Abstract
Background:
Walking with user-driven treadmill control is believed to be more like overground walking than fixed-speed treadmill walking. Walking speed and ground reaction forces differ between overground and fixed-speed treadmill walking, but not between overground and user-driven treadmill walking in healthy and post-stroke subjects. However, studies assessing spatiotemporal gait parameters during user-driven treadmill walking are limited. This information may help confirm that user-driven treadmill walking is more like overground walking than fixed-speed treadmill walking, as well as inform the development of post-stroke gait rehabilitation programs.
Research Question:
How do spatiotemporal gait parameters for individuals post-stroke differ between fixed-speed and user-driven treadmill walking?
Methods:
Eighteen subjects (10M, 8F; 62 ± 12 years; 1.73 ± 0.12 m; 84.9 ± 12.9 kg; 40 ± 30 months post-stroke) with chronic post-stroke hemiparesis participated in this study. Participants walked on an instrumented treadmill in its fixed-speed and user-driven modes at their self-selected and fastest comfortable walking speeds. Subjects wore retroreflective markers for motion capture. Shapiro-Wilk tests were used to assess for normality and one-way repeated measures ANOVAs were used to compare between conditions with α = 0.05. Bonferroni corrections were used for multiple comparisons.
Results:
Step width was significantly smaller with user-driven control (13.7 cm, 95% CI: [0.131, 0.145]) than fixed-speed control (16.8 cm, 95% CI:[0.160, 0.174]), while step length and step time did not differ across treadmill conditions. Step length and step time differed between self-selected and fast walking speeds, but not treadmill control conditions.
Significance:
The results of this study show that user-driven treadmill control encourages healthy gait biomechanics and a greater sense of stability in post-stroke subjects. Individuals post-stroke walked with smaller step width with user-driven treadmill control, which has been associated with increased balance. Post-stroke gait rehabilitation may benefit from programs with user-driven treadmill training paradigms to improve mobility following stroke.
Keywords: User-driven treadmill control, Self-paced treadmill, Stroke, Spatiotemporal parameters, Step width
INTRODUCTION
Each year, 795,000 people experience a stroke, which often impairs gait and balance [1–4]. Spatiotemporal parameters differ between healthy and post-stroke subjects, and step width is adjusted to modulate stability [3]. Post-stroke gait rehabilitation typically uses a fixed-speed treadmill (FSTM) because it is spatially and financially efficient [5]. However, walking speed [6,7] is slower with FSTMs than overground, limiting training because slower speeds increase the risk of falls and mortality [8].
Custom user-driven treadmill (UDTM) control changes the treadmill speed in real time by responding to the user’s step length, propulsion, and center-of-mass position, leading to faster self-selected (SS) walking speeds in healthy [6] and post-stroke [7] subjects compared to FSTM speeds. Kinematic gait variability that is inherent to healthy walking is reduced during FSTM walking in unimpaired individuals, but restored with UDTM walking, suggesting that UDTM conditions may provide an overground-like environment [9–11].
The purpose of this study was to examine the effects of UDTM control on spatiotemporal gait measures in post-stroke subjects compared to FSTM control at matched speeds. We hypothesized that step length would increase and step time would decrease with the increased speeds associated with UDTM control. Step width was expected to be similar between UDTM and FSTM conditions because the UDTM controller does not account for step width.
METHODS
Participants
Twenty individuals with chronic post-stroke hemiparesis were recruited through the University of Delaware’s Stroke Registry database. Two were excluded from analysis for failure to complete all trials due to time constraints. Power analysis determined that 18 participants were sufficient for distinguishing a mean step width difference of 5 ± 4 cm [3] with a power of 0.8. Participant characteristics are shown in Table 1, and all were independent ambulators [7]. Inclusion and exclusion criteria were the same as Ray et al. [7]. The study was approved by the University of Delaware’s Institutional Review Board and participants signed an informed consent.
Table 1:
Participant characteristics. Data are presented as mean ± standard deviation.
| Characteristic | Value |
|---|---|
| Sex | 10 M | 8 F |
| Age | 62 ± 12 years |
| Time since stroke | 40 ± 30 months |
| Height | 1.73 ± 0.12 m |
| Mass | 84.90 ± 12.92 kg |
Experimental Protocol
Subjects were outfitted with 42 retroreflective markers for motion capture with an 8-camera system (Motion Analysis Corp., Santa Rosa, CA, USA) [6,7]. For this analysis, only the heel markers were required. Kinematic data was sampled at 100 Hz.
Subjects performed a series of walking trials on an instrumented split-belt treadmill (Bertec Corp., Worthington, OH, USA) with fixed-speed and user-driven control [6]. Kinetic data was sampled at 2000 Hz from the treadmill and handrails, which the subjects could use with a “light touch.” Participants wore a harness to prevent falls.
After a maximum of five minutes of familiarization with FSTM and UDTM conditions, subjects performed six trials, each for one minute, at their steady-state (± 0.2 m/s) SS and fastest comfortable (Fast) walking speeds [7]. The first four trials were randomized:
| 1. FSTM at SS Speed | 2. FSTM at Fast Speed |
| 3. UDTM at SS Speed | 4. UDTM at Fast Speed |
Then, two speed-matching conditions were performed in a random order, where the UDTM speed was imposed upon the FSTM [7]:
| 5. FSTM at UDTM SS | 6. FSTM at UDTM Fast |
Analysis
Kinetic and kinematic data were filtered with a fourth order low-pass Butterworth filter at 30 Hz and 6 Hz, respectively. Calculations were performed using Visual 3D (C-Motion Inc., Germantown, MD, USA) and MATLAB (MathWorks, Natick, MA, USA). Step width and paretic and non-paretic step length and step time were averaged over the trial for each condition. Step width was the medial-lateral distance between the heel markers at subsequent heel strikes. Step length was the anterior-posterior distance between the heel markers at subsequent heel strikes. Step time was the time between contralateral and ipsilateral heel strikes, determined by the kinetic data. Step length and step time asymmetry were calculated (Equation 1) [3].
| (1) |
The Shapiro-Wilk test assessed normality. One-way repeated measures ANOVAs compared between conditions, with Tukey post-hoc analysis performed as needed (α = 0.05) [7]. Bonferroni corrections compensated for multiple comparisons.
RESULTS
Step width was significantly narrower on the UDTM (13.7 cm) than the FSTM (16.8 cm), even at matched speeds (p < 0.0001, Figure 1A). There were no significant differences in step width between FSTM conditions.
Figure 1:
Mean ± standard deviation for variables of interest. (A) Mean step width was significantly narrower in both UDTM control conditions than in all four FSTM control conditions. The four FSTM conditions did not differ from each other, and the two UDTM conditions did not differ from each other. (B) Mean step length did not differ with treadmill control, only with walking speed. The three SS conditions had shorter step lengths than the three Fast conditions. Paretic and non-paretic step length did not differ from each other for all conditions. (C) Mean step time did not differ between UDTM control and FSTM control but did between SS and Fast speeds. Both paretic and non-paretic step time were shorter in the three Fast conditions than the three SS conditions. Paretic and non-paretic step time differed from each other for all conditions.
Both paretic and non-paretic step length differed between SS and Fast speeds but not between FSTM or UDTM conditions (p < 0.05, Figure 1B). Paretic and non-paretic step length did not differ from each other across any conditions (p > 0.6). Step length asymmetry differed in 4 of 15 comparisons (Table 2).
Table 2:
Step length asymmetry and step time asymmetry data presented as group mean ± standard deviation. Step length asymmetry differed in 4 of 15 comparisons: FSTM at UDTMSS vs. UDTMFast (p = 0.001); FSTMSS vs. UDTMFast (p = 0.007); FSTM at UDTMSS vs FSTM at UDTMFast (p = 0.025); and FSTM at UDTMSS vs FSTMFast (p = 0.036). Step time asymmetry did not differ in any comparisons.
| Condition | Step Length Asymmetry (%) | Step Time Asymmetry (%) |
|---|---|---|
| FSTMSS | 1.385 ± 4.184 | 12.291 ± 2.761 |
| UDTMSS | −0.257 ± 2.244 | 10.715 ± 2.870 |
| FSTMFast | −1.540 ± 2.009 | 11.694 ± 3.122 |
| UDTMFast | −2.899 ± 4.413 | 10.710 ± 2.645 |
| FSTM at UDTMSS | 2.088 ± 2.612 | 11.574 ± 1.857 |
| FSTM at UDTMFast | −1.698 ± 3.335 | 11.735 ± 2.393 |
Paretic and non-paretic step time differed with walking speed, but not treadmill control (p < 0.001, Figure 1C). Paretic and non-paretic step time differed from each other for all conditions (p < 0.05). Step time asymmetry did not differ across any conditions (p > 0.5, Table 2).
DISCUSSION
Step width was consistent across FSTM trials, suggesting that walking speed did not influence step width. However, step width was significantly smaller in UDTM conditions than FSTM conditions, indicating that the control algorithm impacted step width. The FSTM step width (16.8 cm) was similar to post-stroke FSTM values (17.3 cm) in the literature [3], while unimpaired step width was narrower overground (12 cm) [12] and with FSTM control (11.5 cm) [3]. Post-stroke step width in all conditions was larger than reported unimpaired step width, likely due to a decreased sense of stability. Post-stroke UDTM step width (13.7 cm) showed trends toward unimpaired overground and FSTM step width, suggesting that the UDTM encouraged post-stroke participants to walk with narrower steps. Narrower steps may show an increased sense of stability with UDTM control, which is vital after stroke [1,3,13]. By promoting narrower step width in post-stroke participants that is like unimpaired values, UDTM control may improve the gait rehabilitation environment, which could improve rehabilitation outcomes and increase mobility.
Step length differed between walking speeds, but not between treadmill conditions. Step length was not expected to differ for the speed-matched conditions because step length is related to walking speed. However, the controller showed no effect on step length regardless of walking speed. Subject-specific effects of treadmill control on walking speed have been seen [7] and may exist with step length. Step length asymmetry differed in a few comparisons, potentially an effect of walking speed, but those comparisons spanned differing treadmill conditions and walking speeds so a conclusion cannot be drawn from these results.
Step time also differed with walking speed but not treadmill control, indicating no effect of UDTM control. Step time was not expected to differ for the speed-matched conditions, as step time is related to walking speed. Although temporal asymmetry is often targeted by rehabilitation programs [14,15], UDTM control may not show any benefits over FSTM control for addressing temporal asymmetry.
Limitations
The primary limitation of this study is the heterogeneity of stroke survivors. A more severe stroke may increase balance impairments, requiring larger step width and reducing the likelihood of decreasing step width in UDTM conditions. The participants were high-functioning to ensure they could complete all trials, but individuals with more severe hemiparesis may have more potential for improvement and could benefit from UDTM-based training.
Conclusion
This research demonstrates the potential benefits of UDTM-based gait training. UDTM conditions maintain the increased repetitions and controlled environment of FSTMs [5] while encouraging narrower step width to promote stability for more efficient training. Rehabilitation paradigms may consider adopting UDTM control to improve post-stroke gait rehabilitation.
Highlights.
This study used novel user-driven treadmill control with post-stroke participants.
Spatiotemporal variables (step width, step length, and step time) were calculated.
Step width was narrower during user-driven walking than fixed-speed walking.
Step length and step time were unaffected by user-driven control.
Narrower step width may indicate a greater sense of stability after stroke.
ACKNOWLEDGMENTS
This work was funded by NIH P30 103333 and NIH U54-GM104941, University of Delaware Helwig Mechanical Engineering Fellowship & University of Delaware Dissertation Fellowship. The authors would like to thank Brian Knarr, Margaret French, Jessica Galgiani, and Ryan Wilke.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
CONFLICT OF INTEREST STATEMENT
The authors have no conflicts of interest to report.
References
- [1].Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, Chamberlain AM, Chang AR, Cheng S, Das SR, Delling FN, Djousse L, Elkind MSV, Ferguson JF, Fornage M, Jordan LC, Khan SS, Kissela BM, Knutson KL, Kwan TW, Lackland DT, Lewis TT, Lichtman JH, Longenecker CT, Loop MS, Lutsey PL, Martin SS, Matsushita K, Moran AE, Mussolino ME, O’Flaherty M, Pandey A, Perak AM, Rosamond WD, Roth GA, Sampson UKA, Satou GM, Schroeder EB, Shah SH, Spartano NL, Stokes A, Tirschwell DL, Tsao CW, Turakhia MP, VanWagner LB, Wilkins JT, Wong SS, Virani SS, Heart Disease and Stroke Statistics-2019 Update: A Report From the American Heart Association, Circulation. 139 (2019) e56–e528. 10.1161/CIR.0000000000000659. [DOI] [PubMed] [Google Scholar]
- [2].Mulroy S, Gronley JA, Weiss W, Newsam C, Perry J, Use of cluster analysis for gait pattern classification of patients in the early and late recovery phases following stroke, Gait Posture. 18 (2003) 114–125. 10.1016/S0966-6362(02)00165-0. [DOI] [PubMed] [Google Scholar]
- [3].Chen G, Patten C, Kothari DH, Zajac FE, Gait differences between individuals with post-stroke hemiparesis and non-disabled controls at matched speeds, Gait Posture. 22 (2005) 51–56. 10.1016/j.gaitpost.2004.06.009. [DOI] [PubMed] [Google Scholar]
- [4].Patterson KK, Mansfield A, Biasin L, Brunton K, Inness EL, McIlroy WE, Longitudinal changes in poststroke spatiotemporal gait asymmetry over inpatient rehabilitation, Neurorehabil. Neural Repair 29 (2015) 153–162. 10.1177/1545968314533614. [DOI] [PubMed] [Google Scholar]
- [5].Langhorne P, Coupar F, Pollock A, Motor recovery after stroke: a systematic review, Lancet Neurol. 8 (2009) 741–754. 10.1016/S1474-4422(09)70150-4. [DOI] [PubMed] [Google Scholar]
- [6].Ray NT, Knarr BA, Higginson JS, Walking speed changes in response to novel user-driven treadmill control, J. Biomech 78 (2018) 143–149. 10.1016/j.jbiomech.2018.07.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Ray NT, Reisman DS, Higginson JS, Walking speed changes in response to user-driven treadmill control after stroke, J. Biomech 101 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Abellan Van Kan G, Rolland Y, Andrieu S, Bauer J, Beauchet O, Bonnefoy M, Cesari M, Donini LM, Gillette-Guyonnet S, Inzitari M, Nourhashemi F, Onder G, Ritz P, Salva A, Visser M, Vellas B, Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) task force, J. Nutr. Heal. Aging 13 (2009) 881–889. 10.1007/s12603-009-0246-z. [DOI] [PubMed] [Google Scholar]
- [9].Stergiou N, Decker LM, Human movement variability, nonlinear dynamics, and pathology: Is there a connection?, Hum. Mov. Sci 30 (2011) 869–888. 10.1016/j.humov.2011.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Hollman JH, Watkins MK, Imhoff AC, Braun CE, Akervik KA, Ness DK, A comparison of variability in spatiotemporal gait parameters between treadmill and overground walking conditions, Gait Posture. 43 (2016) 204–209. 10.1016/j.gaitpost.2015.09.024. [DOI] [PubMed] [Google Scholar]
- [11].Kempski KM, Ray NT, Knarr BA, Higginson JS, Dynamic structure of variability in joint angles and center of mass position during user-driven treadmill walking, Gait Posture. 71 (2019) 241–244. 10.1016/j.gaitpost.2019.04.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Parvataneni K, Ploeg L, Olney SJ, Brouwer B, Kinematic, kinetic and metabolic parameters of treadmill versus overground walking in healthy older adults, Clin. Biomech 24 (2009) 95–100. 10.1016/j.clinbiomech.2008.07.002. [DOI] [PubMed] [Google Scholar]
- [13].Kao PC, Dingwell JB, Higginson JS, Binder-Macleod SA, Dynamic instability during post-stroke hemiparetic walking, Gait Posture. 40 (2014) 457–463. 10.1016/j.gaitpost.2014.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Ochoa J, Sternad D, Hogan N, Treadmill vs. Overground walking: Different response to physical interaction, J. Neurophysiol 118 (2017) 2089–2102. 10.1152/jn.00176.2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Patterson KK, Parafianowicz I, Danells CJ, Closson V, Verrier MC, Staines WR, Black SE, McIlroy WE, Gait Asymmetry in Community-Ambulating Stroke Survivors, Arch. Phys. Med. Rehabil 89 (2008) 304–310. 10.1016/j.apmr.2007.08.142. [DOI] [PubMed] [Google Scholar]