Secondary Logo

Journal Logo


Individuals with Unilateral Transtibial Amputation and Lower Activity Levels Walk More Quickly when Using a Hydraulically Articulating Versus Rigidly Attached Prosthetic Ankle-Foot Device

Barnett, Cleveland T. PhD; Brown, Olivia H. M Eng; Bisele, Maria M Eng; Brown, Marcus J. M Eng; De Asha, Alan R. PhD; Strutzenberger, Gerda PhD

Author Information
Journal of Prosthetics and Orthotics: July 2018 - Volume 30 - Issue 3 - p 158-164
doi: 10.1097/JPO.0000000000000179
  • Free


Prosthetic ankle-foot devices that incorporate hydraulically damped articulation between the pylon and prosthetic foot are a relatively new development in prosthetic technology, having only been widely commercially available for approximately 10 years. Feet such as the Kinterra (Freedom Innovations, Morgan, CA, USA), Echelon (Chas. A Blatchford & Sons, Basingstoke, UK), and MotionFoot (Fillauer, Chattanooga, TN, USA) combine a hydraulic articulation unit with an energy-storage-and-return (ESR) foot and are primarily intended for use by individuals with higher levels of mobility, such as those classified as being at least K3 on the Medicare Functional Classification Level (MFCL). The hydraulic dashpots present in the articulation mechanism of such prosthetic ankle-foot devices cause the ankle-foot system to absorb more and return less energy during stance than does an identical rigidly attached foot.1 In addition, because of the hydraulic component, such ankle-foot devices also weigh more than comparable rigidly attached feet.

Despite these apparent drawbacks, it has been reported that hydraulic ankle-foot devices provide functional benefits during walking when compared with feet that are either attached without articulation or attached via an elastic articulation device. The primary reported functional benefit when using a hydraulic ankle-foot device in more active (K3) individuals with lower-limb amputation is an increase in the individual's walking speed.1–4 Walking speed is a primary measure of gait function in individuals with lower-limb amputation.5 Furthermore, for individuals with a lower-limb amputation, an increase in walking speed reflects improved gait function during and after rehabilitation,6–8 and is also associated with decreased temporal asymmetry.9 This increase in walking speed when using a hydraulic ankle-foot device seems to be driven by a reduction in inappropriate fluctuations of center of pressure (COP) progression during prosthetic-limb stance,2 where the COP becomes stationary or travels backward beneath the prosthetic hind and/or mid-foot.10,11 In addition, hydraulic articulation has been found to result in increased forward angular velocity of the prosthetic shank during early stance.2 These effects occur despite the devices' hydraulic dashpots dissipating energy during stance, resulting in reduced energetic efficiency compared with that of a rigidly attached foot.1 Accordingly, the increased walking speed seems to be the result of a reduced “braking effect”12 rather than increased propulsion, allowing the transfer of weight onto the prosthetic limb to occur more smoothly.2 Another effect of using a hydraulic ankle-foot device is a reduction in load-bearing asymmetry during walking1 that possibly contributes to a reported reduction in in-socket pressures due to reduced loading rates.13

The effects of using a hydraulic ankle-foot device have been observed in individuals with both unilateral transtibial amputation (UTA) and transfemoral amputation,14 although, again, only in patients who are described as being at least K3 on the MFCL. Individuals who are less mobile are seldom prescribed ESR feet and therefore rarely use feet with hydraulicankle” function. However, the apparent benefits of using hydraulic ankle-foot devices observed in more mobile individuals may also occur in the less active ones. This suggestion is supported by a low-activity group self-reporting improvements in their gait and prosthetic satisfaction when their prosthetic prescription was changed to include a hydraulic ankle-foot device.15

Therefore, the aim of this study was to investigate the effects of using a non-ESR foot with a hydraulic attachment during overground walking compared with an identical rigidly attached foot in individuals with UTA described as being K2 on the MFCL scale. It was hypothesized that (1) when using the hydraulic ankle-foot device, individuals would walk faster compared with when using an identical rigidly attached foot device. It was expected that any increase in walking speed would be due to the same drivers previously reported in more active individuals when using a prosthetic ankle-foot device that incorporates a hydraulically articulating attachment. Thus, it was also hypothesized that (2) there would be an increased minimum forward/peak backward velocity of COP progression beneath the prosthetic foot, increased angular velocity of the prosthetic shank during early stance, and a reduction in stance phase load bearing asymmetry between the intact and residual limbs when using the hydraulic device compared with a rigidly attached ankle-foot device. Finally, it was expected that these effects would occur despite a reduction in efficiency of the ankle-foot device due to the hydraulic unit dissipating energy during stance.



Five individuals with UTA currently assessed as being K2 on the MFCL scale by their prescribing physician were recruited from the same prosthetic limb and rehabilitation center. All provided written informed consent before to participation in the study, which was approved by the Nottingham Trent University Human Research Ethics Committee (Table 1).

Table 1
Table 1:
Individual participant characteristics

Participants were recruited and included if they 1) were community-living adults aged between 18 and 65 years of age, 2) were able to walk without walking aids for periods of at least 2 minutes, 3) partook in physical activity at least once a week for 30 minutes, 4) had good vision (corrected if necessary), and 5) had no unresolved health issues, as determined using a health screening questionnaire. Individuals were excluded, if they 1) had experienced an unintentional fall in the previous 12 months, 2) experienced undue pain while walking, 3) were current smokers, or 4) were currently taking five or more prescribed medications.


Participants were required to complete the below-described walking tasks while using the same habitual socket/liner and same non-ESR foot, attached under two different conditions: 1) using a hydraulically articulating attachment (HYD—AvalonK2) and (2) using a rigid nonarticulating attachment (RIG—Navigator; both Chas A. Blatchford & Sons, Basingstoke, UK; Figure 1). These feet were chosen as they are identical, save for the nature of attachment to the prosthetic pylon. To ameliorate any order effects, the order in which participants completed walking tasks were counterbalanced across participants according to which was their habitual device, regardless of whether this was the RIG or HYD. For each condition, participants initially completed a familiarization trial followed by a measured trial of the two-minute walk test (2MWT). The 2MWT comprised two 15-m straight sections with a 180° turn at either end, to mitigate the effects of turning on walking test performance.16,17 Participants then completed 10 discrete overground walking trials along a 15-m instrumented walkway including two force plates. Participants were instructed for all tests to walk as they would normally. The same highly experienced prosthetist made all necessary adjustments to all participants' prostheses when changing between prosthetic conditions. Other than different ankle-foot device attachment, there was no difference in the prostheses between conditions. Participants were asked to complete the Activities-Specific Balance Confidence Scale (ABC),18 the Houghton Scale of prosthetic use (Houghton),19 and the Prosthetic Limb Users Survey of Mobility (Plus-M 12),20 which are self-report questionnaires providing information on participants' self-perception of balance confidence, prosthetic use, and mobility, respectively. Higher scores on these scales reflect increased balance confidence (ABC), prosthetic use (Houghton), and mobility (Plus-M 12).

Figure 1
Figure 1:
Technical schematic diagrams of the hydraulically articulating attachment (HYD—AvalonK2) rigid nonarticulating attachment (RIG—Navigator; both Chas A. Blatchford & Sons, Basingstoke, UK) used in the current study.


Participants attended data collection sessions wearing comfortable clothing and their normal everyday shoes. To define a seven-segment model of the lower limbs (feet, thighs, and shanks) and pelvis, reflective markers (14 mm diameter) were affixed bilaterally to participants at the following locations: first and fifth distal metatarsal heads, lateral border and anterior aspect of the foot, calcaneus, medial and lateral malleoli, and femoral epicondyles and anterior and posterior superior iliac spines. A rigid cluster of four markers was also affixed to the lateral side of each shank segment. Foot markers were placed over the shoes. Marker placement on the residual/prosthetic limb was estimated from anatomical landmarks on the intact limb,6 with the prosthesis being modeled as a unified deformable segment.21 Participants commenced each 2MWT trial by standing at the end of the walkway and were free to self-select a turning direction. The number of strides taken by each participant during the 2MWT was recorded by an investigator using a hand tally counter, and the two-minute walk distance (2MWD) was recorded. Participants then completed the overground walking trials at a self-selected speed, for which start positions were adjusted to ensure a clean contact with the force platforms without any obvious targeting or adjustment to stride pattern. A nine-camera motion capture system (Oqus 400; Qualisys AB, Gothenburg, SE) and two force plates (OR6-7; AMTI, Watertown, MA, USA) recorded kinematic and kinetic data at 100 Hz and 500 Hz, respectively. A static calibration was performed by collecting the kinematic data of each participant standing in the anatomical position. Participants were afforded rest breaks as and when required.


Each 2MWT trial yielded outcome measures of 2MWD: walking speed (m/second), determined by dividing the recorded 2MWD by 120 seconds, and the number of strides (stride count). To obtain other variables, biomechanical data for the 10 overground walking trials were analyzed. The raw kinematic data were interpolated using a cubic spline algorithm and both the kinematic and kinetic data were smoothed using a zero-lag Butterworth filter with a 6 Hz cutoff frequency (Visual3D, C-Motion, Germantown, MD, USA). Heel strike and toe-off were defined as ascending and descending thresholds of 20 N in the vertical component of the ground reaction force, respectively. The following biomechanical outcome measures were calculated: 1) load bearing symmetry, defined as the ratio of the peak vertical component of the ground reaction force during intact and prosthetic limb stance; 2) peak shank rotational velocity, defined as the peak angular velocity of the prosthetic shank in the sagittal plane from prosthetic heel strike until intact toe-off; 3) minimum COP velocity, defined as the minimum forward or peak backward (in the direction of travel) velocity of the COP during prosthetic limb stance; and 4) prosthetic energetic efficiency, defined as the ratio of energy absorbed and energy returned by the prosthetic foot device during prosthetic limb stance. Energy absorbed and returned was defined as the positive and negative integrals, respectively, of unified deformable segment power during prosthetic limb stance.21 For each participant, the outcome variables were calculated for each trial, in each prosthetic condition, and the mean for each condition was computed using the results from each trial. No inferential statistical analyses were made; rather, the results for each participant in each prosthetic condition are presented. This approach was taken due to participants' reaction to altered prosthetic componentry being an individual response4 and the small group size. Effect sizes (Cohen d) were calculated using group mean and standard deviation differences between prosthetic conditions.22 An effect size of 0.4 or greater was operationally defined as being clinically meaningful in the current study.23



During the 2MWT, participants walked, on average, with a 6.5% increase in self-selected walking speed (d = 0.4; Figure 2) and thus an increased 2MWD (d = 0.4; Table 2) when using the HYD compared with the RIG device. This increase in walking speed and 2MWD using the HYD device was present across all participants. The number of strides taken during the 2MWT also increased using the HYD when compared with the RIG device in all participants, although not to the same extent as the walking speed, with, on average, a 3.9% increase (d = 0.3; Table 2).

Figure 2
Figure 2:
Individual participant and group mean ± SD walking speeds from the two-minute walk test trials.
Table 2
Table 2:
Two-Minute walk test outcome measures for each participant, group mean ± SD, and between-condition effect size (d) when using both the RIG and HYD attachments


All participants' load bearing was more symmetrical between limbs (d = 0.8; Table 3) when using the HYD compared with RIG device. Similarly, peak shank rotational velocity (d = 1.0; Table 3) increased for all participants except one, and minimum forward COP velocity (d = 0.8; Table 3) increased for all when using the HYD compared with RIG device. The HYD device tended to absorb more energy and return less energy during stance phase, which resulted in a reduced prosthetic energy efficiency for all participants when using the HYD device compared with the RIG device (d = 0.7; Table 3).

Table 3
Table 3:
Biomechanical outcome measures for each participant, group mean ± SD, and between-condition effect size (d) when using both the RIG and HYD attachments


The aim of the current study was to investigate the effects of using a hydraulically articulating “ankle” attachment versus a rigid nonarticulating attachment with a non-ESR prosthetic foot on gait performance during level gait in individuals with UTA described as being K2 on the MCFL scale by their physician. The first hypothesis, that when using the hydraulic ankle-foot device, individuals would walk faster compared with when using an identical rigidly attached foot device, was supported. Every participant in the current study walked more quickly, on average 7.8 m further, during the 2MWT when using the hydraulic ankle-foot device compared with when using the rigidly attached foot. In addition, post hoc analysis indicated that during the 10 discrete trials, participants' mean (SD) walking speed was greater using the hydraulic ankle-foot device compared with using the rigidly attached foot (HYD, 1.19 [0.09] m/second; RIG, 1.16 [0.10] m/second). In addition, walking speed was greater in both prosthetic conditions during discrete trials compared with during the 2MWT. This observation of increased walking speeds during discrete trials versus continuous walking in the current study are consistent with previous reports from healthy individuals.24 Increases in walking speed have been previously demonstrated in individuals with UTA with higher levels of physical function.1–4 One cohort study of more active individuals2 reported a 7% increase (d = 0.5) in self-selected walking speed when using a hydraulicankle” device, which is similar to the results of the current study (6.5% increase, d = 0.4). Thus, we feel that, despite the current study being a case series rather than cohort study, it demonstrates that hydraulicankle” function also seems to benefit those defined as having a relatively low level of activity.

Increased walking speed is positively correlated with improved self-efficacy of gait among individuals with lower-limb amputation.25 Every participant in the current study walked more quickly with the HYD and stated a preference for using the HYD rather than the RIG device. This preference corroborated a previous report of improved user satisfaction when using an AvalonK2 ankle-foot device.15 After data collection, all were offered whichever ankle-foot device (HYD or RIG) they preferred (if it was not their currently prescribed device), to be provided to their prosthetist for subsequent fitting. Four of the five participants, whose currently prescribed device was the RIG, opted for the HYD. The fifth participant, who used a HYD before data collection, retained that device. Anecdotally, but interesting nonetheless, after data collection, all participants were asked whether they had felt as if they were walking faster when using one ankle-foot device in particular and all said no. It is well documented that self-selected walking speed is related to minimizing energy expenditure (e.g., McNeill Alexander26). Although joint kinetics were not outcome variables in the current study, previously it has been reported that use of a similar HYD, but with an ESR foot, resulted in a reduction in mechanical work per meter traveled at the intact limb in more active individuals,2 which possibly contributed to a significant reduction in metabolic cost because of the function of the HYD.14 Although no supporting data are presented from the current study, it may be postulated that, given the increase in walking speed and lack of awareness of such among participants, use of a HYD device has similar effects in less mobile individuals too. This suggestion should certainly be the subject of future research.

The second hypothesis related to the biomechanical explanation of the predicted increased walking speed associated with the hydraulically articulating ankle-foot device. The hypothesis that there would be an increased minimum forward/peak backward COP velocity beneath the prosthetic foot, increased angular velocity of the prosthetic shank during early stance, and a reduction in stance phase interlimb load-bearing asymmetry was supported in most of participants. These findings were consistent with those previously reported in higher-activity individuals with the same level of amputation walking using devices with similar functions1,2 and go some way in explaining the increases in walking speed observed in the current study. The increased energy absorbed and dissipated rather than returned by the hydraulic dashpot present in the HYD may have allowed the individuals to load the residual limb to a greater extent. This was reflected in the increase in interlimb loading symmetry, without the requirement for this energy to be attenuated by deformation in the remaining proximal biological joints and/or structures, for example, biological knee joint, residuum-socket interface. There are no supporting data; thus, this is a speculation, but this could be the driver of reduced in-socket pressures reported when using a hydraulic ankle-foot device.13 In addition, the improved forward COP progression and shank angular velocity (displayed in all participants except P1) when using the hydraulic ankle-foot device reflected smoother center of mass progression during prosthetic stance. The increase in minimum forward COP velocity in all participants (including P2, although remaining marginally negative) also reflected a reduction in the “dead spot” reported by some individuals with lower-limb amputation, as progression over the prosthetic limb is interrupted during stance phase. When considered together, these factors point to an overall reduced “braking effect,”12 particularly during early stance, when using the hydraulic ankle-foot device. It would seem that shifting the functional requirements from the biological structures to the mechanical device during early stance is potentially beneficial, where individuals exchange the static stability of the nonarticulating rigid ankle-foot device for the dynamic ability of the hydraulically articulating ankle-foot device. Future research should attempt to investigate whether similar effects are observed in the same patient group when performing other commonly encountered activities of daily living such as stepping, stair ascent/descent, and walking on slopes and uneven surfaces.

There were only five participants in the current study; however, research has shown that reactions to a change of prosthetic device are specific to the individual.4 Often, within a cohort study, a significant group effect is observed between conditions, whereas some individuals within the group display no reaction or the opposite reaction. The increased walking speed when using the HYD was present for all participants. Likewise, the biomechanical differences that occurred between device conditions were consistent, and almost ubiquitous, across participants. Only one individual did not present increased angular velocity of the prosthetic shank when using the HYD. All others responded as hypothesized across all outcome variables. Therefore, despite the small sample size, we feel that the findings from the current study are of clinical relevance at both the individual level and also to national health care providers. The demonstrated increases in walking speed suggest that improved mobility in an individual may be achieved via prescription. This increased mobility could possibly lead to subsequent improved completion of daily tasks and/or engagement in social activities. In addition, given that patients themselves previously reported a perceived benefit of such devices to mobility and prosthetic satisfaction,15 this could suggest that widespread use of such devices may be beneficial to the wider body of less active individuals living with UTA. However, before the widespread adoption of such devices, the long-term effects and potential benefits of such hydraulically articulating ankle-foot devices to both the individual and health care systems must be established and should be the focus of future investigation.


There are a number of limitations to the current study, the most obvious of which is the size of the sample, which was limited to include only individuals who used the specified components to prevent any differences being due to the foot itself rather than the change in attachment. Although the sample size was only five individuals, prosthetic prescription is made on an individual basis. For every participant in the current study, large and consistent effects were observed; thus, the authors feel that the presented results are still valid. The authors do, however, acknowledge that confirmatory future research should attempt to assess whether these magnitudes of effect are maintained in the wider patient population. Also, the effects observed in the current study were acute (same day) and do not speak to any long-term effects. This begs the question as to whether these differences would be maintained over longer periods of time and what the subsequent influences would be on physical activity and quality of life. This is not answerable by the current study but warrants further investigation. Finally, a highly experienced prosthetist with knowledge of all of the components made all adjustments in the current study. However, where this is not possible, it remains to be seen if similar effects would be observed.


Individuals with unilateral transtibial amputation who are described as K2 by their prescribing physician walk faster when using a non-ESR foot with a hydraulically articulating attachment when compared with an identical foot with a rigid, nonarticulating attachment. This improvement in walking performance can be partially explained by a reduced “braking effect” in early stance as a result of the action of the hydraulic component present in the articulating attachment.


The authors thank Chas. A. Blatchford & Sons for providing prosthetic components and prosthetist assistance in the current study. Chas. A. Blatchford & Sons had no role in the study design, data collection, analysis and interpretation, manuscript writing, or decision to submit for publication.


1. De Asha AR, Johnson L, Munjal R, et al. Attenuation of centre-of-pressure trajectory fluctuations under the prosthetic foot when using an articulating hydraulic ankle attachment compared to fixed attachment. Clin Biomech (Bristol, Avon) 2013;28:218–224.
2. De Asha AR, Munjal R, Kulkarni J, Buckley JG. Walking speed related joint kinetic alterations in transtibial amputees: impact of hydraulic ankle damping. J Neuroeng Rehabil 2013;10:107.
3. Johnson L, De Asha AR, Munjal R, et al. Toe clearance when walking in people with unilateral transtibial amputation: Effects of passive hydraulic ankle. J Rehabil Res Dev 2014;51:429–437.
4. De Asha AR, Barnett CT, Struchkov V, Buckley JG. Which prosthetic foot to prescribe?: biomechanical differences found during a single-session comparison of different foot types hold true 1 year later. JPO J Prosthet Orthot 2017;29:39–43.
5. Waters RL, Perry J, Antonelli D, Hislop H. Energy cost of walking of amputees: the influence of level of amputation. J Bone Joint Surg Am 1976;58:42–46.
6. Barnett C, Vanicek N, Polman R, et al. Kinematic gait adaptations in unilateral transtibial amputees during rehabilitation. Prosthet Orthot Int 2009;33:135–147.
7. Barnett CT, Polman RC, Vanicek N. Longitudinal kinematic and kinetic adaptations to obstacle crossing in recent lower limb amputees. Prosthet Orthot Int 2013;38:437–446.
8. Barnett C, Polman RC, Vanicek N. Longitudinal changes in transtibial amputee gait characteristics when negotiating a change in surface height during continuous gait. Clin Biomech (Bristol, Avon) 2014;29:787–793.
9. Nolan L, Wit A, Dudziński K, et al. Adjustments in gait symmetry with walking speed in trans-femoral and trans-tibial amputees. Gait Posture 2003;17:142–151.
10. Ranu HS. An evaluation of the centre of pressure for successive steps with miniature triaxial load cells. J Med Eng Technol 1988;12:164–166.
11. Schmid M, Beltrami G, Zambarbieri D, Verni G. Centre of pressure displacements in transfemoral amputees during gait. Gait Posture 2005;21:255–262.
12. Silverman AK, Neptune RR. Muscle and prosthesis contributions to amputee walking mechanics: a modeling study. J Biomech 2012;45:2271–2278.
13. Portnoy S, Kristal A, Gefen A, Siev-Ner I. Outdoor dynamic subject-specific evaluation of internal stresses in the residual limb: hydraulic energy-stored prosthetic foot compared to conventional energy-stored prosthetic feet. Gait Posture 2012;35:121–125.
14. De Asha AR, Munjal R, Kulkarni J, Buckley JG. Impact on the biomechanics of overground gait of using an ‘Echelon’ hydraulic ankle-foot device in unilateral trans-tibial and trans-femoral amputees. Clin Biomech 2014;29:728–734.
15. Moore R. Patient evaluation of a novel prosthetic foot with hydraulic ankle aimed at persons with amputation with lower activity levels. JPO J Prosthet Orthot 2017;29:44–47.
16. Barnett CT, Bisele M, Jackman JS, et al. Manipulating walking path configuration influences gait variability and six-minute walk test outcomes in older and younger adults. Gait Posture 2016;44:221–226.
17. Brooks D, Parsons J, Hunter JP, et al. The 2-minute walk test as a measure of functional improvement in persons with lower limb amputation. Arch Phys Med Rehabil 2001;82:1478–1483.
18. Powell LE, Myers AM. The Activities-specific Balance Confidence (ABC) Scale. J Gerontol A Biol Sci Med Sci 1995;50A:M28–M34.
19. Devlin M, Pauley T, Head K, Garfinkel S. Houghton Scale of Prosthetic use in people with lower-extremity amputations: reliability, validity, and responsiveness to change. Arch Phys Med Rehabil 2004;85:1339–1344.
20. Hafner BJ, Gaunaurd IA, Morgan SJ, et al. Construct validity of the Prosthetic Limb Users Survey of Mobility (PLUS-M) in adults with lower limb amputation. Arch Phys Med Rehabil 2016.
21. Takahashi KZ, Kepple TM, Stanhope SJ. A unified deformable (UD) segment model for quantifying total power of anatomical and prosthetic below-knee structures during stance in gait. J Biomech 2012;45:2662–2667.
22. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Earlbaum Associates, Inc, Publishers; 1988.
23. Page P. Beyond statistical significance: clinical interpretation of rehabilitation research literature. Int J Sports Phys Ther 2014;9:726–736.
24. Brown MJ, Hutchinson LA, Rainbow MJ, et al. Comparison of self-selected walking speeds and walking speed variability when data are collected during repeated discrete trials and during continuous walking. J Appl Biomech 2017;1–14.
25. Miller WC, Speechley M, Deathe B. The prevalence and risk factors of falling and fear of falling among lower extremity amputees. Arch Phys Med Rehabil 2001;82:1031–1037.
26. McNeill Alexander R. Energetics and optimization of human walking and running: the 2000 Raymond Pearl memorial lecture. Am J Hum Biol 2002;14:641–648.

lower-limb amputation; prostheses; foot device; ankle; hydraulic

Copyright © 2018 American Academy of Orthotists and Prosthetists