Immediate effects of ankle-foot orthosis on gait coordination and kinematics in subacute stroke

Ankle-foot orthoses (AFOs) are commonly prescribed after stroke, yet their immediate effects on bilateral hip-knee coordination and joint-specific contributors to functional gains are unclear.

What are the immediate effects of AFO use on bilateral coordination and joint-level kinematics, and which changes relate to performance gains?

Forty-eight individuals with subacute stroke (mean age 57.9 ± 13.9 years; 28 men/20 women; FAC median 3.0) performed overground 10-m trials under pre-AFO and immediate post-AFO conditions (no acclimatization); eight had two measurements (56 pre-post sets). A prefabricated semi-rigid AFO (UD-Flex, anterior shell) was fitted on the paretic limb. Seven inertial-measurement units recorded spatiotemporal metrics, phase-specific (stance/swing) peak angles, and hip-knee cyclogram metrics. Condition differences were expressed as Δ = post - pre. Elastic-net selection followed by ordinary least squares identified predictors of Δ velocity, Δ cadence, and Δ hip/knee ROM.

AFO use was associated with a small increase in walking speed (+0.01 m/s; p = 0.01), below MCID, whereas cadence and stride time were unchanged On the paretic limb, swing-phase peak hip flexion (-3.1°), knee flexion (-4.4°), and ankle dorsiflexion (-2.3°) decreased; contralateral peaks showed no significant change. Cyclogram geometry was largely preserved. Predictive models linked faster walking to greater non-paretic hip/knee excursions and paretic-ankle change (adjusted R²=0.36). Cadence related positively to non-paretic knee and paretic hip excursions and negatively to paretic-ankle change.

AFO use was associated with reduced paretic swing-phase peak flexion and ankle stabilization, coinciding with a modest, sub-MCID speed gain. Findings support training that pairs paretic-ankle stability with non-paretic propulsion, while recognizing the limited clinical magnitude of the immediate effect.