Background To improve gait, persons with calf muscle weakness can be provided with a dorsal leaf spring ankle foot orthosis (DLS-AFO). These AFOs can store energy during stance and return this energy during push-off, which, in turn, reduces walking energy cost. Simulations indicate that the effect of the DLS-AFO on walking energy cost and gait biomechanics depends on its stiffness and on patient characteristics. We therefore studied the effect of varying DLS-AFO stiffness on reducing walking energy cost, and improving gait biomechanics and AFO generated power in persons with non-spastic calf muscle weakness, and whether the optimal AFO stiffness for maximally reducing walking energy cost varies between persons. Methods Thirty-seven individuals with neuromuscular disorders and non-spastic calf muscle weakness were included. Participants were provided with a DLS-AFO of which the stiffness could be varied. For 5 stiffness configurations (ranging from 2.8 to 6.6 Nm/degree), walking energy cost (J/kg/m) was assessed using a 6-min comfortable walk test. Selected gait parameters, e.g. maximal dorsiflexion angle, ankle power, knee angle, knee moment and AFO generated power, were derived from 3D gait analysis. Results On group level, no significant effect of DLS-AFO stiffness on reducing walking energy cost was found (p = 0.059, largest difference: 0.14 J/kg/m). The AFO stiffness that reduced energy cost the most varied between persons. The difference in energy cost between the least and most efficient AFO stiffness was on average 10.7%. Regarding gait biomechanics, increasing AFO stiffness significantly decreased maximal ankle dorsiflexion angle (− 1.1 ± 0.1 degrees per 1 Nm/degree, p < 0.001) and peak ankle power (− 0.09 ± 0.01 W/kg, p < 0.001). The reduction in minimal knee angle (− 0.3 ± 0.1 degrees, p = 0.034), and increment in external knee extension moment in stance (− 0.01 ± 0.01 Nm/kg, p = 0.016) were small, although all stiffness’ substantially affected knee angle and knee moment compared to shoes only. No effect of stiffness on AFO generated power was found (p = 0.900). Conclusions The optimal efficient DLS-AFO stiffness varied largely between persons with non-spastic calf muscle weakness. Results indicate this is caused by an individual trade-off between ankle angle and ankle power affected differently by AFO stiffness. We therefore recommend that the AFO stiffness should be individually optimized to best improve gait. Trial registration number Nederlands Trial Register 5170. Registration date: May 7th 2015. http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=5170

No abstract