Work

Development of a Novel Method for the Diagnosis of Sarcopenia Using Thigh DXA Imaging

Year: 2020
Type: dissertation
Abstract: Background: Sarcopenia, defined as the loss of muscle mass and function, increases the risk of falls and thus fractures and has unclear/ indistinct thresholds. Historically, sarcopenia is characterized in terms of lean mass index (LMI, lean mass/ ht2) or appendicular lean mass index (ALMI, appendicular lean mass/ ht2). However, recent studies have shown that function declines more rapidly than lean mass with age, leading to a transition in the defining criteria for sarcopenia from mass towards function. Purpose: This research investigates the thigh region of interest within a whole-body DXA scan to calculate muscle volume, area, and density, and evaluates whether these metrics perform better than ALMI and LMI in relation to function. Method: DXA scans were acquired during implementation of the Sarcopenia in Women Across the Lifespan (SarWol) study, a cross-sectional study of 30 healthy premenopausal (age 35.82 ± 7.67), 30 healthy postmenopausal (age 67.04 ± 7.47), and 26 postmenopausal women with recurrent falls (age 71.26 ± 8.82). A Hologic Discovery system was used by ISCD-trained technologists to acquire the scans. Two subregions, a mid-thigh region and a half-thigh region, normalized to the participant’s femur length were placed precisely on each whole body scan. As opposed to the commonly used cylindrical model, two truncated cone models were applied to each region, namely, a cone model and an elliptical model. Sets of parameters were extracted to distinguish the subcutaneous fat from the standard outputs of the DXA software. Then, equations were developed to obtain muscle area, density, and volume in each subregion based on each geometrical model and tissue properties. Knee-extension torque using a Biodex dynamometer was obtained for each participant in a standardized fashion. Correlation coefficients and multivariate linear regression models were used to evaluate DXA-derived parameters (ALMI, LMI and our new thigh muscle parameters) in relation to knee-extension torque and grip strength. Results: The DXA-derived muscle volume was moderately correlated to knee-extension torque while ALMI (r = 0.16, p = .130) and LMI (r = -0.01, p = 0.961) were not significantly correlated. r Mid-thigh Half-thigh Knee-Extension Torque Elliptical model 0.42 0.42 Cone Model 0.37 0.40 p
Language: English
Field: Medicine
Open Access status: green
OpenAlex ID: W3164886592
Cites: 24
Cited by:
Related to: 9