Table 4.
An overview of the literature focusing on gait parameters and measurement devices for post-stroke gait studies considered in this review.
| References | Gait parameters | Sample | Time post-stroke | Measurement/Protocol | Observation |
|---|---|---|---|---|---|
| Moseley et al. (72) | Segmented kinematics | n/a | n/a | n/a | Decreased peak hip extension in the late stance phase; Decreased peak lateral pelvic displacement in stance phase; Increased peak lateral pelvic displacement in stance phase; Decreased knee flexion (or knee hyperextension) in stance phase; Increased knee flexion in stance phase; Decreased ankle plantarflexion at toe-off. |
| Moore et al. (73) | Segmented kinematics | n/a | n/a | n/a | Decreased peak hip flexion and ankle dorsiflexion in swing phase; Reduction in the peak knee flexion in early swing phase; Decreased knee extension prior to heel strike; |
| Nickel (74) | Gait velocity, gait cycle time, cadence, stride length, total double support time, single support time, duration of stance phase, duration of swing phase | 49 stroke patients and 24 controls (controls had either transient ischemic episodes or asymptomatic carotid stenosis, symmetrical gait without walking support); time since stroke | avg 43.4 (range 0.5 to 336) months | Portable stride analyser, an insole system with compression foot switches (B& L Engineering, Santa Fe Springs, CA). | Cadence and velocity improved over time; Asymmetric patterns did not change over time; Age-matched controls in this study showed abnormal gait behavior compared to normal subjects. |
| Olney et al. (75) | Spatiotemporal, joint kinematics, moments, mechanical work and power | 31 hemiplegic stroke patients | avg 11.4 (range 2.0 to 88.0) months | 2D motion capture system (LoCam 51 camera); 3 trials; | Use of principal component analysis (PCA) for clustering of variables. |
| Silver et al. (76) | Walking speed, cadence, gait cycle symmetry (intralimb stance-swing ratio, interlimb stance duration ratio, interlimb swing ratio, overall stance-swing ratio) | 5 post-ischemic stroke patients (mild to moderate gait asymmetries due to residual hemiparesis) | 26 ± 4.6 (range 9 to 70) months | Videotape (Peak Motus Video Analysis system); modified Get-Up and Go task. | Improvements in walking speed and cadence, reduction in time required to complete the task; Sophisticated kinematics and kinetics analysis required to draw further results. |
| Woolley (77) | Distance and temporal parameters, joint kinematics, kinetics, mechanical power, energy expenditure, electromyography | n/a | n/a | n/a | Many gait deviations in the hemiplegic patients may be related to reduced walking velocity. |
| Hesse (78) | Stance and swing time symmetry, ground reaction forces, muscle activity profile, cardiovascular fitness | Hemiparetic subjects | n/a | 10 meter test most commonly used; 2 walking trials | 10-meter test and 6 min test are highly recommended to derive basic gait parameters; Abnormal muscle activity observed in stroke population; trajectory of vertical forces and center of pressure varies between controls and post-stroke patients; appearance of stance and swing time asymmetry. |
| Hsu et al. (20) | Gait velocity, step length asymmetry ratio, single support time asymmetry ratio | 26 stroke patients (those with limited lower-body joint range of motion, joint pain, and history of unstable medical conditions, neurological, and/or musculoskeletal issues were excluded) | avg 10.3 (range 1 to 43) months | GaitMatII (EQ Inc., Plymouth Meeting, PA) (3.8 m); Cybex 6000 isokinetic dynamometer (Cybex International Inc., Medway, MA) to measure isokinetic muscle strength; 6 trials per speed condition; comfortable- and fast-speed | The weakness of the affected hip flexors and knee extensors contribute to a decrease in gait velocity; The spasticity of the affected ankle plantarflexors causes asymmetry. |
| Patterson et al. (79) | Stance time, swing time, double support time, intra-limb ratio of swing-stance time, step length, spatiotemporal symmetry | 161 stroke patients and 81 age-matched healthy subjects | avg 23.7 (SD 32.1) months | GAITRite (10 m); 3 trials | Ratio equation can be used for standardization due to its clinical utility; Swing time, stance time, and step length are the most useful gait parameters |
| Patterson et al. (80) | Velocity, spatiotemporal symmetry | 171 stroke patients data (first-ever unilateral stroke; hemorrhagic or ischemic) | avg 23.3 (SD 31.1) months | GAITRite mat (CIR Systems Inc., New Jersey, USA); 3 trials; preferred/comfortable speed | Swing time, stance time, and step length asymmetries may progress in the long term post-stroke stages; In terms of gait velocity and neurological and motor deficit, no difference is seen across the stages. |
| Laudanski (81) | joint angles of hip, knee, and ankle | 10 chronic hemiparetic stroke patients and 10 healthy controls | 6.5 ± 5.4 years | 7 IMU sensors (Xsens Technology B.V., Netherlands), placed at midthigh, midshank, midfoot, and pelvis; Optotrak 3020 system (Northern Digital Inc., Ontario, Canada) for validation; force plates (AMTI, Newton, MA); 3 trials; self-selected speed | IMU-based systems are suitable for lower limb major joint angle estimation of healthy subjects and range of motion estimation of stroke patients. Additional calibration techniques are required for the application in stroke population. |
| Yang et al. (82) | Walking speed, temporal symmetry (stance ratio, swing ratio, swing-stance ratio, overall symmetry ratio) | 13 stroke patients (with unilateral lower limb weakness; able to walk independently; and could follow instructions) | 23.4 ± 15.1 months | Two IMU sensors (MicroStrain Inc., Williston, USA); shank-mounted; 10 m walking test; 3 trials; self-selected speed | Subjects' walking speed was comparable with other studies on stroke; Gait symmetry measurements were consistent with previous studies. |
| Nadeau et al. (40) | Spatio-temporal parameters, kinematics, kinetics | Provides a comparison with literature in terms of the actual values for healthy | n/a | Optotrak system (Northern Digital Inc., Ontario, Canada) | Kinematics:- lower limb joint motion profiles similar to those of healthy individuals, but with reduced peak amplitudes; Kinetics:- Asymmetric pattern, and reduced peak moment and powers on the affected side. |
| Trojaniello et al. (83) | Gait velocity, stance time, swing time, step time, stride time | 10 hemiparetic subjects, 10 subjects with Parkinson's disease, 10 subjects with Huntington's disease, and 10 healthy elderly subjects | n/a | Single IMU (Opal™, APDM); lower-trunk mounted; GAITRite (12 m); single trial; self-selected, comfortable speed | Temporal parameters measured were less accurate due to the presence of missed/extra gait events; Post-stroke gait analysis using single IMU is found to be challenging. |
| Parisi et al. (84) | Gait cycle time, stance time, swing time, initial double support time, terminal double support duration, cadence, velocity, step length, stride length | 5 hemiparetic stroke patients and 3 healthy controls | n/a | Single IMU (Shimmer, Dublin, Ireland) placed at lower trunk; optoelectronic motion capture system (ELITE 2002, BTS S.p.A., Milano, Italy) for validation; 2 force plates; 12 m hallway; 1-3 trials; self-selected speed | Low-cost system for accurate measurement of spatiotemporal features. |
| Wüest et al. (85) | Gait velocity, cadence, stride length, gait limb phase, gait stance phase, gait peak swing velocity, gait asymmetry | 14 stroke patients (ischemic or hemorrhagic; free from musculoskeletal illness, cardiovascular disorders, or other neurologic diseases) and 25 nondisabled controls | any stage after stroke | 8 body-fixed inertial sensors (Physilog, GaitUp; Lausanne, Switzerland); 2 sessions each with 3 trials; Timed Get-Up and Go task; | Excellent test-retest reliability; IMU-based timed Get-Up and Go can distinguish stroke patients from nondisabled controls. |
| Zhang et al. (86) | Path length, strike angle, lift of angle, maximum angular velocity, stance ratio, load ratio, foot flat ratio, push ratio | 16 stroke patients (ischemic or hemorrhagic) and 9 healthy controls | 5 months to 11 years (median 20 months) | Inertial sensors (MTw Awinda, Xsens Technologies B.V., Enschede, The Netherlands), shoe and lower-back mounted; 6 Minute-Walk-Test | Symmetry assessment using a single 3D accelerometer on low back shows good discriminative power compared to the one based on spatiotemporal parameters derived from two feet sensors. |
| Rastegarpanah et al. (87) | Step speed, step length, step time, joint angles of hip, knee, and ankle, peak ground reaction forces | 4 stroke patients with hemiparesis, and 4 healthy controls (no history of neurological disorders or brain damage) | n/a | VICON MX System; Kistler force plate; 10-meter walk; 6 trials | Effect of targeting motor control on spatiotemporal parameters of gait in healthy controls as well as stroke patients; effect on peak ground reaction forces in stroke patients. |
| Solanki et al. (88) | Stride length, step length, stride time, step time, single support time, swing and stance phase duration, symmetry index | 9 post-stroke patients and 15 healthy controls | 1 to 48 months | Shoe FSR (Force Sensing Resistors), paper walkway, VICON (Vicon Motion Systems Ltd, Oxford, United Kingdom) | Design of a cost-effective and portable Shoe FSR device for gait characterization using spatiotemporal data; applicable for outdoor use. |
| Latorre et al. (89) | spatiotemporal and kinematic | 82 post-stroke (ischemic or hemorrhagic) patients (age≥10, able to walk 10 m with/without assistance, able to understand instructions) and 355 healthy subjects (age≥10, no history of musculoskeletal or vestibular disease and/or prosthetic surgery) | 748.55 ± 785.12 days | Kinect v2; at a comfortable speed | The system showed excellent reliability, validity, and variable sensitivity, thus can be used as alternative to expensive laboratory-based assessment systems, although its sensitivity to kinematic measurements is limited. |
| Wang et al. (90) | plantar pressure difference (PPD), step count, stride time, coefficient of variation, phase coordination index (PCI) | 18 hemiparetic patients and 17 healthy adults | n/a | textile capacitive pressure sensing insole with a real-time monitoring system; 20 m long corridor;at a comfortable speed | In comparison with healthy adults, stroke patients showed higher PPD, larger step count, a larger average stride time and a lower mean plantar pressure on the paretic leg, increased plantar pressure in the toe region and lateral foot, and a threefold higher PCI. This study further confirmed the clinical applicability of textile insole sensors. |
| Rogers et al. (91) | peak plantar pressure and contact area | 21 stroke patients (≥3 months post-stroke, able to walk 10 m independently with or without a walking aid, had no other co-existing neurological condition) | ≥3 months | Tekscan HR Mat (TekScan™ South Boston, USA); 3 walking trials; self-selected comfortable speed; 2 test sessions in 2 weeks apart | Plantar pressure analysis protocol resulted in good to excellent repeatability for foot regions, except for toes. |