The role of augmented feedback in pediatric robotic-assisted gait training: A case series

Issue title: State of the Science for Pediatric Rehabilitation Engineering

Guest editors: Richard Foulds and Sergei Adamovich

Article type: Research Article

Authors: Patritti, Benjamin L.^{a; b} | Sicari, Monica^{a; c} | Deming, Lynn C.^a | Romaguera, Fernanda^a | Pelliccio, Marlena M.^a | Kasi, Patrick^a | Benedetti, Maria Grazia^d | Nimec, Donna L.^a | Bonato, Paolo^{a; e; *}

Affiliations: [a] Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Rehabilitation Hospital, Boston, MA, USA | [b] Department of Rehabilitation and Aged Care, Repatriation General Hospital, Adelaide, Australia | [c] Department of Physical Medicine and Rehabilitation, University of Catania, Catania, Italy | [d] Istituto Ortopedico Rizzoli, University of Bologna, Italy | [e] Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA, USA | Department of Biomedical Engineering, New Jersey Institute of Technology, Newark, NJ, USA

Correspondence: [*] Address for correspondence: Paolo Bonato, PhD, Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Rehabilitation Hospital, 125 Nashua St., Boston, MA 02144, USA. Tel.: +1 617 573 2745; Fax: +1 617 573 2769; E-mail: pbonato@partners.org,

Abstract: Recent work has examined the feasibility of robotic-assisted gait training in pediatric patients, including children with cerebral palsy (CP). Herein we present a case series describing clinical outcomes in four children diagnosed with spastic diplegia due to CP who underwent gait training using a robotic driven gait orthosis (DGO) (Pediatric Lokoma©). The children were paired based on functional abilities and observed gait characteristics. Two children were classified as Gross Motor Function Classification System (GMFCS) level II and displayed a crouch gait pattern. The other two children were classified as GMFCS level III and exhibited a toe-walking gait pattern. Each child participated in a 6-week intervention of robotic-assisted gait training that involved three 30-minute sessions per week. One child of each GMFCS level underwent training using the augmented feedback module of the Pediatric Lokomat© system, while the other child of each pair underwent training without using the augmented feedback module. Pre-training, post-training, and 3-month follow-up evaluations were performed including clinical tests of standing and walking function (Gross Motor Function Measure, GMFM, sections D and E, respectively), walking speed, and walking endurance. Clinical gait analysis was also performed during each evaluation using a motion capture system to assess changes in gait biomechanics. All children showed an improvement in gait speed. For the lower functioning children (i.e. GMFCS level III), this may be mediated by improved trunk control suggested by concomitantly large increases in standing function (i.e. GMFM section D). Conversely, for the higher functioning children (i.e. GMFCS level II), large changes were observed in walking function assessed by the GMFM section E. The use of augmented feedback was associated with larger improvements in walking function (i.e. GMFM section E) and in walking speed. Furthermore, subjects who received robotic-assisted gait training in combination with augmented feedback showed a large increase in endurance while subjects who received robotic-assisted gait training without augmented feedback showed limited or no improvements in endurance. Overall, these results suggest that the use of an augmented feedback module in combination with robotic-assisted gait training leads to better outcomes than robotic-assisted gait training without augmented feedback.

Keywords: Augmented feedback, body weight supported treadmill training, cerebral palsy, gait, virtual reality

DOI: 10.3233/TAD-2010-0306

Journal: Technology and Disability, vol. 22, no. 4, pp. 215-227, 2010