A Comprehensive Review of the Correlations of Measurement Parameters among Modern Technologies for Sarcopenia Assessment.

Abstract

Sarcopenia is characterized by a degeneration of muscle mass and strength, which impairs mobility and causes fragility in older adults. Modern assessment technologies include magnetic resonance imaging (MRI), computed tomography (CT), bioelectrical impedance analysis (BIA), dual-energy X-ray absorptiometry (DXA), and ultrasound (US). Despite the tremendous efforts made in the past to investigate the correlations among measurement parameters of these technologies, inconsistencies in the reported correlations persist, and there is no existing review article considering all these technologies for sarcopenia assessment, resulting in a lack of a common understanding about the correlations of these techniques. Herein, we conduct a comprehensive review to scrutinize the reported correlations between each pair of these five technologies from 51 publications in the literature. We classified these five technologies into two categories: image-based methods (IBMs), including CT, MRI, and US, and non-image-based methods (NIBMs), including DXA and BIA, resulting in ten pairs of correlations analyzed. The averaged correlations for NIBM-NIBM (BIA-DXA: 0.909) and IBM-IBM (MRI-CT: 0.978; MRI-US: 0.861; CT-US: 0.875; overall: 0.905) were high, while those for NIBM-IBM exhibited lower correlations with wider variations (DXA-CT: 0.834; BIA-CT: 0.824; BIA-MRI: 0.715; DXA-MRI: 0.834; DXA-US: 0.497; and BIA-US: 0.463; overall: 0.695). Generally, the correlations within the same category were greater than those between different categories. The dissimilar measurement locations and components could apparently cause lower correlations (e.g., BIA-US and DXA-US); the lower averaged correlations do not necessarily represent their low clinical value for sarcopenia assessments. By contrast, integrating NIBM and IBM may leverage the strengths of each technology, complementing their metrics, and bring a more holistic connection to sarcopenia. We hope to facilitate an enhanced understanding of the reported correlations from the literature, offering the community insightful recommendations for selecting technologies toward further research.