BACKGROUND & AIMS: Accurately measuring nutritional status in children with severe cerebral palsy (CP) is a challenge. This review seeks to assess the validity of skinfold measurements and bioelectrical impedance analysis (BIA) for measuring body composition in children with severe CP. METHODS: We executed a literature search on the validation of both methods in children with severe CP. To be eligible for inclusion, a study had to report on a statistical comparison between these two methods and any method of reference. The QUADAS tool was used for quality assessment. RESULTS: The search strategy resulted in 1549 studies of which 5 studies eventually met the inclusion criteria. When comparing body composition outcomes of skinfold measurements to a reference method, correlation coefficients were found ranging from 0.406 to 0.988. Correlation coefficients between body composition data of BIA and a reference method ranged from 0.515 to 0.95. CONCLUSIONS: Although a number of authors found favorable agreement between skinfold measurements and BIA in comparison with reference methods, the small numbers studied, the lack of methodological quality measured by QUADAS, and the use of inappropriate analytical methods hamper solid conclusions. Copyright 2009 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
BACKGROUND & AIMS: Accurately measuring nutritional status in children with severe cerebral palsy (CP) is a challenge. This review seeks to assess the validity of skinfold measurements and bioelectrical impedance analysis (BIA) for measuring body composition in children with severe CP. METHODS: We executed a literature search on the validation of both methods in children with severe CP. To be eligible for inclusion, a study had to report on a statistical comparison between these two methods and any method of reference. The QUADAS tool was used for quality assessment. RESULTS: The search strategy resulted in 1549 studies of which 5 studies eventually met the inclusion criteria. When comparing body composition outcomes of skinfold measurements to a reference method, correlation coefficients were found ranging from 0.406 to 0.988. Correlation coefficients between body composition data of BIA and a reference method ranged from 0.515 to 0.95. CONCLUSIONS: Although a number of authors found favorable agreement between skinfold measurements and BIA in comparison with reference methods, the small numbers studied, the lack of methodological quality measured by QUADAS, and the use of inappropriate analytical methods hamper solid conclusions. Copyright 2009 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.