Thesis (MScPhysio)--Stellenbosch University, 2012 / ENGLISH ABSTRACT: SYSTEMATIC REVIEW ABSTRACT - BACKGROUND: Ankle Foot Orthoses (AFOs) are considered as the most suitable lower limb orthosis to correct gait deficits related to ankle instability. AFOs are recommended to minimize gait deviations and to correct drop foot or equinus foot in hemiplegic patients. OBJECTIVES - To identify the effectiveness of different ankle orthoses and/or supports on the temporal, spatial, kinetic and kinematic gait parameters. To critically appraise the methodological quality of the included studies and to provide a description of the studies with a view to identify opportunities to improve future research quality. METHODS - Search strategy A comprehensive search was conducted between March and October 2010, and updated in August 2011. Thirteen computerized bibliographic databases were individually searched, namely PubMed Central, Cohrane Library, CINAHL, OT Seeker, SPORTDiscus, PsyARTICLE, PEDro, Proquest, Biomed Central, Science Direct, Clinicaltrials.gov, Web of Science, and Ingenta Connect. All databases were searched since their inception. The following key terms were used: stroke, hemipleg*, assistive device*, ankle foot orthos*, AFO, (splint*), taping, and strapping. A secondary search (pearling) was conducted by screening the reference lists of all eligible full text studies. The authors of the unpublished studies were conducted to minimize publication bias. Selection criteria The following selection criteria applied: all relevant randomized and non-randomized controlled trails published in English; participants were post-stroke patients older than eighteen years; interventions included any type of ankle foot orthosis (AFO), ankle taping or strapping and ankle foot splint without any additional intervention and the comparison/control groups were limited to walking without support, either barefoot or walking with shoes only. Studies were excluded when the outcome measures did not focus on at least one of the following: temporal spatial gait parameters, kinetic gait parameters or kinematic gait parameters. Data collection and analysis Two reviewers independently selected trials for inclusion and assessed methodological quality. The data was extracted by the primary reviewer and validated by a second reviewer. In event of disagreement, a third reviewer was asked to re-evaluate until consensus could be reached. Homogenous data were statistically summarized in sub-group meta-analysis using Revman© Review Manager Software. The results of heterogeneous data were summarized in a narrative form. MAIN RESULTS - The search yielded 11134 initial hits. Sixteen studies met the inclusion/exclusion criteria. The studies investigated the immediate effect of various types of AFOs on a broad range of temporal spatial gait parameters mainly gait speed, cadence, stride and step length. Only two studies reported on the kinetic and six on various kinematic gait parameters. The meta-analysis yielded significant improvement in gait speed (0.06 m/s; 95% CI 0.04, 0.08. p < 00001), walking cadence (5.41; 95% CI 3.79, 7.03. p < 00001), stride length (6.67; 95% CI 3.29, 10.06. p < 00001) and step length (2.66; 95% CI 1.59, 3.72. p < 0.00001). CONCLUSION - AFOs are effective to improve mobility, gait speed, cadence, stride and step length for post-stroke patients and may have a positive impact on the daily function of post-stroke patients. . The long term benefit or adverse effects of AFOs are still inconclusive. The effectiveness of AFOs on the kinetic and the frontal- or transverse- plane joint kinematics is largely unresolved. There is insufficient evidence to either support or refute the effectiveness of taping/strapping and splinting of the ankle on hemiplegic gait. EXPERIMENTAL STUDY ABSTRACT - BACKGROUND: Temporal, spatial and affected ankle kinematic gait parameters of adults with hemiplegia are significantly different from the normal able-bodied population. Enabling hemiplegic patients to walk is a major goal of rehabilitation programs. Taping of the plegic ankle could be utilized by therapists as external support of the ankle to improve foot position and placement during gait rehabilitation. OBJECTIVE - The purpose of the study was to describe the immediate effect of neutral ankle taping on temporal spatial gait parameters and ankle joint kinematics of the affected ankle in ambulant adult hemiplegic patients. METHODS - A clinical trial using a crossover randomized testing order was conducted on a convenient sample of ten ambulant hemiplegic patients at the Physiotherapy and Motion Analysis Clinic, Faculty of Health Sciences, Stellenbosch University, Tygerberg, Cape Town, South Africa. The affected ankle joint was taped in a neutral talocrural dorsiflexion/ plantarflexion and neutral hindfoot inversion/ eversion position using rigid adhesive tape (5 cm). The gait parameters were analysed according to the Plug-In Gait Model using a motion analysis system (Vicon Nexus 1.1.7; Vicon Motion System Limited, Oxford, UK). The analyses were repeated six times for each testing condition and the average values were used for further analysis. The data were analyzed using Least Square Means tests and post hoc Fisher (Least Significant Difference) LSD multiple comparison tests to determine the significant differences at 95% confidence level. RESULTS - The main results of the study indicate that taping of the affected ankle joint in a neutral position does not significantly improve (p>0.5) temporal spatial gait parameters and ankle joint kinematics in ambulant adult hemiplegic patients. The following positive trends were however found and need to be further explored in larger homogeneous study samples: ankle taping of ambulant adult hemiplegic patients has limited benefits on selected temporal parameters as ankle taping could potentially improve cadence. Ankle taping could decrease plantarflexion of the plegic leg at initial contact. CONCLUSIONS - A systematic review revealed no conclusive evidence either to support or refute the beneficial effects of ankle taping on gait parameters of ambulant adult hemiplegic patients. Ankle taping of ambulant adult hemiplegic patients has potential clinical benefits on temporal, spatial and affected ankle kinematics, gait cadence and affected leg swing and stance duration.

Le travail développé dans cette thèse a pour objectif la génération de trajectoires de marches dynamiques optimales qui puissent être validées sur une plateforme expérimentale.Le dispositif mis en oeuvre est le système locomoteur d'un robot humanoïde de petite taille (80cm pour 15kg) nommé Tidom. Ses caractéristiques mécaniques et son architecture de commande sont présentées dans le premier chapitre.Pour générer les trajectoires sur un pas de marche, nous avons utilisé une méthode d'optimisation paramétrique. Cette technique repose sur l'approximation des paramètres de configuration du mouvement par des fonctions splines de classe C3 constituées de polynômes de degré 4 raccordés en des instants équirépartis sur la durée du mouvement jusqu'aux suraccélérations. Les efforts de contact entre le pied balancé et le sol en phase bipodale sont également paramétrés par des fonctions splines, de classe C0. Les accélérations et les couples articulaires sont raccordés aux instants de transition entre les différentes phases du mouvement pour améliorer le contrôle et éviter la détérioration de la mécanique. Le vecteur des paramètres d'optimisation est ainsi composé des coordonnées articulaires aux points de jonction, des vitesses et accélérations aux extrémités des phases de la marche, des efforts de contact pied/sol en double appui auxquels s'ajoutent la longueur de pas et la durée de chaque phase. Il est à noter que la seule donnée d'une vitesse de marche permet d'engendrer un pas optimal cyclique.Plusieurs expérimentations présentées dans le dernier chapitre permettront à terme d'implémenter les trajectoires optimales sur le robot. / The work developed in this thesis aims at generating optimal trajectories for dynamic walking motions that can be validated on an experimental platform.The device used is the locomotion system of a small size humanoid robot (80cm and 15kg) named Tidom. Its mechanical characteristics and control architecture are presented in the first chapter.A parametric optimization method is developed to generate walking step trajectories. It consists in approximating joint motion coordinates using C3-spline functions, made up of 4-order polynomials linked at times equally distributed along the motion time, up to jerk linking. The contact forces between stance foot and ground are approximated using spline functions of class C0. Joint acceleration and joint torques are continuous at the transitions between single and double support phases of a step to improve the robot control and to prevent mechanical damage. The optimization variables are discrete values of joint coordinates at connecting points, joint velocities and accelerations at phase bounds, discrete values of contact forces at connecting points during the double support phase. The step length and the relative length of the step phases are also accounted for. The walking velocity is the only data required for generating an optimal cyclic step.Some experiments presented in the last chapter are the first steps towards the implementation of optimal trajectories on the humanoid robot.

Background: The yearly incidence of people with Spinal Cord Injury (SCI) is between250,000 and 500,000, according to the World Health Organization (WHO). The injury often reduces the ability to walk. Various consequences affect the nervous system and, thus, the entire body. Therefore, the patient population with SCI is highly heterogeneous also in their gait patterns. Multiple tools are used to classify and understand the walking impairments caused by the injury. Objective: To underline the added value brought by the integration of 3D gait analysis to more standard methods (GDI, GPS, GVS, spatiotemporal parameters, ASIAgrade, muscle strength, and spasticity) in the evaluation and interpretation of gait patterns of subjects with SCI. Methods: 3D gait analysis with a passive optical motion capture system (Vicon)and four force plates was performed in 7 control subjects and 3 with SCI. The model used for marker placement and pre-processing was CGM 2.3. Matlab was used to analyze and plot the kinematic and kinetic joints’ data and calculate the GDI, GPS, and GVS indexes and spatiotemporal parameters for subjects with SCI and the control group. A specialized physiotherapist conducted the clinical assessment of the patients with SCI in a rehabilitation center. This included: ASIA grade and review, muscle strength, and spasticity with Daniels Whorthingham scale and Modified Ashworth scale, respectively. The evaluation of the result was qualitative. Results: The integration of 3D gait analysis show further understanding in the assessment of walking impairments. The indexes resumed the impairments and classified the subjects but lacked temporal and functional perspective. Gait phases and spatiotemporal parameters suggested difficulties in stability and balance but could not identify the problem’s origin. Lastly, clinical assessment enlightened the singular motor and sensory function impairments. 3D gait analysis contextualized these results identifying gait patterns and functional implications. Conclusion: Integrating 3D gait analysis might give a deeper understanding of subjects with SCI’s gait strategies and impairments. Indeed this complex technique links the other methods’ results, contextualizing them and gaining information.