P.59 Associations between body composition estimates and motor function in ambulatory individuals with Duchenne muscular dystrophy

Abstract

The heterogeneity of functional decline observed in individuals with Duchenne muscular dystrophy (DMD) presents a long-standing challenge to optimizing clinical care and implementing clinical trials. Ongoing efforts to identify biomarkers and predictors of disease progression may help improve accuracy of prognosis to ameliorate these challenges. The relationship between body composition and motor performance in individuals with DMD has been explored using several estimation methods. Although Body Mass Index (BMI) is a feasible estimate of body composition, it is confounded by the short stature of corticosteroid treated patients. Estimation methods utilizing imaging, techniques, such as dual energy x-ray absorptiometry (DXA) and magnetic resonance imaging (MRI), have shown promise as biomarkers of disease progression. One DXA-based estimation method is appendicular lean mass index (ALMI), the total lean mass in arms and legs divided by height in meters squared, is significantly decreased in children with DMD compared to healthy controls. The relationship between ALMI and motor performance across age has not been well studied. This study will compare the association of body composition estimates (BMI / ALMI) to change in motor outcomes, including the North Star Ambulatory Assessment and Timed Rise from Supine, across childhood using a longitudinal natural history dataset. The sample includes 100+ individuals and 275 unique observations in children with DMD aged 3 -14 years. Preliminary analyses with 97 individuals and 263 observations found ALMI values in the top quartile was associated with a 1 second faster Timed Rise from Supine compared to ALMI values in the bottom quartile with age and corticosteroid treatment duration controlled, p < .001 (-.67, -1.33). Established correlates of motor function, including corticosteroid dose and ankle range of motion, will be included as covariates.