The Poetry of…Feet?

Home > Articles > The Poetry of…Feet?
By Phil Stevens, MEd, CPO

When researchers conduct studies to evaluate different medical interventions, they can observe countless parameters. However, some parameters will be more relevant to the immediate concerns of individual patients than others. Ultimately, studies that target questions of importance to individual patients may be more helpful in practical clinical decision making. The British Medical Journal refers to such studies as Patient-Oriented Evidence that Matters, or POEMs.1

According to the British Medical Journal, a POEM must satisfy these three criteria:

  1. Address a question that physicians encounter.
  2. Measure outcomes that physicians and their patients care about.
  3. Have the potential to change the way that physicians practice.

When research from the prosthetics profession is considered, it would certainly appear that some outcomes satisfy these criteria more than others. For example, patients may be more concerned about walking speed, energy consumption, and day-to-day functionality than peak joint moments, decelerative loading rates, and temporal symmetry. This article presents some of the findings of recent literature on prosthetic foot and ankle components with an eye toward the POEM's criteria-focusing on those parameters that might be more relevant to patients and practitioners alike.

Good News for Active Patients with Transfemoral Amputations

Photograph courtesy of Endolite.

The authors of our first pair of articles used a rather narrow set of inclusion criteria, creating a fairly homogenous group of subjects. The six subjects had the following in common: all had a traumatic etiology to their transfemoral amputations, walked without assistive devices, and were free from significant cardiac or pulmonary disease. Their ages ranged from 38 to 50 years old, and their limb lengths were also remarkably similar, ranging from 10-12 inches. The patients' height showed little variance, ranging from 5 ft. 8 in. to 6 ft. 2 in. In addition to these similarities, they all presented as established users of Endolite's Multiflex foot, with five of the six patients using the Endolite Intelligent Prosthesis, a microprocessor-regulated knee unit.

The methodology for both papers was a simple A-B design as subjects were observed across a range of performance measures in their legacy prostheses, provided with a new foot, allowed to acclimate to the new condition, and then retested across the same battery of performance indices. In this case, the subjects all received the Ossur Vari-Flex foot following their initial evaluations in the Multiflex foot, and were allowed four to six weeks of acclimation before the collection of the second data set.

Vari-Flex. Photograph courtesy of Ossur.

Endolite describes the Multiflex as a multiaxial foot suitable for K2-level ambulators. By contrast, the Vari-Flex represents the quintessential dynamic pylon/foot system as described by Medicare's L-5980 billing code: "flex-foot system." Among the research questions for these studies were whether or not the introduction of the more dynamic Vari-Flex foot would provide active, above-knee amputees with functional benefits regarding such things as gait speed and symmetry, power generation, and oxygen consumption.

With respect to the first paper, subjects were brought to the gait lab under both testing conditions.2 In each data series, they were asked to walk at their preferred walking speed and a "brisk" speed, defined as "as fast as you can safely walk." For each speed in each session, the averages of three trials were collected.

The observed gait velocities epitomize the frustrations of clinicians, manufacturers, and policy makers alike. Despite the tremendous similarities between the subjects, their responses to the new prosthetic foot were quite variable. This was particularly evident in the observed variances in self-selected walking speeds. As a cohort, velocity increased an average of 7 percent with the Vari-Flex. However, this variance ranged from one subject who actually decreased his preferred walking speed by 4 percent with the more dynamic foot, to two subjects who increased their preferred walking speeds 16 percent and 14 percent, respectively. Variance in "brisk" walking was less striking, with all subjects demonstrating about 4-percent velocity improvement with the dynamic pylon foot.

In addition to walking speeds, a second publication from the same cohort of study participants reported on energy consumption.3 As a rationale, the authors asserted that the ability of a prosthetic foot to store energy through mid stance and release it during terminal stance might equate to energy savings for the individual patient.

To test this hypothesis, the cohort underwent energy-consumption tests during both a baseline assessment with the Multiflex prosthetic foot and following acclimation to the new Vari-Flex prosthetic foot. Oxygen consumption was measured while subjects walked on a treadmill with predetermined speed increases every two minutes. While the authors did not report any specific energy-consumption values, they did report that average oxygen-consumption values were less at all tested gait speeds with the Vari-Flex prosthetic foot. Interestingly, the differences between feet reached statistical significance at lower, rather than higher gait speeds, suggesting that the energy efficiency of a dynamic pylon foot can be experienced at slower velocities.

While the similarities of the patients in these papers may limit the transferability of its findings, it appears that active, transfemoral amputees are likely to increase their walking speed and decrease their energy expenditure with the inclusion of more dynamic feet.

Vertical Shock Absorption: A Question of Comfort?

Clinical experience supports the assertion that when clearance permits, patients generally appreciate the inclusion of vertical shock absorption (VSA) in either the pylon or the foot mechanism itself. We know that patients tend to like this feature, but we don't fully understand why. Fortunately, there is a developing body of literature that attempts to answer this question.

Unlike the previously described research efforts, researchers involved in the next two studies chose to expand their inclusion criteria, increasing their number of subjects but sacrificing some control of variables outside of the specific component of interest. For example, in the first study, the authors recruited only those subjects with a unilateral transtibial amputation and at least six months of experience using a prosthesis. As a result, their ages ranged broadly from 31 to 79, as did their amputation etiology (seven traumatic, three vascular) and their legacy prosthetic foot type.4

While the authors collected a great deal of data, our discussion will center on gait speeds and patient-generated feedback. To that end, subjects were asked to come to the gait lab and walk at several different speeds, including their preferred pace and fastest-possible pace. Following the initial data collection, subjects were fitted with an Endolite telescopic-torsion, shock-absorbing pylon (SAP) and sent home for four weeks of acclimation before a second data set was collected.

The anticipated effects of artificial VSA on speed were uncertain. On one hand, the authors reasoned that by reducing impact forces and their associated pain, subjects might walk more quickly. On the other hand, the absorption of momentum might inhibit forward progression and slow down gait. In the end, it appears that both hypotheses may have been correct. Some subjects demonstrated statistically significant increases in their preferred walking speeds with the addition of VSA, while others demonstrated statistically significant decreases. The same phenomenon was observed with fastest-possible walking speeds. The effect of VSA on walking velocities was ultimately inconsistent.

In addition to an abundance of laboratory-measured data sets, the authors used a questionnaire at the study's conclusion to collect subjective feedback from their patient models. Specifically, subjects were shown 25 statements describing functional changes associated with the inclusion of the SAP and asked whether they agreed, disagreed, or experienced no change. Some of the more striking findings are summarized in Table 1.

Table 1: Patient responses to questions regarding the use of a shock-absorbing pylon (SAP). Summarized from Gard and Konz.4

The addition of VSA was generally equated with improved comfort, especially during walking, easier turning and stepping down from stairs or curbs, equivalent stability, and smoother walking. In contrast, there was little data to support the ideas that VSA decreased pain, facilitated faster walking, or enabled longer walking distances or greater activity levels.

In considering those statements that were generally accepted, the authors noted that there were two vascular subjects who did not like the addition of VSA, claiming that the unit made them feel unstable. Because both of these subjects presented with decreased tactile sensation due to vascular sensory deprivation, the researchers suggested that they may have relied upon the impact forces associated with the legacy prostheses to provide feedback regarding limb loading. Adjunctively, both walked at gait speeds that may have been too slow to derive the benefits of VSA.

A second, more recent study has largely confirmed these observations.5 In this effort, a similar transtibial population was assembled with respect to their average age and amputation cause. At the beginning of this study, however, all subjects relinquished their legacy prosthetic feet and were fitted with a Seattle LightFoot. They were randomized into two groups: one was fit with the Endolite Telescopic-Torsion Pylon, while the other was fit with a standard, rigid pylon. Patients were provided with three weeks of acclimation prior to data collection and then brought to the lab for testing. They were then switched to the alternate pylon condition, provided three additional weeks of acclimation, and brought back for final data collection.

Once again, there was no consistent effect on the average walking speed of the subjects with and without the VSA component. As with the previous study, attempts were made to elucidate patient feedback with respect to issues of performance, pain, and fatigue. Whereas the prior study asked the subjects to compare their perceptions after having experienced both test conditions, the second survey asked subjects to rate their perceptions generally after each test condition (rigid pylon and SAP), without specifically comparing the two test conditions. For perceived performance, subjects indicated their responses on a visual analog scale that was converted to a ten-point scale with 0 indicating strong disagreement and 10 indicating strong agreement. Subject responses are summarized in Table 2. While reported differences in performance between the two conditions largely failed to achieve "statistical significance," it is clear that the SAP condition was consistently associated with higher perceived-performance scores, particularly with respect to the ease of completing daily activities and relative comfort while ambulating on hard surfaces; this is despite the possible ceiling affects associated with such broad statements as "easily do my normal day-to-day activities" and "very comfortable."

Table 2: Patient responses (ranging from 0-10, where higher scores indicated improved performance) to various statements under two testing conditions: (1) With a standard, rigid pylon; (2) With a shock-absorbing pylon (SAP). Summarized from Berg et al.5

Like the previous study, patient perceptions of pain and fatigue in the two conditions were far less striking. Perhaps to the credit of the treating prosthetist, perceived pain values were rather low in both conditions with only a slight bias in favor of the SAP condition. Fatigue values were also quite similar, with patients reporting only slight reductions associated with the SAP.

Based on these collective findings, it appears that the addition of VSA should not reasonably be expected to result in increased gait speed among unilateral transtibial amputees. Similarly, large improvements in reported pain and increases in perceived activity level should not be anticipated. However, walking comfort, descending stairs and curbs, turning, and the general performance of day-to-day activities may be improved in this population with the addition of VSA.

Does Dynamic Response Influence Stability and Mobility?

In the final study, comparisons were sought between the SAFE II foot and the Flex-Walk, representative of Medicare's billing categories L-5972 and L-5981, respectively.6 Researchers recruited 11 subjects for this analysis-a relatively young cohort with traumatic unilateral transtibial amputations who were all using one of the two studied feet prior to the study. Subjects were randomly placed into two groups to determine which foot they would test with first. Following the first test session, the subjects were fitted with the alternate foot and given a minimum of 30 minutes to become, as the authors describe, "familiar and comfortable with the properties of each prosthetic foot." During this brief acclimation period, subjects were encouraged to engage in a variety of tasks including walking on even and uneven terrain, turning, stair climbing, and, if possible, running.

Each testing session included an evaluation of a number of spatiotemporal and kinetic variables, the majority of which will not be discussed in this article. Among the collected data were gait-velocity values. Given the incredibly brief acclimation periods, it is not surprising that these values were essentially unchanged. However, such observations are quite limited in their transferability as they only provide information about the most immediate effects associated with the two foot types; they provide very little insight on the self-selected walking speeds that might be associated with sustained use of a given foot.

In addition to the laboratory-derived data, the authors collected patient feedback across various stability and mobility domains, with each subject reporting his or her perceived functionality associated with each foot type on a 1-10 scale (with 10 being the best possible score). These findings are summarized in Table 3. As with the observations on gait speed, these findings need to be considered against the fundamental limitation of a brief acclimation period. While the cohort's average mobility and ability scores assigned to the Flex-Walk condition were consistently higher than those of the SAFE II foot, the differences were often quite small, and individual preferences were variable. Consistent, substantial preferences for the Flex-Walk over the SAFE II were only noted for the ability to walk quickly and remain stable while standing on a foam surface. These activities appeared to almost immediately discriminate between the two foot types.

Table 3: Patient ratings across various functional domains in the SAFE II and Flex-Walk (ranging from 1-10 where higher scores indicated improved ability/stability). Reported foot preferences for given functional domains. Summarized from Underwood et al.6


While the recent literature on prosthetic foot and ankle components is quite limited, fortunately, much of what has been published addresses practical concerns of both patients and their prosthetists.

Phil Stevens, MEd, CPO, is in clinical practice with Hanger Prosthetics & Orthotics, Salt Lake City, Utah. He can be reached at


  1. Smith R. A POEM a week for the BMJ. BMJ. 2002;325(7371):963.
  2. Graham LE, Datta D, Heller B, Howitt J. A comparative study of conventional and energy-storing prosthetic feet in high-functioning transfemoral amputees. Arch Phys Med Rehabil. 2007;88:801-806.
  3. Graham LE, Datta D, Heller B, Howitt J. A comparative study of oxygen consumption for conventional and energy-storing prosthetic feet in transfemoral amputees. Clin Rehabil. 2008;22:896-901.
  4. Gard SA, Konz RJ. The effect of a shock-absorbing pylon on the gait of persons with unilateral transtibial amputation. J Rehabil Res Dev. 2003;40(2):109-124.
  5. Berg JS, Czerniecki JM, Klute GK. Efficacy of shock-absorbing versus rigid pylons for impact reduction in transtibial amputees based on laboratory, field, and outcome metrics. J Rehabil Res Dev. 2005;42(6):795-808.
  6. Underwood HA, Tokuno CD, Eng JJ. A comparison of two prosthetic feet on the multi-joint and multi-plane kinetic gait compensations in individuals with a unilateral trans-tibial amputation. Clin Biomech. 2004;19:609-616