Amputees and non-amputees adapt similarly to split-belt treadmill assistance

Individuals with lower-limb amputation experience higher walking energy cost and discomfort than able-bodied controls when using traditional passive prostheses. Powered prosthetics aim to reduce energy cost and improve comfort, but have not shown empirical metabolic benefits approaching those of comparable exoskeletons or simulation studies. One theorized explanation is that amputees adapt less quickly, and less effectively, to walking assistance than able-bodied participants. Previous work has shown that split-belt walking is an effective test of adaptation to an assistance condition, and that slowing one belt reduces energy cost compared to tied walking at the fast belt speed. The goal of this study was to characterize the metabolic cost of amputee split-belt walking with the prosthetic leg on the slow belt, simulating an assistive device. We compared amputees (N = 10) to matched controls (N = 10) as they walked on a split-belt treadmill at a 5:2 belt speed ratio. Amputees experienced an 18% average reduction in metabolic rate, similar to the 16% reduction for the control group. Similar levels of step length asymmetry were used by both groups. Amputees rated both split and tied walking as more uncomfortable than did the matched controls, but the difference between the ratings was the same for the two groups. These results suggest that amputees are equally capable of exploiting substantial levels of external mechanical assistance. Adaptation ability alone may not explain the low levels of energy cost reduction from current powered prostheses.