- Email: [email protected]
Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking
Parkinsonism and Related Disorders xxx (2017) 1e8
Contents lists available at ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking Gaurav Chenji, MS a, c, 1, Melissa L. Wright, MPT a, e, 1, Kelvin L. Chou, MD a, b, c, Rachael D. Seidler, PhD d, e, Parag G. Patil, MD PhD a, b, c, * a
Surgical Therapies Improving Movement Program, University of Michigan, Ann Arbor, MI, USA Department of Neurology, University of Michigan, Ann Arbor, MI, USA Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA d Deparment of Psychology, University of Michigan, Ann Arbor, MI, USA e School of Kinesiology, University of Michigan, Ann Arbor, MI, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 November 2016 Received in revised form 12 February 2017 Accepted 21 February 2017
Background: Gait impairment in Parkinson's disease reduces mobility and increases fall risk, particularly during cognitive multi-tasking. Studies suggest that bilateral subthalamic deep brain stimulation, a common surgical therapy, degrades motor performance under cognitive dual-task conditions, compared to unilateral stimulation. Objective: To measure the impact of bilateral versus unilateral subthalamic deep brain stimulation on walking kinematics with and without cognitive dual-tasking. Methods: Gait kinematics of seventeen patients with advanced Parkinson's disease who had undergone bilateral subthalamic deep brain stimulation were examined off medication under three stimulation states (bilateral, unilateral left, unilateral right) with and without a cognitive challenge, using an instrumented walkway system. Results: Consistent with earlier studies, gait performance declined for all six measured parameters under cognitive dual-task conditions, independent of stimulation state. However, bilateral stimulation produced greater improvements in step length and double-limb support time than unilateral stimulation, and achieved similar performance for other gait parameters. Conclusions: Contrary to expectations from earlier studies of dual-task motor performance, bilateral subthalamic deep brain stimulation may assist in maintaining temporal and spatial gait performance under cognitive dual-task conditions. © 2017 Published by Elsevier Ltd.
Keywords: Parkinson's disease Deep brain stimulation Dual-task Gait Cognitive impairment
1. Introduction In Parkinson's disease, gait impairments reduce mobility and increase fall risk, leading to patient injury and loss of independence [1,2]. Kinematic alterations in parkinsonian gait include reduced velocity, stride length, increased double-limb support (DLS) time, and stride-to-stride variability [3,4]. Gait impairments worsen under real-world conditions when ambulation is paired with simultaneous cognitive challenges, such as when navigating a complex environment [2,5,6]. Under such “dual-task” conditions,
* Corresponding author. 1500 E, Medical Center Drive, SPC 5338, Ann Arbor, MI 48109-5338, USA. E-mail address: [email protected] (P.G. Patil). 1 G. Chenji and M. L. Wright contributed equally to this work.
patients experience further degradation in gait than when examined without a cognitive challenge [2]. When patients with Parkinson's disease are treated with subthalamic deep brain stimulation (STN DBS), both unilateral and bilateral DBS alleviate changes in gait velocity, stride length, and gait asymmetry under single-task conditions [3,7e10]. Compared with unilateral STN DBS, bilateral STN DBS produces additional gait improvement [10e12]. However, recent studies suggest that the superiority of bilateral DBS may not be maintained under dualtasking. A study of upper-extremity function found that bilateral STN stimulation worsens motor performance under cognitive dualtask conditions [13]. However, these effects were not observed for gait under dual-task conditions [14]. Although degraded performance may be expected, the relative impact of bilateral to unilateral STN DBS on gait performance under cognitive dual-task
http://dx.doi.org/10.1016/j.parkreldis.2017.02.028 1353-8020/© 2017 Published by Elsevier Ltd.
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
2
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
conditions remains unknown. Based on prior studies, we hypothesize that compared to unilateral stimulation, bilateral stimulation results in a decline in gait performance under dual-task conditions. We examined walking kinematics under single- and dual-task conditions in patients with Parkinson's disease treated with bilateral STN DBS. Bilateral, unilateral left-sided and right-sided stimulation states were examined off medications under single- and dual-task walking. Contrary to expectations, we find that bilateral STN DBS exceeds unilateral STN DBS gait performance, under cognitive dual-task conditions.
condition. A rest period of 2e3 min separated each walking trial. Patients first attempted to perform walking trials without an assistive device. If the patient was unable to complete the trials without an assistive device, then the least restrictive device (cane or walker) was allowed throughout the study for safety. Mechanical artifacts due to assistive devices were removed prior to analysis of footfalls. Three walking trials were included in a gait test to calculate the gait variables for each testing condition.
2. Materials and methods
Evaluated gait parameters included step length, step length variability, cadence, normalized velocity, double-limb support (DLS) time, and functional ambulation performance index. Step length (cm) is the measured distance from heel center of the foot to heel center of the opposite foot. Step length variability (cm) is defined as standard deviation of the step length. Cadence is the rate of stepping (steps/min). Velocity (cm/sec) is defined as the distance traveled divided by the time elapsed between first contact of the first and last footfalls. Normalized velocity (sec1) is obtained after dividing the velocity by the average leg length. Gait cycle time (sec) is the elapsed time between the first contacts of two consecutive footfalls of the same foot. DLS is the gait cycle time between heel contact of one footfall to the toe-off of the opposite footfall, in seconds. Hence, right DLS is the percentage of the gait cycle between right heel contact and left toe-off. The functional ambulation performance index quantifies gait performance at a self-selected pace [18,19]. Parameters utilized to calculate the index include standard velocity normalized to leg length, step to leg length ratio, step time, left and right leg asymmetry of step length, and dynamic base of support. Values are deducted from an initial score of 100 points based on deviations in gait parameters. Calculation of the index was performed by software provided with the instrumented walkway system.
2.1. Patients Seventeen patients with advanced Parkinson's disease who had undergone bilateral STN DBS surgery at the University of Michigan were enrolled in this study. Eligibility criteria for DBS at our center include an established diagnosis of Parkinson's disease, 30% improvement in MDS-UPDRS Part III motor scores with levodopa, the presence of motor fluctuations or severe rest tremor not adequately managed by medications, and an absence of significant dementia or depression on formal neuropsychological testing [15]. Inclusion criteria for the study included DBS surgery within 5 years, stable optimized stimulation settings for 3 months, ability to ambulate independently for 25 feet in the on stimulation/off medication state, and stable cognitive ability as evaluated by the Mini-Mental State Examination [16]. Non-native English speakers were excluded from the study. Written informed consent was obtained from all patients. The Medical School Institutional Review Board of the University of Michigan approved this study. 2.2. Experimental design Prior to gait testing, baseline MDS-UPDRS Part III motor scores were collected while patients were on medication with bilateral STN stimulation. Patients were then subsequently examined at least 12 h after stopping dopaminergic medication under three stimulation states (bilateral DBS, unilateral left DBS, unilateral right DBS). The order of stimulation testing conditions was determined through a computer-generated random sequence and programmed by a third-party clinician. Investigators and patients were blinded to stimulation state. Each adjustment of DBS state was followed by 90e120 min of rest to minimize carryover effects between conditions. MDS-UPDRS Part III motor scores and gait were evaluated for each stimulation state. Patients were not evaluated under offmedication/off-stimulation states due to safety concerns related to freezing and falling. For each stimulation state, gait was evaluated under single-task (walking alone) and dual-task (walking while simultaneously meeting a cognitive challenge) conditions. Cognitive tasks included forward counting in English by threes or alternate alphabet letternaming. In order to minimize learning effects, cognitive tasks were counterbalanced for each patient and were different for each walking trial. Starting numbers ranged from zero to ten and letters from A to G; these were provided verbally to the patient just prior to each walking trial. Quantitative gait analysis was performed using a validated, commercially available instrumented walkway system (GAITRite; CIR Systems, Inc., Franklin, NJ) [17]. It measures 16 feet (487 cm) in length and 2 feet (61 cm) in width, and records timing and location of foot strikes through implanted mechanical sensors at a sampling rate of 80 Hz. Patients performed walking trials wearing comfortable flat-soled shoes at a self-selected pace. Three walking trials were performed for each stimulation state and cognitive task
2.3. Data analysis
2.4. Statistical analysis Statistical analysis was performed using SPSS Statistics for Macintosh, Version 21.0 (IBM Corp., Armonk, NY). Spatiotemporal parameters of gait and MDS-UPDRS motor scores were analyzed with repeated measures of ANOVA, with the three stimulation states and the three gait tasks as repeated measures. The Greenhouse-Geisser correction was used when the assumption of sphericity was violated. Post hoc contrasts with Bonferroni adjustment for multiple comparisons were used to evaluate differences between stimulation states, task complexity, and interactions between the two, if significance was achieved. A P-value less than 0.05 is considered significant. 3. Results 3.1. MDS-UPDRS motor score for unilateral and bilateral DBS Seventeen DBS patients met inclusion/exclusion criteria and were enrolled in the study. Table 1 reports patient demographics, Mini-Mental State Examination scores, and MDS-UPDRS motor scores for patients under each stimulation state. Patient MDSUPDRS motor scores with bilateral DBS stimulation, both on and off medication, were improved compared to unilateral stimulation to the right or left side (P ¼ 0.0005 and P < 0.0001, respectively). 3.2. Functional ambulation performance Fig. 1A shows the effects of stimulation state and task condition on functional ambulation performance index. There was a main
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
3
Table 1 Patient demographics and MDS-UPDRS motor scores under various stimulation parameters (N ¼ 17). Patient Characteristics
Mean ± Standard Deviation
Age, years Gender Handedness (Edinburgh Handedness Inventory)
61.7 ± 6.46 14 males; 3 females 15 right-handed 1 left-handed 1 ambidextrous 28.8 ± 0.95 23.9 ± 14.4 28.4 ± 10.3 31.6 ± 7.66 40.5 ± 9.77
Mini Mental State Examination (MMSE) Time after surgery (months) MDS-UPDRS motor scores (ON meds/bilateral DBS) MDS-UPDRS motor scores (OFF meds/bilateral DBS) MDS-UPDRS motor scores (OFF meds/unilateral DBS)
effect of task complexity on functional ambulation performance. Ambulation performance decreased significantly under dual-task conditions [F(2,32) ¼ 13.68; P < 0.0001; h2 ¼ 0.46]. Gait under both cognitive dual-task conditions (numbers and letters) resulted in significantly lowered functional ambulation performance scores compared to the single-task condition of gait alone (P ¼ 0.016 and P ¼ 0.001, respectively). There was no main effect of stimulation state [F(2,32) ¼ 1.72; P ¼ 0.195; h2 ¼ 0.01], indicating that bilateral stimulation was not significantly different than unilateral stimulation, irrespective of task condition. Lastly, interactions between the main effects of stimulation state and task condition were not significant [F(2.-57,41.1) ¼ 1.86; P ¼ 0.158; h2 ¼ 0.10; GreenhouseGeisser]. Overall, we found that as cognitive load increased, functional gait performance decreased in both unilateral and bilateral DBS. 3.3. Normalized velocity Fig. 1B illustrates effects of stimulation state and task condition on normalized velocity. There was a significant main effect with task condition [F(1.49,23.80) ¼ 28.74; P < 0.0001; h2 ¼ 0.46; Greenhouse-Geisser]. Normalized velocity decreased significantly under dual-task conditionsdwalking while counting numbers (P ¼ 0.0004), and walking while naming letters (P < 0.0001)d compared to single-task walking. There was no main effect with stimulation state [F(2,32) ¼ 2.52; P ¼ 0.096; h2 ¼ 0.14] or interaction between stimulation state and task condition [F(4,64) ¼ 1.28; P ¼ 0.287; h2 ¼ 0.07]. Hence, cognitive dual-tasking decreased gait velocity equally in all three stimulation states, with no difference between bilateral vs. unilateral stimulation states. 3.4. Cadence Fig. 2A shows effects of stimulation state and task condition on cadence. The main effect of task condition was significant, with a substantial reduction during cognitive dual-task conditions [F(1.40,22.45) ¼ 11.952; P < 0.0001; h2 ¼ 0.428; GreenhouseGeisser]. The main effect of stimulation state was not statistically significant [F(2,32) ¼ 11.29; P ¼ 0.505; h2 ¼ 0.04; GreenhouseGeisser]. Pairwise comparisons revealed that cadence during cognitive dual-tasking was significantly lowered compared to single-task walking (P ¼ 0.006 for both dual-task conditions). 3.5. Double-limb support time Fig. 2B presents the effects of stimulation state and task condition on DLS time. Stimulation state is indicated as ipsilateral, contralateral, or bilateral. Ipsilateral refers to stimulation of the brain hemisphere on the same side as heel contact initiating the period of DLS. Contralateral refers to stimulation of the brain hemisphere opposite the heel contact initiating the period of DLS.
For each stimulation state, lateralized DLS times are averaged. In contrast to results reported thus far, patients experienced statistically significant main effects for both task condition [F(1.344,44.34) ¼ 15.18; P < 0.0001; h2 ¼ 0.32; GreenhouseGeisser] and stimulation state [F(2,66) ¼ 3.69; P ¼ 0.03; h2 ¼ 0.10]. Pairwise comparison revealed that each cognitive dualtask condition (numbers and letters) resulted in significantly longer DLS times compared to single-task gait (P < 0.0001 and P ¼ 0.0002, respectively). Evaluation of within-patient contrasts for stimulation state revealed that bilateral stimulation resulted in lowered DLS time than either-sided unilateral stimulation [F(1,33) ¼ 7.7; P ¼ 0.009; h2 ¼ 0.19]. 3.6. Step length Fig. 3A depicts the effects of stimulation state and task condition on step length. Stimulation states are indicated either as bilateral, ipsilateral or contralateral to the measured step. As with DLS time, both task condition [F(1.445,47.67) ¼ 18.47; P < 0.0001; h2 ¼ 0.36; Greenhouse-Geisser] and stimulation state [F(1.753,57.88) ¼ 6.75; P ¼ 0.003; h2 ¼ 0.17; Huynh-Feldt] significantly affected step length. Post hoc contrasts for task condition indicate that cognitive dual-task conditions resulted in shorter step length for both number and letter dual-tasks (P ¼ 0.022 and P < 0.0001, respectively. Post hoc contrasts for stimulation state revealed that bilateral stimulation enabled the longest step length, irrespective of task condition, compared to ipsilateral unilateral stimulation (P ¼ 0.0016) or contralateral unilateral stimulation (P ¼ 0.0058). No significant interaction between stimulation state and task condition was apparent. To summarize, step length decreased significantly during both cognitive dual-task conditions and during unilateral stimulation, compared to single-task conditions and bilateral stimulation state. 3.7. Step length variability Fig. 3B reflects the effects of stimulation state and task condition on step length variability. Ipsilateral refers to variability of step length measured during same-side unilateral DBS stimulation. Contralateral refers to variability of step length measured during opposite-side unilateral DBS. For each unilateral stimulation state, results for right and left sides were averaged. There was a statistically significant main effect with task condition on step length variability [F(1.743,57.51) ¼ 11.29; P ¼ 0.0001; h2 ¼ 0.26; Huynh-Feldt]. Post hoc contrasts revealed that dual-task gait during alternate letter-naming resulted in significantly greater step length variability than either single-task gait (P ¼ 0.0014) or dual-task gait during counting by threes (P ¼ 0.0036). There was no significant main effect with stimulation state [F(1.366,45.07) ¼ 1.48; P ¼ 0.237; h2 ¼ 0.04]. Hence, dual-task conditions had a significant effect on step length variability, with
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
4 G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 1. A. Effects of task complexity on functional ambulation under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load (numbers and letters), compared to gait alone. B. Effects of task complexity on normalized velocity under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect with cognitive dual-tasking compared to gait alone.
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 2. A. Changes in cadence on task complexity under three stimulation parameters (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under both cognitive load tasks when compared to gait alone. B. Effects of task complexity on DLS time under varying stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load when compared to gait alone. Bars spanning across a single task indicate a significant main effect for stimulation (specifically, bilateral DBS resulted in a shorter DLS time than ipsilateral DBS).
5
6 G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 3. A. Changes in step length with increasing task complexity across three stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load tasks (numbers and letters) compared to gait alone. Bars spanning across a single task indicate a significant main effect for stimulation (specifically, bilateral DBS resulted in longest overall step length compared to unilateral DBS to either side in all three tasks). B. Effects of task complexity on step length variability under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for both cognitive load tasks when compared to walking only.
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
greater variability during alternate letter-naming than for counting by threes. 4. Discussion We tested the hypothesis, demonstrated for upper extremity function, that bilateral STN DBS degrades gait performance under cognitive dual-task conditions. Consistent with earlier studies, gait performance declined for all parameters under cognitive dual-task conditions (counting or letters), independent of stimulation state. However, bilateral DBS produced greater improvements in step length and DLS time than unilateral DBS. Together, these data suggest that bilateral stimulation may assist in maintaining temporal and spatial gait performance under cognitive dual-task conditions. 4.1. MDS-UPDRS motor scores and DBS Patients experienced dramatic improvements in MDS-UPDRS motor scores with DBS. Bilateral STN DBS, both on and off medication, significantly improved total MDS-UPDRS motor scores compared to unilateral left or right stimulation conditions. There was no difference in total motor scores between bilateral DBS while on or off medication. These improvements in MDS-UPDRS motor scores are consistent with earlier findings of clinical improvement with STN DBS [20]. 4.2. DBS and gait under cognitive dual-task conditions Increased cognitive load during gait led to significant decreases in performance, independent of stimulation state. Our findings demonstrate that all measured temporal and spatial parameters of gait decline with increased task complexity. These results correspond to previous studies that investigated declines in motor and cognitive performances in patients with Parkinson's disease from single-to dual-task conditions [21e23] and are also consistent with the findings by Seri-Fainshtat et al. [14], demonstrating that bilateral DBS improves gait performance independent of task complexity. Bilateral DBS resulted in the longest step length for all tasks, compared to unilateral stimulation. Additionally, during gait while counting alternate numbers, bilateral DBS led to the shortest DLS time, compared to unilateral DBS. These results extend previous studies demonstrating that bilateral DBS improves gait more than either-sided unilateral stimulation, under single-task conditions [3,12]. Notably, though bilateral stimulation significantly improves step length and DLS time compared to either-sided unilateral stimulation, significant differences were not found among the other gait parameters across the three stimulation states. Increased task complexity resulted in decreased gait performance. Given the results of previous studies, we expected bilateral DBS to improve the spatiotemporal parameters of gait more than unilateral DBS during single-task gait. Additionally, based on the findings of Alberts et al. [13], we expected that bilateral stimulation would result in significantly worsened gait performance as task complexity increased, compared to unilateral stimulation. We found no differences in performance between stimulation states for most gait parameters, under single-task and cognitive dual-task conditions. Unexpectedly, step length and DLS time improved with bilateral stimulation. Specifically, bilateral stimulation improved step length and DLS time more than unilateral stimulation to either side under dual-task conditions. These results are contrary to the expectations from Alberts et al. [13], which found that bilateral DBS under dual-task conditions produced significantly degraded cognitive and motor performance compared with
7
unilateral stimulation. However, the specific cognitive tasks in our two studies differ. Alberts et al. investigated performance during a working memory (n-back) and upper limb-grip force task, whereas our study utilizes a working memory and gait task requiring truncal muscles. While specific mechanisms underlying motor performance degradation under cognitive dual-task conditions remain unknown, models of cortical-subcortical activity during DBS provide some insights. Studies of globus pallidus internus (GPi) neuronal activity in parkinsonian primates indicate that DBS replaces pathological signaling with a steadier pattern, thereby producing more normal movements [24,25]. In the STN, DBS stimulation may spread from motor to adjacent associative and affective STN regions, resulting in an “informational lesion.” Under conditions of increased cognitive demand, but not under less taxing conditions, the informational lesion may lead to subtle cognitive deficits and a limited ability to accommodate increased cognitive loads, reducing dual-task performance. Alberts et al. used the STN informational lesion hypothesis to explain why bilateral DBS was worse than unilateral DBS under cognitive load for upper extremities, suggesting that bilateral stimulation could produce a larger informational lesion than unilateral DBS. Even if true, this model may not be as relevant for gait and postural abnormalities, where other deep nuclei, such as the pedunculopontine nucleus and associated mesencephalic locomotor circuits, play a more prominent role in motor performance [26]. Our study has several limitations. First, patients were not tested both off-medication and off-stimulation. Given the advanced symptoms of many patients and the long duration of testing off medications, our patients could not tolerate the off-stimulation state in this study. Second, the testing protocol was quite lengthy (approximately 6 h). As a result, gait performance may have decreased significantly by the end of the session. To counter this potential limitation, stimulation test conditions were counterbalanced, alleviating the impact that fatigue would be expected to have on our results. Another limitation may be that only 90e120 min elapsed between stimulation adjustment and gait testing. Considerably longer durations may be required for the full effects of stimulation to wear off [27]. Finally, our study population was limited to patients treated with STN DBS. While STN is thought to affect gait more than GPi stimulation, we cannot say whether these results extend to patients with GPi DBS. To our knowledge, this study is the first to examine the effects of unilateral and bilateral STN DBS on gait under single and cognitive dual-task conditions in Parkinson's disease. Our findings highlight the importance of assessing performance under cognitive loading, as more representative of everyday life. The inability of STN DBS to elicit changes in the circuitry involved in executive function has led researchers to investigate interventional strategies to cope with gait under cognitive load. Several researchers have started using virtual reality, cognitive therapy to improve gait while multitasking [28,29]. Overall, we find that bilateral STN DBS is superior to unilateral STN DBS for some gait parameters (step length and DLS time), and MDS-UPDRS motor scores. For other gait parameters, there is no significant difference between unilateral and bilateral stimulation.
Author roles All aspects of research project, statistical analysis design, review, and critique (M.L.W., K.L.C., P.G.P.). Research project conception and organization (R.D.S.). All aspects of statistical analysis and manuscript preparation (G.C., M.L.W., R.D.S., P.G.P.).
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
8
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Financial disclosure/conflict of interest None of the authors have any personal financial or institutional interest in any of the drugs, materials, or devices described in this article. Funding sources This work was supported by the A. Alfred Taubman Medical Research Institute Emerging Scholars Program, Blue Cross Blue Shield of Michigan Foundation, Foundation for Physical Therapy, Michigan Parkinson Foundation, and Parkinson's Disease Foundation. Acknowledgements We would like to thank Karen Cummings and Neal Rakesh for technical assistance with recruitment, DBS programming, data collection and/or data analysis for this study. References [1] D. Muslimovic, B. Post, J.D. Speelman, B. Schmand, R.J. de Haan, Determinants of disability and quality of life in mild to moderate Parkinson disease, Neurology 70 (2008) 2241e2247. [2] V.E. Kelly, A.J. Eusterbrock, A. Shumway-Cook, The effects of instructions on dual-task walking and cognitive task performance in people with Parkinson's disease, Park. Dis. (2012) 67e1261. [3] M. Faist, J. Xie, D. Kurz, W. Berger, C. Maurer, P. Pollak, C.H. Lucking, Effect of bilateral subthalamic nucleus stimulation on gait in Parkinson's disease, Brain 124 (2001) 1590e1600. [4] N. Giladi, Gait disturbances in advanced stages of Parkinson's disease, Adv. Neurol. 86 (2001) 273e278. [5] G. Yogev, N. Giladi, C. Peretz, S. Springer, E.S. Simon, J.M. Hausdorff, Dual tasking, gait rhythmicity, and Parkinson's disease: which aspects of gait are attention demanding? Eur. J. Neurosci. 22 (2005) 1248e1256. [6] S. O'Shea, M.E. Morris, R. Iansek, Dual task interference during gait in people with Parkinson disease: effects of motor versus cognitive secondary tasks, Phys. Ther. 82 (2002) 888e897. [7] P. Crenna, I. Carpinella, M. Rabuffetti, M. Rizzone, L. Lopiano, M. Lanotte, M. Ferrarin, Impact of subthalamic nucleus stimulation on the initiation of gait in Parkinson's disease, Exp. Brain Res. 172 (2006) 519e532. [8] E.L. Johnsen, P.H. Mogensen, N.A. Sunde, K. Ostergaard, Improved asymmetry of gait in Parkinson's disease with DBS: gait and postural instability in Parkinson's disease treated with bilateral deep brain stimulation in the subthalamic nucleus, Mov. Disord. 24 (2009) 590e597. [9] H. Stolze, S. Klebe, M. Poepping, D. Lorenz, J. Herzog, W. Hamel, B. Schrader, J. Raethjen, R. Wenzelburger, H.M. Mehdorn, G. Deuschl, P. Krack, Effects of bilateral subthalamic nucleus stimulation on parkinsonian gait, Neurology 57 (2001) 144e146. [10] A.J. Bastian, V.E. Kelly, F.J. Revilla, J.S. Perlmutter, J.W. Mink, Different effects of unilateral versus bilateral subthalamic nucleus stimulation on walking and reaching in Parkinson's disease, Mov. Disord. 18 (2003) 1000e1007.
[11] R. Kumar, A.M. Lozano, E. Sime, E. Halket, A.E. Lang, Comparative effects of unilateral and bilateral subthalamic nucleus deep brain stimulation, Neurology 53 (1999) 561e566. [12] S.D. Tabbal, M. Ushe, J.W. Mink, F.J. Revilla, A.R. Wernle, M. Hong, M. Karimi, J.S. Perlmutter, Unilateral subthalamic nucleus stimulation has a measurable ipsilateral effect on rigidity and bradykinesia in Parkinson disease, Exp. Neurol. 211 (2008) 234e242. [13] J.L. Alberts, C. Voelcker-Rehage, K. Hallahan, M. Vitek, R. Bamzai, J.L. Vitek, Bilateral subthalamic stimulation impairs cognitive-motor performance in Parkinson's disease patients, Brain 131 (2008) 3348e3360. [14] E. Seri-Fainshtat, Z. Israel, A. Weiss, J.M. Hausdorff, Impact of sub-thalamic nucleus deep brain stimulation on dual tasking gait in Parkinson's disease, J. Neuroeng Rehabil. 10 (2013) 38. [15] R.R. Coleman, V. Kotagal, P.G. Patil, K.L. Chou, Validity and efficacy of screening algorithms for assessing deep brain stimulation candidacy in Parkinson disease, Mov. Disord. Clin. Pract. 1 (2014) 342e347. [16] A.J. Mitchell, A meta-analysis of the accuracy of the mini-mental state examination in the detection of dementia and mild cognitive impairment, J. Psychiatr. Res. 43 (2009) 411e431. [17] K.E. Webster, J.E. Wittwer, J.A. Feller, Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait, Gait Posture 22 (2005) 317e321. [18] A.J. Nelson, Functional ambulation profile, Phys. Ther. 54 (1974) 1059e1065. [19] A. Gouelle, Use of functional ambulation performance score as measurement of gait ability: review, J. Rehabil. Res. Dev. 51 (2014) 665e674. [20] K.L. Chou, J.L. Taylor, P.G. Patil, The MDS-UPDRS tracks motor and non-motor improvement due to subthalamic nucleus deep brain stimulation in Parkinson disease, Park. Relat. Disord. 19 (2013) 966e969. [21] R. Marchese, M. Bove, G. Abbruzzese, Effect of cognitive and motor tasks on postural stability in Parkinson's disease: a posturographic study, Mov. Disord. 18 (2003) 652e658. [22] M. Morris, R. Iansek, F. Smithson, F. Huxham, Postural instability in Parkinson's disease: a comparison with and without a concurrent task, Gait Posture 12 (2000) 205e216. [23] L. Rochester, V. Hetherington, D. Jones, A. Nieuwboer, A.M. Willems, G. Kwakkel, E. Van Wegen, Attending to the task: interference effects of functional tasks on walking in Parkinson's disease and the roles of cognition, depression, fatigue, and balance, Arch. Phys. Med. Rehabil. 85 (2004) 1578e1585. [24] W.M. Grill, A.N. Snyder, S. Miocinovic, Deep brain stimulation creates an informational lesion of the stimulated nucleus, Neuroreport 15 (2004) 1137e1140. [25] T. Hashimoto, C.M. Elder, M.S. Okun, S.K. Patrick, J.L. Vitek, Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons, J. Neurosi 23 (2003) 1916e1923. [26] K.A. Follett, D. Torres-Russotto, Deep brain stimulation of globus pallidus interna, subthalamic nucleus, and pedunculopontine nucleus for Parkinson's disease: which target? Park. Relat. Disord. 18 (Suppl 1) (2012) S165eS167. [27] P. Temperli, J. Ghika, J.G. Villemure, P.R. Burkhard, J. Bogousslavsky, F.J. Vingerhoets, How do parkinsonian signs return after discontinuation of subthalamic DBS? Neurology 60 (2003) 78e81. [28] A. Mirelman, I. Maidan, T. Herman, J.E. Deutsch, N. Giladi, J.M. Hausdorff, Virtual reality for gait training: can it induce motor learning to enhance complex walking and reduce fall risk in patients with Parkinson's disease? J. Gerontol. A Biol. Sci. Med. Sci. 66 (2011) 234e240. [29] O. Segev-Jacubovski, T. Herman, G. Yogev-Seligmann, A. Mirelman, N. Giladi, J.M. Hausdorff, The interplay between gait, falls and cognition: can cognitive therapy reduce fall risk? Expert Rev. Neurother. 11 (2011) 1057e1075.
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Contents lists available at ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking Gaurav Chenji, MS a, c, 1, Melissa L. Wright, MPT a, e, 1, Kelvin L. Chou, MD a, b, c, Rachael D. Seidler, PhD d, e, Parag G. Patil, MD PhD a, b, c, * a
Surgical Therapies Improving Movement Program, University of Michigan, Ann Arbor, MI, USA Department of Neurology, University of Michigan, Ann Arbor, MI, USA Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA d Deparment of Psychology, University of Michigan, Ann Arbor, MI, USA e School of Kinesiology, University of Michigan, Ann Arbor, MI, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 November 2016 Received in revised form 12 February 2017 Accepted 21 February 2017
Background: Gait impairment in Parkinson's disease reduces mobility and increases fall risk, particularly during cognitive multi-tasking. Studies suggest that bilateral subthalamic deep brain stimulation, a common surgical therapy, degrades motor performance under cognitive dual-task conditions, compared to unilateral stimulation. Objective: To measure the impact of bilateral versus unilateral subthalamic deep brain stimulation on walking kinematics with and without cognitive dual-tasking. Methods: Gait kinematics of seventeen patients with advanced Parkinson's disease who had undergone bilateral subthalamic deep brain stimulation were examined off medication under three stimulation states (bilateral, unilateral left, unilateral right) with and without a cognitive challenge, using an instrumented walkway system. Results: Consistent with earlier studies, gait performance declined for all six measured parameters under cognitive dual-task conditions, independent of stimulation state. However, bilateral stimulation produced greater improvements in step length and double-limb support time than unilateral stimulation, and achieved similar performance for other gait parameters. Conclusions: Contrary to expectations from earlier studies of dual-task motor performance, bilateral subthalamic deep brain stimulation may assist in maintaining temporal and spatial gait performance under cognitive dual-task conditions. © 2017 Published by Elsevier Ltd.
Keywords: Parkinson's disease Deep brain stimulation Dual-task Gait Cognitive impairment
1. Introduction In Parkinson's disease, gait impairments reduce mobility and increase fall risk, leading to patient injury and loss of independence [1,2]. Kinematic alterations in parkinsonian gait include reduced velocity, stride length, increased double-limb support (DLS) time, and stride-to-stride variability [3,4]. Gait impairments worsen under real-world conditions when ambulation is paired with simultaneous cognitive challenges, such as when navigating a complex environment [2,5,6]. Under such “dual-task” conditions,
* Corresponding author. 1500 E, Medical Center Drive, SPC 5338, Ann Arbor, MI 48109-5338, USA. E-mail address: [email protected] (P.G. Patil). 1 G. Chenji and M. L. Wright contributed equally to this work.
patients experience further degradation in gait than when examined without a cognitive challenge [2]. When patients with Parkinson's disease are treated with subthalamic deep brain stimulation (STN DBS), both unilateral and bilateral DBS alleviate changes in gait velocity, stride length, and gait asymmetry under single-task conditions [3,7e10]. Compared with unilateral STN DBS, bilateral STN DBS produces additional gait improvement [10e12]. However, recent studies suggest that the superiority of bilateral DBS may not be maintained under dualtasking. A study of upper-extremity function found that bilateral STN stimulation worsens motor performance under cognitive dualtask conditions [13]. However, these effects were not observed for gait under dual-task conditions [14]. Although degraded performance may be expected, the relative impact of bilateral to unilateral STN DBS on gait performance under cognitive dual-task
http://dx.doi.org/10.1016/j.parkreldis.2017.02.028 1353-8020/© 2017 Published by Elsevier Ltd.
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
2
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
conditions remains unknown. Based on prior studies, we hypothesize that compared to unilateral stimulation, bilateral stimulation results in a decline in gait performance under dual-task conditions. We examined walking kinematics under single- and dual-task conditions in patients with Parkinson's disease treated with bilateral STN DBS. Bilateral, unilateral left-sided and right-sided stimulation states were examined off medications under single- and dual-task walking. Contrary to expectations, we find that bilateral STN DBS exceeds unilateral STN DBS gait performance, under cognitive dual-task conditions.
condition. A rest period of 2e3 min separated each walking trial. Patients first attempted to perform walking trials without an assistive device. If the patient was unable to complete the trials without an assistive device, then the least restrictive device (cane or walker) was allowed throughout the study for safety. Mechanical artifacts due to assistive devices were removed prior to analysis of footfalls. Three walking trials were included in a gait test to calculate the gait variables for each testing condition.
2. Materials and methods
Evaluated gait parameters included step length, step length variability, cadence, normalized velocity, double-limb support (DLS) time, and functional ambulation performance index. Step length (cm) is the measured distance from heel center of the foot to heel center of the opposite foot. Step length variability (cm) is defined as standard deviation of the step length. Cadence is the rate of stepping (steps/min). Velocity (cm/sec) is defined as the distance traveled divided by the time elapsed between first contact of the first and last footfalls. Normalized velocity (sec1) is obtained after dividing the velocity by the average leg length. Gait cycle time (sec) is the elapsed time between the first contacts of two consecutive footfalls of the same foot. DLS is the gait cycle time between heel contact of one footfall to the toe-off of the opposite footfall, in seconds. Hence, right DLS is the percentage of the gait cycle between right heel contact and left toe-off. The functional ambulation performance index quantifies gait performance at a self-selected pace [18,19]. Parameters utilized to calculate the index include standard velocity normalized to leg length, step to leg length ratio, step time, left and right leg asymmetry of step length, and dynamic base of support. Values are deducted from an initial score of 100 points based on deviations in gait parameters. Calculation of the index was performed by software provided with the instrumented walkway system.
2.1. Patients Seventeen patients with advanced Parkinson's disease who had undergone bilateral STN DBS surgery at the University of Michigan were enrolled in this study. Eligibility criteria for DBS at our center include an established diagnosis of Parkinson's disease, 30% improvement in MDS-UPDRS Part III motor scores with levodopa, the presence of motor fluctuations or severe rest tremor not adequately managed by medications, and an absence of significant dementia or depression on formal neuropsychological testing [15]. Inclusion criteria for the study included DBS surgery within 5 years, stable optimized stimulation settings for 3 months, ability to ambulate independently for 25 feet in the on stimulation/off medication state, and stable cognitive ability as evaluated by the Mini-Mental State Examination [16]. Non-native English speakers were excluded from the study. Written informed consent was obtained from all patients. The Medical School Institutional Review Board of the University of Michigan approved this study. 2.2. Experimental design Prior to gait testing, baseline MDS-UPDRS Part III motor scores were collected while patients were on medication with bilateral STN stimulation. Patients were then subsequently examined at least 12 h after stopping dopaminergic medication under three stimulation states (bilateral DBS, unilateral left DBS, unilateral right DBS). The order of stimulation testing conditions was determined through a computer-generated random sequence and programmed by a third-party clinician. Investigators and patients were blinded to stimulation state. Each adjustment of DBS state was followed by 90e120 min of rest to minimize carryover effects between conditions. MDS-UPDRS Part III motor scores and gait were evaluated for each stimulation state. Patients were not evaluated under offmedication/off-stimulation states due to safety concerns related to freezing and falling. For each stimulation state, gait was evaluated under single-task (walking alone) and dual-task (walking while simultaneously meeting a cognitive challenge) conditions. Cognitive tasks included forward counting in English by threes or alternate alphabet letternaming. In order to minimize learning effects, cognitive tasks were counterbalanced for each patient and were different for each walking trial. Starting numbers ranged from zero to ten and letters from A to G; these were provided verbally to the patient just prior to each walking trial. Quantitative gait analysis was performed using a validated, commercially available instrumented walkway system (GAITRite; CIR Systems, Inc., Franklin, NJ) [17]. It measures 16 feet (487 cm) in length and 2 feet (61 cm) in width, and records timing and location of foot strikes through implanted mechanical sensors at a sampling rate of 80 Hz. Patients performed walking trials wearing comfortable flat-soled shoes at a self-selected pace. Three walking trials were performed for each stimulation state and cognitive task
2.3. Data analysis
2.4. Statistical analysis Statistical analysis was performed using SPSS Statistics for Macintosh, Version 21.0 (IBM Corp., Armonk, NY). Spatiotemporal parameters of gait and MDS-UPDRS motor scores were analyzed with repeated measures of ANOVA, with the three stimulation states and the three gait tasks as repeated measures. The Greenhouse-Geisser correction was used when the assumption of sphericity was violated. Post hoc contrasts with Bonferroni adjustment for multiple comparisons were used to evaluate differences between stimulation states, task complexity, and interactions between the two, if significance was achieved. A P-value less than 0.05 is considered significant. 3. Results 3.1. MDS-UPDRS motor score for unilateral and bilateral DBS Seventeen DBS patients met inclusion/exclusion criteria and were enrolled in the study. Table 1 reports patient demographics, Mini-Mental State Examination scores, and MDS-UPDRS motor scores for patients under each stimulation state. Patient MDSUPDRS motor scores with bilateral DBS stimulation, both on and off medication, were improved compared to unilateral stimulation to the right or left side (P ¼ 0.0005 and P < 0.0001, respectively). 3.2. Functional ambulation performance Fig. 1A shows the effects of stimulation state and task condition on functional ambulation performance index. There was a main
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
3
Table 1 Patient demographics and MDS-UPDRS motor scores under various stimulation parameters (N ¼ 17). Patient Characteristics
Mean ± Standard Deviation
Age, years Gender Handedness (Edinburgh Handedness Inventory)
61.7 ± 6.46 14 males; 3 females 15 right-handed 1 left-handed 1 ambidextrous 28.8 ± 0.95 23.9 ± 14.4 28.4 ± 10.3 31.6 ± 7.66 40.5 ± 9.77
Mini Mental State Examination (MMSE) Time after surgery (months) MDS-UPDRS motor scores (ON meds/bilateral DBS) MDS-UPDRS motor scores (OFF meds/bilateral DBS) MDS-UPDRS motor scores (OFF meds/unilateral DBS)
effect of task complexity on functional ambulation performance. Ambulation performance decreased significantly under dual-task conditions [F(2,32) ¼ 13.68; P < 0.0001; h2 ¼ 0.46]. Gait under both cognitive dual-task conditions (numbers and letters) resulted in significantly lowered functional ambulation performance scores compared to the single-task condition of gait alone (P ¼ 0.016 and P ¼ 0.001, respectively). There was no main effect of stimulation state [F(2,32) ¼ 1.72; P ¼ 0.195; h2 ¼ 0.01], indicating that bilateral stimulation was not significantly different than unilateral stimulation, irrespective of task condition. Lastly, interactions between the main effects of stimulation state and task condition were not significant [F(2.-57,41.1) ¼ 1.86; P ¼ 0.158; h2 ¼ 0.10; GreenhouseGeisser]. Overall, we found that as cognitive load increased, functional gait performance decreased in both unilateral and bilateral DBS. 3.3. Normalized velocity Fig. 1B illustrates effects of stimulation state and task condition on normalized velocity. There was a significant main effect with task condition [F(1.49,23.80) ¼ 28.74; P < 0.0001; h2 ¼ 0.46; Greenhouse-Geisser]. Normalized velocity decreased significantly under dual-task conditionsdwalking while counting numbers (P ¼ 0.0004), and walking while naming letters (P < 0.0001)d compared to single-task walking. There was no main effect with stimulation state [F(2,32) ¼ 2.52; P ¼ 0.096; h2 ¼ 0.14] or interaction between stimulation state and task condition [F(4,64) ¼ 1.28; P ¼ 0.287; h2 ¼ 0.07]. Hence, cognitive dual-tasking decreased gait velocity equally in all three stimulation states, with no difference between bilateral vs. unilateral stimulation states. 3.4. Cadence Fig. 2A shows effects of stimulation state and task condition on cadence. The main effect of task condition was significant, with a substantial reduction during cognitive dual-task conditions [F(1.40,22.45) ¼ 11.952; P < 0.0001; h2 ¼ 0.428; GreenhouseGeisser]. The main effect of stimulation state was not statistically significant [F(2,32) ¼ 11.29; P ¼ 0.505; h2 ¼ 0.04; GreenhouseGeisser]. Pairwise comparisons revealed that cadence during cognitive dual-tasking was significantly lowered compared to single-task walking (P ¼ 0.006 for both dual-task conditions). 3.5. Double-limb support time Fig. 2B presents the effects of stimulation state and task condition on DLS time. Stimulation state is indicated as ipsilateral, contralateral, or bilateral. Ipsilateral refers to stimulation of the brain hemisphere on the same side as heel contact initiating the period of DLS. Contralateral refers to stimulation of the brain hemisphere opposite the heel contact initiating the period of DLS.
For each stimulation state, lateralized DLS times are averaged. In contrast to results reported thus far, patients experienced statistically significant main effects for both task condition [F(1.344,44.34) ¼ 15.18; P < 0.0001; h2 ¼ 0.32; GreenhouseGeisser] and stimulation state [F(2,66) ¼ 3.69; P ¼ 0.03; h2 ¼ 0.10]. Pairwise comparison revealed that each cognitive dualtask condition (numbers and letters) resulted in significantly longer DLS times compared to single-task gait (P < 0.0001 and P ¼ 0.0002, respectively). Evaluation of within-patient contrasts for stimulation state revealed that bilateral stimulation resulted in lowered DLS time than either-sided unilateral stimulation [F(1,33) ¼ 7.7; P ¼ 0.009; h2 ¼ 0.19]. 3.6. Step length Fig. 3A depicts the effects of stimulation state and task condition on step length. Stimulation states are indicated either as bilateral, ipsilateral or contralateral to the measured step. As with DLS time, both task condition [F(1.445,47.67) ¼ 18.47; P < 0.0001; h2 ¼ 0.36; Greenhouse-Geisser] and stimulation state [F(1.753,57.88) ¼ 6.75; P ¼ 0.003; h2 ¼ 0.17; Huynh-Feldt] significantly affected step length. Post hoc contrasts for task condition indicate that cognitive dual-task conditions resulted in shorter step length for both number and letter dual-tasks (P ¼ 0.022 and P < 0.0001, respectively. Post hoc contrasts for stimulation state revealed that bilateral stimulation enabled the longest step length, irrespective of task condition, compared to ipsilateral unilateral stimulation (P ¼ 0.0016) or contralateral unilateral stimulation (P ¼ 0.0058). No significant interaction between stimulation state and task condition was apparent. To summarize, step length decreased significantly during both cognitive dual-task conditions and during unilateral stimulation, compared to single-task conditions and bilateral stimulation state. 3.7. Step length variability Fig. 3B reflects the effects of stimulation state and task condition on step length variability. Ipsilateral refers to variability of step length measured during same-side unilateral DBS stimulation. Contralateral refers to variability of step length measured during opposite-side unilateral DBS. For each unilateral stimulation state, results for right and left sides were averaged. There was a statistically significant main effect with task condition on step length variability [F(1.743,57.51) ¼ 11.29; P ¼ 0.0001; h2 ¼ 0.26; Huynh-Feldt]. Post hoc contrasts revealed that dual-task gait during alternate letter-naming resulted in significantly greater step length variability than either single-task gait (P ¼ 0.0014) or dual-task gait during counting by threes (P ¼ 0.0036). There was no significant main effect with stimulation state [F(1.366,45.07) ¼ 1.48; P ¼ 0.237; h2 ¼ 0.04]. Hence, dual-task conditions had a significant effect on step length variability, with
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
4 G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 1. A. Effects of task complexity on functional ambulation under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load (numbers and letters), compared to gait alone. B. Effects of task complexity on normalized velocity under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect with cognitive dual-tasking compared to gait alone.
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 2. A. Changes in cadence on task complexity under three stimulation parameters (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under both cognitive load tasks when compared to gait alone. B. Effects of task complexity on DLS time under varying stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load when compared to gait alone. Bars spanning across a single task indicate a significant main effect for stimulation (specifically, bilateral DBS resulted in a shorter DLS time than ipsilateral DBS).
5
6 G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
Fig. 3. A. Changes in step length with increasing task complexity across three stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for gait under cognitive load tasks (numbers and letters) compared to gait alone. Bars spanning across a single task indicate a significant main effect for stimulation (specifically, bilateral DBS resulted in longest overall step length compared to unilateral DBS to either side in all three tasks). B. Effects of task complexity on step length variability under various stimulation states (means and standard errors). Bars spanning across various tasks indicate a significant main effect for both cognitive load tasks when compared to walking only.
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
greater variability during alternate letter-naming than for counting by threes. 4. Discussion We tested the hypothesis, demonstrated for upper extremity function, that bilateral STN DBS degrades gait performance under cognitive dual-task conditions. Consistent with earlier studies, gait performance declined for all parameters under cognitive dual-task conditions (counting or letters), independent of stimulation state. However, bilateral DBS produced greater improvements in step length and DLS time than unilateral DBS. Together, these data suggest that bilateral stimulation may assist in maintaining temporal and spatial gait performance under cognitive dual-task conditions. 4.1. MDS-UPDRS motor scores and DBS Patients experienced dramatic improvements in MDS-UPDRS motor scores with DBS. Bilateral STN DBS, both on and off medication, significantly improved total MDS-UPDRS motor scores compared to unilateral left or right stimulation conditions. There was no difference in total motor scores between bilateral DBS while on or off medication. These improvements in MDS-UPDRS motor scores are consistent with earlier findings of clinical improvement with STN DBS [20]. 4.2. DBS and gait under cognitive dual-task conditions Increased cognitive load during gait led to significant decreases in performance, independent of stimulation state. Our findings demonstrate that all measured temporal and spatial parameters of gait decline with increased task complexity. These results correspond to previous studies that investigated declines in motor and cognitive performances in patients with Parkinson's disease from single-to dual-task conditions [21e23] and are also consistent with the findings by Seri-Fainshtat et al. [14], demonstrating that bilateral DBS improves gait performance independent of task complexity. Bilateral DBS resulted in the longest step length for all tasks, compared to unilateral stimulation. Additionally, during gait while counting alternate numbers, bilateral DBS led to the shortest DLS time, compared to unilateral DBS. These results extend previous studies demonstrating that bilateral DBS improves gait more than either-sided unilateral stimulation, under single-task conditions [3,12]. Notably, though bilateral stimulation significantly improves step length and DLS time compared to either-sided unilateral stimulation, significant differences were not found among the other gait parameters across the three stimulation states. Increased task complexity resulted in decreased gait performance. Given the results of previous studies, we expected bilateral DBS to improve the spatiotemporal parameters of gait more than unilateral DBS during single-task gait. Additionally, based on the findings of Alberts et al. [13], we expected that bilateral stimulation would result in significantly worsened gait performance as task complexity increased, compared to unilateral stimulation. We found no differences in performance between stimulation states for most gait parameters, under single-task and cognitive dual-task conditions. Unexpectedly, step length and DLS time improved with bilateral stimulation. Specifically, bilateral stimulation improved step length and DLS time more than unilateral stimulation to either side under dual-task conditions. These results are contrary to the expectations from Alberts et al. [13], which found that bilateral DBS under dual-task conditions produced significantly degraded cognitive and motor performance compared with
7
unilateral stimulation. However, the specific cognitive tasks in our two studies differ. Alberts et al. investigated performance during a working memory (n-back) and upper limb-grip force task, whereas our study utilizes a working memory and gait task requiring truncal muscles. While specific mechanisms underlying motor performance degradation under cognitive dual-task conditions remain unknown, models of cortical-subcortical activity during DBS provide some insights. Studies of globus pallidus internus (GPi) neuronal activity in parkinsonian primates indicate that DBS replaces pathological signaling with a steadier pattern, thereby producing more normal movements [24,25]. In the STN, DBS stimulation may spread from motor to adjacent associative and affective STN regions, resulting in an “informational lesion.” Under conditions of increased cognitive demand, but not under less taxing conditions, the informational lesion may lead to subtle cognitive deficits and a limited ability to accommodate increased cognitive loads, reducing dual-task performance. Alberts et al. used the STN informational lesion hypothesis to explain why bilateral DBS was worse than unilateral DBS under cognitive load for upper extremities, suggesting that bilateral stimulation could produce a larger informational lesion than unilateral DBS. Even if true, this model may not be as relevant for gait and postural abnormalities, where other deep nuclei, such as the pedunculopontine nucleus and associated mesencephalic locomotor circuits, play a more prominent role in motor performance [26]. Our study has several limitations. First, patients were not tested both off-medication and off-stimulation. Given the advanced symptoms of many patients and the long duration of testing off medications, our patients could not tolerate the off-stimulation state in this study. Second, the testing protocol was quite lengthy (approximately 6 h). As a result, gait performance may have decreased significantly by the end of the session. To counter this potential limitation, stimulation test conditions were counterbalanced, alleviating the impact that fatigue would be expected to have on our results. Another limitation may be that only 90e120 min elapsed between stimulation adjustment and gait testing. Considerably longer durations may be required for the full effects of stimulation to wear off [27]. Finally, our study population was limited to patients treated with STN DBS. While STN is thought to affect gait more than GPi stimulation, we cannot say whether these results extend to patients with GPi DBS. To our knowledge, this study is the first to examine the effects of unilateral and bilateral STN DBS on gait under single and cognitive dual-task conditions in Parkinson's disease. Our findings highlight the importance of assessing performance under cognitive loading, as more representative of everyday life. The inability of STN DBS to elicit changes in the circuitry involved in executive function has led researchers to investigate interventional strategies to cope with gait under cognitive load. Several researchers have started using virtual reality, cognitive therapy to improve gait while multitasking [28,29]. Overall, we find that bilateral STN DBS is superior to unilateral STN DBS for some gait parameters (step length and DLS time), and MDS-UPDRS motor scores. For other gait parameters, there is no significant difference between unilateral and bilateral stimulation.
Author roles All aspects of research project, statistical analysis design, review, and critique (M.L.W., K.L.C., P.G.P.). Research project conception and organization (R.D.S.). All aspects of statistical analysis and manuscript preparation (G.C., M.L.W., R.D.S., P.G.P.).
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028
8
G. Chenji et al. / Parkinsonism and Related Disorders xxx (2017) 1e8
Financial disclosure/conflict of interest None of the authors have any personal financial or institutional interest in any of the drugs, materials, or devices described in this article. Funding sources This work was supported by the A. Alfred Taubman Medical Research Institute Emerging Scholars Program, Blue Cross Blue Shield of Michigan Foundation, Foundation for Physical Therapy, Michigan Parkinson Foundation, and Parkinson's Disease Foundation. Acknowledgements We would like to thank Karen Cummings and Neal Rakesh for technical assistance with recruitment, DBS programming, data collection and/or data analysis for this study. References [1] D. Muslimovic, B. Post, J.D. Speelman, B. Schmand, R.J. de Haan, Determinants of disability and quality of life in mild to moderate Parkinson disease, Neurology 70 (2008) 2241e2247. [2] V.E. Kelly, A.J. Eusterbrock, A. Shumway-Cook, The effects of instructions on dual-task walking and cognitive task performance in people with Parkinson's disease, Park. Dis. (2012) 67e1261. [3] M. Faist, J. Xie, D. Kurz, W. Berger, C. Maurer, P. Pollak, C.H. Lucking, Effect of bilateral subthalamic nucleus stimulation on gait in Parkinson's disease, Brain 124 (2001) 1590e1600. [4] N. Giladi, Gait disturbances in advanced stages of Parkinson's disease, Adv. Neurol. 86 (2001) 273e278. [5] G. Yogev, N. Giladi, C. Peretz, S. Springer, E.S. Simon, J.M. Hausdorff, Dual tasking, gait rhythmicity, and Parkinson's disease: which aspects of gait are attention demanding? Eur. J. Neurosci. 22 (2005) 1248e1256. [6] S. O'Shea, M.E. Morris, R. Iansek, Dual task interference during gait in people with Parkinson disease: effects of motor versus cognitive secondary tasks, Phys. Ther. 82 (2002) 888e897. [7] P. Crenna, I. Carpinella, M. Rabuffetti, M. Rizzone, L. Lopiano, M. Lanotte, M. Ferrarin, Impact of subthalamic nucleus stimulation on the initiation of gait in Parkinson's disease, Exp. Brain Res. 172 (2006) 519e532. [8] E.L. Johnsen, P.H. Mogensen, N.A. Sunde, K. Ostergaard, Improved asymmetry of gait in Parkinson's disease with DBS: gait and postural instability in Parkinson's disease treated with bilateral deep brain stimulation in the subthalamic nucleus, Mov. Disord. 24 (2009) 590e597. [9] H. Stolze, S. Klebe, M. Poepping, D. Lorenz, J. Herzog, W. Hamel, B. Schrader, J. Raethjen, R. Wenzelburger, H.M. Mehdorn, G. Deuschl, P. Krack, Effects of bilateral subthalamic nucleus stimulation on parkinsonian gait, Neurology 57 (2001) 144e146. [10] A.J. Bastian, V.E. Kelly, F.J. Revilla, J.S. Perlmutter, J.W. Mink, Different effects of unilateral versus bilateral subthalamic nucleus stimulation on walking and reaching in Parkinson's disease, Mov. Disord. 18 (2003) 1000e1007.
[11] R. Kumar, A.M. Lozano, E. Sime, E. Halket, A.E. Lang, Comparative effects of unilateral and bilateral subthalamic nucleus deep brain stimulation, Neurology 53 (1999) 561e566. [12] S.D. Tabbal, M. Ushe, J.W. Mink, F.J. Revilla, A.R. Wernle, M. Hong, M. Karimi, J.S. Perlmutter, Unilateral subthalamic nucleus stimulation has a measurable ipsilateral effect on rigidity and bradykinesia in Parkinson disease, Exp. Neurol. 211 (2008) 234e242. [13] J.L. Alberts, C. Voelcker-Rehage, K. Hallahan, M. Vitek, R. Bamzai, J.L. Vitek, Bilateral subthalamic stimulation impairs cognitive-motor performance in Parkinson's disease patients, Brain 131 (2008) 3348e3360. [14] E. Seri-Fainshtat, Z. Israel, A. Weiss, J.M. Hausdorff, Impact of sub-thalamic nucleus deep brain stimulation on dual tasking gait in Parkinson's disease, J. Neuroeng Rehabil. 10 (2013) 38. [15] R.R. Coleman, V. Kotagal, P.G. Patil, K.L. Chou, Validity and efficacy of screening algorithms for assessing deep brain stimulation candidacy in Parkinson disease, Mov. Disord. Clin. Pract. 1 (2014) 342e347. [16] A.J. Mitchell, A meta-analysis of the accuracy of the mini-mental state examination in the detection of dementia and mild cognitive impairment, J. Psychiatr. Res. 43 (2009) 411e431. [17] K.E. Webster, J.E. Wittwer, J.A. Feller, Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait, Gait Posture 22 (2005) 317e321. [18] A.J. Nelson, Functional ambulation profile, Phys. Ther. 54 (1974) 1059e1065. [19] A. Gouelle, Use of functional ambulation performance score as measurement of gait ability: review, J. Rehabil. Res. Dev. 51 (2014) 665e674. [20] K.L. Chou, J.L. Taylor, P.G. Patil, The MDS-UPDRS tracks motor and non-motor improvement due to subthalamic nucleus deep brain stimulation in Parkinson disease, Park. Relat. Disord. 19 (2013) 966e969. [21] R. Marchese, M. Bove, G. Abbruzzese, Effect of cognitive and motor tasks on postural stability in Parkinson's disease: a posturographic study, Mov. Disord. 18 (2003) 652e658. [22] M. Morris, R. Iansek, F. Smithson, F. Huxham, Postural instability in Parkinson's disease: a comparison with and without a concurrent task, Gait Posture 12 (2000) 205e216. [23] L. Rochester, V. Hetherington, D. Jones, A. Nieuwboer, A.M. Willems, G. Kwakkel, E. Van Wegen, Attending to the task: interference effects of functional tasks on walking in Parkinson's disease and the roles of cognition, depression, fatigue, and balance, Arch. Phys. Med. Rehabil. 85 (2004) 1578e1585. [24] W.M. Grill, A.N. Snyder, S. Miocinovic, Deep brain stimulation creates an informational lesion of the stimulated nucleus, Neuroreport 15 (2004) 1137e1140. [25] T. Hashimoto, C.M. Elder, M.S. Okun, S.K. Patrick, J.L. Vitek, Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons, J. Neurosi 23 (2003) 1916e1923. [26] K.A. Follett, D. Torres-Russotto, Deep brain stimulation of globus pallidus interna, subthalamic nucleus, and pedunculopontine nucleus for Parkinson's disease: which target? Park. Relat. Disord. 18 (Suppl 1) (2012) S165eS167. [27] P. Temperli, J. Ghika, J.G. Villemure, P.R. Burkhard, J. Bogousslavsky, F.J. Vingerhoets, How do parkinsonian signs return after discontinuation of subthalamic DBS? Neurology 60 (2003) 78e81. [28] A. Mirelman, I. Maidan, T. Herman, J.E. Deutsch, N. Giladi, J.M. Hausdorff, Virtual reality for gait training: can it induce motor learning to enhance complex walking and reduce fall risk in patients with Parkinson's disease? J. Gerontol. A Biol. Sci. Med. Sci. 66 (2011) 234e240. [29] O. Segev-Jacubovski, T. Herman, G. Yogev-Seligmann, A. Mirelman, N. Giladi, J.M. Hausdorff, The interplay between gait, falls and cognition: can cognitive therapy reduce fall risk? Expert Rev. Neurother. 11 (2011) 1057e1075.
Please cite this article in press as: G. Chenji, et al., Parkinsonian gait improves with bilateral subthalamic nucleus deep brain stimulation during cognitive multi-tasking, Parkinsonism and Related Disorders (2017), http://dx.doi.org/10.1016/j.parkreldis.2017.02.028