A comparison of oxygen consumption during walking between children with and without below-knee amputations.

It has been previously reported that the kinematics and kinetics of children with below-knee amputations (BKAs) are significantly different during ambulation when compared with those of children without amputations.[1-4] These differences have been attributed to prosthetic limb mechanical deficiencies.[3-5] Although quantifying mechanical differences provides information useful for the improvement of prosthetic limbs, it does not address how the mechanical differences affect energy costs in children with BKAs.

Studies comparing the energy costs during walking in adults with BKAs have shown that they have the same energy costs as adults without amputations.[6-10] This similarity was achieved, however, because the persons with amputations walked at slower speeds. When the adults with BKAs walked at the same speed as adults without amputations, their energy costs were approximately 32% greater.[8] Gonzalez et al[7] found that adults with BKAs and long residual limbs (greater that 8% of total body height) had a 10% increase in oxygen consumption ([Vo.sub.2]) compared with adults without amputations, whereas adults with BKAs and short residual limbs (less than 6% of total body height) had a 40% increase when the groups performed the same activities.

It is currently unknown whether it is possible to generalize these differences in energy costs between adults with and without BKAs to children. A preliminary study conducted in our laboratory indicated that children with BKAs (n=3) consumed 10% more oxygen than did children without amputations (n=2) when walking at the same speeds.[1] Due to the small numbers in each group, it was not possible to generalize these trends to the larger populations of children with BKAs.

The purpose of our investigation was to compare physiological measures of heart rate and [Vo.sub.2] during walking (1) between children with BKAs and long residual limbs and children with BKAs and short residual limbs and (2) between children with BKAs and children without amputations.

Method

Subjects

Ten children (1 female, 9 male) with BKAs and 14 children (8 female, 6 male) without amputations volunteered as subjects for this study. Descriptive measures for the children with BKAs and the children without amputations are given in Tables 1 and 2, respectively. All children with BKAs had a prosthesis for at least 4 years and said they were comfortable with the fit of the prosthesis. To compare children with different residual limb lengths, the children with BKAs were placed into two groups by a physical therapist according to the international classification of amputations. One group consisted children with BKAs and residual limbs of less than two thirds of intact limb length. The other group consisted of those children who had a stump length equal to or greater than two thirds of intact limb length. These groups were considered short- and long-limbed, respectively, and were similar to the grouping of Gonzalez et al.[7] Medical histories indicated that, except for the amputation in the children with BKAs, all children were in good health and had no known medical problems. Subjects and parents were familiarized with the testing procedures and equipment prior to signing human consent forms.

[TABULAR DATA OMITTED]

Table 2. Characteristics of Children Without Amputations (n=14)

                        Body   Chosen
               Height   Mass   Walking Speed
     Age (y)   (cm)     (kg)   (m/s)           Gender

      6(a)     105.0    15.8   0.81            M
      7        126.5    23.2   0.97            F
      9        127.0    28.0   1.17            M
      8        133.0    34.9   1.97            F
     10        137.0    30.8   1.22            M
     10        145.0    40.0   1.00            F
     11        152.0    36.8   1.00            F
     11        152.5    42.1   1.33            M
     12        154.0    38.6   1.19            F
     13        159.5    56.5   1.33            F
     14        164.0    63.4   1.17            M
     14        170.5    54.1   1.03            F
     17        182.5    68.6   1.17            M
X    11.1      148.4    42.5   1.16
SD    3.0       21.0    16.2   0.28

(a) Subject walked at each speed only once.

Procedure

Oxygen uptake (in milliliters per kilogram per minute) was measured using a Quinton Instruments Model 24-72 Treadmill System[*] and a Horizon Metabolic Measurement Cart System.[dagger] The gas analyzers were calibrated before and after each test. Volume and temperature were calibrated prior to each test. Heart rate (in beats per minute) was monitored during testing using a Polar Sport Tester PE 3000.[double dagger] The Sport Tester system consists of a transmitter that is attached to an electrode belt. The electrode belt was secured around the subject's chest at the level of the fifth intercostal space below the left nipple. The Sport Tester system permitted continuous monitoring of heart rate.

Subjects were familiarized with the treadmill (approximately 3-5 minutes) and then asked to select their freely chosen walking speed (CWS). The CWS for each subject was determined by increasing and decreasing the speed of the treadmill until the subject verbally communicated which speed was the most comfortable and approximated the speed the subject would likely choose if he or she were walking about during an average day. The CWS plus 20% and CWS minus 20% were then calculated. Subjects were fitted with a headgear apparatus and then familiarized with each of the three speeds while wearing the headgear. In addition, a fixed speed of 1.2 m/s was also tested because it had been used in related investigations.[1-3] The subjects then rested until their heart rate returned to the pre-warm-up state.

The headgear system, designed to secure the mouthpiece to the expired air hose, was connected to a metabolic cart and was fit onto each child's head along with the mouthpiece and a noseclip. To record pretest heart rates, the subjects were asked to stand in a relaxed manner for 2 minutes prior to starting the test. They then walked for 2 minutes at each of the four speeds in a randomly assigned order. No rest was given between speeds, and changes to the new speeds were made with the subjects on the treadmill. In this way, the subjects were permitted to gradually adapt to new speeds. All tests were repeated, with the exception of the fixed-speed test, which each subject completed only once. For some smaller subjects (Tabs. 1 and 2), the treadmill protocol was modified such that they were only tested once at each speed. After the treadmill tests were completed, each subject was seated in a chair and heart rate was again monitored for 2 minutes. Heart rate and [Vo.sub.2] were measured every 30 seconds, but only the final 30 seconds in each of the 2-minute intervals was used for calculations. Rose et al[11] determined that the [Vo.sub.2] measured during the final 30 seconds of a 2-minute test for a group of 18 children without known impairments and for a group of 13 children with cerebral palsy reflected a steady-state level. This result was confirmed in a pilot investigation' conducted prior to this study. Finally, to assess the repeatability of the measures, 2 children without amputations who were not involved in the study were retested 1 week after the initial testing session (Tab. 3).

[TABULAR DATA OMITTED]

Data Reduction and Analysis

Oxygen consumption relative to distance ([SVo.sub.2]) (in milliliters per kilogram per meter) and heart rate at the 2-minute marks along with the corresponding walking speed and physical characteristics for each subject were entered into a computer file for statistical analysis. A physiological cost index (PCI) was derived using the following equation:

(1) PCI=(hw-rhr)/s

where "hw" is heart rate while walking (in beats per minute), "rhr" is heart rate at rest (in beats per minute), and "s" is average walking speed (in meters per minute).[12] Percentage of maximum heart rate (%MHR) was derived for each walking speed using the following equation:

(2) %MHR=sshr/age hr X 100

where "sshr" is the steady-state exercise heart rate (in beats per minute) and "age hr" is the age-predicted maximum heart rate (220-age) (in beats per minute).[8] Data for the two trials at the same speeds were averaged for each subject.

Two multivariate analyses of variance (MANOVAs) were used to determine significant differences in physiological measures between children with BKAs and long residual limbs and children with BKAs and short residual limbs and between children with BKAs and those without amputations, The within-group factor was walking speed. The MANOVAs were necessary because the dependent variables (ie, heart rate, PCI, %MHR, [SVo.sub.2]) were correlated. Statistical significance were based on a probability level of <.05. The post hoc test used in this investigation was an F test.

Because the population density in the region where the study was conducted was low, the number of subjects with amputations available for this study was limited. A test to determine the power associated with the investigation was therefore performed.[12]

Results

Comparison Within Children

With Below-Knee Amputations

There were no differences between the children with BKAs and long residual limbs and the children with BKAs and short residual limbs with regard to personal characteristics (ie, CWS, resting heart rate, age, height, body mass, and gender) (Tab. 1). In addition, there were no differences between the two groups of children with an amputation with respect to the physiological characteristics (ie, heart rate, [SVo.sub.2], PCI, %MHR) (Tab. 4).

[TABULAR DATA OMITTED]

Comparison Between Children

With Below-Knee Amputations

and Those Without Amputations

There were no differences between the children with BKAs and the children without amputations for age, height, and weight. Only one female child with a BKA was available for testing. There were also no differences between groups for [Vo.sub.2] as a function of age.

The means and standard errors for the four physiological measures for the children with BKAs and the children without amputations are illustrated in Figures 1 through 4. The two-way repeated-measures MANOVA (groups [subjects with BKAs and subjects without amputations], walking speeds [20% below CWS, CWS, 20% above CWS, and fixed]) conducted for the physiological measures (heart rate, [SVo.sub.2] PCI, and %MHR) provided varying results. The multivariate main effects for groups was significant (F=3.42; df=4,17; P<.05). The group univariate F-test results for heart rate and %MHR were not significant (F=0.00; df=1,20; P<.05 and F=0.00; df=1,20; P<.05, respectively), whereas the group univariate F-test results for [SVo.sub.2] and PCI were significant (F=9.28; df=1,20; P<.05 and F=4.86; df=1,20; P<.05, respectively). The multivariate main effects for walking speed were significant (F=19.69; df=12,9; P<.05). The speed univariate test results were significant for heart rate (F=21.20; df=3,18; P<.05), [SVo.sub.2] (F=28.16; df=3,18; P<.05), and %MHR (F=6.88; df=3,18; P<.05) and not significant for PCI (F=2.37; df=3,18; P<.05). There was no interaction effect (P>.05) multivariately or univariately. The power test indicated that for [SVo.sub.2] and PCI, a power of 0.95 or greater existed ([alpha] <.05).

Discussion

We investigated physiological cost differences of walking between children with long and short residual limbs and between children with BKAs and children without amputations. A number of limitations are associated with this investigation. The small number of children with BKAs was limited by the number of children in the local region. This sample size, however, was greater than that previously reported.[1] In addition, because there was no significant difference between the children with long residual limbs and those with short residual limbs, all children with amputations were combined to form a single group. That group was compared with the group of children without amputations.

A further limitation was with regard to the test-retest evaluation. First, it would have been desirable to retest children with amputations as well as those without amputations. Such retesting, however, was not possible due to the time frame of the investigation and the proximity of the children with amputations to the laboratory. Second, the results for the test-retest evaluation indicated that the differences between days was of the same magnitude as the differences between groups. Because only 2 children were retested, it is impossible to determine how the day-to-day effect would have influenced the group effect. Finally, the age range of the children was 6 to 18 years. Astrand and

Rodahl[14] have reported that [Vo.sub.2] in boys and girls is about the same until about year 14. At that time, [Vo.sub.2] increases at a greater rate in boys than in girls. In our investigation, 6 of the 7 children aged 14 years or older were boys. Of these 6 boys, 4 were children with amputations and 2 had no amputations. No difference was found between groups as a function of age. if an influence of age had existed, however, it would probably have moved the results for the group of children with BKAs closer to those of the group of children without amputations.

Physiological measures of effort have been documented for adults with BKAs There is a paucity of available information, however, for children with BKAs. Nielsen et al[8] and Pagliarulo et al[9] reported that adults with BKAs have a higher energy need than adults without amputations when walking at the same speeds (a fixed speed), and the results of the our study suggests that this difference in energy cost extends to children. Adults appear to choose an optimal energy-efficient speed; therefore, adults with BKAs choose a slower walking speed.[6-10] Our study demonstrated no difference in CWS between children with BKAs and children without amputations. A significant difference in energy needs (15% for all walking speeds combined), however, was found at all walking speeds (Fig. 2). Apparently, children with BKAs are willing to walk at the same speed as children without amputations and to accept the consequences of greater energy expenditure. Cooper et al[15] reported that the anaerobic threshold for a group of children was approximately 25 [mL.kg.sup.-1].[min.sup.-1] during walking. In our study, the children without amputations were at 53% of this anaerobic threshold, whereas the children with BKAs were at about 56% of the threshold. Thus, both groups of children easily fell below the anaerobic threshold value reported by Cooper et al.[15] Whether this difference prevents children with BKAs from performing all the activities of children without amputations during a typical day is impossible to determine from our investigation. This issue, however, warrants further investigation.

The CWSs and heart rates recorded in our investigation are comparable to those previously attained on adults with BKAs. Nielsen et al[8] and Pagliarulo et al[9] reported a CWS of 1.1 m/s (SD=1.3) for adults with BKAs, whereas in our investigation the CWS of the children with BKAs averaged 1.05 m/s (SD=0.22). The heart rate recorded for CWS by Pagliarulo et al[9] was 106 b/min (SD=10), whereas the heart rate recorded in our investigation was slightly higher at 114.9 b/min (SD=13.8) (Fig. 1). This difference was likely due to the higher resting heart rate in children compared with adults.[16]

We recorded an increase of 15% in [SVo.sub.2] for children with BKAs as compared with children without amputations at their CWS (Fig. 2). In studies of adults, the general consensus was an approximate 32% increase in [Vo.sub.2], as reflected in [SVo.sub.2] at the CWS.[8] The higher increase seen in the studies of adults may be a result of the significantly slower walking speed for adults with BKAs, which would result in an increase in [SVo.sub.2] (noting that at a slower speed it would take longer to walk the same distance). The lower [SVo.sub.2] values with the higher speeds in our study agree with the results for adult subjects reported by Nielsen et al.[8]

The PCI was studied by Butler et al[17] for children without amputations (n=72, age range=3-12 years). They reported a mean PCI of 0.4 (SD=0.12). They also stated that PCI is independent of age, height, and gender, but may be affected by puberty. There appear, however, to be no studies reporting the effects of stage of puberty on PCI. In our study, we recorded PCIs of 0.37 (SD=0.03) for children with BKAs and 0.25 (SD=0.01) for children without amputations at CWS (Fig. 3). Stage of puberty was not assessed, however, so the possibility of differences in stage of puberty between groups could not be determined. The lower PCI values for children without amputations may be a reflection of the higher resting heart rate for that group. The differences in resting heart rate, however, were not significant between the two groups of children.

Nielsen et al[8] suggested that %MHR was a good indicator of the relative exercise intensity, and they found a %MHR of approximately 65% for CWS. In our study, we found a %MHR of approximately 55% for children with and without BKAs. Nielsen et al[18] recommended, as a general guideline, that %MHR should be below 80%. The results of our study, therefore, indicate that the children with BKAs and those without amputations were well within their physiological limits while walking (Fig. 4).

We found no difference in CWS in the subjects with BKAs when grouped by residual limb lengths. In contrast, Gonzalez et al[7] and Waters et al[10] found a higher CWS for adult subjects with a short residual limb BKA compared with adult subjects with a longer residual limb. Gonzalez et al[7] suggested that this finding may be due to age, duration of amputation, general physical condition, complicating illnesses, and prosthetic type and fit. The difference between the investigations with adults and children with BKAs warrants further scrutiny. This difference, however, could be related to the percentage of maximum oxygen consumption at which the groups feel comfortable during walking.

Other research[2,3,5] has shown biomechanical differences between children with BKAs and children without amputations. The results of our study support the notion that a relationship between biomechanical differences previously reported between the two groups could have a significant effect on physiological function. Although not the focus of this investigation, it would seem reasonable to hypothesize that the reported differences between the two groups of children are a result of the lack of function of the prosthesis and the prosthetic limb. Thus, if the goal is to enable children with BKAS to walk like children without amputations, then research should be directed toward developing prostheses that permit the prosthetic limb to function more like an intact limb. Such a prosthesis would probably permit dorsiflexion and plantar-flexion movements and provide propulsive forces similar to those generated by children without amputations.

Summary

The results of this study indicate (1) that children with BKAs chose walking speeds similar to those of children without amputations, regardless of residual limb length; (2) that no differences existed in heart rates between children with BKAs and children without amputations or within children with BKAs with respect to residual limb length; and (3) that an increase in [Vo.sub.2] existed for children with BKAs.

Children with BKAs had higher energy needs for walking than children without amputations. Whether the increased energy needs prevent or inhibit children with BKAs from having a lifestyle comparable to that of children without amputations is currently unknown and warrants further research.

References

[1] Engsberg JR, Macintosh BR, Harder JA. Comparison of effort between below-knee-amputee children and normal children. Journal of the Association of Children's Prosthetic-Orthotic Clinics. 1991;26(2):46-52. [2] Engsberg JR, Lee AG, Tedford KG. Normative ground reaction force data for able-bodied and below-knee-amputee children during walking. J Pediatr Orthop. 1993;13:169-173. [3] Engsberg JR, Tedford KG, Harder JA. Center of mass location and segment angular orientation of below-knee-amputee and able-bodied children during walking. Arch Phys Med Rehabil. 1992;73:1163-1168. [4] Lewallen R, Dyck G, Quanbury M, et al. Gait kinematics in below-knee child amputees: a force-plate analysis. J Pediatr Orthop. 1986;6:291-298. [5] Engsberg JR, Tedford KG, Harder JA, Clynch G. Timing changes for stance, swing and double support in a recent child below the knee amputee. Pediatric Exercise Science. 1990;2:255-262. [6] Ganguli S, Datta SR, Chatterjee BB, Roy BN. Metabolic cost of walking at different speeds with patellar tendon-bearing prosthesis. J Appl Physiol. 1974;36:440-443. [7] Gonzalez EG, Corcoran PJ, Reyes RL. Energy expenditure in below-knee amputees: correlation with stump length. Arch Phys Med Rehabil. 1974;55:111-119. [8] Nielsen DH, Shurr DG, Golden JC, Meier K. Comparison of energy cost and gait efficiency during ambulation in below-knee amputees using different prosthetic feet: a preliminary report. Journal of Prosthetics and Orthotics. 1989;1(1):24-31. [9] Pagliarulo MA, Waters RL, Hislop HJ. Energy cost of walking of below-knee amputees having no vascular disease. Phys Ther. 1979;59:538-534. [10] Waters RL, Perry J, Antonelli D, Hislop H. Energy cost of walking of amputees: the influence of level of amputation. J Bone Joint Surg. 1976;58:42-46. [11] Rose J, Gamble JG, Medeiros J, et al. Energy cost of walking in normal children and in those with cerebral palsy: comparison of heart rate and oxygen uptake. J Pediatr Orthop. 1989;9:276-279. [12] MacGregor J. The objective measurement of physical performance with long-term ambulatory physiological surveillance equipment (LAPSE). In: Scoot FC, Raftery EB, Goulding L, eds. Proceedings of the Third International Symposium on Ambulatory Monitoring. London, England: Academic Press Inc (London) Ltd; 1979:29-39. [13] Lieber RL. Statistical significance and statistical power in hypothesis testing. J Orthop Res. 1990;8:304-309. [14] Astrand PO, Rodahl K. Textbook of Work Physiology: Physiological Bases of Exercise. New York, NY: McGraw-Hill Book Co; 1977:319. [15] Cooper DM, Berry C, Lamarra N, Wasserman K. Kinetics of oxygen uptake at the onset of exercise as a function of growth in children. J Applied Physiol. 1985;59:211-217. [16] Moss AJ. Indirect methods of blood pressure measurement. Pediatr Clin North Am. 1978;25:3-4 [17] Butler P, Engelbrecht M, Major RE, et al. Physiological cost index of walking for normal children and its use as an indicator of physical handicap. Dev Med Child Neurol. 1984;26:607-612. [18] Nielsen DH, Amundsen LR. Exercise physiology: an overview with emphasis on aerobic capacity and energy cost. In: Amundsen LR, ed. Cardiac Rehabilitation. New York, NY: Churchill Livingstone Inc; 1981:11-28.