The near-infrared diffuse correlation spectroscopy muscle perfusion index in human calf muscle: accounting for...

During physiological provocations, relative changes in muscle blood flow and O₂ extraction can be estimated using near-infrared diffuse correlation spectroscopy (NIRS-DCS). However, the NIRS-DCS blood flow index (BFI) may underestimate perfusion changes unless changes in hemoconcentration are accounted for. Accordingly, we compared NIRS-DCS calculations of muscle perfusion and O₂ extraction against MRI-derived arterial spin labeling (ASL) and MR susceptometry-based venous oximetry (SvO₂), respectively. Thirteen healthy young adults (5 F) completed an ischemic-reperfusion test and plantarflexion exercise inside the bore of a 3 T MRI scanner. NIRS-DCS BFI was either 1) unadjusted (BFI), or converted into a muscle perfusion index (MPI) by 2) scaling BFI linearly for changes in hemoconcentration (MPI_LIN), or 3) scaling BFI for changes in hemoconcentration squared (MPI_SQR). Compared with ASL MRI, peak DCS perfusion was significantly lower during postischemic reperfusion when using BFI but not when using MPI_LIN or MPI_SQR. Similarly, end-exercise perfusion from BFI was significantly lower than ASL following plantarflexion exercise, whereas no difference was observed when comparing MPI_LIN and MPI_SQR to ASL. NIRS estimations of SvO₂ were lower than MRI after 8 mins of ischemia, but similar following moderate-intensity plantarflexion exercise. However, following high-intensity exercise, NIRS estimations of SvO₂ were similar to MRI only after accounting for changes in hemoconcentration. Collectively, these results highlight the importance of accounting for changes in hemoconcentration when using NIRS-DCS to estimate changes in muscle perfusion during physiological provocations such as ischemic reperfusion and exercise.NEW & NOTEWORTHY We compared changes in skeletal muscle blood flow estimated by NIRS-DCS against arterial spin-labeling MRI during ischemic reperfusion and exercise. NIRS-DCS blood flow index underestimated relative changes in muscle perfusion across all procedures unless the calculations were adjusted for changes in hemoconcentration and total heme content, which resulted in a muscle perfusion index (MPI) that was almost superimposed on ASL-MRI. These results demonstrate that NIRS-DCS MPI noninvasively quantifies changes in muscle perfusion in humans.