The influence of carbon dioxide and body position on near-infrared spectroscopic assessment of cerebral hemoglobin oxygen saturation.

Near-infrared spectroscopy may allow continuous and noninvasive monitoring of regional brain hemoglobin oxygen saturation by measuring the differential absorption of infrared light by oxyhemoglobin and deoxyhemoglobin. We have previously examined the correlation between the spectroscopic signal generated by a prototype cerebral oximeter (Invos 3100; Somanetics, Troy, MI), and global brain hemoglobin oxygen saturation calculated from arterial and jugular venous bulb oxygen saturations. Because the technology does not distinguish between arterial and venous hemoglobin saturation, changes in the proportion of cerebral arterial and venous blood volume, which may result from changes in blood flow or venous distending pressure, may confound measurements. In eight conscious volunteers breathing hypoxic oxygen mixtures, we examined the influence of supine, 20 degrees Trendelenburg, and 20 degrees reverse Trendelenburg positions on the correlation of the spectroscopic measurement of cerebral oxygen saturation in the field assessed by the probe (CSfO2) and the calculated brain hemoglobin oxygen saturation (CScombO2), estimated as 0.25 x arterial saturation plus 0.75 x jugular venous bulb oxygen saturation. We found that changes in position did not influence the association between CSfO2 and CScombO2 (r2 = 0.69-0.885) during hypoxic challenge. In a second set of eight volunteers, we studied the influence of hypercapnia and hypocapnia and body position on the association between CSfO2 and CScombO2, and found that they were less well correlated (r2 = 0.366-0.976) in individual patients. Because changes in body position and Paco2 confound the relationship between CSfO2 and CScombO2, changes in CSfO2 can best be assessed if position and Paco2 are constant.