| | The Effect of Simulating Weight Gain on the Energy Cost of Walking in Unimpaired Children and Children With Cerebral PalsyPresented to the European Society for Movement Analysis in Adults and Children, September 27–29, 2007, Athens, Greece. Abstract Plasschaert F, Jones K, Forward M. The effect of simulating weight gain on the energy cost of walking in unimpaired children and children with cerebral palsy. ObjectiveTo examine the effect of simulating weight gain on the energy cost of walking in children with cerebral palsy (CP) compared with unimpaired children. DesignRepeated measures, matched subjects, controlled. SettingUniversity hospital clinical gait and movement analysis laboratory. ParticipantsChildren (n=42) with CP and unimpaired children (n=42). InterventionsAddition of 10% of body mass in weight belt. Main Outcome MeasuresEnergy cost of walking parameters consisting of walking speed, Physiological Cost Index, Total Heart Beat Index, oxygen uptake (V̇o2), gross oxygen cost, nondimensional net oxygen cost, and net oxygen cost with speed normalized to height were measured by using a breath-by-breath gas analysis system (K4b2) and a light beam timing gate system arranged around a figure 8 track. Two walking trials were performed in random order, with and the other without wearing a weighted belt. ResultsChildren with CP and their unimpaired counterparts responded in fundamentally different ways to weight gain. The unimpaired population maintained speed and V̇o2 but the children with CP trended toward a drop in their speed and an increase in their V̇o2. The oxygen consumption of children with CP showed a greater dependence on mass than the unimpaired group (P=.043). ConclusionsAn increase of a relatively small percentage in body mass began to significantly impact the energy cost of walking in children with CP. This result highlights the need for weight control to sustain the level of functional walking in these children. IN THE PRESENCE OF PATHOLOGY, the human gait mechanism can be disrupted with the inevitable consequence that greater effort is required for ambulation. Pathologic insult can result in structural, mechanical, and control problems, and all of these are prevalent in CP. The incidence of CP has continued to remain relatively unchanged in recent decades despite continued improvement in fetal monitoring and neonatal intensive care, and the condition still presents in 2 out of every 1000 births.1 The natural history of CP and the capacity to walk is dependent on the evolution of appropriate muscle power, selective control, intelligence, and the presence or absence of skeletal deformation.2 Appropriate assessment and consequently targeted treatment intervention can favorably influence prognostic outcome,3 and, as a result, a number of techniques for the evaluation of walking function and efficiency have been developed. Previously, emphasis was placed on mechanical outcome measures such as evaluating changes from normative kinematics and kinetics. Although these measures describe deviations from “normal,” they do not reflect the capacity and the effort required for walking in CP. Other measures such as the calculation of mechanically derived powers4 and the implementation of quantitative electromyographic techniques5 have also been used to evaluate walking effort. However, these are confounded by muscle cocontraction that consumes energy without producing measurable (gross) mechanical movement. Physiologically based assessment techniques are aimed at addressing this problem. Here, however, limitations are associated with calorimetric approaches, with direct calorimetry being impractical in the gait laboratory environment because it requires immersion inside a calorimeter for extended periods of time.6 Thus, physiologic approaches are limited to indirect methods such as PCI,7 oxygen carts, and, latterly, portable ambulatory respiratory gas analysis systems such as the Cosmed K4b2.a Although portable gas analysis systems appear to be the only practical approach, there is no consensus as to which is the most appropriate method for calculating the energy cost of walking using cardiovascular and respiratory physiologic parameters. Some authors promote the use of oxygen cost indices normalized to body surface area8 height (leg length)9 or to a carefully selected combination of parameters designed to produce a nondimensionalized index.10 More recently, a THBI was proposed11 in place of the now less popular PCI.7 Others believe that the ideal measure of gait efficiency should be independent of growth and maturation.9, 10 One method, in particular, uses an index normalized to body mass,10 and this efficiency index should therefore be independent of body mass. However, clinical observation frequently reveals that an increasing body mass challenges the walking capacity of some patients with CP, and it is this very aspect that puts them at risk of further loss of walking mobility and thus greater potential for additional weight gain. The physical activity level of children with CP is known to decrease over time with growth and maturation when compared with an unimpaired peer group,12 and it is acknowledged that weight gain is linked to decreased walking mobility.13, 14 Reductions in activity levels in this cohort are also often exacerbated by problems of spasticity and the development of contractures. Thus, the problem of decreasing walking efficiency in these children becomes multifactorial depending on a combination of other factors such as height gain, altered fitness, and muscle strength. Of these factors, body mass can be modified in a simple and controlled way because artificially added weight allows all of these other variables to be held constant. Although the impact of body mass on walking efficiency is documented in other patient groups,15, 16, 17 there are currently no published data on the effect of simulated weight gain on measures of walking efficiency in children with CP. Therefore, in this study, we aim to assess the effect of adding the equivalent of 10% of additional body weight on the energy cost of walking in unimpaired children and in a group of children who have been diagnosed with CP. This simulated weight-gain study may provide an insight into the impact of increasing weight on the long-term multidisciplinary management that is aimed at maintaining and improving walking efficiency in children with CP. Methods  Participants Children (n=42) who have been diagnosed with a spectrum of CP (26 boys and 16 girls) and unimpaired children (n=42; 21 boys and 21 girls) were randomly identified from the electronic clinical database and asked if they would volunteer to participate in this study. We anticipated that this number would be required to achieve an 80% statistical power based on our previous experience of the SDs of the energy cost parameters used in our study. Power calculations were performed using Graphpad Statmate version 2.0.b The children with CP were matched with the group of unimpaired children for age, sex, height, and weight (table 1). All children with CP were ambulatory community walkers. Although all subjects could walk independently, some also regularly used walking aids (splints, k-walker, elbow crutches, canes) during everyday activities, and, therefore, these subjects were given the choice of using their aids during the tests. All subjects walked barefoot unless footwear was a part of their normalized walking aid regimen. | | |  | Groups | Age (y) | Age range (y) | Height (m) | Weight (kg) | |
|---|
| Unimpaired | 11.4±2.1 (12) | 6–15 | 1.50±0.13 | 42.6±10.7 | | | Cerebral Palsy | 12.0±2.6(12) | 6–16 | 1.47±0.16 | 40.5±12.8 | | | | |
Local ethical approval and informed consent were obtained before testing, and all children were informed that they could withdraw from the study at any time without compromising their future management. The children with CP were classed as either grade 1 or 2 according to the modified Gross Motor Function Clinical Score,18 which has recently been expanded and revised.19 Equipment All of the testing equipment was used regularly within the clinical gait laboratory, and calibrations were performed according to manufacturer's instructions and laboratory practice. The equipment consisted of a portable gas analysis system (Cosmed K4b2) that measured the percentage oxygen and carbon dioxide in expired air, the respiratory rate, the tidal volume, and the heart rate. A facemask was worn that covered both the nose and mouth. The mask contained a turbine with a flow rate sensor and tubing that carried a sample of the inspired and expired air to a small analysis and logging unit that was mounted on the subject's chest. The test area was a dedicated gait laboratory with a figure 8 track of 34-m length and an inhouse, overhead-mounted timing gate system. The track was designed to permit continuous walking along straights and curves that were not too tight for physically impaired subjects to negotiate when using elbow crutches, canes, or k-walkers. The track design also precluded bias to the same leg on corners to allow for balancing of the number of left and right turns. Procedure Testing took place in a thermostatically controlled gait laboratory environment. A period of familiarization with the test area, testing procedure, and equipment was provided for each subject. Two walking tests were performed at least 4 hours after the last meal: 1 test without additional weight and the other in which weight equivalent to 10% of body weight was carried. This additional weight was placed in a waist-mounted belt into which different weight increments ranging between 0.1kg and 1kg could be evenly distributed. The order of the 2 walking tests (with and without added weight) was randomized, and a rest period of at least 1 hour between walks was ensured. Water was made available to all children between testing as required. The phases within each walking test were as follows: (1) resting: a minimum of 5 minutes of supported sitting (on a chair with a backrest) was used to identify the children's baseline heart rate and oxygen consumption; (2) walking: after 5 minutes of sitting, the child walked around the figure 8 track for 8 minutes at his/her own preferred walking speed; and (3) recovery: data collection for 5 minutes (or longer if required to return to baseline values) in supported sitting concluded the testing protocol. Data Recording and Processing Methods Data collection was performed using the K4b2 and timing gate data-logging systems was commenced simultaneously and both recorded throughout and in parallel. The timing gate system was sampled at 20Hz, giving a resolution on timing gate events of .05 seconds. The K4b2 data were recorded on a breath-by-breath basis, which required subsequent interpolation (onto a 1-Hz time base/sampling frequency) to allow combination with the speeds calculated from the timing gate events. These data were then filtered with a low-pass second-order Butterworth filter with a cutoff frequency of .005Hz. This frequency was selected based on the time required to reach steady state during walking, which has been reported to occur between 2 and 4 minutes.20, 21, 22 The processed data for the 2 walking conditions of each subject for oxygen consumption (and carbon dioxide production) together with the derived speed data were then used to calculate the following measures: or In this study, all parameters were calculated on the basis of instantaneous values. For example, the PCI at any time, t, was calculated by using the following equation: This enabled the evaluation of each energy cost parameter throughout the duration of the trial rather than relying on the calculated parameters that used the average data for the whole of the walking section of the trial. Data points at times at which the respiratory quotient was .90 or higher were removed from the analysis. These were considered to represent anaerobic or near anaerobic activity even though in practice they often coincided with the subject speaking (despite instruction not to speak unless necessary). For the purposes of this study, only the values that were calculated during predefined steady-state conditions were used. Theoretically, steady state should be defined when the absolute slope of V̇o2 with respect to time is equal to 0. In practice, because of practical issues such as measurement system (white) noise, a slope of 0 would rarely if ever be measured. Therefore, steady state was considered to occur during periods when the absolute slope of V̇o2 with respect to time was less than .00025mL·kg·s−2. In essence, this near 0 value was considered to be representative of a 0 value in d(V̇o2)/dt to account for the inevitable low-level numeric rounding errors and noise that remain even after filtering. The energy cost parameters reported are the mean of the instantaneous parameters calculated during the steady-state V̇o2 periods defined previously. Data Analysis Analysis was performed by using the paired Student t test and the Wilcoxon matched-pairs signed-rank test to test for differences. The Pearson product-moment correlation coefficient was used to examine correlations between the 2 matched paired groups of children during the 2 test conditions of walking with and without additional weight. The assumptions of normality were tested by using the Kolmogorov-Smirnov test. When data did not satisfy normality, nonparametric tests were applied. The level of statistical significance was set at P less than or equal to .05. The effect of added weight on each of the energy cost parameters was also treated in 2 different ways for analysis as either corrected or uncorrected body mass values. The corrected values were derived by using the subject's body mass + added weight for the added weight trials (ie, just as if the subject had undergone a weight gain). The uncorrected values were calculated by using only the subject's body mass without regard to the additional weight that he/she carried during the added weight trials (ie, just as if the subject was carrying an external weight). It should be noted that although the calculation of V̇o2, gross oxygen cost, nondimensional net oxygen cost, and net oxygen cost with speed normalized to height all incorporate body mass, the calculations of speed, PCI, and THBI do not (table 2). Results A summary of descriptive statistics of the energy cost parameters for both groups in the 3 conditions of no added weight, carrying weight and, in the case of the oxygen-based parameters, the weight-gain scenario are presented in table 2. Although kinematics, step, and stride parameters were not measured alongside speed, no gross change in the overall gait pattern of any of the subjects was observed. More specifically, no increase in crouch or instability was noted. Speed The walking speed in the group of unimpaired children was significantly faster (t=8.530, P<0.001) than the children with CP before any weight was added (1.16±0.14m/s and 0.93±0.21m/s unimpaired and CP means ± SDs, respectively). This remained the case when both groups carried 10% of their body mass. The unimpaired group did not drop their walking speed (t=.07, P=.47), but there was a trend toward decreasing speed in the children with CP (t=1.38, P=0.09). The impact of weight was also reflected by the increase in SD and range of speeds in the group of children with CP that was not observed in the unimpaired group (see table 2). PCI and THBI The PCI significantly increased in both groups when weight was carried (t=8.27, P<0.001 for the unimpaired; t=5.83, P<0.001 for the children with CP). PCI was also significantly higher in the CP group than in the unimpaired group both before (t=8.19, P<0.001) and after (t=7.67, P<0.001) weight addition. The increase in PCI with weight was larger in the group with CP than that of the unimpaired children (t=2.18, P=0.02). The THBI data failed the normality test. Wilcoxon signed-rank tests revealed a similar pattern of significant findings to those of PCI. Oxygen-Based Parameters Normality tests revealed a mix of normal and nonnormally distributed data, and, therefore, we applied the Student t test (in the case of normal data) and the Wilcoxon signed-rank test (in the case of nonnormal data). The V̇o2 of the group of unimpaired children was significantly lower than that of the group of children with CP before any weight was added (W=735, P<0.001). This remained the case when both groups carried 10% of their body mass (W=767, P<0.001). Similar significant findings were obtained for the gross oxygen cost (W=885, P<0.001), the nondimensional net oxygen cost (t=8.39, P<0.001), and the net oxygen cost with speed normalized to height (NOCh) (t=9.09, P<0.001) energy cost parameters, with the children with CP displaying significantly larger values than the unimpaired children for each of these parameters. In the carrying-weight scenario, both groups of children used significantly more oxygen per kilogram than when they did not carry weight (t=4.27, P<0.001 for unimpaired; W=445, P=0.002 for the children with CP). Simulating Weight Gain There were no significant changes with weight gain in any of the oxygen-based energy cost parameters in the unimpaired group of children. The children with CP had significantly higher nondimensional net oxygen cost (W=311, P<0.03) and net oxygen cost with speed normalized to height (W=305, P=0.03) in the weight-gain scenario except for V̇o2 (W=281, P=0.08) and gross oxygen cost (W=49, P=0.38). In the case of gross oxygen cost, the actual increase (ie, weight gain minus no added weight) in energy cost parameters with weight gain was significantly higher in the group of children with CP when compared with the increase in energy cost in the control group (W=895, P<0.001). Relationship of Mass to the Oxygen-Based Energy Cost Parameters A strong linear relationship was identified with mass for both groups of children (64% and 60%, respectively of the shared variance) for the unimpaired children and children with CP (fig 1). The 2 regression lines did not possess homogeneity of regression slopes (P=0.043, F=4.219), confirming that the increase in oxygen required per kilogram of increased body mass was greater in the CP group than in the group of unimpaired children. The relationship between mass and oxygen consumption was as expected; however, in theory, there should have been no relationship between mass and the energy cost parameters nondimensional net oxygen cost and net oxygen cost with speed normalized to height in any of the 3 conditions of no added weight, carrying weight, and weight gain (fig 2). The Pearson product-moment correlation revealed a small (.10≤ |r| ≤ .29) to medium (.30≤ |r| ≤.49) correlation with mass in all cases except in the case of net oxygen cost with speed normalized to height with no added weight in the children with CP (r=.037). However, it was only in the case of the nondimensional net oxygen cost no added weight in the unimpaired group that a significant correlation with mass was found (P=.37, r=–.323). Homogeneity of regression slopes for nondimensional net oxygen cost and net oxygen cost with speed normalized to height with mass was examined. This revealed nonsignificant differences between the slopes of the unimpaired children and those with CP for both nondimensional net oxygen cost and net oxygen cost with speed normalized to height of P equal to .333 and .981, respectively, for the worst-case slope-difference scenario of weight gain. Discussion From a clinical perspective, it is clear that children who are able to walk even with a mild form of CP use more energy to do so than their unimpaired counterparts. It is also frequently observed by those regularly involved with the care of children with CP that as these children become heavier walking becomes more energy intensive. Thus, a potentially vicious cycle of increasing weight and decreasing mobility becomes established. To try to formalize these observations, we simulated the effect of a relatively small weight gain on children with CP and on an age-matched control group of unimpaired children. We then examined what effect this would have on several parameters that reflected the energy expenditure (energy cost) of walking. The energy cost of walking in children with CP was indeed found to be significantly higher than in the age-matched control group. This was true for both test conditions when no additional weight was carried and when 10% of body weight was artificially added in a weight belt that was carried around the waist. In many ways, this result was expected and emphasizes the need to minimize stresses that are already applied to the continually developing neuromuscular skeletal function of children with CP. It was evident that the self-selected walking speed in children with CP was already slower than that of the age-matched control group. Additionally, the children with CP tended to slow their walking speed when weight was added. This reduction in speed was not significant, but it did show a trend toward significance, and this was supported by an increase in the observed spread of speeds in these children. Fitness levels may play an important role in this aspect, with some children being generally more active than others in everyday activities. PCI increased in both groups when weight was added, and this was significantly higher in the children with CP than in the unimpaired children, both before and after weight addition. Similar results were obtained for THBI. It has been reported that unimpaired adult females can carry an extra 20% of their body weight without compromising their metabolic cost.23 However, in our study, in which subjects were required to carry an additional 10% of their body weight, significant rises were observed in V̇o2 consumption in both groups of children. This is in agreement with Hanson24 who found that weight gain in unimpaired adult males resulted in a proportionate increase in oxygen usage during walking, irrespective of whether the additional weight (up to 19%) was in the form of induced obesity or a backpack load. Despite the absolute V̇o2 measures under loaded and unloaded conditions being higher in our group of children with CP than in our unimpaired group, the fractional changes were similar at approximately one tenth (10%) of the added 10% weight. Differences were also observed between our 2 groups. The percentage change in the SD of V̇o2 under loaded conditions revealed that the distribution of V̇o2 in children with CP tended to become more closely clustered. It may be that this group possesses fundamentally different pulmonary physiology or that they were working toward the upper limit of their aerobic capacity, and, hence, further increases in V̇o2 would be less likely. This drop in speed was not observed in the unimpaired children, and this suggests that when weight is added the children with CP may be unable to maintain unloaded self-selected walking speeds. Alternatively, these children may elect to decrease their walking speed to limit the need to increase their oxygen consumption. Unnithan et al25 concluded that higher intersegmental mechanical power flows accounted for the majority of the differences between unimpaired children and children with CP. This seems to imply that these differences are caused by biomechanic inefficiency, and, indeed, this is supported by the findings of Unnithan et al26 who related oxygen consumption to an increased cocontraction in children with CP. The combined effect of a drop in speed and change in V̇o2 on the different energy cost measures that use V̇o2 and speed in their calculation were therefore examined. It was found that with added weight, the percentage increases in gross oxygen cost, nondimensional net oxygen cost, and net oxygen cost with speed normalized to height were approximately twice in the CP cohort when compared with that of the unimpaired group. This result emphasizes the need to use energy-efficiency measures that allow for and are sensitive to speed changes when assessing this group of patients. It also highlights the problem of performing energy-efficiency measurements on a treadmill. Some children with CP do possess self-selected speed adjustment and indeed the ability to walk safely on a treadmill. However, walking at a fixed speed on a treadmill would mask or prevent self-selected speed adjustment in those patients who, as in our study, use speed adjustment to cope with their biomechanic inefficiency. The energy cost of walking index (net oxygen cost with speed normalized to height) is said to be essentially independent of speed at least in unimpaired adults below the self-selected walking speed.9 However, the results of this study show a larger net oxygen cost with speed normalized to height increase in the CP group, with similar changes in V̇o2 for both groups (in the carrying-weight scenario). Because net oxygen cost with speed normalized to height is dependent on V̇o2 and speed, then, by implication, the larger change in net oxygen cost with speed normalized to height in the CP group must be related to speed drop. This would imply that net oxygen cost with speed normalized to height may not be independent of speed in children with CP. Presumably, this may be related to biomechanic differences in the relationship between height (leg length) and step (or stride) length in the children with CP compared with the unimpaired group. This would be expected given the presence of joint contracture in children with CP. The energy cost of walking parameters incorporating oxygen consumption all normalize to mass (in slightly different ways) to allow the comparison of energy cost of walking indices between subjects of different mass. In theory then, when incorporating the additional weight into the body mass (weight-gain scenario), the calculated energy cost of walking indices, nondimensional net oxygen cost, and net oxygen cost with speed normalized to height should show no significant change. Our results for the unimpaired subjects tend to support this theory despite the fact that the data support a weak relationship between the indices and mass. In contrast, nondimensional net oxygen cost and net oxygen cost with speed normalized to height still show significant rises in the weight-gain simulation in children with CP. This is perhaps not surprising because mass is shown to have a stronger relationship with oxygen consumption in the group of children with CP. In essence, it would seem that the nondimensional net oxygen cost and net oxygen cost with speed normalized to height indices do not appear to normalize for mass as effectively in our children with CP as they do in children without pathology. Study Limitations Ten percent of body mass was the amount by which our subjects' weight was increased in the walking trials. Although this fraction was small but imposed by ethical committee constraints, it did reveal significant differences in the energy cost demands for walking between our 2 groups of children. These differences may have been more apparent if a slightly larger percentage had been used to simulate weight gain, particularly in the examination of the recent impact that nondimensional energy cost indices have had over the last half decade. Further study that used a range of percentage incremental body mass increases would also enable a precise weight range to be determined at which point major impact would be observed on the energy requirements for continued ambulation in CP. It is acknowledged that the distribution of the additional weight may also impact our results in that we chose to place this near to the center of mass of our subjects. The location of the center of gravity may indeed vary between individual subjects and between our 2 subject groups, but because the added mass was only 10%, the impact of the lack of mass distribution may have only limited effect on our findings. Obviously, to confirm this, a separate study would be required. It is also acknowledged that a small number of our CP group walked in their shoes with their usual braces and assistive devices such as canes or crutches. Because the focus of our study was to assess the impact of adding weight, the important issue was to assess the impact of weight change under the child's preferred walking conditions. Therefore, the subjects were asked to walk barefoot and unaided if they felt able to do so, but the ultimate choice to use braces and assistive devices was given to the children to facilitate the conditions that they considered necessary to support their walking ability. The devices were used identically for both trials (with and without added weight). Fundamentally, our study was a pre- and postintervention (weight change) model and not a study on the impact of the use of braces and assistive devices. Therefore, this aspect is not considered to be a major limitation to our study. Our study design involved repeated measures on 2 groups. However, despite a relatively large sample size with expected powers of 80% or greater, some datasets failed normality and homogeneity of variance checks. As a result, we chose to apply a series of paired parametric and nonparametric tests rather than analysis of variance throughout. These issues were probably linked to the wide range of functional ability even within our subjects Gross Motor Function Classification System groupings of I and II. Conclusions The 2 populations of children with CP and unimpaired children responded in 2 fundamentally different ways to added weight. The unimpaired population elected to maintain speed but increased their V̇o2. The group of children with CP showed a comparable percentage increase in their V̇o2 values but in combination with a trend toward a drop in speed with the addition of a load. The isolated measurement of oxygen consumption and/or speed does not provide a full description of the different adaptations of walking in CP between loaded and unloaded conditions. There are 3 energy cost parameters that combine speed and O2 consumption, and each of these showed similar percentage changes with added weight in both the unimpaired and in children with CP. Normalization algorithms are intended to provide indices for walking efficiency that should be independent of speed, weight, and even subject height. The response to added weight when using these indices in unimpaired children and in children with CP was different. This raises the question of the effectiveness of comparison of unimpaired and CP walking efficiencies over time. Irrespective of these issues, our results still support the clinical observation that it is important to control weight in children with CP if these children are to maintain their functional walking capacity. This information may assist in the long-term management decision making for this group of patients, but the results should be consolidated with further study on the impact of different percentage body-weight additions. This would enhance our limited understanding of the link between mass and energy expenditure of walking ability in CP. Suppliers References 1. 1Odding E, Roebroeck ME, Stam HJ. The epidemiology of cerebral palsy: incidence, impairments and risk factors. Disabil Rehabil. 2006;28:183–191. MEDLINE |
CrossRef
2. 2Keen M. Early development and attainment of normal mature gait. J Prosthet Orthot. 1993;5:35. 3. 3Gage JR. Gait analysis in cerebral palsy. New York: Cambridge Univ Pr; 1991;. 4. 4Winter DA. Biomechanics of human movement. Toronto: John Wiley and Sons; 1979;. 5. 5Mian OS, Thom JM, Ardigò LP, Narici MV, Minetti AE. Metabolic cost, mechanical work, and efficiency during walking in young and older men. Acta Physiol (Oxf). 2006;186:127–139. MEDLINE |
CrossRef
6. 6Irsigler K, Veitl V, Sigmund A, Tschegg E, Kunz K. Calorimetric results in man: energy output in normal and overweight subjects. Metabolism. 1979;28:1127–1132. MEDLINE |
CrossRef
7. 7MacGregor J. The objective measurement of physical performance with long term ambulatory physiological surveillance equipment (LAPSE). In: Stott FD, Raferty EB, Goulding L editor. Proceedings of 3rd International Symposium on Ambulatory Monitoring. London: Academic Pr; 1979;p. 29–39. 8. 8Bowen T, Cooley SR, Castagno P, Miller F, Richards J. A method for normalisation of oxygen cost and consumption in normal children while walking. J Pediatr Orthop. 1998;18:589–593. MEDLINE |
CrossRef
9. 9Baker R, Hausch A, McDowell B. Reducing the variability of oxygen consumption measurements. Gait Posture. 2001;13:202–209. Abstract | Full Text |
Full-Text PDF (163 KB)
|
CrossRef
10. 10Schwartz MH, Koop SE, Bourke JL, Baker R. A nondimensional normalization scheme for oxygen utilization data. Gait Posture. 2006;24:14–22. Abstract | Full Text |
Full-Text PDF (676 KB)
|
CrossRef
11. 11Hood VL, Granat MH, Maxwell DJ, Hasler JP. A new method of using heart rate to represent energy expenditure: the Total Heart Beat Index. Arch Phys Med Rehabil. 2002;83:1266–1273. Abstract | Full Text |
Full-Text PDF (139 KB)
|
CrossRef
12. 12Maher CA, Williams MT, Olds T, Lane AE. Physical and sedentary activity in adolescents with cerebral palsy. Dev Med Child Neurol. 2007;49:450–457. MEDLINE |
CrossRef
13. 13Andersen RE, Crespo CJ, Bartlett SJ, Cheskin LJ, Pratt M. Relationship of physical activity and television watching with body weight and level of fatness among children: results from the Third National Health and Nutrition Examination Survey. JAMA. 1998;279:938–942. MEDLINE |
CrossRef
14. 14Holm K, Li S, Spector N, Hicks F, Carlson E, Lanuza D. Obesity in adults and children: a call for action. J Adv Nurs. 2001;36:266–269. MEDLINE |
CrossRef
15. 15Agre JC. The role of exercise in the patient with post-polio syndrome. Ann N Y Acad Sci. 1995;753:321–334. MEDLINE |
CrossRef
16. 16Gazzani F, Bernardi M, Macaluso A, et al. Ambulation training of neurological patients on the treadmill with a new Walking Assistance and Rehabilitation Device (WARD). Spinal Cord. 1999;37:336–344. MEDLINE 17. 17Brown JP. Orthopaedic care of children with spina bifida: you've come a long way, baby!. Orthop Nurs. 2001;20:51–58. MEDLINE 18. 18Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214–223. MEDLINE 19. 19Palisano R, Rosenbaum P, Bartlett D, Livingston M. GMFCS - E & R. http://www.canchild.ca/Portals/0/outcomes/pdf/GMFCS-ER.pdf. 20. 20Corry IS, Duffy CM, Cosgrave AP, Graham HK. Measurement of oxygen consumption in disabled children by cosmed K2 portable telemetry system. Dev Med Child Neurol. 1996;38:585–593. MEDLINE 21. 21Waters RL, Hislop HJ, Thomas L, Campbell J. Energy cost of walking in normal children and teenagers. Dev Med Child Neurol. 1983;25:184–188. MEDLINE 22. 22Waters RL, Lunsford BR, Perry J, Byrd R. Energy-speed relationship of walking: standard tables. J Orthop Res. 1988;6:215–222. MEDLINE |
CrossRef
23. 23Charteris J, Scott PA, Nottrodt JW. Metabolic and kinematic responses of African women headload carriers under controlled conditions of load and speed. Ergonomics. 1989;32:1539–1550. MEDLINE |
CrossRef
24. 24Hanson JS. Exercise responses following production of experimental obesity. J Appl Physiol. 1973;5:587–591. 25. 25Unnithan VB, Dowling JJ, Frost G, Bar-Orr O. Role of mechanical power estimates in the O2 cost of walking in children with cerebral palsy. Med Sci Sports Exerc. 1999;31:1703–1709. MEDLINE |
CrossRef
26. 26Unnithan VB, Dowling JJ, Frost G, Bar-Or O. Role of co-contraction in the O2 cost of walking in children with cerebral palsy. Med Sci Sports Exerc. 1996;28:1498–1504. MEDLINE |
CrossRef
a Movement Analysis Laboratory, University Hospital, Ghent, Belgium b Faculty of Engineering, Institute of Biomedical Technology, University of Ghent, Ghent, Belgium  Correspondence to Frank Plasschaert, MD, Department of Orthopedics & Traumatology, University Hospital Ghent, De Pintelaan 185, 9000 Gent, Belgium
Supported by Research Fellowship Funding, Robert Jones & Agnes Hunt Orthopedic Hospital Oswestry, Shropshire, UK. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprints are not available from the authors. PII: S0003-9993(08)00841-1 doi:10.1016/j.apmr.2008.05.023 © 2008 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
00841-1/journalimage?img=PIIS0003999308008411.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00841-1/journalimage?img=PIIS0003999308008411.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)