Impact of adipose and bone tissues in diffuse optical measurements of skeletal muscle: in vivo data and theoretical simulations

This study employs theoretical simulations to investigate the influence of adipose and bone tissues on frequency-domain near-infrared spectroscopy (FD-NIRS) measurements of forearm skeletal muscle (brachioradialis). Our analysis is based on diffusion theory models of two-layered and three-layered media to assess bone contributions to in vivo FD-NIRS data (AC intensity (<i>I</i>) and phase (<i>ϕ</i>)) collected on the human forearm at source-detector separations of 25 and 37 mm. Specifically, the in vivo FD-NIRS data were collected during a venous occlusion protocol to measure blood flow (BF) or an arterial occlusion protocol to measure oxygen consumption (OC). Our results reveal that while two-layer models are unable to qualitatively reproduce the in vivo data, three-layer models can reproduce them for specific selections of the model parameters. Such model parameters include a smaller BF and OC in adipose tissue than muscle tissue, and a greater baseline reduced scattering coefficient (𝜇'_𝑠 ) in adipose tissue than muscle tissue. The three-layer model analysis shows that bone tissue may contribute to FD-NIRS data collected at the source-detector distances used in this study when it is located at relatively shallow depths of ≤20 mm. These findings provide valuable insights into the role of tissue heterogeneity in non-invasive FD-NIRS measurements of skeletal muscle hemodynamics.