Reliability and accuracy in three-dimensional gait analysis: a comparison of two lower body protocols.

The standard Plug-in-Gait (PiG) protocol used in three-dimensional gait analysis is prone to errors arising from inconsistent anatomical landmark identification and knee axis malalignment. The purpose of this study was to estimate the reliability and accuracy of a custom made lower body protocol (MA) compared with the PiG protocol. Twenty-five subjects volunteered to evaluate the intertrial reliability. In addition, intersession reliability was examined in 10 participants. An indirect indicator of accuracy according to the knee varus/valgus and flexion/extension range of motion (ROM) was used. Regarding frontal plane knee angles and moments as well as transverse plane motions in the knee and hip joint, the intersession errors were lower for the MA compared with the standard approach. In reference to the knee joint angle cross-talk, the MA produced 4.7° more knee flexion/extension ROM and resulted in 6.5° less knee varus/valgus ROM in the frontal plane. Therefore, the MA tested in this study produced a more accurate and reliable knee joint axis compared with the PiG protocol. These results are especially important for measuring frontal and transverse plane gait parameters.