The Effects of a Powered Knee Orthosis on Gait Kinematics of Children With Knee Extension Deficiency

Crouch gait/walking, characterized by an 'over-flexed' knee when the leg is supporting body weight, is common in children with diagnoses of cerebral palsy, spina bifida and other incomplete spinal cord injuries. The "Agilik" is a leg exoskeleton device that aims to improve how children with crouch gait walk. In this study the investigators will quantify the improvement that the Agilik facilitates in children with crouch gait in two ways: 1) the difference the Agilik makes when the participants start using it, and 2) any 'training effects' that can be seen in barefoot walking after six sessions of training with the Agilik.

Specific Aims: Improving knee kinematics during gait, in children exhibiting crouch gait due to weakness, would help to mitigate the ensuing deleterious effects - pain, joint deformity, radiological abnormalities and loss of independent gait. The primary goal of this protocol is to assess the effectiveness of a powered knee orthosis (the "Agilik", developed by Bionic Power) for gait in patients exhibiting crouch gait. The investigators will specifically focus on improvement in knee extension during stance in patients with crouch gait. The primary goal is characterized by two hypotheses, and two respective objectives: Hypothesis 1: Use of the Agilik in children with crouch gait will result in an immediate improvement in sagittal plane knee kinematics during gait, with particular focus on knee extension during stance phase. Objective 1: By using motion capture to quantitatively compare sagittal plane knee kinematics and temporal-spatial metrics of gait performed with the Agilik, without the device and also with any braces/orthoses the patient is currently using. This objective will provide insight into the immediate effect of utilizing the Agilik for mitigation of crouch gait characteristics. Hypothesis 2: Use of the Agilik in children with crouch gait over a series of training sessions will result in a sustained improvement of knee extension during stance phase, even when the child is walking barefoot, without the aid of the Agilik. Objective 2: Compare sagittal plane kinematics and temporal-spatial metrics between barefoot walking trials at the onset of the orthosis intervention, with barefoot walking trials at the conclusion of the final orthosis session. This objective will provide insight into any sustained/learned effects of the use of the orthosis that results in a change in gait patterns even when the orthosis is not being used. Background, Significance & Literature Review: Children with diagnoses of cerebral palsy (CP), spina bifida (SB) or other incomplete spinal cord injuries (iSCI) often experience atypical and problematic gait which, if not corrected, can lead to loss of mobility, independence and quality of life. While each child with these diagnoses is unique, and the diagnosis will manifest clinically in varying ways, some specific gait deviations, such as crouch gait, are commonly observed. Crouch gait is characterized by an overly flexed knee during the stance phase of gait. A crouched posture reduces the capacity of muscles to extend the knee and hip and is significantly less efficient. Anti-gravity (extensor) muscle weakness can contribute to crouch gait, particularly towards greater deterioration with increasing age in part because the increase in muscle strength does not keep pace with the rate of body growth. Achieving and maintaining adequate strength levels in those with motor disabilities is challenging. Traditional ankle bracing compensates for weakness by providing passive support and often leads to greater weakness in these muscles over time. Clearly there is a need for more effective training strategies or devices that can preserve or possibly even augment strength on a continuous basis for those with crouch gait, and thereby help to maintain gait function. Wearable devices that provide some type of external assistance and/or support to help a person perform a functional task, often referred to as robotic exoskeletons, are increasingly available as training/assistive devices. However, there is still a lack of evidence supporting their use and for determining optimal methods for specific patient populations. The primary purpose of this study is to evaluate the effectiveness of a powered knee prosthesis - the Agilik - for improving crouch gait in children. Dynamic or assistive knee devices have been shown to be effective in rehabilitation populations such as those with spinal cord injury and may prove to be beneficial for those exhibiting crouch gait, as well. Data Collection: This is a study of outpatients with crouch gait to evaluate the effectiveness of an assistive orthosis (the Agilik) to improve crouch gait. The study includes seven visits for participants within a two month period. The first visit (visit 1) is for consent, eligibility evaluation, medical history and physical examination, casting for brace fabrication which will take 2 hours. Fabrication of the pair of custom braces for each participant will take approximately 4 weeks. The second visit (visit 2) is for initial setup, tuning, and testing of the Agilik during walking, lasting a maximum of 4 hours. To facilitate tuning of the device, the attending physiotherapist (PT) will work with personnel from Bionic Power, during visit 2, to optimize the torque profile of the Agilik with respect to normalizing gait parameters. If the participant is able to complete the ~10 m walkway using the Agilik without assistance from another person (but with their mobility aid if typically used for walking) on the second visit, initial data collection while walking with the Agilik, and the baseline conditions, will take place on visit 2 (Table 1), along with additional walking practice for accommodation with the device, lasting a maximum of 4 hours. 'Baseline conditions' refer to barefoot, and patient use of any current brace/orthosis. If participants are observed to be unstable during their training session with the Agilik, participants will wear a harness (Maxi Sky 2, Arjo, Malmo, Sweden) to continue to train with the Agilik. The Maxi Sky 2 system monitors participants while they walk to protect them from contacting the ground in the event of a fall by supporting their body weight if they become unstable. The ensuing four visits (visits 3-6) will facilitate time for the participant to practice with, and record walking trials for, the Agilik lasting a maximum of 2 hours each. During the Agilik trials, knee extension assist will be provided during late swing and/or stance phase. No assistance will be provided during early and mid-swing phases. The final visit (visit 7) will involve practice with, and recording trials for, the Agilik, and also another recording of the barefoot and brace/orthosis conditions collected in visit 1 (see Table 1 for a summary of activities occurring during each visit). At each visit the participants attend, the attending PT will briefly interview the participant and their family to determine if there are any changes since the previous visit that would indicate that their participation in the study should be halted and/or if the PT should be aware of any heightened risks in their participation. If the participant's condition has changed such that they are unable to meet the aforementioned inclusion criteria, two options will be provided: 1) the participant visit may be postponed to allow for recovery of the patient to original condition at their initial enrollment in the study, or 2) the participant may withdraw from the study. Option #2 will be required if the study team deems there is no reasonable expectation for the participant to return to a health state in which they are able to meet the original inclusion criteria within a month of their previous visit. Gait assessment includes passive reflective markers that are taped to the skin on each segment of the lower limb. Surface electromyography (EMG) electrodes will be used to measure muscle activity during walking. Adhesive surface electrodes (Delsys Inc., Boston, MA) will be placed on the skin over the muscles following skin preparation with an alcohol pad. Electrodes will be placed bilaterally on knee extensor muscles, knee flexor muscles, ankle dorsiflexor muscles, and ankle plantar flexor muscles by a trained therapist. EMG electrode pads placed directly on the skin will not interfere with the brace. During each participant's walking trials (including the barefoot, existing braces/orthoses and Agilik conditions), the investigators will collect 3D kinematic (motion), force plate, and surface EMG data to quantify the gait patterns during overground walking. Gait biomechanics will be collected in the laboratory using a Qualisys motion capture system (Qualisys, Goteborg, Sweden) and force plates (AMTI, Watertown, MA). In all walking trials throughout the duration of the study, the participants will be directed to walk at a self-selected, comfortable speed. They will be instructed to walk the 10 m at their self-selected pace. When they reach the end of the 10m, they will stop, turn 180 degrees, wait for instruction to continue, and then return to the original starting point. This process will be repeated until at least five 10 m walks are successfully performed and recorded. Data Analysis: To accomplish objective 1, the investigators will compare peak knee angle during stance and knee moment profiles during stance in the sagittal plane, between conditions. An improvement in peak knee extension of more than 10° will be considered as the threshold for clinical significance based on a study that separated children with crouch into clinical categories (mild, moderate, or severe) based on 10° increments in knee angle. The investigators will also compare temporal-spatial metrics from each condition, including: gait velocity, step length, stride length, time spent in stance and step width. To complete objective 2, the barefoot trials from visit 1 and visit 6 will be compared across the same metrics listed above for objective 1. To compare the aforementioned gait parameters within subjects but across conditions (barefoot, existing braces/orthoses and the Agilik), the investigators will perform paired t-tests to determine statistical significance between the means of each parameter (knee angle and gait speed). The investigators will employ Bonferroni correction to account for multiple testing of p-values across conditions. The investigators will consider an increase in knee extension greater than 10 degrees during stance to be a clinically significant improvement. While our sample size is limited to 5 participants in this trial, similar device-centered rehabilitation studies have utilized similar sample sizes and were able to observe statistically significant changes due to large effect sizes. Based on these studies, the investigators expect to be able to test the hypotheses and achieve the objectives defined for this study.

The Agilik is a leg orthosis device supporting the knee joint, and is intended to be worn on one or both the legs. The device is customized for single patient use. It is composed of custom-fabricated polypropylene leg shells (thigh, calf and foot components) with joint supports, actuator, sensors, cabling and a battery waist pack. The Agilik is attached to the patient using straps - two on the thigh and two on the shank. A rotational linkage connects the thigh and calf portions of the device, is placed lateral to the knee joint, and is aligned with the knee center of rotation.

Maximum knee extension (measured in degrees) during the stance phase of gait will be compared between barefoot gait and gait while wearing the Agilik.

Maximum knee extension (measured in degrees) during the stance phase of gait will be compared between barefoot gait at the start of the trial and barefoot gait after 6 weeks of weekly training sessions with the Agilik.

Sagittal knee moment profiles (measured in N.m/kg) during the stance phase of gait will be compared between barefoot gait and gait while wearing the Agilik.

Sagittal knee moment profiles (measured in N.m/kg) during the stance phase of gait will be compared between barefoot gait at the start of the trial and barefoot gait after 6 weeks of weekly training sessions with the Agilik.

Gait velocity (measured in m/s) will be compared between barefoot gait and gait while wearing the Agilik.

Gait velocity (measured in m/s) will be compared between barefoot gait at the start of the trial and barefoot gait after 6 weeks of weekly training sessions with the Agilik.

Step-length during gait (measured in m) will be compared between barefoot gait and gait while wearing the Agilik.

Step-length during gait (measured in m) will be compared between barefoot gait at the start of the trial and barefoot gait after 6 weeks of weekly training sessions with the Agilik.

Inclusion Criteria: Age 5-19 years of age Male or female Able to understand and follow simple directions based on parent report and physician observation during history and physical examination Able to provide verbal/written assent. Less than 20 degrees of knee flexion contracture with hip extended in supine position. Less than 10 degrees of plantar flexion contracture in neutral foot alignment. A measured foot-thigh angle of -10 to 25 degrees in prone position. Based on clinical observation of gait pattern, the patient walks with crouch gait. (The exact level of knee extension deficiency, or crouch, will be quantified after inclusion using gait analysis.) Able to walk at least 10 feet without stopping with or without a walking aid. Exclusion Criteria: Any neurological, musculoskeletal or cardiorespiratory injury, health condition, or diagnosis other than cerebral palsy, muscular dystrophy, spina bifida, or incomplete spinal cord injury that would affect the ability to walk as directed for short periods of time. A history of a seizure in the past year.

O'Sullivan R, Horgan F, O'Brien T, French H. The natural history of crouch gait in bilateral cerebral palsy: A systematic review. Res Dev Disabil. 2018 Sep;80:84-92. doi: 10.1016/j.ridd.2018.06.013. Epub 2018 Jun 27.

Hicks JL, Schwartz MH, Arnold AS, Delp SL. Crouched postures reduce the capacity of muscles to extend the hip and knee during the single-limb stance phase of gait. J Biomech. 2008;41(5):960-7. doi: 10.1016/j.jbiomech.2008.01.002. Epub 2008 Mar 4.

Waters RL, Mulroy S. The energy expenditure of normal and pathologic gait. Gait Posture. 1999 Jul;9(3):207-31. doi: 10.1016/s0966-6362(99)00009-0.

Opheim A, Jahnsen R, Olsson E, Stanghelle JK. Walking function, pain, and fatigue in adults with cerebral palsy: a 7-year follow-up study. Dev Med Child Neurol. 2009 May;51(5):381-8. doi: 10.1111/j.1469-8749.2008.03250.x. Epub 2008 Feb 3.

Damiano DL, Arnold AS, Steele KM, Delp SL. Can strength training predictably improve gait kinematics? A pilot study on the effects of hip and knee extensor strengthening on lower-extremity alignment in cerebral palsy. Phys Ther. 2010 Feb;90(2):269-79. doi: 10.2522/ptj.20090062. Epub 2009 Dec 18.

Duffy CM, Hill AE, Cosgrove AP, Corry IS, Mollan RA, Graham HK. Three-dimensional gait analysis in spina bifida. J Pediatr Orthop. 1996 Nov-Dec;16(6):786-91. doi: 10.1097/00004694-199611000-00016.

Greene PJ, Granat MH. A knee and ankle flexing hybrid orthosis for paraplegic ambulation. Med Eng Phys. 2003 Sep;25(7):539-45. doi: 10.1016/s1350-4533(03)00072-9.

Kang J, Martelli D, Vashista V, Martinez-Hernandez I, Kim H, Agrawal SK. Robot-driven downward pelvic pull to improve crouch gait in children with cerebral palsy. Sci Robot. 2017 Jul 26;2(8):eaan2634. doi: 10.1126/scirobotics.aan2634. Epub 2017 Jul 26.

Conner BC, Schwartz MH, Lerner ZF. Pilot evaluation of changes in motor control after wearable robotic resistance training in children with cerebral palsy. J Biomech. 2021 Sep 20;126:110601. doi: 10.1016/j.jbiomech.2021.110601. Epub 2021 Jul 8.