Page 56 - Maximizing the efficacy of ankle foot orthoses in children with cerebral palsy

56
Chapter III
walking speed. Thereafter, subjects performed six walking trials of 2 minutes at this comfortable speed. For each trial, AFO-FC heel height was set into low, medium or high, and combined with either the stiff or flexible footplate. The sequence of these six combinations was randomly applied.
Data processing
Joint flexion-extension angles and net internal moments were calculated using the Human Body Model and D-flow software[18,19]. Joint flexion-extension angles were calculated using the orientation of the distal segment with respect to the orientation of proximal segment and expressed in the sagittal plane of the proximal segment. The SVA, calculated in the global sagittal plane, was defined as the angle between the anterior surface of the tibia and the vertical[13] (see Figure 3.1). Another line was created using the position of the marker at the dorsal shell of the AFO (#14 and #23) and the lateral malleolus marker (#15 and #24) in the global sagittal plane. The angle between the two lines represented changes of the position of the shank with respect to the AFO (i.e. Shank-to-AFO angle). Assuming that this angle would be unchanged with a fully immobilized ankle joint, smaller angles would indicate movement of the shank towards the AFO’s dorsal shell. Calculations of the SVA and Shank-to-AFO angle were done using Matlab 2011 (The Mathworks, USA).
Marker and force plate data were low pass filtered at 6 Hz using the Human Body Model[19]. To select only strides with foot placement on a single belt, a stride was excluded if i) the force of that stride deviated more than 100% from the mean force of all strides, or ii) the length of the stride deviated more than 20 samples from the median length of all strides. For further processing, only correctly recorded strides were selected, based on two criteria i) single-belt foot placement and ii) sufficient marker data (i.e. no occlusion) to calculate the considered parameters. Subsequently, remaining strides were normalized to 100% gait cycle and the SVA, lower limb joint flexion-extension angles and net internal moments, and Shank-to-AFO angles of the right leg were determined at midstance, defined as the moment that the malleolus marker of the contralateral leg (#24) passed the malleolus marker of the ipsilateral leg (#15). The parameters were limited to midstance, as the SVA is clinically used to evaluate the effects of tuning at this stage of the gait cycle[13]. The parameters were averaged over 25 steps, which were selected starting from the end of the trial.