de Kam Digna, Roelofs Jolanda M B, Geurts Alexander C H, Weerdesteyn Vivian
Department of Rehabilitation, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, The Netherlands.
Sint Maartenskliniek Research, Nijmegen, The Netherlands.
PLoS One. 2018 Feb 22;13(2):e0192961. doi: 10.1371/journal.pone.0192961. eCollection 2018.
To determine the predictive value of leg and trunk inclination angles at stepping-foot contact for the capacity to recover from a backward balance perturbation with a single step in people after stroke.
Twenty-four chronic stroke survivors and 21 healthy controls were included in a cross-sectional study. We studied reactive stepping responses by subjecting participants to multidirectional stance perturbations at different intensities on a translating platform. In this paper we focus on backward perturbations. Participants were instructed to recover from the perturbations with maximally one step. A trial was classified as 'success' if balance was restored according to this instruction. We recorded full-body kinematics and computed: 1) body configuration parameters at first stepping-foot contact (leg and trunk inclination angles) and 2) spatiotemporal step parameters (step onset, step length, step duration and step velocity). We identified predictors of balance recovery capacity using a stepwise logistic regression. Perturbation intensity was also included as a predictor.
The model with spatiotemporal parameters (perturbation intensity, step length and step duration) could correctly classify 85% of the trials as success or fail (Nagelkerke R2 = 0.61). In the body configuration model (Nagelkerke R2 = 0.71), perturbation intensity and leg and trunk angles correctly classified the outcome of 86% of the recovery attempts. The goodness of fit was significantly higher for the body configuration model compared to the model with spatiotemporal variables (p<0.01). Participant group and stepping leg (paretic or non-paretic) did not significantly improve the explained variance of the final body configuration model.
Body configuration at stepping-foot contact is a valid and clinically feasible indicator of backward fall risk in stroke survivors, given its potential to be derived from a single sagittal screenshot.
确定中风后患者在单步向后平衡扰动恢复能力方面，迈步脚接触时腿部和躯干倾斜角度的预测价值。
一项横断面研究纳入了24名慢性中风幸存者和21名健康对照者。我们通过在平移平台上让参与者接受不同强度的多方向姿势扰动来研究反应性迈步反应。本文重点关注向后扰动。参与者被要求以最多一步从扰动中恢复。如果根据该指令恢复了平衡，则将一次试验分类为“成功”。我们记录了全身运动学并计算了：1）首次迈步脚接触时的身体构型参数（腿部和躯干倾斜角度）和2）时空步长参数（步长起始、步长、步长持续时间和步速）。我们使用逐步逻辑回归确定平衡恢复能力的预测因素。扰动强度也作为预测因素纳入。
包含时空参数（扰动强度、步长和步长持续时间）的模型能够正确地将85%的试验分类为成功或失败（Nagelkerke R2 = 0.61）。在身体构型模型中（Nagelkerke R2 = 0.71），扰动强度以及腿部和躯干角度正确地对86%的恢复尝试结果进行了分类。与具有时空变量的模型相比，身体构型模型的拟合优度显著更高（p<0.01）。参与者组和迈步腿（患侧或非患侧）并没有显著提高最终身体构型模型的解释方差。
鉴于迈步脚接触时的身体构型有可能从单个矢状面截图中得出，它是中风幸存者向后跌倒风险的一个有效且临床可行的指标。
