Near-infrared spectroscopy and indocyanine green derived blood flow index for noninvasive measurement of muscle perfusion during exercise

Abstract

Near-infrared spectroscopy (NIRS) with the tracer indocyanine green (ICG) may be used for measuring muscle blood flow (MBF) during exercise, if arterial ICG concentration is measured simultaneously. Although pulse dye densitometry allows for noninvasive measurement of arterial dye concentration, this technique is sensitive to motion and may not be applicable during exercise. The aim of this study was to evaluate a noninvasive blood flow index (BFI), which is derived solely from the muscle ICG concentration curve. In 10 male cyclists 5 mg ICG were injected into an antecubital vein at rest and during cycling at 30, 60, 70, 80, 90, and 100% of previously determined maximal work load. Simultaneously blood was withdrawn through a photodensitometer at 20 ml/min from the radial artery to measure arterial ICG concentration. To measure muscle tissue ICG concentrations, two sets of NIRS optodes were positioned on the skin, one over the left seventh intercostal space and the other over the left vastus lateralis muscle. MBF was calculated from the arterial and muscle concentration data according to Fick's principle. BFI was calculated solely from the muscle concentration curve as ICG concentration difference divided by rise time between 10 and 90% of peak. During exercise mean BFI values changed similarly to MBF in both intercostal and quadriceps muscles and showed excellent correlations with MBF: r = 0.98 and 0.96, respectively. Individual data showed some scattering among BFI and MBF values but still reasonable correlations of BFI with MBF: r = 0.73 and 0.72 for intercostal and quadriceps muscles, respectively. Interobserver variability, as analyzed by Bland-Altman plots, was considerably less for BFI than MBF. These data suggest that BFI can be used for measuring changes in muscle perfusion from rest to maximal exercise. Although absolute blood flow cannot be determined, BFI has the advantages of being essentially noninvasive and having low interobserver variability.