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

Changes in Ground Reaction Forces during Prosthetic Training of People with Transfemoral Amputations: A Pilot Study

Christiane Gauthier-Gagnon, MSc, PT
Denis Gravel, PhD, PT
Hélène St-Amand, BSc, PT, MPA, PT
Christian Murie, BSc, PT
Michel Goyette, Eng

ABSTRACT

The purpose of this study was to evaluate changes in floor reaction forces over the prosthesis during stance and gait throughout prosthetic training. Fifteen transfemoral amputation patients (mean age: 62 ± 13 years) were evaluated during prosthetic training. As the patients were standing and walking in parallel bars at free speed, ground reaction forces were measured with a force plate and gait velocity was measured using two photocells. Pretraining and posttraining changes of average peak forces were evaluated by paired t-tests (p < 0.05). In the training period, weightbearing over the prosthesis during standing increased (p < 0.05) and significant decreases (p < 0.05) were observed in the nonamputated limb. From initial to discharge evaluation, gait velocity and vertical ground reaction forces increased significantly. Changes in vertical forces were minimal (< 5%) for three subjects, and four subjects had lower values at the end of training. No significant changes could be detected in the anteroposterior and mediolateral ground reaction forces.

Key Words: leg amputation, transfemoral prosthesis, weightbearing, gait, kinetics

Introduction

Supporting body weight in static and dynamic conditions is one of the main functions of the lower limb. Symmetrical weight shifting over the limbs during stance and gait is a relevant clinical problem for people with a lower limb amputation. Through limb loss, the center of gravity is shifted laterally to the side of the nonamputated limb--a shift that is not fully compensated for by the mass of the prosthesis.¹ Thus, the increase in vertical loading on the nonamputated side is not only related to the difference between the weight of the prosthesis and the weight of the anatomical segment. Other factors, such as pain and/or postural instability are probably responsible for the asymmetrical weightbearing during stance and gait. This is not surprising, because most amputations are performed on elderly people who are afflicted with a multiplicity of age- and disease-related problems that may cause damage to the sensory systems regulating postural control.

The average loading on each foot, as calculated across a sample of normal adult subjects, is symmetrical during stance.^2,3,4 It is important to note that averaging across subjects may be misleading with regard to real asymmetry, because the greater load supported by the right leg in half of the sample can be canceled by the greater load supported by the left leg in the other half of the sample. A better indicator of the weight distribution in a sample is the average difference between the two limbs. For normal subjects, a mean difference of 6 to 10% of body weight has been reported.^1,2,4,5 On average, in the years after prosthetic fitting, people with a transfemoral amputation supported close to 40% of their body weight on the prosthesis during standing (or an equivalent difference of 20% between the two sides).^1,3,4 During gait, significant asymmetries in foot loading were also observed. For both transtibial and transfemoral amputees, vertical ground reaction forces were less over the prosthesis than the characteristic normal forces.^6,7,8

Body weight transference over the prosthesis when standing and walking is an important goal in the rehabilitation of people with a lower limb amputation. Few studies have examined changes in standing and walking parameters during prosthetic training, although considerable variations in the rate of improvement have been observed clinically, specifically with transfemoral amputees. Over the training period, transfemoral amputees showed a marked reduction in the degree of asymmetry in both stride time and double-breaking support time.⁹ Walking speeds also increased dramatically. Baker and Hewison¹⁰ measured a 100% increase in velocity during training for both transtibial and transfemoral amputees. Statistically significant increases in mean weightbearing under the prosthesis from 32% body weight (first session) to 41% body weight (final session) were also demonstrated during stance.¹¹ Changes in ground reaction forces when standing and walking were not investigated during rehabilitation of transfemoral amputees. The objective of this pilot study was to assess progress during prosthetic training of these amputees. More specifically, weightbearing over the prosthesis and nonamputated limb when standing, gait velocity, and changes in vertical, anteroposterior, and mediolateral components of ground reaction forces over the prosthesis during gait were investigated.

Methods

Fifteen transfemoral amputees (12 men and three women) participated in the study. Their mean age (± 1 SD) was 62 ± 13 years. To have a realistic profile of the progress of people with transfemoral amputations during prosthetic training, no exclusion criteria were defined. Of the 15 cases, 13 amputations were the result of peripheral vascular disease, one was traumatic, and one was consecutive to poliomyelitis. Fourteen subjects were fitted with a quadrilateral socket and silesian band, and one subject was fitted with a contoured adduction trochanteric-controlled alignment method suction socket. All had single-axis knee and foot-ankle assemblies. The prosthetic training started 78 ± 43 days after amputation surgery and lasted an average of 40 ± 13 days.

All subjects underwent a physiotherapy evaluation before the first test. Subsequently, clinical evaluations were made weekly to identify specific problems encountered during prosthetic training and before discharge. Information noted included age, sex, indication for amputation, associated medical conditions, type of artificial limb supplied, condition of the nonamputated limb, and ambulatory aids used. All subjects provided written informed consent before participating in the study.

Experimental assessments were performed each week during prosthetic training. Using a force plate (Model OR6-5; Advanced Mechanical Technology, Inc., Watertown, MA), ground reaction forces were measured in three conditions with eyes open, including: 1) standing 10 seconds, arms alongside the body, with the nonamputated limb on the force plate; 2) standing 10 seconds, arms alongside the body, with the prosthetic leg on the force plate; and 3) walking in parallel bars at free speed over a force plate incorporated midway in a 5-m walkway. Three trials were recorded weekly in each condition. During the standing tests, forces acting on the prosthesis and on the nonamputated limb were measured separately. Foot position was standardized using a wooden template. Heels were 10 cm apart, and foot angle relative to the sagittal plane was 10 degrees. The template was removed during data collection. Static weightbearing when standing was the numerical average of vertical force recorded during the 10-second period. Gait velocity was recorded using two photocells placed 134 cm apart. Subjects were asked to walk at a comfortable speed without looking at the force plate. The vertical, anteroposterior (or fore-aft), and mediolateral force components were recorded to assess dynamic weightbearing during gait. Preliminary tests were done on the force plate to weigh the subjects.

The force plate signals were processed by direct current amplifiers with an upper frequency limit set at 1000 Hz and digitized by an analog-to-digital convertor at a sampling rate of 120 Hz (Data Translation board, Model DT2839; Data Translation, Marlboro, MA). During stance, the signal was averaged over the 10-second period and expressed in percentage of body weight. During walking trials, the peak forces in each direction were extracted from the force-time curves, excluding the initial impact forces and trailing forces (Figure 1 ). Processing of the data included normalization of absolute forces by body weight. For all conditions, average peak forces of the three trials normalized to body weight were used in the statistical analyses. Mean walking speed for the three trials were also computed. Pretraining and posttraining changes were evaluated by Student's paired t tests. A p value of < 0.05 was taken as significant.

Results

Figure 2 shows the changes in percentage of weightbearing over the prosthesis and nonamputated limb at the beginning and end of prosthetic training. The horizontal reference line represents the 50% critical value of equal weightbearing distribution on the two limbs. During prosthetic training, weightbearing over the prosthesis increased significantly (p < 0.05); simultaneously, significant decreases were observed on the nonamputated lower extremity (p < 0.05).

From initial to discharge evaluation, gait velocity increased from 0.25 to 0.35 mean walking speed, a 48% increase (Table 1 ). Vertical ground reaction force also increased significantly (p < 0.05) but, as shown in figure 3 , changes were highly variable between subjects. Four subjects (subjects 1 through 4) had lower values posttraining as compared with pretraining, and three subjects (subjects 6 through 8) had increases of less than 5% of body weight. In the remaining subjects, the increases were greater than 10%. No significant changes were noted for the anteroposterior ground reaction forces. In the mediolateral plane, the ground reaction forces applied toward the medial side of the subject increased significantly by approximately 1% (from 3.97% to 5.25%).

Discussion

Static Weightbearing

Weightbearing on the prosthesis improved significantly after prosthetic training, reaching close to normal weight distributions. As reported by Summers and colleagues,¹¹ more weight was shifted over the nonamputated leg. In the present study, the preferential weightbearing on the nonamputated limb was well within the expected range calculated by Arsenault and Valiquette.¹ After the amputation of part of a limb, the body's center of mass is shifted toward the opposite side. After years of prosthetic wear, this shift is only partially corrected, because the mass of the prosthesis is less than that of the amputated segment.¹ Studying young, traumatic amputees who had been walking with prostheses for more than 6 years, Arsenault and Valiquette1 estimated the distribution of body weight to be an average of 14% greater over the nonamputated limb and reported that the actual weight distributions over the nonamputated leg were even greater than that predicted. It is possible that the subjects of the present study shifted more weight on the prosthesis to protect the dysvascular limb. In fact, in all but four of the subjects, their limb was amputated because of peripheral vascular disease and the nonamputated limb had severe vascular problems.

Throughout rehabilitation, 10 to 30% increases in weightbearing were reported.^11,12 The highest values were recorded by Stolov and colleagues¹² whose subjects were using immediate postsurgical prostheses. In the present case, the observed improvements barely reached an average of 6%. This may be because of high initial values. At the beginning of the prosthetic training, all subjects had well-healed and mature stumps and thus could tolerate more pressure at the initial testing. Moreover, three subjects^1,13,14 had been standing with a pneumatic prosthesis in the week before their prosthetic fitting.

Gait Parameters

Increments in gait velocity were 43% compared with 100%, somewhat lower than those reported by Baker and Hewison.¹⁰ As shown in Table 2 , maximal changes in velocity occurred during the first 2 weeks of prosthetic training and minimal progression was obtained after 30 days of treatment. Baker and Hewison¹⁰ observed 55% increases in velocity between day 1 and day 15 of training, followed by a 30% increase over the next 15 days. However, their subject population consisted mainly of transtibial amputees, who walk faster than transfemoral amputees. Normally, rehabilitation programs can be completed in 4.6 weeks (32 days) by transfemoral amputees and in 5.4 weeks (38 days) by transtibial amputees.¹³ Longer prosthetic training periods do not necessarily result in increased prosthetic use.¹⁴ Prolonged prosthetic training may be related to chronic health problems and delayed wound healing. De facto, three subjects (subjects 9, 10, and 11) underwent major health problems (cardiac and pulmonary problems) during their rehabilitation program. They resumed training thereafter, but less intensively and over a longer period of time.

The increases in vertical forces were significantly higher after training, but variable between subjects. Four subjects had lower vertical forces after training than before training, and three subjects had increases in weightbearing of less than 5% body weight. The observed variability in foot loading during gait could be explained by at least two factors. The first factor relates to the gait measures that were obtained when subjects were in the parallel bars, where they could bear weight on their hands. At the end of prosthetic training, most of our subjects were able to walk with either canes or a walker, but were tested in parallel bars. To be closer to reality, it would be more appropriate to adapt the walkway to evaluate the subjects walking with ambulatory devices. The second factor to be considered is that transfemoral amputees are people affected by multiple medical problems that can impair locomotor capabilities with the prosthesis. In fact, one (subject 4) of the four subjects who had lower values after prosthetic training had stump pain that interfered with unilateral weightbearing, two (subjects 1 and 3) had hip contractures, and another (subject 2) was ambulating with a walker.

In contrast to velocity and static weightbearing (Table 2 ), most gains in vertical foot loading during walking were observed after the first 2 weeks, when subjects started to use ambulatory devices. Thereafter, progress was constant. It may be assumed that, as amputees achieved symmetrical, static weightbearing, they started walking outside parallel bars with ambulatory devices, thus indicating less reliance on hand support and greater locomotor control.

Fore-aft and mediolateral ground reaction forces were not affected practically by the prosthetic training. Seliktar and Mizrahi¹⁵ explored the possibilities of using ground reaction forces for identification of specific irregularities in transtibial amputee gait. They also found that the variability of amputee gait was considerable, rendering interpretation of the data meaningless. Hence, they suggested a case-per-case analysis of the force curves. Nevertheless, visual inspection of our force curves was inconclusive because no specific pattern could be identified. It is true that tests were done during training, when the prosthesis may not yet have been comfortable and walking was insecure. When looking at Table 2 from a clinical point of view, however, it can be noted that during the first 2 weeks of training the transfemoral amputee tended to gain more control over the prosthesis at heel strike (posterior forces), as evidenced by less forces generated.

The intent of this study was to observe changes during prosthetic training with regard to forces acting on the prosthesis. Three variables reflected significant improvement of locomotion for the transfemoral amputee gait and could be used as indicators of gait training progression; these variables included static weightbearing, gait velocity, and vertical loading of the prosthesis during gait. Static weightbearing, which quantifies skin tolerance to pressure in the socket and gives evidence of progress in prosthetic training, was demonstrated to be a predictor of walking ability for the vascular transtibial amputee.¹⁶ It also correlated closely with forces acting through the prosthetic limb during locomotion. Gait velocity, which measures forward progression of the body and provides information on locomotor abilities, was found to be an indicator of gait performance and the best single index of walking ability.^8,17 Furthermore, when the amputee attained maximal weightbearing, velocity was found to be a more sensitive measure of progress than weightbearing.⁸ Both static weightbearing and gait velocity can be readily quantified in the course of a treatment using simple and inexpensive tools such as a bathroom scale and a stopwatch. These outcome measures could be used to quantify and predict locomotor abilities during prosthetic training. However, a longitudinal study is required to assess whether static weightbearing, gait velocity, and vertical loading of the prosthesis during gait could be predictors of active use of the prosthesis in the years after discharge from the rehabilitation program.

Conclusion

People with transfemoral amputations bear more weight on the prosthesis when standing and less on the nonamputated limb after prosthetic training. During gait, velocity and vertical ground reaction forces increase signifi-cantly after training. However, some subjects have either absent or less than 5% body weight changes in vertical ground reaction force when walking in parallel bars. Additionally, fore-aft forces were significantly different after rehabilitation.

Acknowledgments

The participation of all subjects is acknowledged. Special thanks go to Stéphane Cardinal and Nathalie Brulé for their assistance in this project. The authors are grateful to the physiotherapy department of the Institut de Réadaptation de Montréal for their kind cooperation throughout this study. The project was supported by a grant from the Réseau de Recherche en Réadaptation de Montréal et de l'Ouest du Québec (RRRMOQ), Québec, Canada.

References:

1. Arsenault AB, Valiquette C. Etude de la statique posturale des amputes du membre inferieur: Correlations théoriques et pratiques de la mise en charge. Physiother Can. 1981;1:17-33.
2. Murray MP, Peterson RM. Weight distribution and weight-shifting activity during normal standing posture. Phys Ther. 1973;53:741-748.
3. Lord M, Smith DM. Foot loading in amputee stance. Prosthet Orthot Int. 1984;8:159-164.
4. Summers GD, Morrison JD, Cochrane GM. Foot loading characteristics of amputees and normal subjects. Prosthet Orthot Int. 1987;1:33-39.
5. Isakov E, Mizrahi J, Ring H, Susak Z, Hakim N. Standing sway and weight-bearing distribution in people with below-knee amputations. Arch Phys Med Rehabil. 1992;73:174-178.
6. Suzuki K. Force plate study of the artificial limb gait. J Jpn Orthop Assoc. 1972;40:503-516.
7. Hermodsson Y, Ekdahl C, Persson BM, Roxendal G. Gait in male transtibial amputees: A comparative study with healthy subjects in relation to walking speed. Prosthet Orthot Int. 1994;8:68-77
8. Jones ME, Bashford GM, Mann JM. Weight bearing and velocity in transtibial and transfemoral amputees. Prosthet Orthot Int. 1997;21:183-186.
9. Cheung C, Wall JC, Zelin S. A microcomputer-based system for measuring temporal asymmetry in amputee gait. Prosthet Orthot Int. 1983;7:131-140.
10. Baker PA, Hewison SR. Gait recovery pattern of unilateral lower limb amputees during rehabilitation. Prosthet Orthot Int. 1990;14:80-84.
11. Summers GD, Morrison JD, Cochrane GM. Amputee walking training a preliminary study of biomechanical measurements of stance and balance. Int Dis Stud. 1988;10:1-5.
12. Stolov WC, Burgess EM, Romano RL. Progression of weight bearing after immediate prosthesis fitting following below-knee amputation. Arch Phys Med Rehabil. 1971;52:491-502.
13. Francis W, Renton CJC. Mobility after major limb amputation for arterial occlusive disease. Prosthet Orthot Int. 1987;11:85-89.
14. Gauthier-Gagnon C, Grisé M-C, Potvin D. Predisposing factors related to prosthetic use by people with a transtibial and transfemoral amputation. J Prosthet Orthot. 1998;10:99-109.
15. Seliktar R, Mizrahi J. Some gait characteristics of below-knee amputees and their reflection on the ground reaction forces. Eng Med. 1986;15:27-34.
16. Jones ME, Steele JR, Bashford GM, Davidson IR. Static versus dynamic prosthetic weight-bearing in elderly transtibial amputees. Prosthet Orthot Int. 1997;21:100-106.
17. Skinner HB, Effeney DJ. Gait analysis in amputees. Amer J Phys Med. 1985;64:82-89.

Website built by oandp.com