A novel method for measuring subcutaneous adipose tissue using ultrasound in children - interobserver consistency

Background: Currently, sufficiently accurate field methods for body composition assessment in children are missing. The ultrasound method for assessing adipose tissue thickness has been used extensively in sport medicine. However, there are no studies looking at the reliability of this method in non-athletic children. This paper aims to determine the inter-observer reliability in measuring the uncompressed subcutaneous adipose tissue thickness using ultrasound, in children. Subjects, Materials and Methods: Forty healthy children (20 males, 20 females), median age 11.85 years (5.3 to 18.1 years) were evaluated. Median body mass index standard deviation score (BMI SDS) was -0.13 (-3.9 to 4). Three observers used a Hosand BX 2000 Ultrasonic Adipometer to measure uncompressed subcutaneous adipose tissue thickness at three sites: triceps, subscapular, supraspinale. A single experienced observer used the three sites to also measure the compressed adipose thickness using a skinfold caliper. Results: Individual observer deviations from the mean value of the three observers in adipometer measurement had a standard deviation of 1.74 mm, 92.8% were less than 3 mm. Analysis separated by individual anatomical sites showed high reliability values for triceps: linear regression R-2=0.84, p=0.000; intraclass correlation coefficient (ICC)=0.92 and standard error of measurement (SEM)=0.63. The values at supraspinale site were R-2=0.82, p=0.000, ICC=0.89 and SEM=1.17, while for subscapular the values were lower: R-2=0.79, p=0.000, ICC=0.78 and SEM=1.02. The body fat percentage (BF%) calculated using skinfold measurements was highly correlated with the BF% calculated by the adipometer (R=0.92, R-2=0.83, p=0.000). The Pearson's correlation between BMI SDS and BF% calculated from skinfold was R=0.52, R-2=0.28, p=0.001, while for the adipometer it was R=0.53, R-2=0.27, p=0.000. Conclusions: This novel ultrasound measurement technique can be used with good accuracy and reliability to measure uncompressed subcutaneous adipose tissue thickness in children, sustaining its application for research and clinical purposes, however larger studies are needed.