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

Mechanical and Metabolic Work of Persons with Lower-Extremity Amputations Walking with Titanium and Stainless Steel Prostheses: A Preliminary Study

Robert F. Scherer, MSc
James J. Dowling, PhD
Gail Frost, PhD
Marty Robinson, BSc, CO
Karen McLean, BSc, CO

ABSTRACT

A total of 15 subjects with unilateral amputations (8 transfemoral and 7 transtibial) performed treadmill walking with prostheses assembled from titanium and stainless steel components to determine if mass differences had an effect on walking. Standardized components (knees, pylons, adapters, feet) made from each material were added below the level of the socket. Metabolic cost from submaximal oxygen consumption and mechanical power estimates allowing transfers within and between segments were calculated during steady-state walking at self-selected velocities. Results showed that despite significant mechanical power differences, the decreased mass associated with the use of titanium materials did not have a significant effect on the metabolic costs of walking, stride rate, or stride length. Further division of subjects by age and experience walking on a prosthesis suggested that older persons and established walkers benefit most from the use of titanium components, both metabolically and mechanically.

Key Words: energy expenditure, lower-extremity amputation, titanium, stainless steel, prosthesis

Introduction

The question of what may be the most efficient combination of mass and materials for use in the design of lower extremity prostheses has yet to be answered. Recently, rapid progress has been made in implementing the use of titanium as a material in the manufacturing of prosthetic componentry. However, there is little scientific evidence that either quantifies the benefit to the user, or justifies the value of the increased price associated with lightweight materials like titanium. Although it would seem logical to expect increased walking efficiency with lower-mass materials, it would also seem prudent to quantify this with experimental data. Such information would be valuable for both manufacturers and practitioners. The various studies used to estimate the mechanical and metabolic properties of the gait of the amputee while he or she is wearing prostheses with different masses have failed to provide any direct information on the current materials available to amputees.^1-8 These studies have all addressed the mass-energy cost issue by using alternative methods such as simulations, or by applying different loading conditions to the prostheses. The mass differences in actual components made from different materials may be great enough to affect the overall energy cost of the amputee. For a majority of individuals who have had lower-extremity amputations, available materials are the conventional, heavier stainless steel and the newer, lighter titanium. In the literature there has yet to be a comprehensive investigation into the inertial characteristics of prostheses assembled from these materials and the effects they have on energy expenditure during walking.

The purposes of the present study were the following: 1) determine the inertial parameters of transtibial (TT) and transfemoral (TF) prostheses assembled from stainless steel and titanium components; 2) quantify the total mechanical power and metabolic costs of walking for persons with TT and TF amputations using such components and identify any differences that may exist between components; and 3) evaluate the relationship of prosthesis mass with mechanical power and metabolic cost.

Methods

Two women and 13 men who had unilateral, lower-extremity amputations (age: 46.9 (13.9 years) participated in the study. All subjects provided informed consent and the study was approved by the McMaster University Ethics Committee. All but one subject were free of any medical problems that limited unassisted walking, and only metabolic data were collected for that subject. Seven of the subjects had transtibial (TT) amputations and eight had transfemoral (TF) amputations. Table 1 contains clinical characteristics of the subjects. Each subject performed two walking trials on a level treadmill. One trial was performed while the subject was wearing a stainless steel prosthesis, and the other, while the subject was wearing a titanium prosthesis, with the order of the materials being randomized for each subject. The treadmill was set at a comfortable (self-selected) speed for each subject for the first trial, and was kept at the same speed for the second trial. Each trial consisted of an habituation period followed by at least 4 minutes of walking, allowing the subject to achieve steady-state conditions.

Each subject was fitted with new prosthetic components manufactured from either titanium or stainless steel. New components were aligned and adapted to the subjects' original sockets by using an Otto Bock 4-hole socket adapter. The original sockets for the TF group were ISNY (Iceland-Sweden- New York) flexible sockets. New PTB (Petellar Tendon Bearing) sockets were worn by the TT subjects. Knee and foot components were standardized for all trials. TF subjects walked with the Otto Bock 3R36 titanium knee and the Otto Bock 3R20 stainless steel knee. Both are polycentric, constant-friction knees with spring-aided extension assists. All subjects were fitted with the Otto Bock 1D10 dynamic foot so that the only differences between the two prosthetic limbs were the inertial differences between the titanium and stainless steel components. Each subject wore comfortable walking shoes of their own choosing. All technical work, clinical fitting and alignment procedures were performed by the same prosthetist, who ensured that the alignments were identical between trials.

After each walking trial, the subject's prosthesis was disarticulated below the socket and the components and socket were subjected to inertial measurements. These measures involved weighing the prosthesis, balancing the prosthesis on a knife edge to determine the location of the center of mass (COM), and oscillating the prosthesis to determine its moment of inertia about the COM. The residual thigh and shank masses, COM locations, and moments of inertia were mathematically determined with similar methods and equations used by Bach et al.,² derived from Hatze.⁹

Kinematic data were collected by attaching infrared light emitting diodes (IREDs) to each subject at the following locations: (sagittal plane view) right mp, or metacarpal phalangeal, (5th), right heel, right ankle, right knee, right hip, right lumbar 4-5, right shoulder, right elbow, right wrist, right ear canal (medial border) left mp (1st), left heel, left ankle and left knee. The IREDs were tracked by using an optoelectronic device (OPTOTRAK/3020, Northern Digital, Waterloo, Ontario) at a rate of 100 samples per second for 5 seconds during steady-state walking. The data were digitally filtered by using a Butterworth low-pass filter with a cutoff frequency of 6 Hz.

An 11-segment, sagittal plane model was mathematically constructed for each subject by using the filtered joint-segment marker data. The segments consisted of a head/neck, two upper arms, two forearm/hands, a trunk, two thighs, a leg, a foot, and a leg/foot. The leg/foot segment was developed to account for the fixed ankle joint that existed in the components used in this study. Customized software was used to calculate joint and segment displacements and velocities (linear and angular) and segmental and system energies. Mechanical work was estimated from the changes in segmental energies allowing energy transfers within and between the segments.¹⁰ Normalized mechanical power was calculated by dividing the work values by the time of a single stride and by dividing this value by the subject's body weight to yield units of watts per kilogram.

Before walking on the treadmill, subjects stood quietly for 1 minute while oxygen uptake (VO₂) was monitored. Subjects were connected to the equipment through a mouthpiece and low-dead-space valve, to a custom-configured, open-circuit computerized system (Ametek /S-3A/I O₂ analyzer, Hewlett Packard 78356A CO₂ analyzer, Vacumed turbine flow meter, Vacumed data acquisition and metabolic measuring system) which was calibrated with gases of known concentration. During the walking trials, expired gas was collected continuously via a mixing chamber, with VO₂ values recorded every 30 seconds. Three VO₂ values taken during the last 90 seconds of steady-state walking were used to calculate an average estimate of VO₂ for each subject. The net submaximal VO₂ value for each trial was converted to watts per kilogram by using the respiratory exchange ratio to convert the milliliter per kilogram per minute value of O₂ to joules and then dividing by 60 seconds.

Fourteen subjects (7 TT and 7 TF) were evaluated for measures of mechanical work and metabolic work. An additional subject (TF) performed the metabolic test only. A one-way repeated measures ANOVA was performed on the metabolic data from all subjects to examine the effects of the different component materials. Also, a two-factor ANOVA for independent groups (7 TT, 8 TF) was performed by using metabolic values [factor 1 = group (TT/TF) and factor 2 = components (titanium/stainless steel)]. A two-factor ANOVA on group (TT/TF) and components (titanium/stainless steel) was performed to examine mechanical power differences. In each case, significance was tested for at the ( < 0.05 level.)

The subject population was subdivided into smaller groups based on age and experience walking on a prosthesis. The subdivisions were old (>50 years), young (<40 years), new (used a prosthesis <2.25 years), and experienced (used a prosthesis >2.25 years). Every attempt was made to obtain the largest subject population possible, and as such, the distribution of age, sex, and experience were not necessarily representative of the normal population of those with amputations. The numbers of subjects in these subdivisions were not sufficient to perform statistical analyses, but mean values for the energetic data were calculated for comparative purposes. Correlations between prosthesis mass and energetics were also calculated as measures of the strength of a relationship between these variables.

Results

As expected, all prostheses assembled from stainless steel weighed more and had segments with moments of inertia and COM (distance from proximal end) values greater than or equal to those assembled from titanium. Consequently, subjects had a greater mass when wearing stainless steel as opposed to when wearing titanium components. The normalized metabolic cost rates, estimated from submaximal VO₂, were greater for the stainless steel trials than for the titanium trials, but the difference was not significant [(F1,14)= 1.45, p < .249] when all 15 subjects were examined (x 1). Comparison of TF and TT subjects revealed no significant metabolic effects caused by the different materials [(F1,13) = 1.31, p < .274], however TF subjects consumed significantly more oxygen than did TT subjects [(F1,13) = 11.34, p < .005] (Figure 1 ).

Mean mechanical power measurements were significantly lower [(F1,12)= 4.85, p < .048)] when subjects walked with titanium components as compared to with stainless steel (see Figure 2 ). Differences between TF and TT groups, however, were not significant.

The data shown in Figures 1 and 2 were subdivided and shown in Figure 3 . Although only descriptive, it appears as though the TF subjects who had less than 2.25 years of experience walking on a prosthesis (new) performed more mechanical work and had higher metabolic costs when walking on the lighter titanium prostheses. This was contrary to the findings for the other three groups. The correlations between prosthesis mass and metabolic cost rate and mechanical power showed almost no relationship, except for the TT prosthesis mass and metabolic work rate (see Figure 4 ).

Discussion

The results of this study indicated that although reductions in prosthesis mass (ie: use of titanium components) may have caused a reduction in mechanical power, the decrease in mechanical power was not accompanied by a significant decrease in the oxygen consumed, as one might expect. One explanation may be that the difference in mechanical power required for walking with the two materials is not great enough to induce physiological consequences measurable via oxygen consumption. It is possible that the muscle mass responsible for the increased mechanical power output with stainless steel components was too small to significantly increase the metabolic cost of walking with these components over titanium. This would support the idea that the relationship between mechanics and metabolics is not a direct one and may be responsible for results found in similar studies of subjects with amputations that found low correlations and poor regressions involving mechanical and metabolic measures.^1,2,6,11

Alternative explanations for the results in this study can be provided when considering prosthetic design. It is possible that differences in the inertial characteristics of the two materials may be compensated for by the mechanisms built into the prosthesis itself. Slight alterations in the spring assist of the knee joint may relieve natural tissues of the body of the role of producing the increased energy necessary to achieve the same speed, stride rate, and stride length (no extra muscular work would have to be done). This would especially benefit TF subjects. In this study, a great deal of time was devoted to attaining an exact duplicate between the two materials for each subject.

Limitations of the Study

To better understand the results and describe the individual differences that existed between the subjects in this study, the interaction of materials (mass) with age and length of time on a prosthesis were examined for both mechanical and metabolic measures. While statistical analysis could not be done with these data, it was apparent that relatively new walkers (2 years with a prosthesis) did not receive the energy savings associated with lighter, titanium components experienced by established walkers (> 2 years) and young (< 40 years) or old (> 50 years) persons with amputations. It seems prudent that future investigations examine the issue of componentry material on walking energetics for those who have been walking on a prosthesis for a considerable length of time. When examining the possible benefits of titanium, we found that the length of time walking on a prosthesis may be a more important variable than the age of the individual.

Lastly, it was evident that mass of the prosthesis alone was a poor correlate with either mechanical or metabolic energy expenditure for all groups. TT subjects exhibited the highest correlation (r = .61) between prosthesis mass and metabolic values, however, the highest masses did not result in the greatest energy costs. This suggests that the relationship between mass and metabolic cost may not be a linear one, and that there may be an optimal prosthesis mass for certain categories of persons with amputations.

One limitation to this type of study is that the persons with amputations that are thought to benefit the most when walking with lightweight componentry (the elderly) have difficulty attaining steady state when walking on a treadmill without the use of an aid. The current measurement and testing protocols would then have to be adapted to handle situations in which energy transfers within and between the aid and the subject existed.

As a reply to industrial and clinical concerns, even though titanium did not display all of the expected energy benefits for subjects in this study, the mechanical properties of its use in walking do appear to be superior. However, it should be noted that statements from industry regarding the expected benefits of economy of movement for different products should be more carefully examined and researched, especially when dealing with the effectiveness of lightweight materials.

Acknowledgements

The authors wish to thank Mark Agro and Otto Bock of Canada and John Moroz for technical support and the University Research Incentive Fund for financial support.

References:

1. Foerster SA, Bagley AM, Mote CD, Skinner, HB. The prediction of metabolic energy expenditure during gait from mechanical energy of the limb: a preliminary study. J Rehabil Res Dev. 1995;32(2):128-134.
2. Bach TM, Barnes LJ, Evans OM, Robinson IGA. Optimization of inertial characteristics of transfemoral limb prostheses: tests of predictions of a computer simulation. Proceedings Eighth Biennial Conference, Canadian Society for Biomechanics. Calgary, Canada, 1994:124-125.
3. Czerniecki JM, Gitter A, Weaver K. Effect of alterations in prosthetic shank mass on the metabolic costs of ambulation in above-knee amputees. Am J Phys Med Rehabil. 1994:73(5):348-52.
4. Gailey RS, Wenger MA, Raya M, Kirk N, Erbs K, Spyropoulos P, Nash S. Energy expenditure of trans-tibial amputees during ambulation at self-selected pace. Prosthet Orthot Int. 1994;18:84-91.
5. Hale SA. Analysis of the swing phase dynamics and muscular effort of the above-knee amputee for varying prosthetic shank loads. Prosthet Orthot Int.1990;14:125-135.
6. Skinner HB, Mote CD. Optimization of amputee prosthesis weight and weight distribution. Rehabil R D Progr Rep. 1989;26:32-33.
7. Tashman S, Hicks R, Jendrzejczyk DJ. Evaluation of a prosthetic shank with variable inertial properties. Clin Prost Orthot. 1985;9:23-28.
8. Menkveld S, Mansour JM, Simon SR. Mass distribution in prosthetics and orthotics: quantitative analysis of gait using a biomechanical model simulation. Ortho.Trans. 1981;5:228, (Abstract).
9. Hatze H. A model for the computational determination of parameter values of anthropometric segments. National Research Institute for Mathematical Sciences, Pretoria. 1979.
10. Pierrynowski MR, Winter DA, Norman RW. Transfers of mechanical energy within the total body and mechanical efficiency during treadmill walking. Ergonomics. 1980;23:147-156.
11. Beck JC, Czerniecki J. A method for optimization of above-knee prosthetic shank-foot inertial characteristics. Gait Posture. 1994;2:75-84.