CYTOCHROME OXIDASE (CYTOX) IS A MITOCHONDRIAL ENZYME THAT IS RESPONSIBLE FOR THE FINAL METABOLISM OF MOLECULAR OXYGEN TO PRODUCE ADENOSINE TRIPHOSPHATE (ATP) IN THE RESPIRATORY CHAIN. Because its absorption spectrum changes between the oxidized and reduced state, the concentration of the oxidized form can be monitored using near-infrared spectroscopy (NIRS).1 This provides a unique method of imaging the effects of physiologic insults on the intracellular redox environment, both noninvasively and with good time resolution. During obstructive sleep apnea (OSA), the body may suffer repetitive hypoxic and hemodynamic insults many times a minute. Desaturations to 60% to 70% SaO2 are not uncommon, and significant swings in cerebral blood flow velocity (CBFV) also occur.2,3 There is evidence from previous studies that cerebral oxygenation is not stable during OSA,4,5 but the effects of changes in cerebral oxygenation on intracellular metabolism merit further study because changes in the intracellular redox state have been postulated as a cause for neuropsychologic dysfunction in chronic hypoxia.6 The use of NIRS allows noninvasive monitoring of cerebral oxygenation, cerebral hemodynamics, and metabolism. The method was first described by Jobsis1 and has been further developed for clinical use in neonates7 and adults.8,9 The NIRS measures concentration changes in the chromophores: oxygenated hemoglobin (OHb), deoxygenated hemoglobin (HHb), and the oxidation state of CytOx. Absolute quantification of regional tissue saturation in the form of the tissue oxygenation index (TOI), using spatially resolved spectroscopy, has been developed more recently as a feature of a new model, the NIRO-300 (Hamamatsu, Japan).10,11 This measurement has been validated against a blood-gas analyzer in vitro, against timeresolved spectroscopy in the human arm,11 and as a clinical cerebral-saturation measurement in patients undergoing neurosurgical procedures.12 The CytOx measurement from NIRS has been validated in animal models13 and has been shown to correlate with phosphocreatine and nucleoside triphosphate measurements using magnetic resonance spectroscopy (MRS) in animals.14 In previous work, we compared changes in the TOI with changes in SaO2 during 1036 apneas and hypopneas in 13 subjects with OSA, during spontaneous daytime sleep.4 We confirmed that decreases in cerebral oxygenation occur during respiratory events and that the magnitude of the TOI dip depended on the SaO2 dip, apnea duration, and rapid eye movement sleep stage. These results were obtained by recording the TOI output into a free channel on our polysomnography system. At the same time, we also recorded sections of the NIRO-300 digital output, and here we report the results of analysis of CytOx redox changes related to changes in cerebral oxygenation measured as TOI. We have designated this recording period study session 1. Because experience with the clinical use of TOI as a measure of cerebral oxygenation is limited, we also examined the relationship between factors known to influence cerebral oxygenation and CytOx redox changes. In a second daytime study session (study session 2), we measured changes in CBFV, SaO2, and CytOx oxidation state. The overall aim was to quantify changes in CytOx during apnea in patients with OSA and to identify any association with cerebral oxygenation changes.