Assessment of skinfold thickness equations in estimating body composition in children with inflammatory bowel disease.

Abstract

To assess the agreement of commonly used skinfold thickness equations to estimate percentage of Fat Mass (FM%) in paediatric patients with IBD, in comparison with dual energy X-ray absorptiometry (DEXA). Twenty-one paediatric IBD patients were included: 11 females and 10 males; mean age for the entire group: 14.3 years, range 12-16 years, 16 with Crohn's disease, five with ulcerative colitis. The FM% was calculated using six established paediatric skinfold thickness equations and compared to the results obtained by DEXA. The statistical analysis was performed using Spearman's correlation, Lin's concordance correlation coefficient and corresponding 95% confidence interval, Bland-Altman's limits of agreement method and the Bradley-Blackwood test. Correlation between skinfold and DEXA values ranged between 0.85 (Deurenberg) and 0.92 (Durnin and Rahaman and Johnston), all P < 0.001. Lin's concordance correlation coefficients and (95% confidence interval) ranged between 0.702 (0.512-0.891) for the Deurenberg equation and 0.876 (0.779-0.972) for the Brook equation. Average differences between skinfold and DEXA values ranged between -3.6% (Deurenberg) and 2.5% (Weststrate). Bland-Altman limits of agreement were wide, spanning over 10%. Finally, the Bradley-Blackwood test of equality of means and variances was significant in all but the Durnin and Rahaman equation. In adolescents with IBD, fat mass calculated from six skinfold thickness equations showed good correlation but poor agreement with reference values from DEXA. Assessment of body composition using skinfold thickness equations cannot be recommended in paediatric patients with IBD.