Study in Parkinson disease of exercise (SPARX): translating high-intensity exercise from animals to humans, by CG Moore, M Schenkman, WM Kohrt, A Delitto, DA Hall, and D Corcos. Contemp Clin Trials. 2013;36:90-98.
Background and Purpose: The purpose of this exploratory trial was to examine whether people with de novo Parkinson disease (PD) are able to tolerate 30 minutes of treadmill exercise at a moderate- or high-intensity exercise; whether either treadmill training intervention has a clinically promising effect on PD motor symptoms (nonfutility); and to determine safety and attrition. Since large, rigorous trials are challenging and require extensive resources, an advantage of the exploratory, nonfutility design used in this trial is that it offers an efficient way to determine whether a moderate- or high-intensity treadmill exercise warrants large-scale studies to assess the impact of endurance exercise on improvement of PD symptoms.
Method: This phase II, multicenter, assessor-masked, randomized controlled trial enrolled 128 participants with early-stage PD aged 40 to 80 years at Northwestern University, Rush University Medical Center, the University of Colorado, and the University of Pittsburgh. Enrolled participants were not taking medications, ensuring the results of the study were related to effect of treadmill exercise and not confounded by effects of medication and dosage changes. Participants were randomized to treadmill training 4 days/week for 6 months at either (a) high-intensity (80%-85% maximum heart rate [HRmax]), (b) moderate-intensity (60%-65% HRmax), or (c) nonexercise usual care group (wait-list control). The feasibility measure for achieving intensity targets was derived from the mean HR during each treadmill session for weeks 9 to 26, expressed as a percentage of the measured maximum. Adherence was determined by calculating the exercise frequency. Adverse events (AEs) were tracked. The primary measure of futility was the change in motor score of the Unified Parkinson Disease Rating Scale (UPDRS)—a measure of disease severity—at 6 months.
Results: The moderate- and high-intensity groups exercised at 65.9% and 80.2% heart rate maximum for an average of 3.2 and 2.8 days per week, with a slight negative trend over time (α0.02 days/week, P < 0.001); attrition was 8.6%. Exercise-related AEs were more frequent during high-intensity training (high-intensity: 9 AEs; moderate-intensity: 4 AEs). Fall events were comparable between the high-, moderate-intensity, and usual care groups (n = 6, n = 5, n = 9, respectively). For the high-intensity group, the UPDRS motor score change was 0.3 ± 6.3 points after 6 months (statistically nonsignificant) compared with 2.0 ± 5.3 points for the moderate-intensity group, and 3.2 ± 5.6 points for the control group (positive change score for moderate-intensity and usual care groups indicates worsening of motor symptoms).
Discussion and Conclusions: Participation in the high-intensity but not the moderate-intensity group proved nonfutile, indicating that high-intensity treadmill endurance exercise for motor symptom relief in unmedicated persons with PD warrants further investigation. This trial demonstrates persons with de novo PD are capable of tolerating a 6-month high- or moderate-intensity treadmill exercise protocol. Additional takeaways include worsening of motor symptoms (eg, statistically significant increase in UPRDS motor score) among the wait-list control group, and the relatively low occurrence of AEs. Further research on treadmill training exercise in persons with de novo PD is warranted to determine the optimal dosing of aerobic exercise.
Mark A. Hirsch, PhD
Department of Physical Medicine and Rehabilitation, Carolinas Medical Center; Charlotte, North Carolina
Postural motor learning in Parkinson's disease: the effect of practice on continuous compensatory postural regulation, by K. Van Ooteghem, JS Frank, and FB Horak. Gait Posture. 2017;57:299-304.
Background and Purpose: Falls are common in people with Parkinson disease (PD), increase in incidence with disease progression, and are associated with increased morbidity. For these reasons, balance training is commonly used as an intervention when a goal of therapy is to reduce postural instability and the incidence of falls. Yet, despite the ubiquity of balance training programs for persons with PD, little is known about whether they can learn to improve their postural control strategies. This study addressed the question of whether participants with PD have the capacity to learn a task requiring continuous compensatory postural adjustments.
Methods: Eleven participants with PD were tested in the “ON”-medication state and compared with 21 older adults. Participants performed a balance task requiring them to stand on a servo-controlled platform that produced sinusoidal horizontal translations in the forward and backward directions at a fixed frequency of 0.5 Hz, as well as a complex sequence of variable amplitudes from ±0.5 cm to ±15.0 cm. Forty-two trials were completed during the training phase. Unbeknownst to the subject, the perturbation sequence was the same for all trials, thus retention of improvements in performance would reflect motor learning. Retention and transfer of learning to a novel sequence of perturbations were assessed by retesting 24 hours later. Differences in the peak amplitude (gain) and timing (relative phase) of the displacement of the whole-body center of mass (COM) relative to the platform were used to evaluate performance.
Results: Ten of the 11 participants with PD demonstrated practice-related improvements in the spatial (COM gain) and temporal (COM phase) control of balance over the course of training. Moreover, the rate of improvement in the PD group was comparable to the control group. Improvements in COM gain and phase lag were retained over 24 hours and could be transferred to a novel platform perturbation sequence. Importantly, many of the highest rates of practice-related improvements in phase lag were observed in participants with more advanced motor disease, and were retained after 24 hours.
Discussion and Conclusions: This study provides empirical evidence that people with PD, in the on-medication state, retain the capacity for short-term learning of postural responses to continuously imposed perturbations, and improvements can be transferred to a novel sequence. These findings complement prior work by the same group showing that people with PD can also learn protective stepping responses. Demonstration of the capacity for short-term learning of compensatory postural control and reactive protective stepping provides support for the use of balance training programs for persons with PD; however, important questions remain. For example, it is unknown whether improved postural responses following training are retained over the long term, and whether postural control learning translates to real-world scenarios in which the timing, location, direction, amplitude, and rate of application of the postural disturbance are unexpected. Future work using body-worn movement sensors in the home environment over long periods may reveal whether the incidence and severity of falls can be reduced with balance training.
Colum D. MacKinnon, PhD
Department of Neurology, University of Minnesota, Minneapolis, Minnesota
Gait initiation is influenced by emotion processing in Parkinson's disease patients with freezing, by G Lagravinese, E Pelosin, G Bonassi, F Carbone, G Abbruzzese, and L Avanzino. Mov Disord. 2018;33(4):609-617.
Background and Purpose: Freezing of gait (FOG) has a significant impact on safe mobility and quality of life for people with Parkinson disease (PD). The mechanisms for FOG are not well understood. The purpose of this study was to investigate emotional stimuli, and possible limbic system influence, on gait initiation parameters among persons with PD who freeze and who do not freeze compared with age-matched controls.
Methods: The study consisted of 3 groups: persons with PD with freezing of gait (FOG+, n = 15), persons with PD without freezing of gait (FOG−, n = 15), and age-matched controls (n = 14). Other criteria for participants with PD included Hoehn and Yahr stage 3 or less, right side more severely affected, current treatment with dopaminergic therapy, and participating in testing during the “ON”-medication state. All testing was conducted with participants standing on an instrumented mat facing a video screen. Participants were instructed to take a single step forward or backward, leading with the right foot, in response to a stimulus projected on the screen. The experiment was conducted over 2 sessions, (≥15 days in between sessions). In each session, participants performed the stepping task in a baseline condition (neutral stimulus) and a protocol condition. For the “congruent” condition (low cognitive load), participants were instructed to take a forward step in response to a pleasant image, and a backward step in response to an unpleasant image. For the “incongruent” condition (high cognitive load), instructions were given to take a forward step in response to an unpleasant image and a backward step in response to a pleasant image. Spatial and temporal measures of automatic response, step initiation, and step execution were calculated, including sway path, reaction time, anticipatory postural adjustment, and step size. The primary focus was to determine the effect of valence, the emotional value associated with a stimulus, on reaction time and step length.
Results: With forward stepping, the FOG+ group demonstrated longer reaction time and shorter step length with the incongruent task compared with the FOG− and control group. With backward stepping, the FOG+ group demonstrated increased reaction time with both the incongruent and congruent tasks compared with the control group. The FOG+ group also demonstrated increased reaction time with backward stepping in the congruent task compared with the FOG− group. In the FOG+ group, a positive correlation was found between greater frequency of freezing and greater reaction time with the incongruent forward stepping task.
Discussion and Conclusions: The authors suggest that the association between emotional load and impaired gait initiation may implicate limbic system contribution to FOG. The authors further suggest that competing motor, cognitive, and emotional stimuli may result in disordered gait response for persons with PD FOG+. The authors hypothesize that, in the presence of FOG, limbic system influence may heighten symptoms. Although further understanding of the pathways influencing freezing mechanism is needed, the current findings may have interesting implications for physical therapy treatment, possibly suggesting role of cognitive-based treatment strategies or conditioning with gait training.
Teresa C. Baker, PT, DPT, NCS
Center for Neurorehabilitation, Boston University, Boston, Massachusetts