Near-infrared spectroscopy (NIRS) is a noninvasive monitoring technique that measures regional cerebral oxygen saturation (rSO2).Objectives:
The primary aim was to compare the output of 2 NIRS-based cerebral oximetry devices, FORESIGHT (CAS Medical Systems Inc, Branford, Connecticut) and INVOS (Covidien, Boulder, Colorado), to venous oxygen saturations from the jugular venous bulb at cannulation and decannulation of the superior vena cava (SVC). Secondary objectives included evaluating correlations of cerebral saturation, as measured by the NIRS devices, with mean arterial blood pressure (MAP), measured by an invasive arterial line, and end-tidal CO2 (ETCO2).Methods:
Near-infrared spectroscopy, MAP, and ETCO2 data were collected at 13 defined events during each case when hemodynamic instability was expected. At SVC cannulation and decannulation, a 0.1 mL sample of blood was collected from the jugular bulb by the surgeon using a long angiocatheter. The oxygen saturation of these blood samples was measured using an AVOX device and compared with contemporaneous readings from the NIRS probes. Mixed-effects linear regression was used to correlate MAP or ETCO2 with cerebral oxygen saturation (by NIRS) at each time point.Results:
Children undergoing cardiopulmonary bypass for congenital heart surgery (n = 34) were enrolled in the study. At SVC cannulation, both INVOS (r = .78) and FORESIGHT (r = .59) were correlated with AVOX data at P < .001, although the correlation with INVOS was significantly stronger (P = .003). At SVC decannulation, INVOS (r = .68; P < .001) and FORESIGHT (r = .60; P < .001) were similarly correlated with jugular venous rSO2. Correlations of rSO2 (by NIRS) with MAP and ETCO2 levels were stronger than correlations between rSO2 change and change in MAP or ETCO2.Conclusion:
INVOS correlated more strongly than FORESIGHT with the jugular bulb rSO2 at SVC cannulation but may have underestimated oxygen saturation at low rSO2 values. Data from both NIRS devices were correlated with MAP and ETCO2 over the case duration.