Normative ground reaction force data for able-bodied and trans-tibial amputee children during running
J. R. Engsberg *
A. G. Lee *
K. G. Tedford *
J. A. Harder *
The purpose of this investigation was to develop normative ground reaction force data for able-bodied (AB) and trans-tibial amputee (TTA) children during running. Two hundred AB (mean age 9.4 years, range 7-12) and 21 TTA (mean age 11.1 years, range 5-17) children ran (2.2 m/s±10%) over a force platform. Ground reaction force data were normalized, averaged within groups and plotted to produce force-time curves characterizing the different leg types (i.e. able-bodied, non-prosthetic and prosthetic). In addition, discrete variables characterizing the leg type differences were determined. One way ANOVA determined significant differences between variables and a TukeyB Post Hoc analysis defined which variables were significantly different (p<0.05). Results generally indicated differences between the three leg types with the non-prosthetic leg indicating greater forces than the prosthetic and AB legs. The results of this investigation provide normative ground reaction force data for both AB and TTA children during running and can be used for comparison with other groups of children.
In order to assess the benefit, detriment, or irrelevance of a particular change in a prosthesis (e.g. new socket design or new terminal device) it is often desirable to compare measured variables with established norms. If normative data do not exist then the effects of the particular change are more difficult to judge. In the case of a trans-tibial amputee (TTA) child, two sets of normative data appear desirable: data from able-bodied (AB) children and data from other TTA children. In both cases comparisons can be made to determine how closely the TTA child matches the respective groups. While normative data for both AB and TTA children exist for walking (Engsberg et al., 1993; Sutherland et al., 1988), none exist for running. The purpose of this investigation was to develop normative ground reaction force data for AB and TTA children during running.
Two hundred AB (mean age 9.4 years, range 7-12) and 21 TTA (mean age 11.1 years, range 5-17) children volunteered to participate in this investigation. The characteristics describing the AB and TTA children are presented in Table 1 and Table 2, respectively. All children were given a physical examination by a pediatric orthopaedic surgeon and children not deemed able-bodied were excluded from the investigation. AB children were tested over a two week period during which approximately 20 children visited the Human Performance Laboratory each day. TTA children were tested over a one week period during which from 2-5 children visited the laboratory each day.
Ground reaction force data (1,000Hz) were collected during running support. For the AB children one trial was collected from each foot whereas for the ITA children at least three trials were collected from each foot. The same nominal rate of running (2.2 m/sec±10%) was enforced for all trials and was monitored by photocells spaced 2.4 m apart.
To allow for intersubject comparisons ground reaction force data were normalized by dividing by subject weight and contact time on the plate was normalized to a value of 1 (time 0 was touch down of the foot and time 1 was take off) (Engsberg et al., 1991). These data were then averaged within groups and plotted to produce force-time curves characterizing the different leg types (i.e. able-bodied, non-prosthetic and prosthetic). Since the time of occurrence of relative maxima and minima forces for the leg types were not constant, the averaging process did not highlight the differences between leg types. Hence discrete variables characterizing these differences were determined (Andriacchi et al., 1977; Engsberg et al., 1991). For the vertical force-time curves, a slope, two local maxima, a local minimum, local impulses, total impulse, and local times of support were determined. For the anteroposterior force-time curves, two local maxima, impulses for the retarding (i.e. force applied by the foot in the anterior direction) and propulsive (i.e. force applied by the foot in the posterior direction) phases, associated time components, and total impulse were recorded. Impulses for the medial and lateral force components, respective maxima, and durations were compiled from the medio-lateral force-time traces.
One way ANOVA determined any significant differences between variables and a TukeyB Post Hoc analysis defined which variables were significantly different (p<0.05). Differences between variables for:
- the right and left legs of the AB children;
- ages of the AB children;
- gender of the AB children;
- legs of the AB children and the non-prosthetic legs of the TTA children;
- legs of the AB children and prosthetic legs of the TTA children;
- non-prosthetic and prosthetic legs of the TTA children; and
- foot type and suspension of the TTA children were compared.
Fig. 1, Fig. 2 and Fig. 3 present the average force-time curves (i.e. vertical, anteroposterior and mediolateral) for the different leg types during running (i.e. able-bodied, prosthetic, non-prosthetic). Means and standard deviations of the vertical, anteroposterior and mediolateral discrete variables for the AB and TTA children appear in Table 3, Table 4, and Table 5, respectively. For the AB children there were no significant differences for the discrete variables between:
- right and left legs;
- males and females; and
Thus the respective values for the AB children were averaged to form one set of data (n=400) (Rosner, 1982). In addition, no significant differences were found for foot types or suspensions and the data for the TTA children were also grouped to form a single set of values. Significant differences existed between the non-prosthetic and prosthetic values of the TTA children, and between the TTA children and the able-bodied children. Table 3, Table 4 and Table 5 indicate numerically the similarities and differences between the leg types of the two groups of children that could not be accomplished through Fig. 1, Fig. 2 and Fig. 3.
The figures clearly indicate differences between the leg types of the children. For example, in the vertical-force time curves (Fig. 1) the first peak that generally characterizes heel-toe running in adults is prominent in the non-prosthetic leg of the ITA children, becomes less prominent for the AB children and is not evident for the prosthetic leg. For the anteroposterior force-time curves (Fig. 2) there appears to be more abrupt changes in magnitude for the initial phase of the curve for the TTA children than the AB children. In addition, during the final phase, results for the AB children appear to fall between the non-prosthetic and prosthetic legs. For the medio-lateral force-time curves (Fig. 3) the non-prosthetic legs of the TTA children appear to have the same general shape as that of the AB children, however the magnitudes are substantially greater. During the first 25% of the mediolateral curve the prosthetic legs of the amputee children are different from the other leg types since they produce a predominantly lateral force. During the remainder of support the curve is similar to that of the non-prosthetic leg.
For the vertical force-time variables (Table 3) the maximum force (ZMax2) during the latter phase of support was not significantly different between leg types, but the total impulse for the non-prosthetic leg was significantly different from the legs of the AB children. In the anteroposterior direction (Table 4) the absolute maximum in the posterior direction (PMax) and the impulse in that direction (PImp) indicate significant differences between groups. The non-prosthetic limb was significantly greater than the prosthetic limb and the limbs of the AB children for both variables. The prosthetic limb was significantly less than the limbs of the AB children for both variables. For the mediolateral variables (Table 5) it can be observed that the results for the prosthetic legs indicated significant differences when compared to the AB legs. Greater values were obtained for the maximum force in the lateral direction (LMax) and for the time spent applying force in the lateral direction (LTime).
The purpose of this investigation was to determine normative ground reaction force data for AB and TTA children during running. Four limitations appear noteworthy. The first is that the externally measured ground reaction force data may not reflect the internal loading of the joints of the lower limbs. Other forces such as those of muscles and ligaments also contribute to this load. The second limitation was with respect to the running speed. A number of methods exist for comparing the running speeds of the subjects (e.g. fixed speed, fixed cadence, freely chosen speed) with each method having its own advantages and limitations. Since it has been shown that rate of walking can influence force variables for children with trans-femoral amputation, (Zernicke et al., 1985) the same rate of running was used for all subjects in the present investigation. This chosen speed was based upon preliminary pilot investigations performed in the laboratory.
The third limitation was that this investigation determined normative data for TTA children, disregarding inter-subject prosthetic differences (i.e. foot types or suspension). While the design of this study was not well suited for considering the potential effects of these factors, the results were however, examined. In general, no significant differences were found between the variables for foot type and suspension.
The fourth limitation relates to the differences between the average force curves presented in Fig. 1, Fig. 2 and Fig. 3 and the quantification of discrete variables derived from individual subject force curves presented in Table 3, Table 4 and Table 5. The apparent discrepancy of results between the figures and the tables arises from the averaging process used to derive the force curves. Since the maximum values for a particular variable do not occur at the same instant in time for all subjects and leg types the averaged curves will not reflect the peak value displayed in the Table. On the other hand, the discrete variables determined from each individual curve and presented in the Tables will provide the true value since time of occurrence was not used in the determination of the value. For example, the ZMax1 variable in Table 3 reports values of 1.90, 2.11 and 1.89 for the AB, non-prosthetic and prosthetic legs, respectively. However, Fig. 1 indicates a ZMax1 value for AB children to be about 1.5, ZMax1 value for non-prosthetic leg of the TTA to be about 2.0, and Zmax1 value for the prosthetic leg to be about 1.9 (i.e. if the levelling of the curve at about 0.25 of the cycle is used as the average ZMax1 value). Thus it is important to view the curves as providing average shapes characterizing general differences between leg types and the tables as providing accurate discrete information for particular variables associated with the curves.
Prince et al. (1992) presented local maxima in the vertical, anterior, and posterior directions for a group of young TTA adults (n=9, mean age of 16 years). They found significantly greater values for the ZMax2 variable for the non-prosthetic leg when compared to both the prosthetic legs and legs of the AB. The present investigation did not find those differences and indicated no differences between the three leg types. Even though Miller (1987) did not present discrete values for the prosthetic and non-prosthetic limbs of TTA adults (n=4, mean age of 40 years) her figures concur with the results of the present investigation suggesting that approximately the same values occurred for the non-prosthetic and prosthetic legs. In support of the results presented by Prince et al. (1992) and in contrast to the results of the present investigation and of Miller (1987) the authors of this paper have reported greater ZMax2 values for the children of this investigation during walking (Engsberg et al., 1993). The results indicated that the non-prosthetic leg had a significantly greater ZMax2 value than the prosthetic leg and the legs of AB children. In addition, it was reported that the prosthetic leg force was significantly less than the AB leg force for the ZMax2 variable. Further investigation appears warranted in this regard.
In the anteroposterior directions Prince et al. (1992) reported significant differences for the RMax variable between the non-prosthetic legs and the AB legs and between the prosthetic legs and the non-prosthetic legs. The present investigation found no significant differences in RMax values between leg types. For the PMax variable Prince et al. (1992) reported significant differences between the prosthetic leg and both the non-prosthetic and AB legs. The present investigation supported these relationships and also found significant differences between the non-prosthetic and the AB legs. The differences in the results between the two investigations may be explained by subject age, subject numbers, and different prostheses.
Prince et al. (1992) reported that the ZMax2 and the ZImpTo variables were significantly greater (rigid keel only) in non-prosthetic legs when compared to similar values of AB controls. Similar results for the same variables and for approximately the same group of subjects as in the present investigation have been reported for walking (Engsberg et al., 1991 and 1993). In contrast to these results the present investigation did not concur with these findings. A potential explanation could be related to the possible effects of foot types since Prince et al. (1992) reported no significant differences for the two variables for flexible keel feet. Further investigation is necessary in this regard.
Prince et al. (1992) reported that the PMax and Pimp variables were significantly different between the prosthetic legs and the legs of the AB subjects. These loading differences, also occurring in the present investigation, have been reported for walking (Engsberg et al., 1991 and 1993). The similarity of these results appears to indicate that despite the type of terminal device used, the prosthetic legs do not generate propulsive forces similar to those produced by intact legs. The objective of the authors' research in this area is to enable TTA children to walk and run in the same way as AB children. The accomplishment of this objective would however require the development of a prosthesis which allows the prosthetic leg to produce the same forces as those of intact legs.
The results of this investigation provide normative ground reaction force data for both AB and TTA children during running. The results for the AB children can be used for comparison with TTA children and with any other groups of children (e.g. children with cerebral palsy) if similar data are determined. The results for the TTA children can be used to determine if TTA children are functioning similarly to other TTA children. Since the results indicate basic differences between TTA and AB children during running, research should be directed towards eliminating these differences.
Funding provided by the George Reed Foundation for the Handicapped, Hospital for Sick Children Foundation, and the Variety Club of Southern Alberta-Tent 61.
- Andriacchi TP, Ogle JA, Galante JO (1977). Walking speed as a basis for normal and abnormal gait measurements. J Biomech 10, 261-268.
- Engsberg JR, Lee AG, Patterson JL, Harder JA (1991). External loading comparisons between able-bodied and below-knee-amputee children during walking. Arch Phys Med Rehabil 12, 657-661.
- Engsberg JR, Lee AG, Tedford KG, Harder JA (1993). Normative ground reaction force data for able-bodied and below-knee-amputee children during walking. J Pediatr Orthop 13, 169-173.
- Miller DI (1987). Resultant lower extremity joint moments in below-knee amputees during running stance. J Biomech 20, 529-541.
- Prince F, Allard P, Therrien RG, McFadyen BJ (1992). Running gait impulse asymmetries in below-knee amputees. Prosthet Orthot Int 16, 19-24.
- Rosner B (1982). Statistical methods in ophthalmology: an adjustment for the intraclass correlation between eyes. Biometrics 38, 105-114.
- Sutherland DH, Olshen RA, Biden EN, Wyatt MP (1988). The Development of Mature Walking. - Oxford: MacKeith Press.
- Zernicke RF, Hoy MG, Whiting WC (1985). Ground reaction forces and centre of pressure patterns in the gait of children with amputations: preliminary report. Arch Phys Med Rehabil 66, 736-741.