Journal of Orthopaedic & Sports Physical Therapy

l999;29 (9):526-533

Physiological Measurements of Walking

and Running in People With Transtibial
Amputations With 3 Different Prostheses
Miao-ju Hsu, MA, PT1
David H. Nielsen, PhD, PT2
H. john Yack, PhD, P P
Donald G. Shurr, MA, P7; CP04
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

Study Design: A 3-factor (foot type, speed, and mode of ambulation) repeated-measures

P
ersons afflicted with
experimental design was used. lowerextremity ampu-
Objectives: To compare the differences in energy expenditure, gait efficiency, and relative tation are compelled to
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

exercise intensity in persons with transtibial amputations with various prostheses. endure physical disabil-
Background: There is a need for improved prosthetic designs to accommodate physically ity as well as cope with
active persons with lower-extremity amputations. numerous Gnctional limi&tions.
Methods and Measures: We used progressive speeds of treadmill walking (53.64, 67.05, Adaptation has been facilitated
80.46, 93.87, and 107.28 rnlmin) and running (120.69, 134.1, and 147.51 mlmin) with 3 with the evolution of new designs
different types of prostheses: the Solid Ankle Cushion Heel (SACH) foot, the Flex-Foot (FF), in lowerextremity prostheses. The
and the Re-Flex Vertical Shock Pylon NSP) prosthesis. Five physically active men with ultimate goal is to improve motor
unilateral transtibial amputations served as subjects (aged 31.6 5 4.28 years). function and enhance general gait
Results: The following statistically significant differences (improvements)between the Re-Flex performance while minimizing the
Journal of Orthopaedic & Sports Physical Therapy

VSP versus the FF and the SACH foot were found. Energy cost: walking (5%), running notably increased physiological de-
(11%); gait efficiency: walking (6%), running (9%); relative exercise intensity: walking (4%), mands associated with ambulation
running (5%). However, we found no significant differences between the FF and the SACH. after lower-limb amputation. With
Conclusions: The Re-Flex VSP appears to have a positive effect on energy cost, efficiency, heightened interest in physical fit-
and relative exercise intensity compared with the other prosthetic foot types during walking ness and increased participation
and running. ) Orthop Sports Phys Ther 1999;29:52&533. in sports, the young or physically
active person with amputation is
Key Words: energy expenditure, Flex-foot and Re-Flex VSP prostheses, gait challenged to achieve a high level
efficiency, SACH Foot of function. There is a need for
improved prosthetic designs to ac-
commodate the person with limb
amputation who has an active life-
style.
Master of Arts degree student, Physical Therapy Graduate Program, Collegeof Medicine, University
Because of loss of the lower ex-
of lowa, lowa Ci& lowa.
Director and professor, Physical Therapy Graduate Program, College of Medicine, University of tremity, people with transtibial am-
lowa, lowa city, lowa. putations must deal with the lack
Associate professor, Physical Therapy Graduate Program, College of Medicine, University of lowa, of active dorsiflexion plantar flex-
lowa City, lowa. ion, important components of hu-
District manager, American Prosthetics Inc., lowa Ci& lowa, and Adjunct lecturer, Physical Therapy man gait. The dorsiflexors gener-
Graduate Program, College of Medicine, University of lowa, lowa City lowa.
Institutional Review Board: Human Subjects Review Committee of the College of Medicine at the
ate eccentric forces immediately
University of lowa. after heel contact and provide
Send correspondence to David H. Nielsen, Physical Therapy Graduate Program, University of lowa, foot clearance during midswing."
2600 Steindler Building, lowa City, IA 52242-1008. E-mail: david-nieIsen@uiowa.edu The plantar flexors provide active
 push-off and forward acceleration of the recovering
limb4J0-35 and function as a stabilizer for the knee
during m i d s t a n ~ e .It
~ .has
~ ~ been suggested that the
ankle plantar flexors are responsible for the majority
of the energy generated during walking.s6 The gait
of a person with transtibial amputation is compro-
mised by the lack of dorsiflexor plantar flexor func-
Pistan-cylinda Type Pylon
tion.
Various biomechanical factors have been identified
that distinguish the gait of people with transtibial
amputations from control populations. These factors
may be related to the increased metabolic costs expe-
rienced by people with transtibial amputations.
Among the factors that have been identified are the
decreased positive power at the ankle during push- FIGURE 1. Photograph of the Re-flex VSP.
off: decreased range of motion at the ankle,l6vz5a
bias toward a flexor moment at the knee with exten-
sive cocontra~tion,~~ increased work of the ipsilateral plate during heel contact and early stance with sub-
hip flexors,37and increased demands placed on the sequent energy release during late stance and push-
contralateral lower limb.7J5These factors can influ- ~ f f . ~ . ~ .Hence,
' ~ . " use of the dynamic prosthetic foot
ence the energy expended by disrupting the smooth may enhance physical performance and enable the
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

movement of the trunk and placing greater demands person with amputation to participate in sports and
on muscles that may not work as efficiently to move recreational and other physical activities.
the trunk and lower limb. A recent innovation is the Re-Flex Vertical Shock
Variables commonly used in a physiological analy- Pylon (Re-Flex VSP) prosthesis, which uses a piston-
sis of gait include energy cost or oxygen consump cylinder-type pylon with a vertical leaf spring at-
tion, gait efficiency, and relative exercise intensi- tached to the shank in combination with the FF (Fig-
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

The results of these studies indicated

ty.6~12-14J6~22~233' ure 1). Theoretically, this design should provide in-
that the gait of the person with transtibial amputa- creased energy storing and releasing capacity, trans-
tion is less efficient, requires increased energy cost, lating to enhanced push-off during gait If proven
and produces a higher relative exercise intensity. effective, the Re-Flex VSP should assist accommoda-
These factors usually result in slower ambulation tion to more vigorous physical activity, such as jog-
speeds and, in some cases, may determine if ambu- ging and running. Potentially, the Re-Flex VSP, in ad-
lation is possible, especially distance walking or dition to accommodatingjogging and running, may
higher-velocity ambulation such as jogging or run- also have the advantage of improved gait efficiency
ning. Participation in recreational or sport-related and decreased energy cost and exercise intensity.
Journal of Orthopaedic & Sports Physical Therapy

activities may also be prohibitive. However, to date no research is available to test these
The type of prosthesis is an important factor when hypotheses.
considering ambulation for the person with transtibi- The purpose of this study was to investigate and
a1 amputation, especially for persons with more ac- compare the differences in energy cost, gait efficien-
tive lifestyles. Foot type influences on physiological cy, and relative exercise intensity of multiple-speed
measurements showed controversial results. Several walking and running with 3 different types of pros
studies found no significant difference in energy cost thetic feet: the SACH foot, the FF, and the Re-Flex
between the dynamic foot and the conventional Sol- VSP.
id Ankle Cushion Heel (SACH) foot.5J6.31.32 However,
other studies comparing several different types of dy- METHODS
namic prosthetic feet (the Seattle Foot, the Flex-Foot
[FF], and the Proteor) with the conventional SACH Subjects
foot have shown reduced energy cost and improved
gait efficiency at higher walking vel~cities.~-~:" The Five men with nonvascular unilateral transtibial
reduction in energy cost may be attributed to the amputation were recruited as subjects (see Table 1
simulated active plantar flexion provided by the dy- for descriptive data). According to the Day Activity
namic foot. Scores Inventory? all subjects were classified as physi-
The SACH foot uses a static design, which, al- cally active (10 is the criterion score for being classi-
though providing structural support, does not pro- fied as physically active; the lowest value for current
vide any dynamic function. In contrast, the dynamic- subjects was 12). The subjects were proficient walkers
designed prosthetic foot, such as the FF, allows ener- with the SACH, the FF, and Re-Flex VSP and were
gy to be stored through compression of an internal able to run up to 147.51 m/min.

J Orthop Sports Phys Ther .Volume 29. Number 9. September 1999 527
 TABLE 1. Descriptive data for the 5 male subjects with transtibial ampu- tem, Medgraphics Cardio2 metabolic cart (Medical
tations. Graphics Corp, St Paul, Minn), was used to deter-
Variable* Mean SD Range mine oxygen consumption based on 6O-second aver-
Age (y) 31.6 4.28 27-36 aging of individual breath by breath analysis. The
Body weight (kg) 84.3 10.2 74.5-1 00.9 metabolic cart, CO, and 0,gas analyzer, and the air-
Stump length (cm) 15.3 2.70 11.4-19.1 flow tube, pneumotachograph, were calibrated be-
Prosthesis experience (y) 13.1 7.18 1.5-20 fore each test session according to the manufactur-
SACH experience (y) 9.3 9.07 0.08-20
3.8 3.40 1-8
er's specified calibration protocols. By accepted con-
FF experience (y)
Re-Flex VSP experience (y) 1.5 1.42 0.17-3 vention, gas volume measurements were corrected to
Mass of SACH (kg) 1.7 0.16 1.5-1.95 standard reference values of standard temperature,
Mass of FF (kg) 1.8 0.38 1.3-2.25 pressure, and dry air based on existing laboratory
Mass of Re-Flex VSP (kg) 1.8 0.22 1.58-2.08 environmental conditions. The energy cost per meter
Day Activity Scale (units)+ 32.0 12.43 1245
traveled (mL O,/kg/m) was used as a criterion for
SACH indicates Solid Ankle Cushion Heel; FF, Flex-Foot; and Re-Flex gait efficiency.= The percentage of age-predicted
VSP, Re-Flex Vertical Shock Pylon. maximum heart rate (%APMHR),expressed using
t Activity scale range from 0 to 50. The criterion for being classified as
physically active is a score 2 1 0 . ~ the formula (exercise heart rate/age-predicted maxi-
mum heart rate X 100), served as the index of rela-
tive exercise intensity. Heart rate was determined by
All subjects were screened to assess general health. an electrocardiograph radiotelemetry system, includ-
No subjects with known medical problems (cardiovas- ing a model 78101A Hewlett-Packard FM receiver, a
cular, neuromuscular, or other significant abnormali- model 78100A Hewlett-Packard miniature battery-op
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

ties except amputation) were enrolled. Before test- erated transmitter, a model 78330A Hewlett-Packard
ing, written informed consent was obtained from cardioscope (Hewlett-Packard,Cupertino, Calif), and
each subject in accordance with the Human Subjects a model 611 Quinton singlechannel cardiotachome-
Review Committee of the College of Medicine at the ter (Quinton Instrument Co, Seattle, Wash). The
University of Iowa. electrocardiograph recording system was interfaced
with the Medgraphics metabolic cart. A modified
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

Experimental Design and General Procedures chest manubrium V5 (CM5) recording electrode
lead system was used. Procedures to record the heart
A Sfactor repeated-measures experimental design rate were conducted as previously reported.,,
was used in which all treatments were administered
to each subject. The 3 factors were type of foot pros- Statistical Analysis
thesis (the SACH, the FF, and the Re-Flex VSP);
mode of ambulation (walking and running); and The statistical analysis was performed using the
speed (for walking, 53.64, 67.05, 80.46, 93.87, and Statistical Analysis System for Windows (WINSAS) li-
107.28 m/min; for running, 120.69, 134.1, and brary program provided by the University of Iowa. A
Journal of Orthopaedic & Sports Physical Therapy

147.51 m/min). The progressive incremental ambu- probability level of P < .05 was selected for the de-
lation test was performed on a motordriven tread- termination of statistical significance. Means and
mill (model TM310 Trackmaster; JAS Manufacturing, standard deviations were calculated for each variable.
Carrollton, Tex). Three 3Sminute test sessions (one A 2-way repeated analysis of variance (ANOVA) was
for each foot type) on separate days were required. used to test for main effects and interaction of foot
Testing order was randomly assigned with counterbal- type (the SACH foot, the FF, and the Reflex VSP)
ancing across the subjects. versus speed of ambulation. The analyses on the
The test protocol involved 2 minutes of resting walking and running data were performed separately.
data collection, followed by a 5-minute practice ses- The Tukey adjusted multiple comparison procedure
sion walking at 53.64 m/min and then 7 4minute was used for post hoc follow-up analysis.
walking/running exercise stages. The physiological
measurements were monitored continuously; howev- RESULTS
er, during the last 1 minute, steady-state values were
averaged and used for the data analysis. Testing was For this study, energy cost, gait efficiency, and rela-
followed by a Sminute cooldown recovery period, tive exercise intensity were measured during func-
walking at 67.05 m/min. Testing was terminated if tional walking (from 53.64 to 107.28 m/min) and
the subject felt fatigued or felt he could no longer running (from 120.69 to 147.51 m/min) speeds in 5
safely walk or run at the next higher speed. men with unilateral transtibial amputations with the
The dependent variables included energy cost, gait SACH, the FF, and the Re-Flex VSP. Because of the
efficiency, and relative exercise intensity. Energy cost physical challenges associated with running, 2 of the
was quantified through direct measurement of oxy- subjects were unable to complete the 134.1 and
gen uptake. An on-line automated computerized sys- 147.51 m/min SACH foot tests. For the FF and Re-

J Onhop Sports Phys Ther .Volume 29. Number 9. September 1999

 TABLE 2. Mean energy cost (mL OJkglmin) for the 5 subjects with trans- TABLE 4. Mean %APMHR for the 5 subjects with transtibial amputation
tibial amputation for the walking and running tests.' for the walking and running tests.*
SACH FF Re-Flex VSP SACH FF Re-Flex VSP
speed speed
(kmin) Mean (SD) N Mean fSDI N Mean ISD) N (mlmin) Mean (SD) N Mean (SD) N Mean (SD) N
Walking Walking
53.64 11.40 (1.45) 5 11.54 (1.29) 5 11.02 (1.92) 5 53.64 46.74 (6.33) 5 47.47 (7.60) 5 45.65 (4.50) 5
67.05 13.00(1.72) 5 12.92 (1.52) 5 12.48(1.76) 5 67.05 50.40(7.59) 5 50.1 1 (7.74) 5 48.70(5.37) 5
80.46 15.40 (1.85) 5 15.29 (1.87) 5 14.64 (2.33) 5 80.46 54.22 (8.36) 5 52.93 (8.32) 5 51.86 (5.89) 5
93.87 19.42(2.00) 5 18.68(2.57) 5 17.61(2.77) 5 93.87 58.72 (8.84) 5 58.09 (8.98) 5 55.84 (6.46) 5
107.28 23.27 (2.59) 5 22.77 (3.16) 5 21.83 (3.60) 5 107.28 66.10 (9.37) 5 64.76 (9.57) 5 61.98 (7.86) 5
Running Running
120.69 33.35 (3.70) 5 31.95 (4.20) 5 29.97(4.20) 5 120.69 80.72 (14.09) 5 78.56 (14.02) 5 74.49(9.55) 5
134.1 37.07 (5.50) 3 35.98 (4.96) 4 34.21 (4.08) 4 134.1 78.21 (11.13) 3 78.70 (10.21) 4 77.14(8.89) 4
147.51 40.73 (5.29) 3 39.44 (5.37) 4 36.83 (5.07) 4 147.51 82.80(9.98) 3 83.58 (10.44) 4 81.62 (9.19) 4
SACH indicates Solid Ankle Cushion Heel; FF, Flex-Foot; and Re-Flex * %APMHR indicates percentage of age-prediaed maximum heart rate;
VSP, Re-Flex Vertical Shock Pylon. SACH, Solid Ankle Cushion Heel; FF, Flex-Foot; and Re-Flex VSP, Re-Flex
Vertical Shock Pylon.

Flex VSP feet, 1 subject was unable to complete the

134.1 and 147.51 m/min tests. The statistical analy- foot type by speed interaction. The main effect Fsta-
ses, however, were still valid because incomplete tistics for differences of foot type and of walking
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

block ANOVA was used to accommodate for the un- speed were significant (Table 5). Tukey follow-up
balanced design. The energy cost, gait efficiency, and analysis of foot type indicated that the Re-Flex VSP
relative exercise intensity group walking and running was statistically different compared with the SACH
test data for the subjects are numerically (means and and the FF, and the FF and the SACH were not sta-
standard deviations) presented in Tables 2, 3, and 4, tistically significant.
and graphically presented in Figures 2, 3, and 4, r e As illustrated in Figure 3, gait efficiency (mL O,/
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

spectively. The walking and running results will be kg/m) during walking showed a parabolic (down-
described separately to be consistent with the statisti- wardly oriented concave) curve in all subjects. Re-
cal analysis. garding gait efficiency, the lower the numerical val-
ue, the more efficient the gait. Between foot-type
comparisons for the amputee subjects showed pro-
Walking Tests gressive separation of the gait efficiency values
Energy cost increased with walking speed (53.64 to (SACH > FF > Re-Flex VSP) with increasing walking
107.28 m/min) in a curvilinear fashion for all sub- speed. That is, the Reflex VSP appeared to be the
jects (Figure 2). Between foot-type comparisons most efficient, followed by the FF, and then the
SACH. Once again, ANOVA showed main effect dif-
Journal of Orthopaedic & Sports Physical Therapy

showed progressive separation of the energy cost val-

ues (SACH > FF > Re-Flex VSP) with increasing ferences for foot type and walking speed, with no sig-
walking speed. The differences appeared negligible nificant foot type by speed interaction (Table 5). TU; u-
for the 2 lower walking speeds (53.64 and 67.05 m/
min). However, ANOVA revealed a nonsignificant 0 SACH
A n?
Q Re-Flex VSP
TABLE 3. Mean gait efficiency (mL OJkglm) for the 5 subjects with trans-
tibial amputation for the walking and running tests:
SACH FF Re-Flex VSP
speed
(mlmin) Mean (SD) N Mean (SD) N Mean (SD) N
Walking
53.64 0.21 (0.03) 5 0.22 (0.02) 5 0.21 (0.04) 5
67.05 0.19 (0.03) 5 0.19 (0.02) 5 0.19 (0.03) 5
80.46 0.19 (0.02) 5 0.19 (0.02) 5 0.18 (0.03) 5
93.87 0.21 (0.02) 5 0.20 (0.03) 5 0.19 (0.03) 5
107.28 0.22 (0.02) 5 0.21 (0.03) 5 0.20 (0.03) 5
Running
120.69 0.28 (0.03) 5 0.26 (0.03) 5 0.25 (0.03) 5 Speed (m/min)
134.1 0.28 (0.04) 3 0.27 (0.04) 4 0.26 (0.03) 4
FIGURE 2. Group means for energy cost (mL OJkglmin) for the SACH,
147.51 0.28 (0.04) 3 0.27 (0.04) 4 0.25 (0.03) 4
the FF, and the Re-Flex VSP in 5 subjects with transtibial amputations for
SACH indicates Solid Ankle Cushion Heel; FF, Flex-Foot; and Re-Flex walking (53.64,67.05, 80.46, 93.87, and 107.28 dmin) and running
VSP, Re-Flex Vertical Shock Pylon. (120.69, 134.1,and 147.51 dmin) tests.

J Onhop Sports Phys Ther .Volume 29. Number 9. September 1999 529
 'I
0 s m 0 SRCH
A EE A FF
e Re-Flex VSP ~3 Re-Flex VSP

-- .
40 60 80 100 120 140 160
Speed W m i n
FIGURE 3. Group means for gait efficiency (mL OJkglm) for the SACH,
the FF, and the Re-Flex VSP) in 5 subjects with transtibial amputations for FIGURE 4. Group means for relative exercise intensity (%APMHR) for the
walking (53.64, 67.05, 80.46, 93.87, and 107.28 mlmin) and running SACH, the FF, and the ReFlex VSP in 5 subjects with transtibial amputa-
(120.69, 134.1, and 147.51 mlmin) tests. tions for walking (53.64, 67.05, 80.46, 93.87, and 107.28 mlmin) and
running (120.69, 134.1, and 147.51 mlmin) tests.

key post hoc analysis of foot type indicated that the

Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

Re-Flex VSP was statistically different from the SACH The between foot-type differences for the running
and the FF, but the FF and the SACH were not signif- tests were similar but more exaggerated compared
icantly different. with those seen for the walking tests. Statistical analy-
As illustrated in Figure 4, %APMHR increased with sis showed that foot type and speed had a significant
speed in a curvilinear fashion for all subjects. Be- influence on energy cost, gait efficiency, and
tween foot-type comparisons showed progressive s e p %APMHR, with nonsignificant interactions between
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

aration of the gait efficiency values (SACH > FF > foot type and speed (Table 6). Tukey follow-up analy-
Re-Flex VSP) with increasing walking speed. The dif- sis showed statistically significant main effect differ-
ferences between foot type appeared negligible for ences between the Re-Flex VSP compared with the
the 2 lower (53.64 and 67.05 m/min) walking SACH and the FF. However, the FF versus SACH dif-
speeds. Finally, as with energy cost and gait efficien- ferences were not statistically significant.
cy, there were significant main effect differences in
foot type and walking speed and a nonsignificant
foot type by speed interaction (Table 5). Follow-up DISCUSSION
analysis of foot type demonstrated that the Re-Flex
Journal of Orthopaedic & Sports Physical Therapy

VSP was statistically different from the SACH and the Active plantar flexion is one of the primary losses
FF and that the FF and the SACH were not signifi- experienced in persons with transtibial amputations.
cantly different. Various types of lowerextremity prostheses have been
developed to enable a more functional mode of am-
Running Tests bulation. Foot type design received little attention
during the early stages of prosthetic development.
For the running tests (120.69 to 147.51 m/min), Initially, most lowerextremity prostheses used a static-
the energy cost and relative exercise intensity group type anklefoot design such as the SACH foot. With
data appeared to increase in a linear manner with in- advances in technology and a heightened interest of
creases in running speed (Figures 2 and 4). Gait effi- the amputee to pursue more physically challenging
ciency appeared to be relatively stable (Figure 3). activities, such as physical fitness training and recrea-
However, each variable (energy cost, gait efficiency, tional sports, the energy storing-releasing-designed
and %AF'MHR) was elevated for all running speeds. foot evolved.

TABLE 5. Analysis of variance summary for walking tests.

Energy cost Gait efficiency Relative exercise intensity
Degrees of
Variable freedom F P F P F P
- -

Foot 2 7.74 .0011 10.83 .0001 7.15 .0017*

speed 4 373.32 .0001 21.94 .OW1 127.13 .0001
Foot by speed 8 0.58 .7928 0.41 .go75 0.37 .9303
Statistically significant based on the adopted alpha Iwel, P < .05.

530 J Orthop Sports Phys Ther Volume 29. Number 9 September 1999
 TABLE 6. Analysis of variance summary for running
- tests.
Degrees of
Energy cost Gait efficiency Relative exercise intensity
Variable feedom F P F P f P
Foot 2 10.05 .0007* 19.8 .OOOl 10.76 .0005*
speed 2 27.91 .0001 2.14 .I394 54.68 .0001
Foot by speed 4 0.42 .7933 0.18 ,9442 0.93 .4658
Statistically significant based on the adopted alpha level, P < .05.

Walking Tests subjects. The subjects recruited in our present study

were physically active and adept in adjusting to the
Oxygen uptake normalized to body weight (ex- handicaps of walking with a lowerextremity amputa-
pressed as milliliter of oxygen per kilogram of body tion.
weight per minute) served as the criterion measure Comparing the foot type influence on gait efficien-
of the energy cost and represented the physiological cy, the Re-Flex VSP was significantly more efficient
workload demand of physical exercise. Similar to than the FF and the SACH. It is of interest to note
nonimpaired gait, a positive curvilinear relation be- that the greatest FF versus SACH difference on gait
tween energy cost and speed was seen in the person
efficiencyoccurred at the 2 faster speeds. Again, a
with transtibial amputation. The energy cost in the
possible explanation for the lack of overall FF versus
person with transtibial amputation became more a p
SACH significant difference may relate to the physi-
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

parent with increasing walking speeds (Figure 2).

These results were consistent with the findings of cal adeptness of this particular group of subjects.
Macfarlane et all8 and Nielsen et a1.22 Heart rate expressed as %APMHR was adopted as
Regarding the effect of foot type on energy cost, the criterion index of relative exercise intensity.= Be-
Nielsen et dmshowed that the FF was superior to cause of ease of assessment, relative exercise intensity
the SACH at speeds above 53.64 m/min. The pres has been used to evaluate gait performance in many
ent study showed a similar trend; however, the FF studies. In this study, the graphical profiles of the rel-
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

versus the SACH foot differences were not statistical- ative exercise intensity in persons with transtibial am-
ly significant. A possible explanation was that subjects putation showed a curvilinear relation between the
in the present study were able to adapt to the short- relative exercise intensity and speed, which was in
comings of the SACH foot. Thus, the differential agreement with Nielsen et al.n
functional advantages between the dynamic FF and It has been stated that relative exercise intensity
the static SACH were minimized. The significantly re- for ambulation at the self-selected walking velocity
duced energy cost for the Re-Flex VSP compared to should not exceed 7040% maximal aerobic pow-
the FF and the SACH suggest enhanced design bene- er.I7.= In this study, the %APMHR for the transtibial
fits. These results support the improved energy stor-
Journal of Orthopaedic & Sports Physical Therapy

amputee with the SACH was 46% at 53.64 m/min

ing-releasing properties of the Re-Flex VSP described and 66% at 107.28 m/min, indicating that the stress
in the manufacturer's manual. associated with the transtibial amputee during walk-
Gait efficiency was expressed as energy cost divid- ing was well within tolerable physiological limits. The
ed by v e l o ~ i t yAccording
.~ to this convention, the identified speeds correspond to a functional range of
lower the computed numerical value, the more effi- walking velocities for amputees as well as those with
cient the gait. Our study supports the findings of nonimpaired gait. Similar to the results of energy
previous studies indicating that the relation between
cost, the Re-Flex VSP statistically reduced relative ex-
gait efficiency and speed is a parabolic
ercise intensity compared to the SACH and the FF.
Compared to the nonimpaired subjects studied in
This indicated that the Re-Flex VSP reduced the lev-
previous published research, gait efficiency in a per-
el of stress the most.
son with transtibial amputation is decreased, that is,
the values of the energy cost per meter traveled were On the basis of previous resear~h,2'.~~ the energy
higher.4J2.m.24 The results of previous studies indicat- storing-releasing properties of the dynamic feet (the
ed that the optimal gait efficiency in nonimpaired FF and the Re-Flex VSP) appeared to be speed de-
gait usually occurs at a higher speed (83.14 m/min) pendent. The results of our study support this find-
compared to the transtibial amputee gait, with values ing. The FF and the Re-Flex VSP tended to minimize
reported in the literature ranging from 48.28 to the speeddependent increase of energy cost seen at
69.73 m/min.12J4~27~3Wptimal gait efficiency (self-se- the higher walking speeds with the SACH foot. How-
lected walking speed) in persons with transtibial am- ever, the FF appeared to be less adapting at higher
putations in our study occurred at approximately walking speeds compared with the Re-Flex VSP. The
80.46 m/min. This inconsistency in findings may be Re-Flex VSP appeared to be well suited for the physi-
attributed to differences in the activity level of the cally active persons with amputation.

J Orthop Sports Phys Ther.Volume 29-Number 9.September 1999

 Running Tests Although the %APMHR values were high (74-83%),
persons with transtibial amputations still met the ac-
Historically, the primary objective in designing a cepted guidelines for tolerable exercise recommen-
foot prosthesis was to enable the person with ampu- dations (6590%).2
tation to regain function well enough to accomme Biomechanical analysis studies in the literature
date to the demands of slow walking. The ability to have shown that ambulation with the static-type foot
accommodate to the demands of running did not produces displacement of body center of gravity and
evolve until recent years. With advances in technolo- an asymmetrical gait compared with the dynamic
gy and a heightened interest of the amputee to pur- foot.3.4J8-263'
Previous studies that have examined the
sue more physically challenging activities involving biomechanics of different prosthetic feet may help
recreational sports and competitive athletics, various explain the findings of the current study. It has been
types of dynamic feet have been developed. The Re- shown that during running, the FF is able to gener-
Flex VSP is one of the newest of these types of feet; ate 2 to 3 times more energy than the SACH foot.8
however, research on its efficacy is lacking. In addition, limited testing of the &-Flex VSP has
As seen in Figure 2, energy cost increased in a lin- shown that greater movement of the pylon occurs as
ear manner with increases in running speed for all gait speed increases, with the greatest effect seen
subjects. These results corroborate previous studies with j~gging.~' This motion translates into greater en-
that also showed a similar linear relation.11.19The ergy storage and release.21Hence as the gait speed
general foot type differences of energy cost during increases there is also an increase in the amount of
running were similar to those seen with high-speed energy that is stored and recovered with the dynamic
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

walking but were more dramatic. Again, the SACH feet. The greater energy generated by the dynamic
appeared to be the least accommodating and the Re- feet at push+ff, particularly during fast walking and
Flex VSP the most accommodating to running. As running, could potentially contribute to the energy
with walking, the running results suggested that in- needed to move the body forward, thereby reducing
creased speed enhanced the energy storing-releasing the requirements of the lower-limb musculature and
properties of the dynamic foot, which was the most hence the metabolic cost
pronounced in the Re-Flex VSP.
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

Although the present study did not include a b i e

Considering that gait efficiency was computed mechanical analysis, the physiological data corrobo-
from the ratio of the oxygen consumption divided by rate the above findings. On the basis of our empiri-
the ambulation speed, and that the oxygen consump cal findings, compared to the conventional static
tion response curve for running should be linear prosthetic foot (the SACH foot), the energy storing-
(Figure 3) ,l 1~19.20the observed relatively constant gait releasingdesigned dynamic prosthetic feet appeared
efficiency values for the running tests appeared to be to accommodate persons with amputation to a less
logical. The SACH foot was the least efficient and energy, lower exercise intensity and more efficient
the Re-Flex VSP the most efficient during running, gait for both walking and running. These results sug-
which mirrored the trend observed during walking.
Journal of Orthopaedic & Sports Physical Therapy

gest that the design benefits are speed dependent

Lehmann et all6 investigated the metabolic de- and become more apparent at speeds above 67.05
mands in the transtibial amputee during running m/min. In all cases, the Re-Flex VSP was superior to
(139.46 to 201.15 m/min) and found no energy sav- the SACH and the FF. In this context, the Re-Flex
ing in the FF compared with the SACH. Torburn et VSP appears to have the most advanced dynamic de-
alJ' compared the FF and the SACH in subjects with sign characteristics and to be the prosthetic foot of
transtibial amputations at self-selected walking speed choice of the 3 designs studied for the physically ac-
and showed no statistical significance in gait efficien- tive person with transtibial amputation.
cy between the FF (0.219 2 0.02 mL 02/kg/m) and
the SACH (0.227 2 0.03 mL 02/kg/m). In the pres
ent study, the findings showed no statistically signifi- Limitations
cant difference in gait efficiency between the FF and
the SACH; however, the trends of mean values on The sample size in the present study was small,
gait efficiency at each running speed (120.69 to and additional research is recommended with more
147.51 m/min) suggested that the FF was more effi- subjects to corroborate the findings of the current
cient. These results substantiate the benefit of the dy- study. Moreover, in this study the results indicated
namic foot in accommodating the persons with am- that the Re-Flex VSP was superior to the FF and the
putation to running. SACH foot. One can only speculate on the underly-
The foot type influence on relative exercise inten- ing reasons for the observed results. A biomechanical
sity during running was similar to that seen with study involving a combined kinetic and kinematic
walking. Again, %APMHR for the ReFlex VSP had analysis would be necessary to identify the important
the lowest value, suggesting that the Re-Flex VSP bet- factors necessary to more definitively explain the re-
ter accommodates to running than the other 2 feet. sults.

532 J Orthop Sports Phys Ther .Volume 29 Number 9 .September 1999

 tad K, deLateur BJ. Comprehensive analysis of energy
CONCLUSION storing prosthetic feet: Flex foot and Seattle foot versus
T h e results suggested that the energy storing-re- standard SACH foot. Arch Phys Med Rehabil. 1993;74:
1225-1231.
leasingdesigned prostheses are more energyefficient 7. Lough LK, Nielsen DH. Ambulation of children with mye-
than the conventional prosthesis. T h e Re-Flex VSP lomeningocele: parapodium versus parapodium with
appears t o have a positive effect o n energy cost, gait Swivel modification. Dev Med Child Neurol. l986;28:
efficiency, and relative exercise intensity compared 489497.
w i t h other prosthetic foot types during walking and 18. Madarlane PA, Nielsen DH, Shurr DG, et al. Transfemo-
ral amputee physiological requirements: comparisons be-
running.
tween SACH foot walking and flex-foot walking. / Pros-
thet Orthot 1997;9:138-143.
ACKNOWLEDGMENT Margarita R, Cerretelli P, Aghemo P, Sassi G. Energy cost
of running. ) Appl Physiol. 1963;18:367-370.
This study was supported by a grant f r o m Flex- 20. McArdle WD, Katch FI, Katch VL. Exercise Physiology En-
Foot Inc, Aliso Veigo, Calif, with Dr David H. Nielsen ergy, Nutrition, and Human Performance. 2nd ed. Phila-
delphia, Pa: Lea & Febiger; 1986.
as the principal investigator.
21. Miller LA, Childress DS. Analysis of a vertical compliance
prosthetic foot. / Rehab Res Dev. 1997;34:52-57.
22. Nielsen DH, Shurr DG, Golden JC, Meier K. Comparison
REFERENCES of energy cost and gait efficiency during ambulation in
Barth DG, Schumacher L, Thomas SS. Gait analysis and below-knee amputees using different prosthetic f e e t 4
energy cost of below-knee amputees wearing six different preliminary report. 1Prosthet Orthot 1988;1:24-3 1.
prosthetic feet. / Prosthet Orthot. 1992;4:63-75. Pagliarulo MA, Waters R, Hislop H. Energy cost of walk-
Downloaded from www.jospt.org at on September 16, 2017. For personal use only. No other uses without permission.

Blair SN, Gibbons LW, Painter P, Pate RR, Taylor CB, Will ing of below-knee amputees having no vascular disease.
J.American College of Sports Medicine's Guidelines for Phys Ther. 1979;59:538-543.
Exercise Testing and Prescription. 3rd ed. Philade1phia, Ralston HJ. Comparison of energy expenditure during
Pa: Lea & Febiger; 1986. treadmill walking and floor walking. Appl Physiol. 1960;
Bowker JH, Kazim M. Biomechanics of ambulation. In: l5:ll56.
Moore WS, Malone JM, eds. Lower Extremity Amputation. Rao SS, Boyd LA, Mulroy SJ, Bontrager EL, Gronley JK,
Philadelphia, Pa: WB Saunders Co; 1989:261-273. Perry J. Segment velocities in normal and transtibial am-
Copyright 1999 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

Breakey J. Gait of unilateral below-knee amputees. Orthot putees: prosthetic design implications. IEE Trans Rehabil
Prosthet. 1976;30(3): 17-24. Engineer. l998;6:219-226.
Casillas J, Dulieu V, Cohen MI Marcer I, Didier J. Bioen- Robinson JL, Smidt GL, Arora IS. Accelerographic, tem-
ergetic comparison of a new energy-storing foot and poral, and distance gait factors in below-knee amputees.
SACH foot in traumatic below-knee vascular amputa- Phys Ther. 1977;57:898-904.
tions. Arch Phys Med Rehabil. 1995;76:3944. Rohrig W. Submaximal Exercise Testing: Floor and Tread-
Colborne GR, Naumann S, Eng P, Longmnir PE, Berbrayer mill Walking [thesis]. University of lowa, lowa City, lowa,
D. Analysis of mechanical and metabolic factors in the 1978.
gait of congenital below knee amputees: a comparison of Saunders JB, Dec M, lnman VT, Eberhart HD. The major
the SACH and Seattle feet. Am] Phys Med Rehabil. 1992; determinants in normal and pathological gait.) Boneloint
71:272-278. Surg Am. 1958;35A:543-558.
Journal of Orthopaedic & Sports Physical Therapy

Czerniecki JM, Gitter A, Beck JC. Energy transfer mecha- Schneider K, Hart T, Zernicke RF, Setoguchi Y, Oppen-
nisms as a compensatory strategy in below knee amputee heim W. Dynamics of below-knee child amputee gait:
runners. ) Biomech. 1996;29:717-722. SACH foot versus Flex foot. ) Biomech. 1993;26:1191-
Czerniecki JM, Gitter A, Munro C. Joint moment and mus- 1204.
cle power output characteristics of below knee amputees Skinner HB, Effeney DJ. Gait analysis in amputees. Am I
during running: the influence of energy storing prosthetic Phys Med. 1985;64:82-89.
feet. / Biomech. 1991;24:63-75. Torburn L, Perry J, Ayyappa E, Shanfield SL. Below-knee
Day HJB.Assessment and description of amputee activity. amputee gait with dynamic elastic response prosthetic
Prosthet Orthot Int. 1981;5:23-28. feet: a pilot study. / Rehab Res Dew 1990;27:369-384.
Dillingham TR, Lehmann JF, Price R. Effect of lower limb
Torburn L, Powers C, Guiterrez R, Perry J. Energy expen-
on body propulsion. Arch Phys Med Rehabil. 1992;73:
diture during ambulation in dysvascular and traumatic be-
647-65 1.
low-knee amputees: a comparison of five prosthetic feet.
Falls HG, Humphrey LD. Energy cost of running and
walking in young women. Med Sci Sports. 1976;8:9-13. ] Rehab Res Dev. 1995;32:111-119.
Fisher SV, Gullickson G. Energy cost of ambulation in Waters RL, Perry J, Antonelli D, Hislop H. Energy cost of
health and disability: a literature review. Arch Phys Med walking of amputees: the influence of level of amputa-
Rehabil. 1978;59:124-133. tion. / Bone joint Surg. 1976;58A:42-46.
Gonzalez EG, Corcoran PJ, Reyes RL. Energy expenditure Wing DC, Hitenberger DA. Energy-storing prosthetic feet.
in below-knee amputees: correlation with stump length. Arch Phys Med Rehabil. 1989;70:330-335.
Arch Phys Med Rehabil. 1974;55:111-119. Winter DA. Concerning the scientific basis for the diag-
Huang CT, Jackson JR, Moore NB, Fine PR, Kuhlemeier nosis of pathological gait and for rehabilitation protocols.
KV, Traugh GH. Amputation: energy cost of ambulation. Physiother Can. 1985;37:245-252.
Arch Phys Med Rehabil. 1979;60:18-24. Winter DA, Eng P. Energy generation and absorption at
Hurley GRB, McKenny R, Robinson M, Zadravec M, Pier- the ankle and knee during fast, natural and slow cadenc-
rynowski MR. The role of the contralateral limb in below- es. Clin Orthop. 1983;175:147-154.
knee amputee gait. Prosthet Orthot Int. 1990;14:3342. Winter DA, Sienko SE. Biomechanics of below-knee am-
Lehmann JF, Price R, Boswell-Bessete S, Dralle A, Ques- putee gait. Biomechanics. 1988;21:361-367.

J Onhop Sports Phys Ther -Volume 29. Number 9. September 1999