Powered Prosthetic Feet: The Second Chapter

Home > Articles > Powered Prosthetic Feet: The Second Chapter
By Phil Stevens, MEd, CPO, FAAOP

The early clinical reports of the first commercially available powered prosthetic ankle-foot were incredibly promising. Mancinelli et al. first announced their findings in 2011.1 Reporting upon five users of transtibial prostheses during level-ground walking, oxygen consumption was reduced by an average of 8.4 percent and peak late-stance power generation increased by 54 percent.1 Their work was confirmed by a second study of seven transtibial prostheses users, demonstrating a similar decline in metabolic demands of level-ground ambulation
of 8 percent, increasing the positive work of the prosthesis as the trailing leg by 57 percent, increasing preferred walking velocities by 23 percent, and generally bringing measures of metabolic energy costs, preferred walking velocities, and biomechanical patterns in line with able-bodied walkers.2 A short time later, a third clinical trial with 11 subjects suggested the ability of powered ankles to begin to normalize the biomechanics of stair ascent, providing increased peak ankle plantarflexion and push-off power.3

In a fourth publication, published a short time later and reporting upon the same cohort of 11 subjects walking
on level ground, powered ankles were observed to increase ankle range of motion and provide peak ankle power values 125 percent greater than those observed with non-powered carbon feet.4 Increases in self-selected walking speeds over level ground were only slightly lower than those reported in the earliest trials, reported at an average of 6 percent.4 A fifth publication soon followed, reporting upon the ground reaction forces acting upon the unaffected leg of seven users of a powered ankle prosthesis. Peak forces were found to reduce by 2-11 percent at slow and moderate speeds, while external adduction moments acting upon the sound side knee reduced by 12-21 percent at elevated walking velocities, both observations suggesting a role for powered ankle prostheses in reducing the risk of comorbid osteoarthritis of the contralateral knee.5 A sixth publication largely supported these findings, reporting that the use of a powered ankle prosthesis reduced the peak ground reaction forces, loading rates, and external knee flexion moments acting upon the sound side limb during faster walking speeds.6

During these early years, the field welcomed these fairly consistent reports of the benefits associated with the use of powered ankle prostheses. Subsequently a number of more recent reports on this technology have begun to fine-tune the reasonable expectations that should be associated with these devices as they begin to see broader utilization in the prosthetic community. This article reviews recent findings as they pertain to energetic efficiency, self-selected walking speed, and impact forces acting upon the sound side limb.

Push-off Power Fails to Significantly Reduce the Metabolic Cost of Slope Ascent

In one subsequent study, six young military subjects were observed in level-ground ambulation and during navigation of a 5-degree walkway.7 In both exercises, the powered ankle-feet were observed to generate substantially more power than the non-powered carbon fiber feet. During level-ground walking, powered ankles generated 63 percent greater work than the carbon fiber feet (0.275 ± 0.068 J/kg versus 0.168 ± 0.025 J/kg). Notably, the work recorded with the powered ankle also exceeded the work measured on matched controls by 28 percent. This pattern persisted during slope ascent, with the powered ankle prostheses generating 53 percent greater work as the trailing limb relative to carbon feet (0.336 ± 0.118 J/kg versus 0.240 ± 0.106 J/kg).7

While this increased propulsive power led to measured decreases in oxygen consumption for both activities, the improvements were significant for only level-ground walking. In this application, oxygen consumption values were 16 percent lower with the powered ankle prosthesis compared to carbon fiber feet. This change effectively normalized the metabolic rates of the prosthesis users compared to the able-bodied controls. (By contrast, in the carbon fiber foot condition, subjects were found to have 9 percent higher metabolic rates than the same controls).7

However, while the increased push-off work of the powered prosthesis reduced the metabolic demands of slope ascent, the difference failed to normalize oxygen consumption values to those of the controls, and the difference failed to reach significance. In looking at individual subjects, the authors observed two subjects for whom the metabolic costs of slope ascent increased with the use of the powered ankle-feet. In one instance the unit appeared to be underpowered, and in a separate instance the unit appeared to be overpowered. Reflecting on this observation the authors noted that "optimal or near-optimal tuning of the [powered prosthesis] is crucial to obtain a positive effect on [oxygen consumption].7

Powered Push-off Reduced Vertical Force Peaks on the Sound Side Limb During Slope Ascent

In a subsequent study of young transtibial prosthesis users, the impact of powered ankle-feet was considered with respect to peak power generation and the vertical ground reaction forces acting upon the intact limb during slope ascent.8 Consistent with previous reports, the peak plantarflexion power generation at pre-swing with the use of powered ankle-foot prostheses was substantially greater (103 percent) than that obtained with carbon fiber feet and consistent with the values observed in able-bodied controls. By contrast, the peak plantarflexion power generation values recorded with carbon fiber feet were 37 percent lower than those recorded among able-bodied controls.8

With the use of carbon fiber feet, slope ascent required the prosthesis user to generate knee and hip extension to power the body uphill. However, when transitioning off the powered prosthetic limb, the push-off power provided by the powered ankle-foot reduced the demands upon the knee extensors of the intact limb.8

Metabolic Rates Unaffected by Prosthesis Work Rates of Powered Prostheses

A separate publication from the same year is unique among the publications shared in this article in that the powered ankle-foot unit was not commercially available.9 However, like the commercially available unit used in the other cited trials, this technology had the ability to tune the amount of propulsive power generated by the unit.

After allowing acclimation over several days, six subjects were monitored during treadmill ambulation while the push-off power of their prostheses was progressively manipulated from negative and low values to high and highest values. Ultimately, this equated to push-off values that ranged from roughly half to roughly twice the values experienced during able-bodied ambulation. Surprisingly, researchers observed that across this broad variation, net prosthesis work rate did not significantly affect metabolic rate.9 Indeed, none of the observed
participants trended toward lower metabolic rate with increasing ankle push-off work. Also concerning, the collision rate acting upon the intact limb appeared unaffected by the push-off work rate of the prosthesis.9

The restoration and augmentation of ankle push-off was not enough to improve energy economy. The authors identified a number of possible explanations for their observations that conflict with some earlier research findings. These included the push-off timing, subject training, and other subject characteristics. Of note, while most of the subjects of previous clinical trials were extremely young, active, often military subjects, the subjects in this study averaged 47 years of age.9

Metabolic Rates and Preferred Walking Speeds Improved in K4, Not K3 Ambulators

The question of the relationship between powered push-off and improved metabolic efficiencies was further explored in a subsequent clinical trial returning to commercially available powered prosthetic ankle-feet.10 As with the preceding study, the population was much older than those of the early literature on powered feet, with the average age of the ten subjects reported at 46 ± 14.9 years. Metabolic data was collected during sustained ambulation on a level treadmill.10

The authors reasonably hypothesized that energy costs would decrease, and preferred walking velocity would increase with the powered ankle-foot prostheses. This was not the case. Rather, while the authors observed modest decreases in mean metabolic costs (14.5 ± 1.9mL/min/kg versus 14.3 ± 1.7mL/min/kg) and modest mean increases in preferred walking velocities (1.28 ± 0.12m/s versus 1.31 ± 0.14m/s), these differences failed to reach statistical significance.10

Fortunately, the authors then took the time to reexamine their observations by K-level. K4 users demonstrated a mean decrease in their cost of transportation
of 4 percent and a mean increase in their preferred walking velocity of 5.4 percent. By contrast, K3 users experienced a 5.4 percent reduction in their metabolic efficiency with the powered ankle-foot prosthesis and a 1.4 percent decrease in their preferred walking velocity. These findings begin to suggest that some patient types may be more likely than others to experience the improvements in energetics and walking speeds identified in the earliest clinical trials on powered ankle-foot prostheses.10

Push-off Power Provided Variable Benefits on Metabolics During Level Ground and Slope Ascension

The complex relationship between powered push-off and the metabolic cost of walking was further confounded by a final clinical trial.11 As with the last two trials, the cohort was older than those of the earliest studies with a mean age of 42 years among the ten subjects. In contrast to earlier published data, the addition of powered push-off reduced the metabolic costs of ambulation during stair assent but had no measurable effect on the energetic efficiency of level-ground ambulation.11


The most recent literature related to the impact of powered prosthetic ankle-foot mechanisms on self-selected walking speed, the energetic costs of ambulation, and sparing of the sound side limb are complex and occasionally contradictory. Authors repeatedly assert that the restoration of push-off in and of itself does not appear to immediately equate to improved metabolic efficiencies. Several authors were clear in pointing out that the restoration of late-stance propulsive plantarflexion replicates the function of the uniarticular soleus, propelling the body, but not the function of the biarticular gastrocnemius to propel the swing leg forward. In addition, issues associated with tuning the magnitude and timing of that propulsion seem critical. Finally, certain patient factors seem to be in play. While the greatest benefits thus far have been observed in the youngest, most fit, and active users, it would likely be an oversimplification to ascribe candidacy to age. However, general fitness and appropriate training with the technology appear to be key considerations.

Ultimately, this growing body of literature appears to support the common practice of facilitating patient trials to confirm candidacy for the technology prior to final provision of services. Continued research into the promising line of study is needed to better understand how the promise of powered propulsion can be translated into consistent benefits across a broader range of end users.


Phil Stevens, MEd, CPO, FAAOP, is a director with Hanger Clinic's Department of Clinical and Scientific Affairs. He can be contacted at philmstevens@hotmail.com.


1.       Mancinelli, C., B. L. Patritti, and P. Tropea, et al. 2011. Comparing a passive-elastic and a powered prosthesis in transtibial amputees. Annual International Conference of the IEEE Engineering in Medicine and Biology Society 8255-8.

2.       Herr, H. M., and A. M. Grabowski. 2012. Bionic ankle-foot prosthesis normalizes walking gait for persons with leg amputation. Proceedings of the Royal Society B: Biological Sciences 279(1728):457-64.

3.       Aldridge, J. M., J. T. Sturdy, and J. M.Wilken. 2012. Stair ascent kinematics and kinetics with a powered lower leg system following transtibial amputation. Gait & Posture 36(2):291-5.

4.       Ferris, A. E., J. M. Aldridge, C. A. Rábago, and J. M. Wilken. 2012. Evaluation of a powered ankle-foot prosthetic system during walking. Archives of Physical Medicine and Rehabilitation 93(11):1911-8.

5.       Grabowski, A. M., and S. D'Andrea. 2013. Effects of a powered ankle-foot prosthesis on kinetic loading of the unaffected leg during level-ground walking. Journal of Neuroengineering and Rehabilitation 10(1):1-2.

6.       Esposito, E. R., and J. M. Wilken. 2014. Biomechanical risk factors for knee osteoarthritis when using passive and powered ankle-foot prostheses. Clinical Biomechanics 29(10):1186-92.

7.       Esposito, E. R., J. M. Aldridge Whitehead, and J. M. Wilken. 2016. Step-to-step transition work during level and inclined walking using passive and powered ankle–foot prostheses. Prosthetics and Orthotics International 40(3):311-9.

8.       Rábago, C. A., J. Aldridge Whitehead, and J. M. Wilken. 2016. Evaluation of a powered ankle-foot prosthesis during slope ascent gait. PloS One 11(12):e0166815.

9.       Quesada, R. E., J. M. Caputo, and S. H. Collins. 2016. Increasing ankle push-off work with a powered prosthesis does not necessarily reduce metabolic rate for transtibial amputees. Journal of Biomechanics 49(14):3452-9.

10.    Gardinier, E. S., B. M. Kelly, J. Wensman, and D. H. Gates. 2018. A controlled clinical trial of a clinically-tuned powered ankle prosthesis in people with transtibial amputation. Clinical Rehabilitation 32(3):319-29.

11.    Montgomery, J. R., and A. M. Grabowski. 2018. Use of a powered ankle-foot prosthesis reduces the metabolic cost of uphill walking and improves leg work symmetry in people with transtibial amputations. Journal of The Royal Society Interface 15(145):20180