Meta-analysis of metabolic cost of amputee walking shows only a marginal improvement in cost when using ESR feet compared to conventional SACH feet

Background: Energy storage and return (ESR) prosthetic feet are designed to store strain energy during stance and release that energy during push-off and, hence, promote a more energy efficient gait. However, despite ESR feet being prescribed for approximately thirty years, most previous studies have not provided statistically significant evidence that they reduce the metabolic cost of amputee walking. Aim: We report a meta-analysis of previously published data with the aim of providing a definitive answer to whether ESR feet result in more energy efficient amputee walking than conventional solid ankle cushioned heel (SACH) feet. Method: We normalised data from previously published studies so that these data could be combined to provide a larger sample and, hence, achieve statistically significant results. Since the majority of amputee metabolic cost studies don’t include a non-amputee control group, we normalised the data by dividing the cost of amputee walking with ESR feet by the cost with SACH feet, using data from the same study. This removed protocol bias and thus allowed statistical tests using the combined data from the previous studies. Additionally the normalised data was tested for a relationship with speed; since we hypothesised that ESR feet might perform better at higher speeds. Results: The raw data from previous studies for any given foot type is difficult to interpret due to variations in protocol causing considerable variability amongst the results from different studies (figure A). This highlights the importance of normalising the cost of walking when combining results from different studies (figure B). After doing this, we found a statistically significant difference between ESR and SACH feet. However the magnitude of this difference was relatively minor, with the cost of walking with ESR feet (figure B: dashed line) being 96.8% (p < 0.001) of that with SACH feet (figure B: solid line). No relationship between normalised cost of walking and speed was found. Discussion & Conclusion: These results suggest that, in general, ESR feet are not effective at improving the metabolic cost of amputee walking to a clinically relevant degree. The failure of passive ESR feet to improve the cost of walking is likely related to their inability to plantarflex beyond their neutral ankle angle; so that they are unable to provide significant push-off at the end of stance. Future designs of prosthetic feet should focus on providing push-off in a manner similar to a healthy intact ankle and foot