Kinematic signatures underlying freezing of gait: a possible pathophysiological bridge between Parkinson’s Disease phenotypical spectrum and Progressive Supranuclear Palsy

Introduction: Gait analysis has emerged as a valuable tool for the objective assessment of motor dysfunction in neurodegenerative diseases. While spatiotemporal gait parameters are well established as markers of disease and progression, the combination of kinematic assessment may aid to disentangle subtle differences in motor control strategies, especially when assessing episodic symptoms, like freezing of gait (FOG). Although FOG represents one of the most disabling symptoms in Parkinson Disease (PD) and Progressive Supranuclear Palsy (PSP), the extent to which the presence of FOG alters gait patterns, even beyond FOG episodes, remains overlooked. Objective To quantitatively compare gait patterns among PD patients with (PD + FOG) and without FOG (PD-FOG) and PSP patients, combining spatiotemporal and kinematic analyses with the aim of offering insight about both the nature of such phenomenon and the issue of the phenotypical heterogeneity in parkinsonism. Methods Seventy-six patients with parkinsonism (32 PD-FOG, 20 PD + FOG, and 24 PSP, all comparable on both gender and age) underwent optoelectronic gait analysis during a single-task walking and a dual cognitive task. Spatiotemporal and kinematic parameters were extracted using a standardized motion capture protocol. Group differences were investigated using ANCOVA, controlling for disease duration. Results Almost all spatiotemporal parameters significantly differentiated PSP from both PD groups, whereas no differences emerged between PD + FOG vs PD-FOG. Kinematic analysis revealed significant alterations at the pelvic, knee and ankle levels, distinguishing again PSP from PD, but also highlighting several differences between PD + FOG and PD-FOG. In particular, kinematic variables stood along a spectrum spanning across PD-FOG, PD + FOG and PSP, where pelvic tilt, hip flexion-extension, knee rotation and ankle plantarflexion emerged as key discriminative features. Dual-task acquisitions seem to provide information that is coherent, but for fewer features and of reduced magnitude, highlighting the potentially flattening effect of interferences in patients with neurodegenerative disorders. Conclusions Our findings suggest that spatiotemporal gait parameters primarily reflect relevant disease-related differences between PD and PSP, while kinematic variables capture more subtle alterations, hypothetically related to motor control strategies. Therefore, kinematic analysis may offer insight about the pathophysiological mechanism underlying FOG in PD patients, whose gait strategy would resemble an approach more similar to that of PSP patients. These results, combined with a deeper anamnestic profiling, could allow for developing integrated biomarkers and a novel phenotyping approach.