• Citer

Gait Patterns in Individuals with Low Back Pain: An Integrated Analysis of Spatiotemporal Parameters, Centre of Pressure, and Plantar Pressure Across Walking Speeds, with Consideration of Sex and Age

par Leyh, Clara
Président du jury Bourguignon, Mathieu
Promoteur Feipel, Véronique
Publication Non publié, 2026-09-04

Thèse de doctorat

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Introduction and Objectives - Human gait is a complex motor behaviour frequently reported as altered in individuals with low back pain (LBP). However, the interpretation of modified gait patterns remains challenging due to the confounding effects of walking speed, as well as individual determinants such as sex and age. Moreover, most previous studies have focused on isolated gait variables (e.g., spatiotemporal, kinetic or kinematic parameters, ground reaction forces), neglecting often the complementary information provided by the centre of pressure (COP) and plantar pressure distributions. The primary aim of this work was to provide an integrative characterisation of gait in patients with LBP. Specifically, this work sought to (1) quantify the influence of walking speed, LBP, sex, and age on spatiotemporal parameters, COP dynamics, and plantar pressure distributions; (2) investigate the evolution of spatiotemporal parameter distributions as a function of walking speed (study of so-called “Velocity Profiles); and (3) determine whether LBP is associated with specific gait alterations beyond speed-related effects. Methods - A total of 364 participants (179 women), including 204 subjects with LBP (95 women), were recruited. Gait was assessed at self-selected slow, preferred and fast walking speeds using an instrumented walkway (GAITRite or Zeno). Spatiotemporal parameters (e.g., cadence, step and stride length, step and cycle time, cycle gait phases, step width, foot progression angle), COP trajectories and velocities, and plantar pressure variables (including pressure-time integrals, peak pressures, contact areas, and timing parameters across different foot regions) were collected and analysed. To evaluate the speed-dependant evolution of spatiotemporal parameters, linear and polynomial regression equations were used to compute these parameters at measured and normalised velocities (slow, preferred and fast versus 90 cm/s, 125 cm/s and 190 cm/s). Results - Walking speed emerged as the main determinant of all gait parameters investigated, preferentially impacting spatiotemporal parameters in a curvilinear manner. As gait velocity increased, cadence, step/stride lengths and COP velocities increased while temporal metrics and foot progression angle decreased. Marked changes were also noticed in plantar pressure distribution. At fast speed, second order polynomial reconstructed velocity profiles displayed increased outliers and extreme values requiring special attention in comparing results between subjects and gait analyses.Women tended to regulate walking speed through cadence adjustments, whereas men relied more on step and stride length modulation. These differences were reflected in COP dynamics and plantar pressure characteristics and were largely independent of LBP status. Age showed statistically significant associations with several parameters; however, effect sizes were minimal, and coefficients of determination were low. Subjects with LBP walked more slowly with reduced step and stride lengths, lower cadence, and attenuated increases in COP velocity. These differences were most pronounced under fast walking conditions. The combined analysis of COP and plantar pressure measures provided important complementary insights. In particular, the medial midfoot regions emerged as sensitive to LBP-related adaptations, with reduced loading and earlier peak timing observed in patients with LBP compared to controls while pressure distribution in the forefoot tended to be displaced to the lateral aspect of the foot. Additionally, parameters such as step width and foot progression angle, despite limited standalone effects, proved informative when interpreted alongside COP and plantar pressure data, especially for understanding mediolateral control and gait stability. Conclusion - This thesis demonstrates that walking speed is the dominant determinant of gait biomechanics and must be rigorously controlled or normalised in both research and clinical practice. Fast walking conditions should therefore also be included to improve the sensitivity and interpretability of gait assessments, particularly in pathological subjects. Sex also contributes to distinct biomechanical gait strategies, while age plays a relatively minor role when speed is accounted for. LBP is associated with subtle and largely velocity-dependent gait adaptations rather than profound biomechanical dysfunction. These findings support the interpretation of LBP-related gait alterations as adaptive motor strategies aimed at reducing mechanical demand and enhancing stability, rather than as indicators of structural or neuromotor impairment. However, heterogeneity and low pathological specificity of included subjects with LBP suggests that specific subgroups may present distinct mechanisms.This work emphasizes the importance of integrating multiple levels of gait analysis, systematically reporting effect sizes to help distinguish statistical significance from clinical relevance. Clinically, these findings support an approach to rehabilitation focused on improving movement efficiency, adaptability, and tolerance to dynamic demands, rather than strictly normalising gait parameters. The identification of subtle midfoot and forefoot loading alterations further suggests potential avenues for targeted interventions.Overall, this thesis contributes to a more nuanced and integrative understanding of gait in LBP and provides practical methodological and clinical recommendations for future research and application.