Kinematic changes in severe hip osteoarthritis measured at matched gait speeds

Abstract

Kinematic differences between patients with osteoarthritis (OA) and control participants have been reported to be influenced by gait speed. The purpose of this study was to experimentally detect the effect of walking speed on differences in spatiotemporal parameters and kinematic trajectories between patients with hip OA and age‐matched asymptomatic participants using wearable sensors and statistical parametric mapping (SPM). Twenty‐four patients with severe unilateral hip OA and 48 control participants were included in this study. Patients walked at a self‐selected normal speed and control participants at self‐selected normal and slow speeds. Spatiotemporal parameters and kinematic trajectories were measured with the inertial sensor system Rehagait®. Gait parameters were compared between patients with hip OA and control participants for normal and matched speed using SPM with independent sample t‐tests. At self‐selected normal speed, the patient group walked slower (−0.20 m/s, p < .001) and at lower cadence (−5.0 steps/minute, p < .001) as well as with smaller hip flexion (−7.4°, p < .001) and extension (−4.1°, p = .001), higher knee flexion during terminal stance (+8.0°, p < .001) and higher ankle dorsiflexion and plantarflexion (+7.1°, p < .001). While differences in spatiotemporal parameters and the ankle trajectory disappeared at matched speed, some clinically relevant and statistically significant differences in hip and knee trajectories remained. Most differences in sagittal plane gait kinematics between patients with hip OA and control participants were present for matched speed, and therefore appear to be associated with a disease rather than gait speed. Nevertheless, studies investigating hip kinematics in patients with hip OA should involve trials at matched speeds.

Effects of Hip Bracing on Gait Biomechanics, Pain and Function in Subjects With Mild to Moderate Hip Osteoarthritis

“…joint loading on a daily basis ( Astephen Wilson, 2012 ). Additionally, it has been shown that gait modifications in people with HOA persist even at matched gait speeds ( Ismailidis et al, 2021 ). The application of ANCOVA neglects the fact that differences in gait speeds between groups are not a source of random error variability but rather representative of the population characteristics ( Astephen Wilson, 2012 ).…”

Assessing Site Specificity of Osteoarthritic Gait Kinematics with Wearable Sensors and Their Association with Patient Reported Outcome Measures (PROMs): Knee versus Hip Osteoarthritis

“…Of these, 29 had unilateral knee OA and were scheduled to undergo total knee arthroplasty, 30 had unilateral hip OA and were scheduled total hip arthroplasty, and 54 were age-matched asymptomatic persons ( Table 1 ). Subsets of data have been published previously in studies focusing on differences in gait kinematics between limbs in patients with knee OA and compared to healthy controls [ 21 , 23 ] and on differences in gait kinematics between limbs in patients with hip OA and compared to healthy controls [ 22 , 24 ]. For the current analysis, we expanded all groups of participants and focused on detecting differences in joint kinematics between patients with knee OA and patients with hip OA.…”

“…For instance, for patients with knee OA, differences in stride duration, knee flexion range of motion (ROM) in swing and stance have been assessed using the GaitWalk system [ 18 ], kinematic differences at the hip, knee and ankle in the sagittal plane using the H-Gait system [ 19 ], and thigh and shank sagittal and coronal, knee sagittal kinematics as well as temporal gait parameters using the GaitSmart TM system [ 20 ]. In a series of studies, Ismailidis et al have shown that known differences in joint kinematics between patients with knee OA [ 21 ] or patients with hip OA [ 22 ] and healthy controls can be measured using the RehaGait ® system, reported kinematic differences not only for the index joint but also to adjacent joints, and confirmed that some of these differences depend on walking speed [ 23 , 24 ].…”

Inertial Measurement Units and Application for Remote Health Care in Hip and Knee Osteoarthritis: Narrative Review

“…The findings from these studies suggest that information extracted from the raw acceleration or angular velocity signals, even from a single sensor, may be useful to discriminate between people with knee OA and controls and could be related to clinically meaningful participant-reported outcomes. Barrois et al, 2016 [23] Trained machine learning pipeline to estimate hip and knee joint loading; error too large for clinical use Single IMU c (Samsung Galaxy J5 2017, Samsung) inside cell phone, attached to hip Hip OA (n=20) De Brabandere et al, 2020 [31] Automatically extracted features gave best machine learning accuracy in discriminating THA from healthy individuals 7 IMU c (Awinda, Xsens Technologies BV) on feet, shanks, thighs, and back Healthy (n=27) and THA d (n=20) Dindorf et al, 2020 [32] Significant changes in hip and knee kinematics exist between hip OA and healthy individuals in speed matched conditions 7 IMU b (RehaGait, Hasomed) on pelvis, feet, shanks, and thighs Healthy (n=48) and hip OA (n=24) Ismailidis et al, 2020 [61] Significant differences in all spatiotemporal parameters between groups when walking at self-selected speed 7 IMU b (RehaGait) on pelvis, feet, shanks, and thighs Healthy (n=28) and knee OA (n=23) Ismailidis et al, 2020 [26] Linear acceleration (significant) and jerk (insignificant) negatively associated with self-reported instability 5 IMU c (3D myoMOTION, Noraxon) on pelvis, thighs, and shanks Healthy (n=13) and knee OA (n=26) Na and Buchanan, 2021 [25] Stance and double support ratio 2 most consistent discriminating features between OA and controls 2 IMU b (Shimmer3, Shimmer Sensing) on feet Healthy (n=10) and knee OA (n=10)…”

“…Other studies have used more computationally complex approaches to extract spatiotemporal parameters and joint kinematics during walking using IMU data. Ismailidis et al [26,61] published 2 studies, one each in people with end-stage hip OA and those with end-stage knee OA, in which they compared spatiotemporal and sagittal plane kinematics from IMUs between OA and control populations. Using statistical parametric mapping, they observed differences in multiple parameters (eg, cadence, knee, and hip kinematics) between each OA population and controls.…”