Overground versus self-paced treadmill walking in a virtual environment in children with cerebral palsy

- We compared self-paced treadmill walking to walking in a conventional gait lab and in a courtyard.
- Subjects with and without CP walked at similar speed in both labs, but faster in the courtyard.
- Stride width was wider and stride length shorter on the treadmill than overground.
- Kinematic curves were quite similar, but some increased deviation was found in CP on the treadmill.
- Self-paced treadmill walking is a promising tool for kinematic gait analysis in patients with CP.

Treadmill walking offers several advantages for clinical gait analysis and gait training, but may affect gait parameters. We compared walking on a self-paced treadmill in a virtual environment (TM+) with overground walking in a conventional gait lab (CGL), and with natural walking (NW) outside a lab environment on a GaitRite measurement mat, for 11 typically developing (TD) children and 9 children with cerebral palsy (CP). Spatiotemporal parameters and subjective scores on similarity to normal walking were compared between all three conditions, while kinematic parameters and Gait and Motion Analysis Profile Scores (GPS and MAP) were compared between CGL and TM+. Subjects walked slower and with shorter strides in both lab conditions compared to NW. Stride width was 3-4 cm wider in TM+ than in CGL and NW. Mean kinematic curves showed a few differences between CGL and TM+: on the treadmill children with CP walked with on average 2° more pelvic tilt, 7° more knee flexion at initial contact, and more deviating knee and ankle kinematics as indicated by the MAP scores. These differences may in part be due to increased fatigue in TM+ as a result of longer continuous walking time. Our results indicate that differences between self-paced treadmill walking in a VR and walking in a conventional gait lab are generally small, but need to be taken into account when performing gait analysis on a treadmill.