Cerebral palsy (CP) is the most common physical disability in childhood, affecting two to three out of every 1000 children born each year in Australia (Stanley et al. 2000). Nutrition and growth abnormalities are common in this group, with both under nutrition and growth failure, as well as overweight and obesity frequently reported (Reilly and Skuse 1992; Stallings et al. 1993a; Stallings et al. 1993b; Stevenson et al. 1994). In order to appropriately prevent and treat these problems a thorough understanding of the nutritional needs and energy requirements of children with CP is essential. However, data regarding the energy requirements of children with CP are limited as there have been few studies that have measured total energy expenditure (TEE) in this group (Azcue et al. 1996; Bandini et al. 1991; Stallings et al. 1996; van den Berg Emons et al. 1995).
The energy expenditure during walking of children with CP has been investigated on many occasions as an objective measure of gait inefficiency. There is good evidence to suggest that children with CP expend more energy during walking than similarly aged non-disabled children (Bowen et al. 1998b; Bowen et al. 1999; Boyd et al. 1999; Unnithan et al. 1996). The primary aim of the research detailed in this thesis was to investigate what impact the increased energy expended during walking has on physical activity levels (PAL), TEE and energy requirements. Secondary aims included investigation of the relationships between resting energy expenditure (REE), TEE and body composition, and comparison of measured REE with that predicted by Schofield equations (Schofield 1985).
Nineteen ambulatory children with CP, aged five to 12 years, and 25 similarly aged, non-disabled children participated in this study. Due to technical difficulties, isotopic data relating to measurements of TEE were available for only 16 children with CP and 16 non-disabled children.
The children with CP expended more energy when walking than the control group (13.8 ± 4.9kJ/min versus 10.3 ± 2.3kJ/min). In addition, the children with CP walked at a lower velocity than the non-disabled children (61 ±10m/min versus 72 ± 8m/min). TEE was decreased in the children with CP compared to the control group (7012 ± 1268kJ/day and 8309 ± 2088kJ/day respectively) due to decreased activity related energy expenditure (AEE) (1816 ± 858kJ/day versus 2845 ± 1280kJ/day), with a resultant reduction in PAL (1.57 ± 0.23 versus 1.79 ± 0.26). These results show that children with CP, who expend more energy during walking, have decreased PAL and decreased energy requirements. These findings have important implications when estimating the energy requirements of this group.
Schofield equations accurately predicted REE in the children with CP at the group level, however, due to large limits of agreement, use of these equations at the individual level may lead to unacceptable error. The results of this research also suggest that it is not necessary to measure height when predicting REE in children with mild CP. This is advantageous as the high frequency of scoliosis and joint contractures can make measurement of height difficult, if not impossible, in this group.
Oxygen consumption during walking is frequently investigated, in children with CP, to provide quantitative measurement of gait inefficiency. A number of different techniques have been used to adjust oxygen consumption for body weight and walking velocity in this group, including, (1) division by body weight, (2) division by both body weight and walking velocity (i.e. oxygen consumption per kg body weight per metre walked, this is often referred to as oxygen cost), and (3) the use of net oxygen consumption (oxygen consumption during walking minus the oxygen consumed at rest). The third aim of this research was to evaluate these methods for the adjustment of oxygen consumption during walking in children with CP. The results show that typical methods used to adjust oxygen consumption for body weight and velocity may result in erroneous conclusions being drawn. Other more suitable forms of adjustment are described in this thesis.
The final aim of this research was to validate bioelectrical impedance analysis (BIA) as a method for predicting total body water (TBW) in children with CP. There are few body composition assessment methods validated for use in this group. The assumptions inherent in the methodologies may not be valid in children with CP (Reilly and Skuse 1992).
TBW was predicted using impedance measured with a Bodystat 1500, by three previously published equations (Davies and Gregory 1991; Kushner et al. 1992; Pencharz et al. 1997) and the Bodystat algorithm and compared to TBW measured using oxygen-18. The Bodystat algorithm overestimated TBW with the largest bias (CP group 5.8 ± 1.1kg, control group 4.6 ± 1.5kg). The Kushner equation predicted TBW with the least bias (CP group 0.3 ± 1.3kg, control group-0.6 ± 1.3kg), however, the limits of agreement remained large (CP group -2.3 to 2.9kg, control group -3.2 to 2.0kg) suggesting that this equation was accurate for use at the group level but not at the individual level.