Department of Research, Sint Maartenskliniek, Hengstdal 3, 6574 NA Ubbergen, Nijmegen, The Netherlands.
Department of Rehabilitation, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, The Netherlands.
BMC Musculoskelet Disord. 2021 Mar 3;22(1):242. doi: 10.1186/s12891-021-04074-2.
Although it is well-established that osteoarthritis (OA) impairs daily-life gait, objective gait assessments are not part of routine clinical evaluation. Wearable inertial sensors provide an easily accessible and fast way to routinely evaluate gait quality in clinical settings. However, during these assessments, more complex and meaningful aspects of daily-life gait, including turning, dual-task performance, and upper body motion, are often overlooked. The aim of this study was therefore to investigate turning, dual-task performance, and upper body motion in individuals with knee or hip OA in addition to more commonly assessed spatiotemporal gait parameters using wearable sensors.
Gait was compared between individuals with unilateral knee (n = 25) or hip OA (n = 26) scheduled for joint replacement, and healthy controls (n = 27). For 2 min, participants walked back and forth along a 6-m trajectory making 180° turns, with and without a secondary cognitive task. Gait parameters were collected using 4 inertial measurement units on the feet and trunk. To test if dual-task gait, turning, and upper body motion had added value above spatiotemporal parameters, a factor analysis was conducted. Effect sizes were computed as standardized mean difference between OA groups and healthy controls to identify parameters from these gait domains that were sensitive to knee or hip OA.
Four independent domains of gait were obtained: speed-spatial, speed-temporal, dual-task cost, and upper body motion. Turning parameters constituted a gait domain together with cadence. From the domains that were obtained, stride length (speed-spatial) and cadence (speed-temporal) had the strongest effect sizes for both knee and hip OA. Upper body motion (lumbar sagittal range of motion), showed a strong effect size when comparing hip OA with healthy controls. Parameters reflecting dual-task cost were not sensitive to knee or hip OA.
Besides more commonly reported spatiotemporal parameters, only upper body motion provided non-redundant and sensitive parameters representing gait adaptations in individuals with hip OA. Turning parameters were sensitive to knee and hip OA, but were not independent from speed-related gait parameters. Dual-task parameters had limited additional value for evaluating gait in knee and hip OA, although dual-task cost constituted a separate gait domain. Future steps should include testing responsiveness of these gait domains to interventions aiming to improve mobility.
虽然骨关节炎（OA）会影响日常生活中的步态已得到充分证实，但步态的客观评估并非常规临床评估的一部分。可穿戴惯性传感器为在临床环境中常规评估步态质量提供了一种便捷快速的方法。然而，在这些评估中，日常步态中更复杂和有意义的方面，包括转弯、双重任务表现和上半身运动，往往被忽视。因此，本研究的目的是使用可穿戴传感器除了更常见的时空步态参数外，还研究膝关节或髋关节 OA 患者的转弯、双重任务表现和上半身运动。
比较了 25 名单侧膝关节（n=25）或髋关节 OA（n=26）患者和 27 名健康对照者的步态。参与者在 6 米的轨迹上往返行走 180°转弯，同时进行和不进行次要认知任务。使用脚和躯干上的 4 个惯性测量单元收集步态参数。为了测试双重任务步态、转弯和上半身运动是否比时空参数具有更高的附加值，进行了因子分析。计算 OA 组与健康对照组之间的标准化平均差异作为效应大小，以确定这些步态域中的参数对膝关节或髋关节 OA 敏感。
得到了四个独立的步态域：速度-空间、速度-时间、双重任务成本和上半身运动。转弯参数与步频一起构成了一个步态域。在获得的域中，步长（速度-空间）和步频（速度-时间）对膝关节和髋关节 OA 都有最强的效应大小。与健康对照组相比，上半身运动（腰椎矢状面运动范围）显示出很强的效应大小。反映双重任务成本的参数对膝关节或髋关节 OA 不敏感。
除了更常见的时空参数外，只有上半身运动提供了非冗余且敏感的参数，代表髋关节 OA 患者的步态适应性。转弯参数对膝关节和髋关节 OA 敏感，但与速度相关的步态参数不独立。双重任务参数对评估膝关节和髋关节 OA 的步态仅有有限的附加价值，尽管双重任务成本构成了一个单独的步态域。未来的步骤应包括测试这些步态域对旨在改善活动能力的干预措施的反应性。
