatively the acute and chronic effect of the 2 types of cues used individually. An attempt to indirectly compare the efficacy of the two cues on gait parameters was made by Spaulding in the meta-analysis of 2002, where he concluded that auditory cues provided a more consistent and positive effect on gait parameters of PD patients when compared to visual cueing [
In the present study we investigated the effect of visual or auditory cues upon gait parameters both acutely (walking under cueing) and chronically (walking without cueing after a four-week rehabilitation program). We also investigated whether there was any retention of the effect at 3 months.
Regarding the acute effect, we found a significant increase in stride duration and in stride length when patients were exposed to acoustic cues, while a decrease in the number of strides and a reduction in gait speed were observed in patients exposed to visual cues. The worsening of some features of gait, such as the increase of stride duration with the acute acoustic cue and the reduction of speed with the acute visual cue, was not totally surprising because we realized that a proper use of cues by PD patients requires supervision by the therapist and a learning process by the patient to integrate the cue in the automaticity of gait. At the time of acute evaluation the patients had met the therapist only once and they were not familiar with cues. This observation seems relevant for the practical approach to gait training with cues, because it suggests that the adoption of the cueing strategy in the home-unassisted rehabilitation requires an adequate assisted training to avoid that the patients fail to internalize the cues aid or do so with a less functional motor pattern.
At the end of the rehabilitation program, the patients were tested under the uncued paradigm to evaluate the chronic effects of cues. The number of strides was significantly reduced only in the patients that underwent cued rehabilitation. This finding represents an important goal in the rehabilitation of PD gait, typically characterized by a tendency to an increase in the number of steps. It is known that in PD patients the activity of the internal rhythm pacemaker is dysfunctional and therefore we speculate that the observed reduction in the number of strides was promoted by the pacing effect of the external cues adopted [
It is important to underline that in all the three groups there was a significant increase in gait speed, without any statistical differences between groups at T1. Speed of gait represents one of the most comprehensive features of gait in Parkinson’s disease which may become in certain cases an independent indicator of disease severity [
Despite the chronic effect observed at T1, we could not detect any retention after 3 months in none of the groups. This feature probably results as a combination of the progression of neurodegeneration, typical of PD, with the well-known deficit of implicit learning in PD subjects [
In conclusion, our study further supports the usefulness of rehabilitation in improving gait disorders in PD. The selective impact of different kinds of cues on gait parameters suggests the usefulness of evaluating individually the gait pattern of the patients with gait analysis and testing their performance with different type of cues before the beginning of the rehabilitation programme, in order to optimize efficacy. The tendency to lose effects over months underlines once more the need for a continuative and multidisciplinary approach, characterized by serial visits and repeated rehabilitation cycles over the years.
The authors declare that there is no conflict of interests regarding the publication of this paper.
This study was supported by a grant of the Italian Ministry for Health. The authors are grateful to Mauro Fresia and to the group of physical therapists for their dedicated and skillful work.