Near-Infrared Spectroscopy

Cerebral oximetry uses transcranial near-infrared spectroscopy (NIRS) to evaluate changes in cerebral oxygenation noninvasively and continuously. Its operation relies on two basic principles. First, near-infrared light has the capacity to penetrate human tissue, including bone. Second, in these tissues, hemoglobin is the predominant absorbing substance (i.e., chromophore) in the near-infrared range [1]. The binding of oxygen to hemoglobin alters its infrared absorption spectrum. As a result, the concentrations of oxy- and deoxyhemoglobin may be determined by measurement of light absorption at two or more wavelengths. Determination of the absolute concentrations is based on the familiar Lambert–Beer equation and requires the knowledge of the optical path length within the tissue sample volume. Claims regarding the accuracy of time-of-flight and phase-shift technology estimates of path length remain controversial due to the confounding variations in the thickness of skull and cerebrospinal fluid layer, which influence the partial path length of light in the brain, as do blood volume and tissue water content. Despite the continuing uncertainty associated with absolute measurements, a recent study found close agreement between brain oxygen level dependent functional magnetic resonance imaging (BOLD-fMRI) and NIRS estimates of brain stimulation-induced change in deoxyhemoglobin dynamics [2]. In the absence of independent direct measures of path length and cerebral chromophore concentration as well as the influence of skull thickness on photon migration, the putative “absolute” concentrations of oxy- and deoxyhemoglobin produced by some cerebral oximeters should be viewed as unvalidated estimates [3].