Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Kinetic Patterns During Stair Ascent in Patients with Transtibial Amputations Using Three Different Prostheses

H. John Yack, PhD, PT
David H. Nielsen, PhD, PT
Donald G. Shurr, MA, PT, CPO

ABSTRACT

The purpose of the present research was to identify the kinetic strategy used by patients with unilateral transtibial amputations during stair ascent, and to determine the influence of three different prostheses on these kinetics. Five physically active transtibial amputees and five matched control subjects participated in this study. Amputee subjects were randomly assigned each of three prostheses, Flex-Foot, ReFlex VSP, and solid-ankle cushion-heel (SACH) foot. Subjects ascended a portable three-step staircase, with an isolated portion of the first step bolted to a forceplate. Three-dimensional marker positions were collected for the foot, leg, thigh, pelvis, and trunk, bilaterally on the amputee subjects and on the right side of the control subjects.

Marker position data were combined with anthropometric data and forceplate data to perform the kinetic analysis. Net joint moment, power, and work values at the ankle, knee, and hip were determined for three trials for each subject. Peak moment, power, and work values were not different between the nonamputated side and the control subjects with the exception of the ankle, where the amputee subjects had greater power and work values.

A comparison of the amputated and nonamputated sides showed a dominant-knee strategy on the nonamputated side and a dominant-hip strategy on the amputated side. The type of prosthesis was shown to make a difference at the hip, where greater work values were observed when subjects were wearing the SACH foot. The findings of this study may help to guide the rehabilitation of amputees and make a case for selecting an energy-storing prosthesis.

Key Words: prosthetic feet, kinetic strategy, unilateral transtibial amputees

Introduction

Ambulating on stairs is an important aspect of daily activities for many patients with amputations; however, little is known about the kinetic strategies that are used to accomplish this task.¹ During stair ascent, the lower limb functions to not only support and balance the superincumbent weight, but also raise that weight onto the supporting step. For patients with transtibial amputations who use a reciprocal gait pattern, the primary responsibility for raising this weight on the amputated side must be divided between the knee and hip, with the potential for some additional help from the nonamputated side during periods of double support. The foot segment probably has a minimal contribution to the lower-limb energy, however, when energy-storing prostheses are used, there is the potential for some energy recovery. Compared to the traditional solid ankle cushion heel (SACH) foot, the energy-storing feet use an energy-storing flexible keel (FlexFoot) that may be used in combination with a vertical leaf spring and telescoping pylon as part of the shank (ReFlex VSP).^2,3 Theoretically, the ReFlex VSP design offers the possibility of storing and recovering greater energy during movement. One purpose of the present research was to document the net joint moment and power requirements during the stance phase of stair ascent in patients who had unilateral transtibial amputations. A second purpose was to distinguish differences in these patterns that may be attributed to the use of different prostheses.

Review of Literature

A number of studies have investigated kinetics during normal stair ambulation.^4–6 McFadyen and Winter⁵ partitioned the stance phases of stair ascent into three parts: weight acceptance, pull-up, and forward continuance. The patterns for normal stair gait show the dominant role of the knee during weight acceptance and pull-up, with supporting roles played by the hip and ankle. During forward continuance, the ankle has the major role, with relatively little contribution from the knee and hip. Andriacchi et al.⁴ showed somewhat similar patterns for the net joint moment data, but the hip moment had the greatest magnitude during the weight acceptance and pull-up phases.

Two studies from the Pathokinesiology Laboratory at Rancho Los Amigos have examined the stair gait of patients with transtibial amputations.^7,8 In one study, the temporal aspects of stair ambulation were investigated in amputees by using five different prosthetic terminal devices, four energy-storing feet, and the SACH foot.⁷ The only significant difference was less asymmetry during initial double support when subjects used the Flex-Foot compared to the SACH foot. The authors speculated that the greater asymmetry when using the SACH foot was because of the inability of the subjects to advance their body weight over the foot. The authors concluded that patients do not appear to benefit from the use of energy-storing prostheses during stair ambulation.

In the second study from this group, the temporal, kinematic, and electromyographic (EMG) patterns for a trans-tibial group using a Seattle LightFoot on the amputated side were compared to a nonamputee population.⁸ These authors found that ankle angular displacement was reduced in the amputees. They speculated that the decreased ankle range of motion required the amputees to compensate such that there was increased muscle activation patterns seen in the hip (increased 20%) and knee (increased 40%) muscles. EMG data, however, are difficult to interpret and do not allow direct inferences regarding the kinetic strategy that is being used.

We believed that a kinetic analysis would allow for the assessment of the underlying strategy that distinguishes between the lower limbs of amputee subjects, as well as distinguishing between amputee and nonamputee groups. Knowing the moments and powers would permit the responsibility for the work to be distributed among the different joints, thereby enabling a more complete understanding of stair ambulation. It is also possible that a kinetic analysis will enable any potential differences from the prostheses to be distinguished.

Methods

A repeated-measures design was used to compare stair gait with three different prostheses, Flex-Foot (FLX), Re-Flex VSP (RFX), and SACH, during ascent in unilateral transtibial amputees. Biomechanical assessments were used to describe the net joint moment, power patterns, and work values in the lower limbs on the amputated and nonamputated sides. In addition, measurements on the amputee group were compared to a nonamputee control group matched for age and mass.

Subjects

Five male unilateral transtibial traumatic amputees participated in the present study, with a mean age of 31.6 years (range 27–36 years), a mean mass of 85.2 kg (range 75–105 kg), a mean height of 1.81 m (range 1.73–1.88 m), and a mean duration since amputation of 13.1 years (range (1.5–20 years). All subjects were physically active and participated in some recreational or competitive sport. Subjects were proficient at walking with each of the three prostheses and were encouraged to rotate use of the prostheses during the month prior to data collection. A group of five nonamputee control subjects also participated in the present study. Their mean age was 30.6 years (range 24–36 years), mean mass was 81.6 kg (range 71–100 kg), and mean height was 1.81 m (1.78–1.88 m). Informed consent was obtained from all subjects in accordance with University IRB procedures.

Instrumentation

A portable three-step staircase (18 cm riser, 25 cm tread, 76 cm width) with a top platform and an additional low riser bottom step (2.5 cm x 25 cm) was used. The first step was cut out (39 cm width) to allow for an isolated section that was bolted to the forceplate (Kistler). Three-dimensional kinematic data were collected (Optotrak?, Northern Digital Inc., Waterloo, Ontario, Canada) for both sides for the amputee patients, and the right side for the control subjects, as they ascended the staircase. Three noncollinear infrared markers were used to track the foot, leg, thigh, pelvis, and trunk. Marker coordinate data and forceplate data were collected online at 60 Hz and 300 Hz, respectively.

Procedures

Amputee subjects performed the gait activities by using the three prostheses, FLX, RFX, and SACH, that were assigned in a random order. As subjects approached the first step, they stepped on the low riser that was positioned on the floor adjacent to the first step. The low riser simulated an additional step and ensured a more consistent movement pattern as subjects stepped onto the first stair.⁹ The handrail was not used by any of the subjects. After several practice trials, data were collected for a minimum of three trials.

Once stair data were collected, a subject calibration was performed. Data captured with the subject standing (subject calibration) were used to define the transformation matrix between the external marker reference system and the principal axes of each of five body segments used to represent the skeletal system. By using the three infrared emitting diodes on each segment, rigid body representations were created. These representations were then used in digitizing bony landmarks for each segment that represented the skeletal system relative to the locations of the infrared markers.

Data Analysis

Marker position data were combined with anthropometric data for a five-segment model (foot, leg, thigh, pelvis, and trunk). Kinematic data were filtered at 6 Hz by using a fourth-order Butterworth, zero-lag low-pass filter to obtain kinematic and kinetic data (KinGait3, Waterloo). A three-dimensional inverse dynamic solution was obtained, starting with the foot segment, to calculate net joint moments of force and powers at the ankle, knee, and hip.^10,11 Positive work done at each joint was determined by integrating the positive power data.¹² Additional work done by the pylon on the RFX prosthesis was calculated from the linear power (axial force in the leg multiplied by velocity of telescoping pylon). Three gait cycles on both sides for the amputee subjects and right-side data for the nonamputee control subjects were collected. Peak values were determined for the unimodal hip, knee, and ankle moments, and hip and knee powers, and for the bimodal ankle powers. Data for all trials were time normalized. Intrasubject ensemble averages were calculated for the three trials. Intersubject ensemble averages were then calculated.

A two-way repeated measures analysis of variance (ANOVA) was used to test for main effects on prosthesis (FLX, RFX, and SACH) and side (amputated and nonamputated), as well as interactions with an alpha level of .05. A one-way ANOVA was used to compare measures between each lower limb and the control subjects (P = .05). Where appropriate, multiple comparisons were made by using the Student-Newman-Keuls posthoc test.

Results

No difference was found in stance time between the amputated and nonamputated limbs. The mean data over the stance phase for the amputated and nonamputated sides for the ankle, knee, and hip joint moments of force and powers are shown in Figure 1 . Whereas the patterns for both sides are similar, there are marked amplitude differences (Tables 1 and 2 ). At the ankle, the peak magnitude of the moments are reduced on the amputated side (P = .008). Ankle peak power amplitudes on the nonamputated side were greater than the amputated (P = .003) and the control subjects (P = .011), and the control subjects had greater peak power amplitudes then those seen on the amputated side (P = .009).

There were nonsignificant differences at the knee and hip in the peak moments and powers between the nonamputated side and control subjects. At the knee, both the moments and powers are decreased in the amputated compared to the nonamputated lower limb (P = .008, P = .006, respectively). At the hip, both the moments and powers are increased for the amputated lower limb compared to the nonamputated lower limb and the limbs of the control subjects (P = .057, P = .048, respectively).

Work values for both lower limbs over the stance phase are shown in Figure 2 . These data are consistent with the power plots and show greater work being done by the uninvolved lower limb at the ankle and knee (respective differences P = .003, P = .015). At the hip, greater work was done by the involved lower limb (P = .044).

Peak moment values were different for the SACH foot compared to the other two feet (P < .001), but no foot differences were found for the power values (P = .089). On the amputated side greater work was done by the hip extensors when the SACH prosthesis was used (P = .012).

Discussion

The motor strategy for unilateral trans-tibial amputees during the stance phase of stair ascent reveals a dominant-knee extensor strategy on the nonamputated side, similar to the nonamputee control group. The strategy on the amputated side is a hip-extensor dominant strategy. These differences are apparent in both the net knee joint moments and the joint powers and work values calculated from the power data. Differences resulting from the use of energy-storing prostheses were seen at the ankle during late stance and also at the hip, where the moments and work values were greater when the SACH foot was used.

No differences were found in the stance time for the amputated and nonamputated sides or when comparing the amputees to the control subjects. This is in contrast to the work of Torburn et al.,⁷ who found temporal differences between sides, and between amputee and control subjects, during stair gait. For the current study, amputee subjects and controls were more closely matched in age and activity level than in the previous study. Age and fitness level may potentially account for the differences between our study and that of Torburn et al.⁷

The moment and power patterns obtained in the present study for the nonamputee control subjects are consistent with those obtained by McFadyen and Winter.⁵ During the pull-up phase (20% to 52% of stance), the dominant moment was the knee-extensor moment, followed by the ankle and then the hip, with the peaks occurring about the same time (approximately 26% of stance). During the forward continuation phase (52% to 100% of stance), the dominant moment was at the ankle, with the other joint moments having minimal amplitudes. The power data show the knee power during pull-up to be dominant, with an average magnitude of 2.5 W/kg occurring just past 30% of stance. The peak for the hip powers is coincident, but with a much smaller magnitude (0.7 W/kg). The initial positive power peak for the ankle during this interval tends to be even smaller than the hip power (0.5 W/kg), and occurs at a later time (50% of stance). At the ankle, there is a second power peak (2.1 W/kg) during forward continuance (85% of stance).

The moment and power patterns for the nonamputated side are very similar to the patterns obtained for the nonamputee control subjects. The main difference occurred at the ankle, where greater power was generated on the nonamputated side towards the end of the stance phase, as weight transfer occurred. This additional energy generated by the plantarflexors could be used to facilitate weight transfer and help propel the trunk onto the prosthetic limb.

The moment data at the ankle on the amputated side was similar to the nonamputated side and the control subjects. This reflects the fact that weight transfer onto the foot is very similar for both lower limbs. As this weight transfer occurs, it must be countered at the ankle to prevent the ankle from moving into too much dorsiflexion. For the intact limbs, this moment is generated by the plantarflexors. On the amputated side, this moment results from the structural integrity of the prosthesis and is a passive phenomenon. However, the powers at the ankle are different between the amputated and nonamputated sides and control subjects. The greatly reduced power on the amputated side reflects the reality that ankle movement is limited in the prosthesis and the amount of energy generated cannot exceed the amount of energy absorbed, that is, the area under the negative power curve.¹ Thus, whereas it is possible to demonstrate greater energy generation via the energy-storing prostheses, compared to the SACH foot, this still only accounts for between 25% to 40% of the energy that is normally generated.

At the knee, major differences are seen in both the moments and powers when comparing the amputated and nonamputated sides. On the amputated side, the knee does not have a very large role during weight transfer and pull-up. This reflects an apparent strategy by the amputees to keep the superincumbent weight balance over the knee, thereby minimizing the extensor moment. Powers et al.⁸ found that the magnitude of the EMG increased for muscles about the knee in their amputee population. Our analysis shows that the greater activation of these muscles probably reflects an extensive amount of cocontraction at the knee as subjects attempt to balance the superincumbent weight and maintain a substantial amount of joint stiffness. The role of the knee on the amputated side in stair ambulation is, therefore, not dissimilar to the role it has during over ground walking. Here, too, the extensor moment is dramatically reduced or nonexistent.¹³ Amputees appear to have established stability of the knee as the primary goal, leaving the role of energy generation (and in the case of walking, absorption) to the hip and nonamputated side.

At the hip, the consequence of the diminished role of the knee muscles on the amputated side is seen in the increased magnitude of the moments and powers during weight acceptance and pull-up (respective increases of 67% and 133%). This means that the amputees are relying much more on the hip extensor muscle to raise the superincumbent weight. The increased demands placed on the hip extensor muscle group may explain the inability of some amputees to use a reciprocal stair gait pattern. The increased demands at the hip also have implications for strengthening strategies used during rehabilitation of amputees.

Examination of the positive work done at each of the joints substantiates what has been concluded from the moment and power profiles. The work values on the nonamputated side at the hip and knee correspond quite closely with the values for the control group. The work values at the ankle are somewhat higher on the nonamputated side, reflecting the increased work done in late stance.

The positive work values on the amputated side reinforce the obvious differences seen in the kinetic patterns between the two lower limbs and also demonstrate differences that are associated with the different prosthetic devices. The between-prosthetic differences at the ankle are consistent with previous research on walking, showing that there is minimal power generated with the SACH foot.¹⁴ Differences were also seen at the hip, where the work done by the hip extensors on the amputated side was significantly greater when subjects were wearing the SACH foot. Similar but nonsignificant trends were seen on the nonamputated side that might be distilled in a study with more subjects. The differences between the two energy-storing prostheses (FLX and RFX) were also not significant, however, trends in the data suggest the possibility of advantages for using the RFX that might be substantiated with a larger population. These findings document the advantage of energy-storing prostheses in a very concrete way by showing that energy stored by the prosthesis contributes to the work done by the lower limb, thereby reducing the muscular requirements at the hip. This is a most desirable effect that has implications for reducing the physiologic cost of stair ambulation.

It is worth noting that the total energy generated at all three joints is substantially less on the amputated side (Figure 2 ). It is also apparent that the difference between the amputee and control is only partially accounted for by the increased energy demands on the nonamputated side. From a purely mechanical perspective, it would appear that amputees are able to negotiate stairs in a more economic manner than are the control subjects. It must be kept in mind, however, that there are metabolic costs that are transparent to our mechanical analysis. Increased metabolic costs that do not show up on a mechanical analysis are associated with phenomenon such as isometric contractions or cocontractions. As has been pointed out in other research, the activity level of the knee extensors (as measured by EMG) is increased above that of nonamputee subjects.⁸ Such an increase, whereas not evident from our mechanical analysis (indeed, the knee extensor moments and powers are substantially reduced), would be consistent with cocontractions about the knee that would have consequences in a higher metabolic cost. In spite of the evidence from our mechanical analysis, we would anticipate seeing similar trends in stair ambulation as those that have been found for walking, with the metabolic cost of performing the task being higher for the amputee population.

Conclusion

In conclusion, it was found that unilateral trans-tibial amputees use a hip-dominant strategy on the amputated side to ascend stairs, which is different than the knee-dominant strategy used on the opposite side. This results in a substantial increased demand placed on the hip extensor muscles. The additional hip work required by subjects when using the SACH foot means that energy-storing prostheses offer a potential advantage for amputees who must negotiate stairs on a regular basis.

References:

1. Czerniecki JM. Rehabilitation in limb deficiency. 1. Gait and motion analysis. Arch Phys Med Rehabil. 1996;77:S-3–S-8.
2. Lehmann JF, Price R, Boswell-Bessette S, et al. Comprehensive analysis of dynamic elastic response feet: Seattle LightFoot versus SACH foot. Arch Phys Med Rehabil. 1993;74:853–861.
3. Miller LA, Childress DS. Analysis of a vertical compliance prosthetic foot. J Rehabil Res Dev. 1997;34(1):52–57.
4. Andriacchi TP, Andersson GBJ, Fermier RW, et al. A study of lower-limb mechanics during stair-climbing. J Bone Jt Surg. 1980;62-A(5):749–757.
5. McFadyen BJ, Winter DA. An integrated biomechanical analysis of normal stair ascent and descent. J Biomech. 1988;21(9): 733–744.
6. Duncan JA, Kowalk DL, Vaughn CL. Six degree of freedom joint power in stair climbing. Gait Posture 1997;5:204-210.
7. Torburn L, Schweiger GP, Perry J, Powers CM. Below-knee amputee gait in stair ambulation. Clin Orthop. 1994;303:185-192.
8. Powers CM, Boyd LA, Torburn L, Perry J. Stair ambulation in persons with transtibial amputation: An analysis of the Seattle LightFoot. J Rehabil Res Dev. 1997; 34(1):9–18.
9. Yu B, Kienbacher T, Growney ES, Johnson ME. An KN. Reproducibility of the kinematics and kinetics of the lower extremity during normal stair-climbing. J Orthop Res. 1997;15:348–352.
10. Bresler B, Frankel JP. The forces and moments in the leg during level walking. Trans Am Soc Mech Eng. 1950;72:27–36.
11. Brand RA, Yack HJ. Effects of leg length discrepancies on the forces at the hip joint. Clin Orthop. 1996, 333:172-180.
12. Winter DA, Robertson GD. Joint torque and energy patterns in normal gait. Biol Cybern. 1978;29:137–142.
13. Sanderson DJ, Martin PE. Lower extremity kinematic and kinetic adaptations in unilateral below-knee amputees during walking. Gait Posture 1997;6:126–136.
14. Postema K, Hermens HJ, DeVries J, et al. Energy storage and release of prosthetic feet. Part I: biomechanical analysis related to user benefits. Prosthet Orthotic Intl. 1997; 21:17–27.