Gait Analysis on Split-Belt Force Treadmills: Validation of an Instrument

Tesio L, Rota V: Gait analysis on split-belt force treadmills: validation of an instrument. Am J Phys Med Rehabil 2008;87:515–526. Objective:Gait analysis (GA) is usually performed during overground walking, with ground reactions being recorded through force platforms embedded in the floor. Body kinematics and surface electromyography (EMG) are also recorded. GA on treadmills resting on force sensors (GAFT) is rather uncommon and has been criticized on various grounds (less reliability of force records; biomechanical differences in walking kinematics and dynamics). On the other hand, the treadmill allows very fast recording of data from many successive strides, at known and constant average speeds. The goals of this study were (a) to validate a system for GAFT based on a commercially available “split-belt” force treadmill, (b) to provide evidence that data can be compared with established norms from overground GA, and (c) to lend support to the potential clinical validity of the GAFT approach. Design:The treadmill adopted here allows subjects to walk on two parallel independent treadmills. Each of them is mounted on four 3D force sensors. Eight healthy adults (four women; ages 22–35 yrs) were tested on the treadmill, each at the average walking speed he or she adopted overground (average across subjects: 1.35 ± 0.05 m·sec−1). Ground reactions, hip, knee, and ankle sagittal rotations, torques, power, and surface EMG from four thigh and leg muscles were recorded simultaneously. Results:Results were compared with those relating to a sample of 40 healthy adults (20 women; manufacturer's data, replicating published data) walking on floor-embedded force platforms (age 20–40 yrs; speed 1.33 ± 0.06 m·sec−1). An 8% shorter stride length was adopted on the treadmill. All of the other results matched those obtainable overground. Conclusions:The results suggest that GAFT performed on the tested treadmill is a promising method of GA in a clinical setting.