Discordant Classification of Pediatric Overweight and Obesity according to BMI Percentile and Percent Body Fat in Youth with Type 1 Diabetes

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SAT 677-696-Obesity Physiology & Epidemiology
Clinical
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-689
Lori M Laffel*1, Charu Baskaran1, Marcy L Hudson1, Tonja R Nansel2, Leah M Lipsky2, Lisa K Volkening1 and Sanjeev N Mehta1
1Joslin Diabetes Center, Boston, MA, 2Eunice Kennedy Shriver National Institute of Child Health & Human Development, Bethesda, MD
Youth with type 1 diabetes (T1D) are not immune from the current epidemic of childhood overweight and obesity. Based upon BMI determinations, ~1/3 of youth with T1D are overweight or obese, rates that match those reported for American youth in general.

To assess the classification of youth with T1D as overweight and obese using BMI %iles (≥85 to <95th %ile and ≥95th %ile, respectively), we compared against adiposity measured by Dual-energy X-ray Absorptiometry (DXA, Hologic Delphi™ A) in 126 youth, ages 8-17, with T1D. Youth (49% male, 89% white) were 12.9±2.5 (mean±SD) years old and had T1D for 5.9±3.2 years; mean A1c was 8.1±1.1% and 72% received insulin pump therapy. CDC standards from 2000 provided BMI %iles based on youth height and weight. Body composition, reported as percent body fat (%BF), was measured by DXA using standardized protocols. We calculated age- and sex-based cut-points for overweight and obesity by %BF derived from Taylor et al, 2002. Statistical analyses included Pearson correlations and chi-square.

Mean BMI %ile was 71±23 (range 5-99.5); mean %BF was 27.7±8.0% (range 11.5-47.5%). BMI %ile and %BF were moderately correlated (r=0.52, p<.0001), with strongest correlations in pre-pubertal youth (n=40, r=0.72, p<.0001) and weakest in post-pubertal youth (n=32, r=0.30, p=.09). According to BMI %iles, 67% of T1D youth were normal weight, 20% were overweight, and 13% were obese. According to %BF, 48% of T1D youth were normal weight, 43% were overweight, and 9% were obese. The weight status of 38% of T1D youth was differentially classified by BMI %ile Vs DXA measurements of %BF, with 2/3 of discordant classifications resulting from an insensitivity of BMI %ile to detect overweight/obesity; BMI %ile classified fewer youth as overweight/obese Vs %BF. While rates of discordant classification were similar by sex and across pubertal stages, differential classification differed by pubertal status (p<.02). In pre-pubertal youth, discordant classification was mainly due to insensitivity of BMI %ile to detect overweight/obesity (89%); in post-pubertal youth, discordant classification was mainly due to overestimation of adiposity by BMI %ile compared with %BF (58%).

Overall, 49% of overweight/obese T1D youth determined by DXA were classified as normal weight by BMI %ile, potentially resulting in missed opportunities to intervene with diet and exercise to manage weight and reduce cardiovascular risk in the already vulnerable population of T1D youth.

Taylor RW et al., Am J Clin Nutr 2002;76:1416–21

Disclosure: LML: Advisory Group Member, Bristol-Myers Squibb, Advisory Group Member, Sanofi, Consultant, Johnson &Johnson, Consultant, LifeScan/Animas, Consultant, Eli Lilly & Company, Consultant, Menarini. Nothing to Disclose: CB, MLH, TRN, LML, LKV, SNM

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm

Sources of Research Support: NICHD contract HHSN267200703434C; NIH Training Grant 5 T32 DK007260-36