The research team, based at Delft University of Technology and Amsterdam UMC in the Netherlands, recruited seven adults with bilateral calf muscle weakness caused by neuromuscular conditions, including Charcot-Marie-Tooth disease, chronic idiopathic axonal polyneuropathy, and post-polio syndrome. The participants, four men and three women, averaged about 66 years of age, and all but one were already experienced users of dorsal leaf AFOs. Because unilateral AFO users tend to favor their unaffected leg during tasks such as stepping up, the researchers deliberately restricted the study to people with weakness in both legs, ensuring that the measured trailing limb genuinely reflected the effect of the orthosis.

The experimental setup was elegantly modular. Participants wore standardized footwear in a shoes-only condition and then the same shoes combined with a carbon-fiber AFO whose posterior leaf spring could be swapped to produce five distinct stiffness levels, ranging from a very flexible 2.8 Nm/degree to a considerably firmer 6.6 Nm/degree. Stiffness, in this context, describes how strongly the orthosis resists ankle rotation in the sagittal plane. A stiffer spring limits how far the ankle can dorsiflex, which supports the shin as it progresses over the foot but also restricts the range of motion the calf muscles and tendons need to generate positive mechanical work during push-off.

Each participant performed three movement tasks under every condition: gait initiation from quiet standing, obstacle crossing over a rigid beam during walking, and a step-up onto a 15-centimeter platform. Motion capture cameras tracked reflective markers at 100 hertz while force plates embedded in the laboratory floor recorded ground reaction forces at 1,000 hertz. From these data the researchers computed two families of outcomes for the trailing limb during its propulsive phase: forward and upward impulse, which capture the total push-off force delivered to the body, and the positive mechanical work performed at the ankle, knee, and hip, expressed both in absolute terms per kilogram of body mass and as a relative share of total lower-limb work.

The first key finding concerned the effect of simply wearing an AFO. Compared with shoes alone, the most flexible orthosis at 2.8 Nm/degree did not change forward or upward impulse in any of the three tasks, meaning the total mechanical output that propels the body was preserved. However, the device significantly reduced absolute ankle work and absolute knee work, while leaving hip work untouched. When the researchers looked at the relative distribution of work across the three joints, a subtle shift emerged: the knee’s share of total work fell and the hip’s share rose, while the ankle’s relative contribution stayed statistically unchanged, apparently because knee work dropped even more than ankle work did.

The second analysis addressed the central question of stiffness. Treating stiffness as a continuous variable across the five AFO conditions, the team found that each additional Nm/degree of stiffness reduced absolute positive ankle work by roughly 0.005 to 0.008 joules per kilogram, depending on the task, a statistically robust effect confirmed with false-discovery-rate correction. Yet this progressive stiffening produced no significant change in forward or upward impulse, nor in absolute or relative knee and hip work. In other words, the total propulsive output of the body remained remarkably stable even as the ankle’s biological contribution to that output steadily shrank with firmer springs.

Perhaps the most interesting result concerned whether these stiffness effects differed between tasks. For absolute joint work, no interaction between stiffness and movement task reached significance, indicating that the per-degree reduction in ankle work was essentially uniform across gait initiation, obstacle crossing, and step-up. For relative ankle work, however, the interaction fell just short of the corrected significance threshold, with a p-value of 0.010 against a threshold of 0.05 after multiple-testing correction. The pattern in the data suggested that stiffer orthoses reduced the ankle’s relative share more strongly in tasks where ankle push-off dominates propulsion, such as gait initiation and obstacle crossing, and less strongly in step-up, where the knee and hip carry much of the burden of lifting the body onto the platform.

The authors offer a mechanical explanation for this pattern. During gait initiation, the trailing limb must accelerate the body forward from standstill, a job that leans heavily on ankle push-off power. Obstacle crossing similarly demands extra ankle power to achieve both forward acceleration and sufficient foot clearance. Step-up, by contrast, requires substantial knee and hip extensor work to elevate the center of mass, so even a meaningful absolute reduction in ankle work translates into only a small change in the ankle’s relative contribution. The researchers also note that the apparent increase in the hip’s relative share after donning an AFO does not constitute true proximal compensation, because absolute hip work did not increase; a genuine shift of mechanical demand to the hip would require a measurable rise in its absolute output, as has been observed when patients walk faster with AFOs on level ground.

Several limitations temper these conclusions. With only seven participants, statistical power is limited, and the authors explicitly frame the findings as preliminary. All participants were relatively high functioning, able to perform the tasks independently, so the effects in less mobile patients could differ. The study also did not measure energy cost or muscle stress, outcomes that may well respond to stiffness changes even when joint work and impulse do not, and tasks such as fast walking or walking on uneven ground were not included. The absence of a dedicated familiarization period for each stiffness condition is another caveat, though prior work suggests biomechanical responses to AFOs stabilize quickly, and randomized condition order plus the participants’ extensive orthosis experience likely limited learning effects.

Even so, the study carries practical weight. It demonstrates that the ankle-work-reducing effect of AFO stiffness, first documented in level walking, extends across a family of everyday movements, which helps explain why activities such as stair and incline walking feel harder with firmer orthoses. The suggestion that relative work distribution may respond to stiffness in a task-dependent way hints that testing patients on multiple movements, not just flat walking, could yield richer insights for tuning orthosis stiffness to individual lives. Future work combining three-dimensional motion analysis with electromyography and direct measurement of AFO deformation should clarify whether stiffness-related reductions in ankle work reflect reduced muscle demand, altered tendon energy return, or changes in the orthosis’s own contribution, and whether an optimal stiffness exists for the full repertoire of daily movement.

Subject of Research: Effects of ankle-foot orthosis stiffness on lower-limb joint work during everyday movement tasks in people with calf muscle weakness

Article Title: Effects of ankle-foot orthosis stiffness on lower-limb joint work during various movement tasks in people with calf muscle weakness—an explorative study

Article References: Roelfsema, J. R., Mugge, W., & Waterval, N. F. J. (2026). Effects of ankle-foot orthosis stiffness on lower-limb joint work during various movement tasks in people with calf muscle weakness—an explorative study. Frontiers in Rehabilitation Sciences, 7, 1994258. https://doi.org/10.3389/fresc.2026.1994258

Image Credits: AI Generated

DOI: 10.3389/fresc.2026.1994258

Keywords: ankle-foot orthosis, calf muscle weakness, AFO stiffness, joint work, gait initiation, obstacle crossing, step-up, biomechanics, propulsion, rehabilitation, neuromuscular disorders, push-off