Assessment of the validity of bioelectrical impedance for measuring body fat in children with cerebral palsy

Assessment of the validity of bioelectrical impedance for measuring body fat in children with cerebral palsy
DJ OEFFINGER PHD1, MJ GURKA PHD2, RD STEVENSON MD2, CM TYLKOWSKI MD1
1 Shriners Hospital for Children, Lexington, Lexington, KY;
2 School of Medicine, University of Virgina, Charlottesville, VA, USA


Background/Objectives: Accurate assessment of body composition is important in the care of children with cerebral palsy (CP). Bioelectric Impedance Analyzers (BIA) are relatively inexpensive and provide a quick and easy way to measure body fat in the clinical setting. The validity of BIA measurements has not been established for ambulatory children with CP. The study objective was to validate BIA measurements of body fat in ambulatory children with CP by comparing body fat percentage (%BF) measured using a Body Stat Quadscan 4000 BIA device to % BF measured from a Lunar dual energy x-ray absorptiometry (DXA) machine. DXA is the current gold standard for the clinical measurement of % BF.
Design: Cross-sectional validation study.
Participants and Setting: Fifty subjects (33 male) with CP (GMFCS Level I (28), II (16), III (6)); ages 10-18 years (mean 12.6); diplegia (36) and hemiplegia (14) participated from a single pediatric orthopedic hospital.
Materials/Methods: Total body % BF was computed from the BIA (left side and right side) and compared to %BF from DXA. Concordance Correlation Coefficients (CCC) were used to determine agreement between measures. Bland-Altman Analyses were used to investigate the differences. ANOVA and t-tests were used for comparisons between and among groups.
Results: Mean (±SD) %BF for the study population using DXA was 29.0% (±13.2), compared to 26.6% (±9.0) for the right side BIA measure and 26.0% (±9.0) for the left side BIA. Agreement between repeated BIA trials was excellent (CCC>0.99). No differences were found between %BF measured when tested on the left or right side using BIA. Moderate correlations were found between BIA (left side) and DXA (CCC=0.78 (CI: 0.67, 0.89)) and BIA (right side) and DXA (CCC=0.81 (CI: 0.72, 0.91)) (Figure 1A). BIA underestimated %BF (mean difference with respect to DXA %BF=3.1; 95% CI= (-5.2, -0.9); P<0.05). GMFCS level nor gender influenced the performance of BIA compared to DXA. The Bland-Altman graphs demonstrated progressively worse performance by the BIA in predicting %BF for larger values of %BF (Figure 1B).
Conclusions/Significance: BIA does a moderate job estimating body fat in ambulatory children with CP. BIA does better measuring BF for children less than 30% BF. BIA can be used easily in a clinical setting and is less costly than DXA assessments. Given the observed underperformance of BIA for individuals with larger amounts of fat, a correction factor to the BIA equations may improve the ability to estimate %BF from BIA in larger children with CP.

 
Gallery
Gal 1 Gal 2
Supporting Organizations
 
   
   
   
 
Online Feature Partner
 
   
   
 
Locations of visitors to this page
 
 

© 2010 all rights reserved. maintained by Classic Infomedia