- Email: [email protected]
Gait & Posture Special Issue: Gait adaptations in response to obstacle type in fallers with Parkinson’s disease
Gait & Posture 61 (2018) 368–374
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
Gait & Posture Special Issue: Gait adaptations in response to obstacle type in fallers with Parkinson’s disease Lisa Alcocka, Brook Galnaa,b, Jeffrey M. Hausdorffc,d,e, Sue Lorda,f, Lynn Rochestera,g,
T
⁎
a
Institute of Neuroscience/Institute for Ageing, Newcastle University, Newcastle upon Tyne, United Kingdom School of Biomedical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom c Center for the Study of Movement, Cognition and Mobility, Neurological Institute, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel d Rush Alzheimer’s Disease Center and Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, IL, USA e Sagol School of Neuroscience and Department of Physical Therapy, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel f School of Clinical Sciences, Auckland University of Technology, New Zealand g The Newcastle upon Tyne NHS Foundation Trust, Newcastle upon Tyne, United Kingdom b
A R T I C L E I N F O
A B S T R A C T
Keywords: Obstacle negotiation Tripping Older adults Approach phase Crossing phase
Background: Gait impairment places older adults and people with Parkinson’s disease (PD) at an increased risk of falls when walking over obstacles. Increasing the height of obstacles results in greater challenge to balance however little is known about the demands encountered when negotiating obstacles of greater depth which may be greater for PD who often walk with a short, shuffling gait. Research question: To describe gait adaptation in older adults and people with PD when walking over long and tall obstacles. Methods: 20 people with PD and 13 older adults with a history of falls walked across an instrumented walkway under four conditions: level walking, and over a small, long and tall obstacle. Differences due to group, step and obstacle type were evaluated using General Linear Models. Results: An increased step duration, longer single limb support and a wider step (p < .033) were observed when crossing the tall obstacle for both older adults and PD. The PD group completed the crossing steps more slowly than controls, with a shorter step and longer single limb support (p < .043). Obstacle type did not significantly influence step length. Significance: The greatest temporal-spatial adaptations were elicited when participants negotiated the tall obstacle. Electing a wider step when crossing the tall obstacle was a strategy common to both faller groups (older adults and PD). The tall obstacle presented added challenge for PD who spent longer in single limb support during the crossing steps compared to controls. The long obstacle did not cause a disproportionate change in step length in people with PD, and we suggest that the obstacle may have acted as a visual cue in this group.
1. Introduction
Walking over obstacles increases balance demands on the motor system and thus poses an increased risk of trips and falls. The dimensions of an obstacle are associated with the magnitude of gait adaptations required to step over it [7]. In young adults, there is a graded response to obstacle height whereby stance, swing and double limb support duration of the lead crossing step increased with greater obstacle height [8]. However, the spatial components (step length and width) of the crossing steps are less susceptible to change with variation in obstacle height [8]. People with PD approach tall obstacles more slowly [9,10] and adopt a wider base of support when crossing a tall obstacle compared to controls [9,11]. They also spend a longer time in single limb support during the lead crossing step [9,10] and demonstrate a greater mediolateral excursion of their centre of mass [10,12].
Accidental falls are a leading cause of mortality and injury in older adults, especially in people with Parkinson’s disease (PD). PD affects approximately 1% of adults aged ≥60 years [1] and falls prevalence is reportedly double that of the general older population with up to two thirds of people with PD reporting falls in a given year [2,3]. Despite differences in the underlying aetiology of falls in older adults with and without PD, falls commonly occur during ambulation [4] and tripping constitutes a common pre-fall event in both older adults [5] and people with PD [2]. Moreover, tripping is the most common cause of near falls in people with PD who have previously fallen and this was significantly higher than those who had not previously fallen [6]. ⁎
Corresponding author at: Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, NE4 5PL E-mail address: [email protected] (L. Rochester).
https://doi.org/10.1016/j.gaitpost.2018.01.030 Received 22 December 2017; Received in revised form 22 January 2018; Accepted 25 January 2018 0966-6362/ © 2018 Elsevier B.V. All rights reserved.
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 1. Schematic detailing the obstacle dimensions (A) and the approach (A2 and A1) and crossing (lead and trail) steps extracted from the instrumented walkway (B).
randomised control trial (n = 282) evaluating the potential of two 6week treadmill training exercise programmes to improve gait and reduce falls in older adults [16,17]. Briefly, participants were recruited providing that they: reported at least two falls in the 6-months prior to participation; were aged 60–90 years; had adequate hearing and vision; and had no changes to their prescribed medications in the month prior to recruitment [17]. Common exclusion criteria included moderate-tosevere cognitive impairment, and any considerable visual or auditory deficit and orthopaedic or musculoskeletal disease. PD participants had a diagnosis of idiopathic PD according to the UK Parkinson’s Disease Brain Bank Criteria, mild to moderate symptoms (defined as Hoehn and Yahr stage II–III) and were taking anti-Parkinsonian medication (Deep brain stimulation was included). The PD group were tested whilst optimally medicated approximately 1-h post dopaminergic medications. Data presented here were collected at a single site pre-intervention and approved by the local NHS ethics committee (Ref: 12-NE-0249). Written informed consent was obtained according to the Declaration of Helsinki [18]. The Movement Disorder Society revised Unified Parkinson’s Disease Rating Scale [19] was administered by a trained examiner. Freezing of gait was quantified using the new Freezing of Gait Questionnaire (nFOG) [20]. The modified Falls Efficacy Scale (FES-I) was used to assess fear of falling [21]. The four square step test (FSST) [22] was chosen as a measure of dynamic mobility [23] and the Mini-Balance Evaluations Systems Test (Mini-BESTest) [24] which evaluates anticipatory and reactive postural control, sensory orientation and gait stability was used as a measure of general physical functioning. Each of the outcome measures have been shown to be reliable and valid for use with older populations and people with PD [20,23,25–28]. Basic
Similar to younger adults [8], increasing the height of an obstacle results in greater gait adaptations in mild to moderately severe PD, with one study reporting an increased step length and greater limb elevation when negotiating a taller obstacle [13]. Stepping over obstacles of greater depth (requiring adaptations to step length) occurs regularly during everyday life, such as walking over puddles or stepping over the gap when boarding a train. However, little is known about this type of obstacle crossing in older adults and people with PD, with previous studies manipulating obstacle height and not depth. This may be pertinent as step length decreases with age [14] and is further reduced in people with PD [15]. We were interested to explore the strategies made by older adults who have experienced previous falls for whom gait impairments are likely to be more pronounced. Therefore, the aim of this study was to describe gait adaptation in two groups of fallers: older adults and people with PD, when walking over obstacles of increased height and depth. We hypothesised that tall obstacles (increased height) would result in similar gait adaptations in people with PD and controls, but that a long obstacle (greater depth) would be challenging for people with PD and result in greater temporal spatial adaptations compared to controls.
2. Methodology 2.1. Participants A group of older adults with PD and without PD participated in this study which was a collaborative sub-study of the V-TIME (Virtual reality treadmill training to reduce falls and enhance mobility in the elderly) project. The V-TIME project was an international multi-centre 369
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Table 1 Demographic data. Control (n = 13)
PD (n = 20)
Group differences
Age (years) Height (m) Mass (kg) Sex (m/f; n) Education (years) MOCA (score/30) Bilateral visual acuity (LOGMAR) Bilateral contrast sensitivity (Mars CS) Fear of falling (FES-I) General physical function (Mini Best) Dynamic mobility (FSST)
76.4 [7.0] 1.65 [0.11] 73.8 [12.4] 4 m/9f 12.3 [3.9] 24 [5] 0.134 [0.12] 1.47 [0.16] 30.6 [10.6] 23.2 [5.6] 11.4 [7.7]
70.5 [7.7] 1.68 [0.09] 82.6 [16.1] 12 m/8f 13.3 [3.9] 28 [2] 0.003 [0.12] 1.53 [0.15] 36.8 [11.0] 23.0 [4.7] 9.6 [7.7]
p = .029 * p = .457 p = .084 p = .101 p = .484 p < .001 * p = .006 * p = .325 p = .119 p = .911 p = .525
Disease specific outcomes Hoehn & Yahr disease stage II (n) Hoehn & Yahr disease stage III (n) Disease duration (years) UPDRS III (Severity of motor symptoms; score/132) Freezing of gait (freezer/non-freezer; n)
– – – – –
15 5 6.5 (1, 37) 20.6 [7.9] 10/10
– – – – –
Temporal-spatial gait parameters ^ Step velocity (m/s) Step length (m) Step width (m) Step duration (msec) Single limb support (msec) Double limb support (msec)
0.91 [0.22] 0.54 [0.11] 0.11 [0.05] 609 [78] 396 [64] 817 [114]
0.83 [0.21] 0.51 [0.11] 0.11 [0.03] 627 [71] 391 [48] 859 [113]
p = .290 p = .019 * p = .943 p = .487 p = .834 p = .302
*denotes significant group difference p < .05. Disease duration was not normally distributed and is presented Median (minimum, maximum). Categorical data are presented as counts. All other data are presented Mean [SD]. ^ Statistical differences were computed using ANCOVA adjusting for age, global cognition (MOCA) and visual acuity.
ophthalmic measures of binocular visual acuity (LogMAR, Sussex-Vision, UK) and contrast sensitivity (Mars letter chart, Mars Perceptrix™, New York, USA) were obtained.
Heel) and the obstacle were also calculated.
2.2. Protocol
Sample size calculations from prior studies indicated that we required a minimum of 13 people in each group (d 1.2, α two-tailed < .05, β > 0.80; e.g. step velocity difference reported by Galna et al. [9]). Data were generally normally distributed and parametric statistics were used unless otherwise stated. Group differences in demographics were evaluated using independent t-tests with statistical significance accepted at p < .05. Group differences identified from demographic data were entered as covariates in subsequent analyses of group-based effects. General Linear Models were developed for each of the gait characteristics to evaluate the within (obstacle type and step) and between subject (group) effects. Change scores between steps (i.e. difference between A2 and A1 [A2-A1], A1 and Lead [A1-Ld], and Lead and Trail [Ld-Tr]) were calculated to investigate interaction effects. The Toe-Obstacle and Obstacle-Heel distances were the only outcomes that were not normally distributed and non-parametric statistics were used.
2.4. Statistical analysis
Participants began each trial with their eyes closed while the obstacle was placed on the walkway to prevent preparatory motor planning prior to the start of the trial [29]. Upon the word ‘go’, they opened their eyes and walked over a 7-m instrumented walkway (GAITRite®, Platinum model v.4.5, USA, 240 Hz). Participants walked under four experimental conditions: Level-ground walking and walking over a small, long and tall obstacle (see Fig. 1 for obstacle dimensions). Each obstacle was covered in a highly visible yellow poly-vinyl chloride material and was constructed of foam with a wooden base allowing it to compress in the event of contact whilst standing independently. Three trials were collected for each walking condition but only the first trial was selected for further analysis since this was considered to be the most natural response when participants were naïve to the environmental condition. Condition order was standardised with three level gait trials completed first followed by the obstacle crossing trials (small, long then tall). All participants wore a safety harness attached to an overhead gantry to ensure participant safety and prevent falls.
3. Results The PD participants were significantly younger, with better global cognition (MOCA) and visual acuity (p < .03) (Table 1). The groups were matched for other outcomes including: fear of falling; general physical function and dynamic mobility (p > .05). The PD group were of mild to moderate disease severity with half reporting freezing (nFOG score > 1). There were no FOG episodes observed and no obstacle contacts recorded for the first trial for either group or condition. Gait impairment was evident with 75% of the PD group (n = 15) and 62% of the control group (n = 8) walking at < 1.0m/s. Only one significant group difference was observed during level gait indicating that the PD group walked with a significantly reduced step length compared to controls (p = .019). One participant in the PD group could not negotiate the long obstacle condition unaided and one participant from the control group could not negotiate the tall obstacle unaided. Cases
2.3. Data analysis Footfall data obtained from the instrumented walkway were processed within the manufacturer’s software. Temporal-spatial gait characteristics (velocity, step length, step width, step duration, single and double limb support) were collated using Microsoft Access database 2007. Ensemble averages were compiled from step data for the level gait condition. A custom macro implemented within Visual Basic Applications for Microsoft® Excel extracted gait characteristics for each of the approach and crossing steps measured during the obstacle crossing conditions (depicted in Fig. 1). The distance between the toe of the trail limb (Toe-Obstacle) and the heel of the lead limb (Obstacle370
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Table 2 General Linear Model results for the effects of group and obstacle type. Gait characteristic
Between group effects ^ (Group, Group x Step, Group x Obstacle type)
Within group effects (Obstacle type, Step, Obstacle type x Step)
Step velocity
Group x Step (p = .026) No significant post hoc comparisons
Step length
Group (p = .043) Control > PD (p = .043) Group x Step (p = .004) Control Ld-Tr > PD Ld-Tr (p = .015) No significant main or interaction effects
Obstacle type (p < .001) Small > Tall (p = .002) Step (p = .005) A2-A1 < Ld-Tr (p = .001) A1-Ld > Ld-Tr (p = .002) Obstacle type x Step (p < .001) Long A1-Ld > Tall A1-Ld (p < .001) Long Ld-Tr > Tall Ld-Tr (p = .031) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld > Ld-Tr (p < .001) Step (p = .001) A2-A1 > A1-Ld (p = .012) A1-Ld > Ld-Tr (p = .006) Obstacle type x Step (p = .033) Small A1-Ld < Tall A1-Ld (p = .009) Obstacle type (p < .001) Tall > Small & Long (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld < Ld-Tr (p < .001) Obstacle type x Step (p < .001) Small A1-Ld < Long A1-Ld (p = .019) Small A1-Ld < Tall A1-Ld (p = .001) Long A1-Ld < Tall A1-Ld (p < .001) Obstacle type (p < .001) Tall > Small & Long (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld > Ld-Tr (p < .001) Obstacle type x Step (p < .001) Small A1-Ld < Long A1-Ld (p = .016) Small A1-Ld < Tall A1-Ld (p < .001) Long A1-Ld < Tall A1-Ld (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A1-Ld > Ld-Tr (p < .001)
Step width
Step duration
Group x Step (p = .043) PD A1-Ld > Control A1-Ld (p = .038)
Single limb support
Group x Step (p = .043) PD A1-Ld > Control A1-Ld (p = .040)
Double limb support
Group x Obstacle type (p = .035) No significant post hoc comparisons
Results from post hoc tests are italicised. Ld denotes lead step and Tr denote trail step. ^ Group differences are reported whilst controlling for age, MOCA and visual acuity as covariates.
stepping over and their heel (lead step) further from the tall obstacle compared to the small obstacle (Fig. 3).
were excluded pairwise within the General Linear Models leaving 19 people with PD and 12 controls. Group mean, median and dispersion values for the gait data collected during the obstacle crossing trials is included in the Supplementary material.
3.2. Gait adaptations made by people with Parkinson’s disease during obstacle crossing
3.1. Gait adaptations in response to obstacles type People with PD walked with a shorter step length compared to controls (Table 2 and Fig. 2; Main effect: p = .043) which was more pronounced during the lead crossing step (Group x Step interaction; p = .015). Group x Step interactions also showed that people with PD increased their step duration (p = .038) and single limb support duration (p = .040) more than controls for the lead crossing step (A1-Ld, p = .038). There were no statistically significant group differences in the distance between the foot and the obstacle (Fig. 3), although the PD group generally positioned the toe of their trail step further away from the obstacle than controls (∼5 cm) and placed the heel of their lead step closer to the far edge of the obstacle compared to controls (∼3 cm).
All participants walked with a slower velocity, longer step duration and single limb support duration when negotiating the tall obstacle compared to the small and long obstacles (Main effect; p < .001; Table 2 and Fig. 2) . Main effects for Step showed that participants completed the lead crossing step (A1-Ld) more quickly (p = .002) and with a longer step duration (p < .001), double limb support (p < .001), step length, (p < .001) and increased step width (p = .006) compared to the other steps. Obstacle type x Step interactions were identified for step velocity, step width, duration and single limb support. Participants slowed down the most and widened their step width during the lead limb crossing step when walking over the tall obstacle (p < .009). The tall obstacle required the longest single limb support duration, followed by the long and small obstacle (tall > long > small, p < .016). Single limb support time changed significantly with each step for all participants, with the longest single limb support duration observed at the lead step (A1Ld, p < .001) for all obstacle types. There was a significant difference in the Toe-Obstacle and Obstacle-Heel distances such that both groups positioned their toe (trail step) closer to the tall obstacle prior to
4. Discussion The results of the present study confirm that motor symptoms due to PD are evident during obstacle crossing, even when compared to a control group who have previously fallen, with differences highlighted particularly during the more challenging crossing steps. Taller obstacles evoked the largest modifications, indicating they challenged the system 371
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 2. Group mean data for each of the obstacle conditions and steps. a denotes a significant Group x Step interaction and b denotes a significant Obstacle type x Step interaction.
372
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 3. Distance from the toe and heel to the obstacle when negotiating the small, long and tall obstacles.
long obstacle may have acted as an external visual cue prompting an increased step velocity and step length during the lead crossing step. We know that the neurodegeneration associated with PD pathology affects basal ganglia function resulting in difficulty initiating and selecting appropriate motor outputs. To overcome this, physiotherapists train patients with PD to use visual cues as they have been shown to be effective in reconfiguring the spatial scaling of movement [34]. Whilst not of statistical significance, it is clinically of interest that one participant with PD could not physically negotiate the long obstacle unaided. Further investigation revealed that this participant was 1.65 m in height, had the shortest step length of the cohort during level gait (0.24m) and the most severe motor symptoms (UPDRSIII 43/132). It appears that with appropriate motor planning and foot placement before and after the obstacle, the long obstacle did not overly disturb the natural cyclical motion of gait compared to the tall obstacle which we suggest likely caused a greater vertical displacement of the centre of mass. However, the long obstacle still required that both groups significantly increased their step length during the lead crossing step compared to that observed during gait (PD > 31% increase, Control > 24% increase, p < .001). There are some limitations of the present study which must be acknowledged when interpreting the results. Participants were aware of the purpose of the study prior to assessment and perhaps greater group differences may occur in the community when obstacles are not always anticipated. We would expect that obstacles encountered in unfamiliar or changeable environments likely pose greater threat and require quicker temporal-spatial adaptations. Lastly, smaller obstacles encountered in a natural setting may pose added threat than observed in the present study if they are unexpected and perhaps less visually prominent. It is also possible that in environments outside of the laboratory, older adults with a history and/or fear of falling may opt to avoid the obstacle as a safer motor strategy. This was evident in the present study with two participants unable to compete the long and tall obstacle conditions unaided. While the groups were well matched for many functional outcomes, there were distinct differences between the groups in age, global cognition and visual acuity which we accordingly controlled for in the present analyses. Further complementary work will characterise foot trajectories to explore how this is modified with respect to obstacle type to further understand the level of trip risk imposed compared to a non-falling cohort. Understanding the temporal spatial adaptations made during the approach and crossing of different types of obstacles will be informative for guiding falls prevention and physiotherapy management. This study has identified important differences in obstacle crossing performance in
most, as they were associated with participants standing longer in single limb support and eliciting a wider step. The tall obstacle prompted the largest change in the temporal gait parameters. This is in agreement with previous research in young adults [8]. We found that both groups took a wider step when crossing the tall obstacle compared to the small and long obstacles. This has been observed previously in people with PD [9,11] and likely relates to the maintenance of mediolateral balance [30]. Corroborative evidence from studies investigating older adults with and without balance impairment and another in people with PD showed that the horizontal displacement of the centre of mass was significantly larger in those with balance problems and those with PD [12,31]. Moreover, electing a wider step was a strategy employed by older adults when visual information about the proximity of an obstacle was reduced [32]. The findings in the present study concur that older adults with a history of falls elect to use the same strategy as people with PD with a history of falls when crossing tall obstacles. Given that the groups were of similar height (p = .457), the variation in obstacle height ought to have caused a comparable disturbance thus posing similar biomechanical demands for all participants. We observed no differences in step width across the obstacle conditions due to group, which suggests that increasing the base of support when crossing a tall obstacle may be a common strategy employed by both faller groups regardless of PD pathology. The tall obstacle presented an added challenge for the PD group in terms of balance control. During the lead limb crossing step (A1-Ld), the swing limb moves anteriorly from behind the centre of mass to in front of the centre of mass once over the obstacle. This single limb support phase, during which there is a redistribution of weight transfer, may be considered the most challenging step when the body hovers over the obstacle. The findings in the present study show that people with PD crossed the tall obstacle with a longer step duration which was mediated by a longer single limb support time. This may place people with PD at an increased risk of tripping or loosing balance when walking over obstacles, particularly tall obstacles. Although the underlying mechanisms behind this increased single limb support time is unclear, we speculate that interventions that train balance during single limb support and strengthen the lower limbs may improve obstacle crossing performance and safety. Evidence suggests that improving obstacle crossing performance in older adults may be achieved through task-specific interventions targeting obstacle crossing gait using a variety of obstacle types, and reassuringly these improvements were associated with a reduced risk of falls [33]. Surprisingly, there were few significant alterations observed in the PD group when negotiating the long obstacle and we suggest that the 373
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
people with PD. Considering that experiencing a previous fall remains a common feature underpinning an increased risk of future falls in older adults with PD [35], there is a paucity of evidence documenting the temporal-spatial gait alterations made by cohorts composed of fallers exclusively, particularly during a complex and risky task such as obstacle crossing. This study has taken a novel approach to understanding obstacle crossing performance by including a group of older adult fallers matched for fear of falling, general physical function and dynamic mobility as controls and has identified gait adaptations with respect to obstacle type and faller group. Given the exploratory nature of this study we elected to not correct for multiple comparisons and this may be considered a limitation. Further work is consequently required to characterise gait alterations in larger cohorts with obstacles of varying height and depth.
factors and association with falling, Clin. Neurol. Neurosurg. 161 (2017) 51–55. [7] I. Maidan, S. Eyal, I. Kurz, N. Geffen, E. Gazit, L. Ravid, et al., Age-associated changes in obstacle negotiation strategies: does size and timing matter? Gait Posture 59 (2018) 242–247. [8] S. Yamaji, S. Demura, H. Sugiura, Influence of degraded visual acuity from lightscattering goggles on obstacle gait, Health (N. Y.) 3 (2011) 99–105. [9] B. Galna, A.T. Murphy, M.E. Morris, Obstacle crossing in people with Parkinson’s disease: foot clearance and spatiotemporal deficits, Hum. Mov. Sci. 29 (2010) 843–852. [10] E.L. Stegemöller, T.A. Buckley, C. Pitsikoulis, E. Barthelemy, R. Roemmich, C.J. Hass, Postural instability and gait impairment during obstacle crossing in Parkinson’s disease, Arch. Phys. Med. Rehabil. 93 (2012) 703–709. [11] R. Vitório, E. Lirani-Silva, A.M. Baptista, F.A. Barbieri, P.C.R. dos Santos, C. Teixeira-Arroyo, et al., Disease severity affects obstacle crossing in people with Parkinson’s disease, Gait Posture 40 (2014) 266–269. [12] B. Galna, A.T. Murphy, M.E. Morris, Obstacle crossing in Parkinson’s disease: mediolateral sway of the centre of mass during level-ground walking and obstacle crossing, Gait Posture 38 (2013) 790–794. [13] R. Vitório, F. Pieruccini-Faria, F. Stella, S. Gobbi, L.T.B. Gobbi, Effects of obstacle height on obstacle crossing in mild Parkinson’s disease, Gait Posture 31 (2010) 143–146. [14] H.B. Menz, S.R. Lord, R.C. Fitzpatrick, Age-related differences in walking stability, Age Ageing 32 (2003) 137–142. [15] B. Galna, S. Lord, D.J. Burn, L. Rochester, Progression of gait dysfunction in incident Parkinson’s disease: impact of medication and phenotype, Mov. Disord. 30 (2015) 359–367. [16] A. Mirelman, L. Rochester, M. Reelick, F. Nieuwhof, E. Pelosin, G. Abbruzzese, et al., V-TIME: a treadmill training program augmented by virtual reality to decrease fall risk in older adults: study design of a randomized controlled trial, BMC Neurol. 13 (2013) 15. [17] A. Mirelman, L. Rochester, I. Maidan, S. Del Din, L. Alcock, F. Nieuwhof, et al., Addition of a non-immersive virtual reality component to treadmill training to reduce fall risk in older adults (V-TIME): a randomised controlled trial, Lancet 388 (2016) 1170–1182. [18] World Medical Association, Ethical principles for medical research involving human subjects, Bull. World Health Organ. 79 (2001) 373. [19] C.G. Goetz, B.C. Tilley, S.R. Shaftman, G.T. Stebbins, S. Fahn, P. Martinez-Martin, et al., Movement disorder society-sponsored revision of the unified parkinson’s disease rating scale (MDS-UPDRS): scale presentation and clinimetric testing results, Mov. Disord. 23 (2008) 2129–2170. [20] A. Nieuwboer, L. Rochester, T. Herman, W. Vandenberghe, G.E. Emil, T. Thomaes, et al., Reliability of the new freezing of gait questionnaire: agreement between patients with Parkinson’s disease and their carers, Gait Posture 30 (2009) 459–463. [21] L. Yardley, N. Beyer, K. Hauer, G. Kempen, C. Piot-Ziegler, C. Todd, Development and initial validation of the falls efficacy scale-international (FES-I), Age Ageing 34 (2005) 614–619. [22] W. Dite, V.A. Temple, A clinical test of stepping and change of direction to identify multiple falling older adults, Arch. Phys. Med. Rehabil. 83 (2002) 1566–1571. [23] R.P. Duncan, G.M. Earhart, Four square step test performance in people with Parkinson disease, J. Neurol. Phys. Ther. 37 (2013) 2–8. [24] F. Franchignoni, F. Horak, M. Godi, A. Nardone, A. Giordano, Using psychometric techniques to improve the Balance Evaluation Systems Test: the mini-BESTest, J. Rehabil. Med. 42 (2010) 323–331. [25] N. Giladi, J. Tal, T. Azulay, O. Rascol, D.J. Brooks, E. Melamed, et al., Validation of the freezing of gait questionnaire in patients with Parkinson’s disease, Mov. Disord. 24 (2009) 655–661. [26] F.B. Horak, D.M. Wrisley, J. Frank, The balance evaluation systems test (BESTest) to differentiate balance deficits, Phys. Ther. 89 (2009) 484–498. [27] A.L. Leddy, B.E. Crowner, G.M. Earhart, Functional gait assessment and balance evaluation system test: reliability, validity, sensitivity, and specificity for identifying individuals with Parkinson disease who fall, Phys. Ther. 91 (2011) 102–113. [28] N. Dewan, J.C. MacDermid, Fall efficacy scale-international (FES-I), J. Physiother. 60 (2014) 60. [29] A.E. Patla, J.N. Vickers, Where and when do we look as we approach and step over an obstacle in the travel path, Neuroreport 8 (1997) 3661–3665. [30] P.M. McAndrew Young, J.B. McAndrew Young, Voluntary changes in step width and step length during human walking affect dynamic margins of stability, Gait Posture 36 (2012) 219–224. [31] L.-S. Chou, K.R. Kaufman, M.E. Hahn, R.H. Brey, Medio-lateral motion of the center of mass during obstacle crossing distinguishes elderly individuals with imbalance, Gait Posture 18 (2003) 125–133. [32] S. Kunimune, S. Okada, The effects of object height and visual information on the control of obstacle crossing during locomotion in healthy older adults, Gait Posture 55 (2017) 126–130. [33] Y.-Y. Liao, Y.-R. Yang, Y.-R. Wu, R.-Y. Wang, Factors influencing obstacle crossing performance in patients with Parkinson’s Disease, PLoS One 9 (2014) e84245. [34] G.N. Lewis, W.D. Byblow, S.E. Walt, Stride length regulation in Parkinson’s disease: the use of extrinsic, visual cues, Brain 123 (2000) 2077–2090. [35] S.S. Paul, C.G. Canning, C. Sherrington, S.R. Lord, J.C. Close, V.S. Fung, Three simple clinical tests to accurately predict falls in people with Parkinson’s disease, Mov. Disord. 28 (2013) 655–662.
5. Conclusions Obstacles of greater height (tall obstacles), rather than depth (long obstacles), prompt the largest alterations in temporal-spatial gait and as such pose increased challenge to older adults with a history of falls. Increasing step width during the lead crossing step to increase stability was an adaptation common to both falling groups when crossing the tall obstacle, regardless of pathology. An increased time spent in single limb support was observed in people with PD when crossing the tall obstacle which may challenge balance more so than long or small obstacles. Conflict of interest statement The authors have no conflicts of interest to declare. Acknowledgements The authors would like to thank Aodhán Hickey for his help with data collection. This research was funded by the V-TIME project which is a European Union 7th Framework Programme (FP7) under the Health theme (FP7–278169, HEALTH.2011.2.2.2-1). This research was supported by the NIHR Newcastle Biomedical Research Unit based at The Newcastle Hospitals NHS Foundation Trust and Newcastle University. The research was also supported by NIHR Newcastle CRF Infrastructure funding. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gaitpost.2018.01.030. References [1] R.L. Naussbaum, C.E. Ellis, Alzheimer's disease and Parkinson's disease, N. Engl. J. Med. 348 (2003) 1356–1364. [2] H. Stolze, S. Klebe, C. Zechlin, C. Baecker, L. Friege, G. Deuschl, Falls in frequent neurological diseases, J. Neurol. 251 (2004) 79–84. [3] A. Ashburn, E. Stack, R.M. Pickering, C.D. Ward, A community-dwelling sample of people with Parkinson’s disease: characteristics of fallers and non-fallers, Age Ageing 30 (2001) 47–52. [4] K. Mactier, S. Lord, A. Godfrey, D. Burn, L. Rochester, The relationship between real world ambulatory activity and falls in incident Parkinson’s disease: influence of classification scheme, Parkinsonism Relat. Disord. 21 (2014) 236–242. [5] W.P. Berg, H.M. Alessio, E.M. Mills, C. Tong, Circumstances and consequences of falls in independent community-dwelling older adults, Age Ageing 26 (1997) 261–268. [6] T. Gazibara, D. Kisic Tepavcevic, M. Svetel, A. Tomic, I. Stankovic, V.S. Kostic, et al., Near-falls in people with Parkinson’s disease: circumstances, contributing
374
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
Gait & Posture Special Issue: Gait adaptations in response to obstacle type in fallers with Parkinson’s disease Lisa Alcocka, Brook Galnaa,b, Jeffrey M. Hausdorffc,d,e, Sue Lorda,f, Lynn Rochestera,g,
T
⁎
a
Institute of Neuroscience/Institute for Ageing, Newcastle University, Newcastle upon Tyne, United Kingdom School of Biomedical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom c Center for the Study of Movement, Cognition and Mobility, Neurological Institute, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel d Rush Alzheimer’s Disease Center and Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, IL, USA e Sagol School of Neuroscience and Department of Physical Therapy, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel f School of Clinical Sciences, Auckland University of Technology, New Zealand g The Newcastle upon Tyne NHS Foundation Trust, Newcastle upon Tyne, United Kingdom b
A R T I C L E I N F O
A B S T R A C T
Keywords: Obstacle negotiation Tripping Older adults Approach phase Crossing phase
Background: Gait impairment places older adults and people with Parkinson’s disease (PD) at an increased risk of falls when walking over obstacles. Increasing the height of obstacles results in greater challenge to balance however little is known about the demands encountered when negotiating obstacles of greater depth which may be greater for PD who often walk with a short, shuffling gait. Research question: To describe gait adaptation in older adults and people with PD when walking over long and tall obstacles. Methods: 20 people with PD and 13 older adults with a history of falls walked across an instrumented walkway under four conditions: level walking, and over a small, long and tall obstacle. Differences due to group, step and obstacle type were evaluated using General Linear Models. Results: An increased step duration, longer single limb support and a wider step (p < .033) were observed when crossing the tall obstacle for both older adults and PD. The PD group completed the crossing steps more slowly than controls, with a shorter step and longer single limb support (p < .043). Obstacle type did not significantly influence step length. Significance: The greatest temporal-spatial adaptations were elicited when participants negotiated the tall obstacle. Electing a wider step when crossing the tall obstacle was a strategy common to both faller groups (older adults and PD). The tall obstacle presented added challenge for PD who spent longer in single limb support during the crossing steps compared to controls. The long obstacle did not cause a disproportionate change in step length in people with PD, and we suggest that the obstacle may have acted as a visual cue in this group.
1. Introduction
Walking over obstacles increases balance demands on the motor system and thus poses an increased risk of trips and falls. The dimensions of an obstacle are associated with the magnitude of gait adaptations required to step over it [7]. In young adults, there is a graded response to obstacle height whereby stance, swing and double limb support duration of the lead crossing step increased with greater obstacle height [8]. However, the spatial components (step length and width) of the crossing steps are less susceptible to change with variation in obstacle height [8]. People with PD approach tall obstacles more slowly [9,10] and adopt a wider base of support when crossing a tall obstacle compared to controls [9,11]. They also spend a longer time in single limb support during the lead crossing step [9,10] and demonstrate a greater mediolateral excursion of their centre of mass [10,12].
Accidental falls are a leading cause of mortality and injury in older adults, especially in people with Parkinson’s disease (PD). PD affects approximately 1% of adults aged ≥60 years [1] and falls prevalence is reportedly double that of the general older population with up to two thirds of people with PD reporting falls in a given year [2,3]. Despite differences in the underlying aetiology of falls in older adults with and without PD, falls commonly occur during ambulation [4] and tripping constitutes a common pre-fall event in both older adults [5] and people with PD [2]. Moreover, tripping is the most common cause of near falls in people with PD who have previously fallen and this was significantly higher than those who had not previously fallen [6]. ⁎
Corresponding author at: Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, NE4 5PL E-mail address: [email protected] (L. Rochester).
https://doi.org/10.1016/j.gaitpost.2018.01.030 Received 22 December 2017; Received in revised form 22 January 2018; Accepted 25 January 2018 0966-6362/ © 2018 Elsevier B.V. All rights reserved.
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 1. Schematic detailing the obstacle dimensions (A) and the approach (A2 and A1) and crossing (lead and trail) steps extracted from the instrumented walkway (B).
randomised control trial (n = 282) evaluating the potential of two 6week treadmill training exercise programmes to improve gait and reduce falls in older adults [16,17]. Briefly, participants were recruited providing that they: reported at least two falls in the 6-months prior to participation; were aged 60–90 years; had adequate hearing and vision; and had no changes to their prescribed medications in the month prior to recruitment [17]. Common exclusion criteria included moderate-tosevere cognitive impairment, and any considerable visual or auditory deficit and orthopaedic or musculoskeletal disease. PD participants had a diagnosis of idiopathic PD according to the UK Parkinson’s Disease Brain Bank Criteria, mild to moderate symptoms (defined as Hoehn and Yahr stage II–III) and were taking anti-Parkinsonian medication (Deep brain stimulation was included). The PD group were tested whilst optimally medicated approximately 1-h post dopaminergic medications. Data presented here were collected at a single site pre-intervention and approved by the local NHS ethics committee (Ref: 12-NE-0249). Written informed consent was obtained according to the Declaration of Helsinki [18]. The Movement Disorder Society revised Unified Parkinson’s Disease Rating Scale [19] was administered by a trained examiner. Freezing of gait was quantified using the new Freezing of Gait Questionnaire (nFOG) [20]. The modified Falls Efficacy Scale (FES-I) was used to assess fear of falling [21]. The four square step test (FSST) [22] was chosen as a measure of dynamic mobility [23] and the Mini-Balance Evaluations Systems Test (Mini-BESTest) [24] which evaluates anticipatory and reactive postural control, sensory orientation and gait stability was used as a measure of general physical functioning. Each of the outcome measures have been shown to be reliable and valid for use with older populations and people with PD [20,23,25–28]. Basic
Similar to younger adults [8], increasing the height of an obstacle results in greater gait adaptations in mild to moderately severe PD, with one study reporting an increased step length and greater limb elevation when negotiating a taller obstacle [13]. Stepping over obstacles of greater depth (requiring adaptations to step length) occurs regularly during everyday life, such as walking over puddles or stepping over the gap when boarding a train. However, little is known about this type of obstacle crossing in older adults and people with PD, with previous studies manipulating obstacle height and not depth. This may be pertinent as step length decreases with age [14] and is further reduced in people with PD [15]. We were interested to explore the strategies made by older adults who have experienced previous falls for whom gait impairments are likely to be more pronounced. Therefore, the aim of this study was to describe gait adaptation in two groups of fallers: older adults and people with PD, when walking over obstacles of increased height and depth. We hypothesised that tall obstacles (increased height) would result in similar gait adaptations in people with PD and controls, but that a long obstacle (greater depth) would be challenging for people with PD and result in greater temporal spatial adaptations compared to controls.
2. Methodology 2.1. Participants A group of older adults with PD and without PD participated in this study which was a collaborative sub-study of the V-TIME (Virtual reality treadmill training to reduce falls and enhance mobility in the elderly) project. The V-TIME project was an international multi-centre 369
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Table 1 Demographic data. Control (n = 13)
PD (n = 20)
Group differences
Age (years) Height (m) Mass (kg) Sex (m/f; n) Education (years) MOCA (score/30) Bilateral visual acuity (LOGMAR) Bilateral contrast sensitivity (Mars CS) Fear of falling (FES-I) General physical function (Mini Best) Dynamic mobility (FSST)
76.4 [7.0] 1.65 [0.11] 73.8 [12.4] 4 m/9f 12.3 [3.9] 24 [5] 0.134 [0.12] 1.47 [0.16] 30.6 [10.6] 23.2 [5.6] 11.4 [7.7]
70.5 [7.7] 1.68 [0.09] 82.6 [16.1] 12 m/8f 13.3 [3.9] 28 [2] 0.003 [0.12] 1.53 [0.15] 36.8 [11.0] 23.0 [4.7] 9.6 [7.7]
p = .029 * p = .457 p = .084 p = .101 p = .484 p < .001 * p = .006 * p = .325 p = .119 p = .911 p = .525
Disease specific outcomes Hoehn & Yahr disease stage II (n) Hoehn & Yahr disease stage III (n) Disease duration (years) UPDRS III (Severity of motor symptoms; score/132) Freezing of gait (freezer/non-freezer; n)
– – – – –
15 5 6.5 (1, 37) 20.6 [7.9] 10/10
– – – – –
Temporal-spatial gait parameters ^ Step velocity (m/s) Step length (m) Step width (m) Step duration (msec) Single limb support (msec) Double limb support (msec)
0.91 [0.22] 0.54 [0.11] 0.11 [0.05] 609 [78] 396 [64] 817 [114]
0.83 [0.21] 0.51 [0.11] 0.11 [0.03] 627 [71] 391 [48] 859 [113]
p = .290 p = .019 * p = .943 p = .487 p = .834 p = .302
*denotes significant group difference p < .05. Disease duration was not normally distributed and is presented Median (minimum, maximum). Categorical data are presented as counts. All other data are presented Mean [SD]. ^ Statistical differences were computed using ANCOVA adjusting for age, global cognition (MOCA) and visual acuity.
ophthalmic measures of binocular visual acuity (LogMAR, Sussex-Vision, UK) and contrast sensitivity (Mars letter chart, Mars Perceptrix™, New York, USA) were obtained.
Heel) and the obstacle were also calculated.
2.2. Protocol
Sample size calculations from prior studies indicated that we required a minimum of 13 people in each group (d 1.2, α two-tailed < .05, β > 0.80; e.g. step velocity difference reported by Galna et al. [9]). Data were generally normally distributed and parametric statistics were used unless otherwise stated. Group differences in demographics were evaluated using independent t-tests with statistical significance accepted at p < .05. Group differences identified from demographic data were entered as covariates in subsequent analyses of group-based effects. General Linear Models were developed for each of the gait characteristics to evaluate the within (obstacle type and step) and between subject (group) effects. Change scores between steps (i.e. difference between A2 and A1 [A2-A1], A1 and Lead [A1-Ld], and Lead and Trail [Ld-Tr]) were calculated to investigate interaction effects. The Toe-Obstacle and Obstacle-Heel distances were the only outcomes that were not normally distributed and non-parametric statistics were used.
2.4. Statistical analysis
Participants began each trial with their eyes closed while the obstacle was placed on the walkway to prevent preparatory motor planning prior to the start of the trial [29]. Upon the word ‘go’, they opened their eyes and walked over a 7-m instrumented walkway (GAITRite®, Platinum model v.4.5, USA, 240 Hz). Participants walked under four experimental conditions: Level-ground walking and walking over a small, long and tall obstacle (see Fig. 1 for obstacle dimensions). Each obstacle was covered in a highly visible yellow poly-vinyl chloride material and was constructed of foam with a wooden base allowing it to compress in the event of contact whilst standing independently. Three trials were collected for each walking condition but only the first trial was selected for further analysis since this was considered to be the most natural response when participants were naïve to the environmental condition. Condition order was standardised with three level gait trials completed first followed by the obstacle crossing trials (small, long then tall). All participants wore a safety harness attached to an overhead gantry to ensure participant safety and prevent falls.
3. Results The PD participants were significantly younger, with better global cognition (MOCA) and visual acuity (p < .03) (Table 1). The groups were matched for other outcomes including: fear of falling; general physical function and dynamic mobility (p > .05). The PD group were of mild to moderate disease severity with half reporting freezing (nFOG score > 1). There were no FOG episodes observed and no obstacle contacts recorded for the first trial for either group or condition. Gait impairment was evident with 75% of the PD group (n = 15) and 62% of the control group (n = 8) walking at < 1.0m/s. Only one significant group difference was observed during level gait indicating that the PD group walked with a significantly reduced step length compared to controls (p = .019). One participant in the PD group could not negotiate the long obstacle condition unaided and one participant from the control group could not negotiate the tall obstacle unaided. Cases
2.3. Data analysis Footfall data obtained from the instrumented walkway were processed within the manufacturer’s software. Temporal-spatial gait characteristics (velocity, step length, step width, step duration, single and double limb support) were collated using Microsoft Access database 2007. Ensemble averages were compiled from step data for the level gait condition. A custom macro implemented within Visual Basic Applications for Microsoft® Excel extracted gait characteristics for each of the approach and crossing steps measured during the obstacle crossing conditions (depicted in Fig. 1). The distance between the toe of the trail limb (Toe-Obstacle) and the heel of the lead limb (Obstacle370
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Table 2 General Linear Model results for the effects of group and obstacle type. Gait characteristic
Between group effects ^ (Group, Group x Step, Group x Obstacle type)
Within group effects (Obstacle type, Step, Obstacle type x Step)
Step velocity
Group x Step (p = .026) No significant post hoc comparisons
Step length
Group (p = .043) Control > PD (p = .043) Group x Step (p = .004) Control Ld-Tr > PD Ld-Tr (p = .015) No significant main or interaction effects
Obstacle type (p < .001) Small > Tall (p = .002) Step (p = .005) A2-A1 < Ld-Tr (p = .001) A1-Ld > Ld-Tr (p = .002) Obstacle type x Step (p < .001) Long A1-Ld > Tall A1-Ld (p < .001) Long Ld-Tr > Tall Ld-Tr (p = .031) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld > Ld-Tr (p < .001) Step (p = .001) A2-A1 > A1-Ld (p = .012) A1-Ld > Ld-Tr (p = .006) Obstacle type x Step (p = .033) Small A1-Ld < Tall A1-Ld (p = .009) Obstacle type (p < .001) Tall > Small & Long (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld < Ld-Tr (p < .001) Obstacle type x Step (p < .001) Small A1-Ld < Long A1-Ld (p = .019) Small A1-Ld < Tall A1-Ld (p = .001) Long A1-Ld < Tall A1-Ld (p < .001) Obstacle type (p < .001) Tall > Small & Long (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A2-A1 < Ld-Tr (p < .001) A1-Ld > Ld-Tr (p < .001) Obstacle type x Step (p < .001) Small A1-Ld < Long A1-Ld (p = .016) Small A1-Ld < Tall A1-Ld (p < .001) Long A1-Ld < Tall A1-Ld (p < .001) Step (p < .001) A2-A1 < A1-Ld (p < .001) A1-Ld > Ld-Tr (p < .001)
Step width
Step duration
Group x Step (p = .043) PD A1-Ld > Control A1-Ld (p = .038)
Single limb support
Group x Step (p = .043) PD A1-Ld > Control A1-Ld (p = .040)
Double limb support
Group x Obstacle type (p = .035) No significant post hoc comparisons
Results from post hoc tests are italicised. Ld denotes lead step and Tr denote trail step. ^ Group differences are reported whilst controlling for age, MOCA and visual acuity as covariates.
stepping over and their heel (lead step) further from the tall obstacle compared to the small obstacle (Fig. 3).
were excluded pairwise within the General Linear Models leaving 19 people with PD and 12 controls. Group mean, median and dispersion values for the gait data collected during the obstacle crossing trials is included in the Supplementary material.
3.2. Gait adaptations made by people with Parkinson’s disease during obstacle crossing
3.1. Gait adaptations in response to obstacles type People with PD walked with a shorter step length compared to controls (Table 2 and Fig. 2; Main effect: p = .043) which was more pronounced during the lead crossing step (Group x Step interaction; p = .015). Group x Step interactions also showed that people with PD increased their step duration (p = .038) and single limb support duration (p = .040) more than controls for the lead crossing step (A1-Ld, p = .038). There were no statistically significant group differences in the distance between the foot and the obstacle (Fig. 3), although the PD group generally positioned the toe of their trail step further away from the obstacle than controls (∼5 cm) and placed the heel of their lead step closer to the far edge of the obstacle compared to controls (∼3 cm).
All participants walked with a slower velocity, longer step duration and single limb support duration when negotiating the tall obstacle compared to the small and long obstacles (Main effect; p < .001; Table 2 and Fig. 2) . Main effects for Step showed that participants completed the lead crossing step (A1-Ld) more quickly (p = .002) and with a longer step duration (p < .001), double limb support (p < .001), step length, (p < .001) and increased step width (p = .006) compared to the other steps. Obstacle type x Step interactions were identified for step velocity, step width, duration and single limb support. Participants slowed down the most and widened their step width during the lead limb crossing step when walking over the tall obstacle (p < .009). The tall obstacle required the longest single limb support duration, followed by the long and small obstacle (tall > long > small, p < .016). Single limb support time changed significantly with each step for all participants, with the longest single limb support duration observed at the lead step (A1Ld, p < .001) for all obstacle types. There was a significant difference in the Toe-Obstacle and Obstacle-Heel distances such that both groups positioned their toe (trail step) closer to the tall obstacle prior to
4. Discussion The results of the present study confirm that motor symptoms due to PD are evident during obstacle crossing, even when compared to a control group who have previously fallen, with differences highlighted particularly during the more challenging crossing steps. Taller obstacles evoked the largest modifications, indicating they challenged the system 371
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 2. Group mean data for each of the obstacle conditions and steps. a denotes a significant Group x Step interaction and b denotes a significant Obstacle type x Step interaction.
372
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
Fig. 3. Distance from the toe and heel to the obstacle when negotiating the small, long and tall obstacles.
long obstacle may have acted as an external visual cue prompting an increased step velocity and step length during the lead crossing step. We know that the neurodegeneration associated with PD pathology affects basal ganglia function resulting in difficulty initiating and selecting appropriate motor outputs. To overcome this, physiotherapists train patients with PD to use visual cues as they have been shown to be effective in reconfiguring the spatial scaling of movement [34]. Whilst not of statistical significance, it is clinically of interest that one participant with PD could not physically negotiate the long obstacle unaided. Further investigation revealed that this participant was 1.65 m in height, had the shortest step length of the cohort during level gait (0.24m) and the most severe motor symptoms (UPDRSIII 43/132). It appears that with appropriate motor planning and foot placement before and after the obstacle, the long obstacle did not overly disturb the natural cyclical motion of gait compared to the tall obstacle which we suggest likely caused a greater vertical displacement of the centre of mass. However, the long obstacle still required that both groups significantly increased their step length during the lead crossing step compared to that observed during gait (PD > 31% increase, Control > 24% increase, p < .001). There are some limitations of the present study which must be acknowledged when interpreting the results. Participants were aware of the purpose of the study prior to assessment and perhaps greater group differences may occur in the community when obstacles are not always anticipated. We would expect that obstacles encountered in unfamiliar or changeable environments likely pose greater threat and require quicker temporal-spatial adaptations. Lastly, smaller obstacles encountered in a natural setting may pose added threat than observed in the present study if they are unexpected and perhaps less visually prominent. It is also possible that in environments outside of the laboratory, older adults with a history and/or fear of falling may opt to avoid the obstacle as a safer motor strategy. This was evident in the present study with two participants unable to compete the long and tall obstacle conditions unaided. While the groups were well matched for many functional outcomes, there were distinct differences between the groups in age, global cognition and visual acuity which we accordingly controlled for in the present analyses. Further complementary work will characterise foot trajectories to explore how this is modified with respect to obstacle type to further understand the level of trip risk imposed compared to a non-falling cohort. Understanding the temporal spatial adaptations made during the approach and crossing of different types of obstacles will be informative for guiding falls prevention and physiotherapy management. This study has identified important differences in obstacle crossing performance in
most, as they were associated with participants standing longer in single limb support and eliciting a wider step. The tall obstacle prompted the largest change in the temporal gait parameters. This is in agreement with previous research in young adults [8]. We found that both groups took a wider step when crossing the tall obstacle compared to the small and long obstacles. This has been observed previously in people with PD [9,11] and likely relates to the maintenance of mediolateral balance [30]. Corroborative evidence from studies investigating older adults with and without balance impairment and another in people with PD showed that the horizontal displacement of the centre of mass was significantly larger in those with balance problems and those with PD [12,31]. Moreover, electing a wider step was a strategy employed by older adults when visual information about the proximity of an obstacle was reduced [32]. The findings in the present study concur that older adults with a history of falls elect to use the same strategy as people with PD with a history of falls when crossing tall obstacles. Given that the groups were of similar height (p = .457), the variation in obstacle height ought to have caused a comparable disturbance thus posing similar biomechanical demands for all participants. We observed no differences in step width across the obstacle conditions due to group, which suggests that increasing the base of support when crossing a tall obstacle may be a common strategy employed by both faller groups regardless of PD pathology. The tall obstacle presented an added challenge for the PD group in terms of balance control. During the lead limb crossing step (A1-Ld), the swing limb moves anteriorly from behind the centre of mass to in front of the centre of mass once over the obstacle. This single limb support phase, during which there is a redistribution of weight transfer, may be considered the most challenging step when the body hovers over the obstacle. The findings in the present study show that people with PD crossed the tall obstacle with a longer step duration which was mediated by a longer single limb support time. This may place people with PD at an increased risk of tripping or loosing balance when walking over obstacles, particularly tall obstacles. Although the underlying mechanisms behind this increased single limb support time is unclear, we speculate that interventions that train balance during single limb support and strengthen the lower limbs may improve obstacle crossing performance and safety. Evidence suggests that improving obstacle crossing performance in older adults may be achieved through task-specific interventions targeting obstacle crossing gait using a variety of obstacle types, and reassuringly these improvements were associated with a reduced risk of falls [33]. Surprisingly, there were few significant alterations observed in the PD group when negotiating the long obstacle and we suggest that the 373
Gait & Posture 61 (2018) 368–374
L. Alcock et al.
people with PD. Considering that experiencing a previous fall remains a common feature underpinning an increased risk of future falls in older adults with PD [35], there is a paucity of evidence documenting the temporal-spatial gait alterations made by cohorts composed of fallers exclusively, particularly during a complex and risky task such as obstacle crossing. This study has taken a novel approach to understanding obstacle crossing performance by including a group of older adult fallers matched for fear of falling, general physical function and dynamic mobility as controls and has identified gait adaptations with respect to obstacle type and faller group. Given the exploratory nature of this study we elected to not correct for multiple comparisons and this may be considered a limitation. Further work is consequently required to characterise gait alterations in larger cohorts with obstacles of varying height and depth.
factors and association with falling, Clin. Neurol. Neurosurg. 161 (2017) 51–55. [7] I. Maidan, S. Eyal, I. Kurz, N. Geffen, E. Gazit, L. Ravid, et al., Age-associated changes in obstacle negotiation strategies: does size and timing matter? Gait Posture 59 (2018) 242–247. [8] S. Yamaji, S. Demura, H. Sugiura, Influence of degraded visual acuity from lightscattering goggles on obstacle gait, Health (N. Y.) 3 (2011) 99–105. [9] B. Galna, A.T. Murphy, M.E. Morris, Obstacle crossing in people with Parkinson’s disease: foot clearance and spatiotemporal deficits, Hum. Mov. Sci. 29 (2010) 843–852. [10] E.L. Stegemöller, T.A. Buckley, C. Pitsikoulis, E. Barthelemy, R. Roemmich, C.J. Hass, Postural instability and gait impairment during obstacle crossing in Parkinson’s disease, Arch. Phys. Med. Rehabil. 93 (2012) 703–709. [11] R. Vitório, E. Lirani-Silva, A.M. Baptista, F.A. Barbieri, P.C.R. dos Santos, C. Teixeira-Arroyo, et al., Disease severity affects obstacle crossing in people with Parkinson’s disease, Gait Posture 40 (2014) 266–269. [12] B. Galna, A.T. Murphy, M.E. Morris, Obstacle crossing in Parkinson’s disease: mediolateral sway of the centre of mass during level-ground walking and obstacle crossing, Gait Posture 38 (2013) 790–794. [13] R. Vitório, F. Pieruccini-Faria, F. Stella, S. Gobbi, L.T.B. Gobbi, Effects of obstacle height on obstacle crossing in mild Parkinson’s disease, Gait Posture 31 (2010) 143–146. [14] H.B. Menz, S.R. Lord, R.C. Fitzpatrick, Age-related differences in walking stability, Age Ageing 32 (2003) 137–142. [15] B. Galna, S. Lord, D.J. Burn, L. Rochester, Progression of gait dysfunction in incident Parkinson’s disease: impact of medication and phenotype, Mov. Disord. 30 (2015) 359–367. [16] A. Mirelman, L. Rochester, M. Reelick, F. Nieuwhof, E. Pelosin, G. Abbruzzese, et al., V-TIME: a treadmill training program augmented by virtual reality to decrease fall risk in older adults: study design of a randomized controlled trial, BMC Neurol. 13 (2013) 15. [17] A. Mirelman, L. Rochester, I. Maidan, S. Del Din, L. Alcock, F. Nieuwhof, et al., Addition of a non-immersive virtual reality component to treadmill training to reduce fall risk in older adults (V-TIME): a randomised controlled trial, Lancet 388 (2016) 1170–1182. [18] World Medical Association, Ethical principles for medical research involving human subjects, Bull. World Health Organ. 79 (2001) 373. [19] C.G. Goetz, B.C. Tilley, S.R. Shaftman, G.T. Stebbins, S. Fahn, P. Martinez-Martin, et al., Movement disorder society-sponsored revision of the unified parkinson’s disease rating scale (MDS-UPDRS): scale presentation and clinimetric testing results, Mov. Disord. 23 (2008) 2129–2170. [20] A. Nieuwboer, L. Rochester, T. Herman, W. Vandenberghe, G.E. Emil, T. Thomaes, et al., Reliability of the new freezing of gait questionnaire: agreement between patients with Parkinson’s disease and their carers, Gait Posture 30 (2009) 459–463. [21] L. Yardley, N. Beyer, K. Hauer, G. Kempen, C. Piot-Ziegler, C. Todd, Development and initial validation of the falls efficacy scale-international (FES-I), Age Ageing 34 (2005) 614–619. [22] W. Dite, V.A. Temple, A clinical test of stepping and change of direction to identify multiple falling older adults, Arch. Phys. Med. Rehabil. 83 (2002) 1566–1571. [23] R.P. Duncan, G.M. Earhart, Four square step test performance in people with Parkinson disease, J. Neurol. Phys. Ther. 37 (2013) 2–8. [24] F. Franchignoni, F. Horak, M. Godi, A. Nardone, A. Giordano, Using psychometric techniques to improve the Balance Evaluation Systems Test: the mini-BESTest, J. Rehabil. Med. 42 (2010) 323–331. [25] N. Giladi, J. Tal, T. Azulay, O. Rascol, D.J. Brooks, E. Melamed, et al., Validation of the freezing of gait questionnaire in patients with Parkinson’s disease, Mov. Disord. 24 (2009) 655–661. [26] F.B. Horak, D.M. Wrisley, J. Frank, The balance evaluation systems test (BESTest) to differentiate balance deficits, Phys. Ther. 89 (2009) 484–498. [27] A.L. Leddy, B.E. Crowner, G.M. Earhart, Functional gait assessment and balance evaluation system test: reliability, validity, sensitivity, and specificity for identifying individuals with Parkinson disease who fall, Phys. Ther. 91 (2011) 102–113. [28] N. Dewan, J.C. MacDermid, Fall efficacy scale-international (FES-I), J. Physiother. 60 (2014) 60. [29] A.E. Patla, J.N. Vickers, Where and when do we look as we approach and step over an obstacle in the travel path, Neuroreport 8 (1997) 3661–3665. [30] P.M. McAndrew Young, J.B. McAndrew Young, Voluntary changes in step width and step length during human walking affect dynamic margins of stability, Gait Posture 36 (2012) 219–224. [31] L.-S. Chou, K.R. Kaufman, M.E. Hahn, R.H. Brey, Medio-lateral motion of the center of mass during obstacle crossing distinguishes elderly individuals with imbalance, Gait Posture 18 (2003) 125–133. [32] S. Kunimune, S. Okada, The effects of object height and visual information on the control of obstacle crossing during locomotion in healthy older adults, Gait Posture 55 (2017) 126–130. [33] Y.-Y. Liao, Y.-R. Yang, Y.-R. Wu, R.-Y. Wang, Factors influencing obstacle crossing performance in patients with Parkinson’s Disease, PLoS One 9 (2014) e84245. [34] G.N. Lewis, W.D. Byblow, S.E. Walt, Stride length regulation in Parkinson’s disease: the use of extrinsic, visual cues, Brain 123 (2000) 2077–2090. [35] S.S. Paul, C.G. Canning, C. Sherrington, S.R. Lord, J.C. Close, V.S. Fung, Three simple clinical tests to accurately predict falls in people with Parkinson’s disease, Mov. Disord. 28 (2013) 655–662.
5. Conclusions Obstacles of greater height (tall obstacles), rather than depth (long obstacles), prompt the largest alterations in temporal-spatial gait and as such pose increased challenge to older adults with a history of falls. Increasing step width during the lead crossing step to increase stability was an adaptation common to both falling groups when crossing the tall obstacle, regardless of pathology. An increased time spent in single limb support was observed in people with PD when crossing the tall obstacle which may challenge balance more so than long or small obstacles. Conflict of interest statement The authors have no conflicts of interest to declare. Acknowledgements The authors would like to thank Aodhán Hickey for his help with data collection. This research was funded by the V-TIME project which is a European Union 7th Framework Programme (FP7) under the Health theme (FP7–278169, HEALTH.2011.2.2.2-1). This research was supported by the NIHR Newcastle Biomedical Research Unit based at The Newcastle Hospitals NHS Foundation Trust and Newcastle University. The research was also supported by NIHR Newcastle CRF Infrastructure funding. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gaitpost.2018.01.030. References [1] R.L. Naussbaum, C.E. Ellis, Alzheimer's disease and Parkinson's disease, N. Engl. J. Med. 348 (2003) 1356–1364. [2] H. Stolze, S. Klebe, C. Zechlin, C. Baecker, L. Friege, G. Deuschl, Falls in frequent neurological diseases, J. Neurol. 251 (2004) 79–84. [3] A. Ashburn, E. Stack, R.M. Pickering, C.D. Ward, A community-dwelling sample of people with Parkinson’s disease: characteristics of fallers and non-fallers, Age Ageing 30 (2001) 47–52. [4] K. Mactier, S. Lord, A. Godfrey, D. Burn, L. Rochester, The relationship between real world ambulatory activity and falls in incident Parkinson’s disease: influence of classification scheme, Parkinsonism Relat. Disord. 21 (2014) 236–242. [5] W.P. Berg, H.M. Alessio, E.M. Mills, C. Tong, Circumstances and consequences of falls in independent community-dwelling older adults, Age Ageing 26 (1997) 261–268. [6] T. Gazibara, D. Kisic Tepavcevic, M. Svetel, A. Tomic, I. Stankovic, V.S. Kostic, et al., Near-falls in people with Parkinson’s disease: circumstances, contributing
374