Dual X-Ray Absorptiometry as a Measure of Leg Length in Children and Adolescents

Abstract

INTRODUCTION: Leg length has been found to be a reliable and useful means for determining adolescent risk factors of metabolic disease (MetD) and type 2 diabetes (T2D). Dual X-ray absorptiometry (DXA) is often used to measure body composition in both adults and children and might provide another, accurate method to determine leg length in children than current methods utilized. Therefore, the purpose of the present study was to determine the relationship between DXA-derived leg length (DXA-LL) and traditionally-derived leg length. Also, to determine the relationship between DXA-LL and measures of cardiometabolic health, heart rate variability (HRV), arterial stiffness, and bone mineral density (BMD) when interacting with subject group status.METHODS: A total of 84 children and adolescents (n = 44 females, 40 males; age: 12.7±2.8 yrs.) had their leg length determined using the standard method of measuring seated height and then subtracting this from standing height. Subjects were scanned on a GE Healthcare Lunar iDXA and analyzed using enCore software version 16.2. The DXA-LL was determined by creating a custom region of interest (ROI) line drawn vertically from the greater trochanter of the femur to the calcaneus. In all participants supine blood pressure and pulse wave analysis (PWA) were determined using an automated blood pressure cuff, pulse wave velocity (PWV) was determined using an automated blood pressure cuff alongside an application tonometer, HRV measures were recording using an electrocardiogram (ECG), and fasting blood samples were collected for the determination of blood biomarkers of MD and T2D risk. The homeostatic model of insulin resistance (HOMA-IR) was calculated from measurements of glucose and insulin. Paired T-test assessed similarity between DXA-LL and traditionally measured leg length. Pearson’s correlation coefficients fit to determine correlation between variables of interest. Linear regression analyses were performed to determine associations between the DXA-LL with blood biomarkers, HRV metrics, and BMD measures. RESULTS: DXA-LL was not significantly different than traditionally-derived leg length (86.05±5.31 vs 87.21±10.24 cm, p=0.26). There was a significant correlation between DXA-LL with group status (r2 = -0.26, p<0.05), total cholesterol (r2 = -0.25, p<0.01), high-density lipoprotein (HDL) (r2 = -0.25, p<0.05), and pulse pressure (PP) (r2 = 0.36, p<0.01), as well as significant associations between cholesterol (Estimate = -1.50, 95% confidence interval= [-2.60, -0.39], p<0.01) and HDL (-0.85 [-1.43, -0.26], p<0.01) with DXA-LL, but not when interacting with group status. Glucose was significantly associated (-1.03 [-1.64, -0.41], p<0.01) with DXA-LL when interacting with group status. PP was found to be significantly associated with DXA-LL alone (0.55 [0.02, 0.45], p < 0.01) as well as when interacting with group status (1.01 [0.36, 1.67], p=0.01). There were significant correlations (r2=0.27-0.47, p<0.05) between total body bone mineral density (BMD), regional BMD, and individual leg BMD and DXA-LL. There were significant associations between all BMD measures (Total BMD: 0.01 [0.001, 0.011], Regional BMD: 0.01 [0.006, 0.018], right leg BMD: 0.01 {0.005, 0.016], left leg BMD: 0.01 {0.008, 0.019], p<0.01 for all) and DXA-LL alone, but not when interacting with group status. CONCLUSION: DXA-LL was not significantly different than traditionally-derived leg length. Therefore, DXA-LL can be used as an accurate and reliable way to measure leg length. In addition, DXA-LL was correlated with total cholesterol, HDL, PP and all measures of BMD. However, no significant association was found when interacting with group status and adjusting for covariates, suggesting that while there may be a connection between DXA-LL and these variables of interest, this sub-analysis did not reveal the predictive nature of this relationship.