SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

Aim  The aim of this study was to investigate the effect of gastrostomy tube feeding on body protein and bone mineralization in malnourished children with cerebral palsy (CP).

Method  Children aged between 4 and 18 years with spastic quadriplegic CP (Gross Motor Function Classification System level V) were recruited from the Children’s Hospital at Westmead to participate in this prospective cohort study. The children had measurements of anthropometry (weight, height, and skinfold), bone mineral content (BMC) by dual-energy X-ray absorptiometry, and total body protein (TBP) by neutron activation analysis before and after gastrostomy tube feeding. Comparison data were collected prospectively from age-matched healthy children and extracted from databases for this study.

Results  A total of 21 children (nine females, 12 males) participated in the study (median age 8y 5mo; interquartile range [IQR] 6y 9mo–11y 10mo). The median length of time of gastrostomy feeding was 19.4 months (IQR 7.7–29.9mo). Significant (p<0.05) improvements were found in the median values for weight (15.4–23.3kg), weight standard deviation scores (SDS; −4.8 to −3.0), height (105.4–118.3cm), per cent body fat (10.7–16.3), TBP (2.4–3.4kg), TBP per cent predicted for height (83.4–99.0), and BMC (469–626g). No significant increases were found in height SDS, TBP per cent predicted for age, or BMC SDS for age or height.

Interpretation  Malnourished children with quadriplegic CP showed significant increases in body fat and protein with gastrostomy tube feeding. No significant change in bone mineralization predicted for age or height was observed.

List of Abbreviations
BMC

Bone mineral content

TBP

Total body protein

SDS

Standard deviation score

Despite the many studies reporting a high incidence of malnutrition in children with severe cerebral palsy (CP), there have been limited studies investigating the effects of nutritional intervention on body composition parameters. The few studies that have been carried out have investigated only changes in growth parameters (weight and height) and body fat, with no exploration into the effects of weight gain on body protein and bone mineralization.1–7

Patients with severe CP have severe muscle atrophy due to neurological injury and disuse, but this is also compounded by malnutrition and poor growth. Moreover, neurological injury may limit protein accretion associated with nutritional rehabilitation. We have previously published studies demonstrating a severe reduction in body protein in children with spastic quadriplegic CP.8,9 Bone disease is also an important consideration in this group, as previous studies have demonstrated large reductions in bone density with an increased fracture rate.10–15

An important question in the nutritional restitution of malnourished patients with CP is whether improved weight gain results in improvement in protein mass and bone mineralization and/or whether it just leads to accumulation of body fat. Thus, the aims of this study were to determine if nutritional rehabilitation via gastrostomy feeding of patients with spastic quadriplegic CP improved total body protein and increased bone mineralization.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

Participants

Twenty-one children (nine females, 12 males; age range, 19y), with a diagnosis of quadriplegic CP were recruited through the Dysphagia Clinic at the Children’s Hospital at Westmead between 2000 and 2008. All children were reliant on wheelchairs for mobility and were dependent on their parent or carer for their everyday needs (Gross Motor Function Classification System [GMFCS] level V). The Dysphagia Clinic is staffed by a paediatric gastroenterologist, a general paediatrician, a nurse, speech therapist, physiotherapist, and dietitian. The purpose of the clinic is to assess and treat children with developmental disabilities and complex feeding and nutritional problems. Children aged less than 4 years were excluded from this study because of the lack of available comparison data in this age range.

The data on the comparison children were from two separate existing comparison databases, as outlined below. All comparison children were healthy with no known medical problems and were of a similar age range to the children with CP (4–19y). The comparison group for the anthropometric and absorptiometry measurements consisted of 172 children (84 females, 88 males) from an existing dual-energy X-ray absorptiometry (DXA) comparison database.16 All children from this database had undergone a DXA measurement and measurements of height, weight, and skinfold anthropometry. Comparison group data for total body protein (TBP) consisted was taken from a group of 72 children (43 females, 29 males) who had undergone TBP measurements at the Children’s Hospital at Westmead.17

This study was approved by the ethics committee of The Children’s Hospital at Westmead. Written informed consent was obtained from the parent or carer of each study participant.

Anthropometry

Two trained observers performed all measurements using standardized techniques.18 In the children with CP, standing height was estimated using knee height. Knee height (SD 0.1cm) was measured with an anthropometer (Holtain Ltd, Crosswell, Crymmych, Dyfed, Wales), with the knee and ankle each bent to a 90° angle. The distance from the heel to the anterior surface of the thigh over the femoral condyles was measured.19 The knee height measurement was then placed into the equations of either Stevenson19 (for children ≤12y) or Chumlea et al.20 (for children >12y) to give an approximate height. Weight measurements (SD 0.1kg) were made with electronic scales (A&D Mercury Pty Ltd, South Australia, Australia) with children wearing minimal clothing. Because the children with CP were unable to stand independently on the scales, the carer stood on the scales while holding the child, then the carer was weighed alone and the difference between the two weights was calculated to give the weight of the child. Weight and height measurements were converted to standard deviation scores (SDS) and were compared with the United States Center for Disease Control growth reference data.21

Skinfold thicknesses were measured in duplicate on the right side of the body at four sites (triceps, biceps, subscapular, and suprailiac) with a Harpenden calliper (British Indicators Ltd, St Albans, Hertfordshire, UK) and compared with Frisancho22 reference data, as well as the comparison group. Owing to the abnormal distribution of the reference data for the triceps skinfold thicknesses, this measure was converted to an SDS according to the methods of Davies et al.23 An estimate of per cent of body fat was calculated from the four skinfold thicknesses using the equations of Brook24 for prepubertal children, and Durnin and Rahaman25 for pubertal children. The pubertal status of each participant was determined by asking the parent or carer to select the pubertal status of their child from a pictorial representation of the five Tanner26 stages of puberty. All anthropometric measures were compared with the comparison group as well as with reference data.

Total body protein by neutron activation analysis

Total body protein was measured by the method of prompt gamma neutron activation analysis (NAA) as previously described by this unit.17,27 The basis of this technique is that the individual is bilaterally irradiated with neutrons from two 252Cf sources while lying supine. 14N is converted to 15N with the emission of a 10.8MeV gamma ray which is specific for nitrogen. Total body nitrogen is calculated by measuring the integral under the nitrogen peak centred on the 10.8MeV of the gamma ray spectrum. By measuring total body nitrogen, total body protein (TBPNAA) can then be determined using the following relationship: mass of protein=6.25 × mass of nitrogen.27,28 The total exposure time is approximately 15 minutes, and the effective dose equivalent delivered during a scan is less than 0.15mSv (quality factor=20 for fast neutrons). The technique has a precision and accuracy of between 1.4% and 5.4% and between 97% and 101.5% respectively, in child-sized phantoms.27 The TBP measurements of the children with CP were compared with TBP predicted for age and height using equations based on the comparison data.

Bone mineral content by DXA

All DXA measurements were performed and analysed by trained staff in the Department of Nuclear Medicine at the Children’s Hospital at Westmead with a commercial radiograph bone densitometer (DPX Lunar Radiation Corp, Madison, WI, USA). Whole-body scans were performed on the fast scan mode using adult total-body software, version 4.7 (Lunar Radiation Corp.), with the child lying supine. The total scan time was approximately 10 minutes, with a total radiation dose of around 0.2μSv. Most of the children with CP were lightly sedated with 0.35mg/kg oral midazolam 30 minutes before the scan to help reduce involuntary or uncontrollable movements during the test. Some of the children with CP were also carefully wrapped in a light cotton bed sheet to keep them in the supine position during testing. The precision of the technique in children as assessed at the Children’s Hospital at Westmead is 1.2% for bone mineral content (BMC).16 Measurements for BMC were converted to age and height standard deviation scores based on data derived from the comparison group.

Gastrostomy tube feeding regimens

The feeding regimens of the gastrostomy tube-fed children were determined by the Dysphagia Clinic dietitian. Some of the gastrostomy tube-fed children continued to eat orally, therefore their feeding regimens were designed to provide between 75% and 100% of their energy requirements through gastrostomy tube feeds using a range of commercially available micronutrient complete formulas. The energy requirements of the gastrostomy tube-fed children were initially estimated to be 1.2 times their measured resting energy expenditure, which was measured at the time of this study (data not presented).9 Weight gain was monitored at regular intervals through the Dysphagia Clinic, and the feeding regimens adjusted according to weight gain. The children were invited to undergo repeat body composition measures after significant weight gain (defined as a 25% increase in body weight).

Statistical analysis

All data were analysed using the Statistical Package for the Social Sciences (SPSS) program (version 15; SPSS Inc, Chicago, IL, USA). The significance level was set at p<0.05. The data were not normally distributed and are presented as medians with interquartile ranges (IQR). A Wilcoxon signed-rank test was used to compare differences between the paired baseline and repeat tests of the body composition parameters.

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

The baseline and repeat measures of study variables of the 21 children (nine females, 12 males) are shown in the tables. Baseline measures of height – height SDS (Table I), TBP percentage of that predicted for height (Table II), and BMC for height SDS (Table III) – were unavailable for seven of the children (four females, three males) because the measurement of knee height was not part of the study protocol at the time of their participation in the study. Paired skinfold measurements were unavailable in two patients. Therefore, the tables depict participants with both baseline and repeat measures for each parameter.

Table I.   Baseline and repeat measures of body composition after nutritional rehabilitation
 BaselineRepeatp value
  1. Data are median (interquartile range), n=21 (nine F, 12M). At baseline the pubertal status of all except two children (one F, one M) were Tanner stage 1, whereas, at repeat testing, 15 (five F, 10M) children were Tanner stage 1 and the remaining six (four F, two M) were Tanner stage 2 or more. an=14. bn=19. SDS, standard deviation score; ns, not significant; PBFskin, per cent of body fat by skinfold anthropometry. M, male; F, female.

Age, y:mo8:5 (6:9 to 11:10)10:4 (8:9 to 14:0)<0.05
Weight, kg15.4 (12.0 to 20.5)23.3 (16.6 to 25.9)<0.05
Weight SDS−4.8 (−6.7 to −3.1)−3.0 (−5.2 to −1.7)<0.05
Height, cma105.4 (99.3 to 121.5)118.3 (114.2 to 127.9)<0.05
Height SDSa−3.9 (−4.7 to −2.0)−3.5 (−4.2 to −1.7)ns
PBFskin,%b10.7 (6.4 to 12.9)16.3 (12.3 to 24.2)<0.05
Table II.   Effect of nutritional rehabilitation on total body protein (TBP)
 BaselineRepeatp value
  1. Data are median (interquartile range). an=17. bn=11. ns, not significant.

TBP, kga2.4 (1.6–3.2)3.4 (2.8–4.0)<0.05
TBP for age, %a51.9 (44.9–62.9)58.0 (45.0–75.0)ns
TBP for height, %b83.4 (72.3–96.0)99.0 (92.0–117.9)<0.05
Table III.   Effect of nutritional rehabilitation on bone mineral content (BMC)
 BaselineRepeatp value
  1. Data are median (interquartile range). an=14. ns, not significant; SDS, standard deviation score.

BMC, g 469 (374 to 632) 626 (509 to 736)<0.05
BMC for age SDS−2.3 (−3.3 to −1.7)−2.5 (−3.6 to −1.7)ns
BMC for height SDSa−0.6 (−1.0 to −0.1)−1.1 (−1.5 to −0.3)ns

The median time difference between baseline and repeat testing was 20.6 months (IQR 11.3–34.4mo). In 13 of the children, there was a median time delay of 5.3 months (IQR 3.9–12.8mo) between having the baseline measurement and insertion of a gastrostomy tube; therefore, the median length of time of actual gastrostomy tube feeding was 19.4 months (IQR 7.7–29.9mo).

Anthropometry

Baseline characteristics indicate significant degrees of stunting, weight deficit, and reduced body fat in this group of children with CP. However, after an extended period of gastrostomy feeds, there was a significant increase in all body composition parameters measured by anthropometry, including body fat. Mean body weight increased by approximately 50%. There was also a significant increase in weight SDS, although height standard deviation did not alter. These results indicate that gastrostomy feeding did improve nutritional status.

Total body protein by neutron activation analysis

Nutritional rehabilitation resulted in a significant increase in TBP between baseline and repeat testing (Table II). A significant increase was also demonstrated in TBP as a percentage of that predicted from comparison data for height between baseline and repeat testing. However, there was no significant change in TBP expressed as percentage of that predicted from comparison data for age.

Bone mineral content by DXA

Nutritional rehabilitation resulted in a significant increase in BMC between baseline and repeat testing; however, there was no significant change in BMC SDS for age or height.

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

This prospective nutritional rehabilitation study revealed that significant weight gain in gastrostomy tube-fed children with CP produced significant increases in height, measures of fat mass, TBP, and BMC. These children also demonstrated significant improvements in their SDS for weight and for TBP expressed as a percentage of that predicted for height from comparison data, indicating that nutritional rehabilitation of malnourished patients with CP does result in increased body protein accretion.

There have been several longitudinal nutritional rehabilitation studies of children with CP,1–7 although only three of these studies were prospective.2,6,7 Three studies were retrospective reviews of the medical records,1,3,5 and one did not state how the data were collected.4 All of the above studies reported measures of weight and/or height, with only three including some basic skinfold anthropometry.2,6,7 None of the studies included a direct measure of body protein or BMC. The use of simple anthropometry alone to assess nutritional status in children and adults with CP is problematic, given the difficulties of obtaining accurate height measurements as well as the presence of reduced skeletal muscle mass from central nervous system injury and disuse.

Interestingly, nutritional rehabilitation in our study did not result in an increase in height SDS. This has also been reported in several other retrospective studies in children with neurological disabilities.1,3–5 The lack of improvement in height SDS is most likely a result of multiple factors, including the lack of weight-bearing activity and poor growth secondary to central nervous system injury. Immobilization, or disuse, during growth can also lead to reduced growth in bone length and, subsequently, a lack of growth in height.29

In three prospective studies of nasogastric and gastrostomy tube feeding in children with severe CP, it was found that, in addition to significant weight gain, the children demonstrated significant increases in their triceps skinfold thickness and, in one study, upper arm muscle area.2,6,7 This finding of an increase in upper arm muscle area suggests an increase in body protein; however, because upper arm muscle area is an indirect measure of body protein and is derived from the measurement of the triceps skinfold thickness and mid-upper arm circumference, this may have simply reflected an increase in fat stores.

The current study is, to our knowledge, the only study to report longitudinal data on TBP in children with CP. However, our centre has measured longitudinal changes in TBP in healthy comparison children.30 Compared with the healthy children, TBP gain per year was lower in the gastrostomy tube-fed children with CP (481g/y vs 406g/y respectively), yet the latter showed significant improvements in their TBP as a per cent of that predicted for height, demonstrating a catch-up in body protein for their size. Although there was no statistically significant increase in TBP as a percentage of that predicted for age, our previously published cross-sectional study demonstrated a significant negative correlation between age and TBP as a percentage of that predicted for age (i.e. as children with CP aged, there was an increasing divergence in body protein between the children with CP and the comparison group).8,9 It may be hypothesized from the results of this study that this divergence in body protein for age may be halted with adequate nutrition. Thus, this study demonstrated that significant increases in body protein can be achieved by gastrostomy tube feeding in children with severe CP.

Aside from the current study, there has been only one other longitudinal study of bone mineralization in children with CP.31 Although the authors were not conducting a nutritional rehabilitation study, they did find that a better nutritional status, as assessed by the triceps standard deviation score, was associated with greater change in BMD per year in 69 children with moderate to severe spastic CP. The study also reported that BMD SDS decreased over time in spite of overall increases in BMD, whereas, in the current study, it was found that there was no significant change in BMC SDS for age or height with gastrostomy tube feeding. However, the results are difficult to compare with the current study because the authors measured rate of change in areal BMD in the distal femur and lumbar spine, rather than in total body BMC, as in this study, no adjustments were made for body size, and they conducted an observational rather than an interventional study.

Despite overall increases in BMC, no improvements in BMC SDS for age or height were observed in our study. This is probably because of the need for both physical activity and nutrition for normal growth and development of bone.29 Consequently, in children with CP, it is unlikely that any significant improvements in BMC SDS will be seen with adequate nutrition alone, but longer-term studies will be required to clarify this. It is important to note, however, that, although there were no significant improvements in BMC SDS, these scores did not significantly decrease over the course of nutritional rehabilitation. Previous studies of bone mineralization in children with severe CP have demonstrated a decline in BMC and BMD SDS with increasing age,9,13,31,32 a pattern that is similar to what we describe for TBP (see above). Furthermore, a significant positive relation between BMC and measures of body protein (lean tissue mass) has been shown to exist in typically developing children as well as in children with severe CP.9,16 This implies that low body protein may predispose children with CP to low bone mineralization.

There are a number of limitations to this study. A small number of patients were recruited with CP, although data from a large number of typically developing children were available for comparison. Interpretation of the results is also confounded by the dilemma of how to normalize the study parameters given the difficulties of attaining accurate measurements of height and bone density in children with CP and the contribution of central nervous system injury and lack of physical movement to the lack of development of lean mass. However, these are the realities of working in such a clinic. Further studies of longer duration will be required to determine if maintenance of adequate nutritional status in spastic quadriplegic CP will ameliorate or halt the decline in BMC standard deviation seen with increasing age. Ideally, these questions would be best answered in a randomized controlled trial. We attempted to perform such as study but failed to recruit any patients over a 12-month period because of lack of caregiver consent to randomization.

Conclusion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

This study demonstrates that nutritional rehabilitation in children with severe CP can result in significant improvements not only in body fat, but also in body protein. Although we did not observe improvements in BMC, further studies will be required to determine the contribution of malnutrition to the age-related decrease in BMC in patients with CP.

What this paper adds

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References
  • • 
    Nutritional rehabilitation in cerebral palsy increases total body protein.
  • • 
    Nutritional rehabilitation does not increase bone mineralization.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References

Dr Arrowsmith undertook this study as part of her requirements for a PhD at Sydney University. She recruited patients and undertook data collection, analysis, and write-up. Professor Gaskin and Drs Allen, Somerville, and O’Loughlin were involved in study design, supervision, and write-up; Dr Allen also assisted with the total body protein measurements. Ms Clarke was involved with initial patient studies and write-up of the manuscript. The study was supported by the National Health and Medical Research Council of Australia, the James Fairfax Institute of Paediatric Nutrition, and Nutricia Australia Pty Ltd.

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. What this paper adds
  8. Acknowledgements
  9. References