Curved Walking Rehabilitation with a Rotating Treadmill in Patients with Parkinson's Disease: A Proof of Concept

Abstract

Training subjects to step-in-place eyes open on a rotating platform while maintaining a fixed body orientation in space [podokinetic stimulation (PKS)] produces a posteffect consisting in inadvertent turning around while stepping-in-place eyes closed [podokinetic after-rotation (PKAR)]. Since the rationale for rehabilitation of curved walking in Parkinson's disease is not fully known, we tested the hypothesis that repeated PKS favors the production of curved walking in these patients, who are uneasy with turning, even when straight walking is little affected. Fifteen patients participated in 10 training sessions distributed in 3 weeks. Both counterclockwise and clockwise PKS were randomly administered in each session. PKS velocity and duration were gradually increased over sessions. The velocity and duration of the following PKAR were assessed. All patients showed PKAR, which increased progressively in peak velocity and duration. In addition, before and at the end of the treatment, all patients walked overground along linear and circular trajectories. Post-training, the velocity of walking bouts increased, more so for the circular than the linear trajectory. Cadence was not affected. This study has shown that parkinsonian patients learn to produce turning while stepping when faced with appropriate training and that this capacity translates into improved overground curved walking.

Keywords: Parkinson’s disease; circular treadmill; curved walking; gait rehabilitation; podokinetic after-rotation.

Figure 1

Scheme of the protocol. It was composed of the clockwise (CW) podokinetic stimulation (PKS) [(A), PKS] and its aftereffect [(B), podokinetic after-rotation (PKAR)] (Part 1), the Stretching phase, and the CCW PKS (C) and its PKAR (D) (Part 2). (A,C) PKS. (B,D) PKAR. In panel (A), a patient steps in place on the center of a platform rotating in CW direction, with the position of the trunk frontal plane roughly perpendicular to a virtual line (red) connecting patient’s eyes to a 3 m distant target (red dot). Panel (B) represents the patient, who steps in place blindfolded (blue bar) on the motionless platform, exhibiting the PKAR. During this aftereffect, the patient inadvertently rotates while stepping-in-place, and the direction of body rotation is opposite to the rotation of the platform in (A). Panel (C) is similar to panel (A), but the platform rotates in the opposite direction. Panel (D) is similar to panel (B), but the patient turns in the opposite direction. The middle panel simply shows images of the “Stretching” phase.

Figure 2

Drawing of the experimental set showing a patient ready for the training session on the rotating platform.

Figure 3

(A) Degrees covered during the 60 s “control” stepping phase (mean values of 15 patients). (B) Cadence during the same trials (n = 14: one patient had episodes of gait hesitation and was not included in the calculation of mean cadence).

Figure 4

Top panel shows the actual velocity (ordinate) and duration (abscissa) of the platform rotation during podokinetic stimulation (PKS) at the 1st (A), 5th (B), and 10th day (C) for the counterclockwise (CCW) direction trials. Bottom panel (D–F) represents the same variables at the 1st, 5th, and 10th day for the clockwise (CW) direction trials. Each patient is identified by a different color. Note the overall progression in duration and velocity of rotation of PKS within sessions and across days.

Figure 5

(A) Mean speed of the rotating platform during the subsequent training sessions. The blue and red symbols indicate the mean velocity in counterclockwise (CCW) and clockwise (CW) direction, respectively, in each training sessions. The black dashed line shows the regression of the “training” curves (CCW and CW collapsed, since no differences were found between CW and CCW data within the same day). Panel (B) shows the mean duration of the podokinetic stimulation (PKS) of the subsequent sessions. The maximum time of training, preliminarily established, was 600 s. No differences were found between CW and CCW within the same day.

Figure 6

Mean cadence of 14 subjects in the 10 days of podokinetic stimulation (PKS) training. Blue and red columns refer to counterclockwise (CCW) and clockwise (CW) direction, respectively. There was no significant difference in cadence between CCW and CW or across days.

Figure 7

Time-course of the angular velocity of one representative patient in the podokinetic after-rotation (PKAR) phase, at the 1st, 5th, and 10th day. Blue values are the PKAR angular velocities after podokinetic stimulation (PKS) in counterclockwise (CCW) direction, red values after PKS in clockwise (CW) direction. There was an increase in both duration and angular velocity across the sessions. In this patient, PKAR was more conspicuous and more variable in CW than CCW direction.

Figure 8

(A) Duration of podokinetic after-rotation (PKAR) increased across days for both rotation directions. (B) Body rotation (cumulative degrees) covered during the whole PKAR period. The rotation extent regularly increased across days. (C) Maximum angular velocity during PKAR also increased steadily. (D) Cadence did not change across the 10 sessions of training. Blue and red columns represent PKAR following counterclockwise (CCW) and clockwise (CW) podokinetic stimulation, respectively. There were no significant differences between CCW and CW. Data are means of 15 patients for each of the 10 days of training except for (D) (n = 14). Asterisks refer to comparison of subsequent days to the first day of training; ***p < 0.0005.

Figure 9

Each subject is represented with a different color. The difference between mean maximum speed between clockwise (CW) and counterclockwise (CCW) podokinetic after-rotation (PKAR) is in ordinate. The corresponding asymmetry scores of patients’ clinical severity are reported in abscissa. There is no significant relationship between the two variables.

Figure 10

Data are means of 15 patients in the 10 days of training, mediated for the two directions (counterclockwise and clockwise). Maximum speed of podokinetic after-rotation (PKAR) (ordinate) and mean speed of podokinetic stimulation (PKS) (abscissa) are positively correlated.

Figure 11

Data are means of 15 patients in the 10 days of training (counterclockwise and clockwise collapsed). Durations of podokinetic stimulation (PKS) and of podokinetic after-rotation (PKAR) are positively correlated. Duration of PKS was higher than the duration of PKAR everyday. Toward the end of the sessions, PKAR could still relatively increase even when the preceding PKS was discontinued on reaching 600 s.

Figure 12

Mean values of gait speed (A1,A2), cadence (B), and stride length (C) during linear and curved overground walking at the beginning (T1) and at the end (T2) of the 10 training sessions. Panel (A2) represents gait speed in the single subjects (identified by a different color). The curved walking condition data are the grand average of both directions (counterclockwise and clockwise): in all patients but two, velocity increased at T2 (**p < 0.005; *p < 0.05).

Figure 13

The plot shows the correlation between the pre-training/post-training change in speed of overground curved walking (ordinate) and the change in podokinetic stimulation (PKS) angular velocity from the first to the last session (abscissa). The regression line was drawn only for the postural instability/gait difficulty (PIGD) group (n = 10), represented by yellow circles. The tremor-dominant (TD) patients represented by green triangles (n = 5) were excluded from the regression.