Associations among preoperative transthoracic echocardiography variables and cerebral near-infrared spectroscopy values at baseline before anesthesia in patients undergoing cardiac surgery: a retrospective observational study.

Cerebral tissue oximetry with near-infrared spectroscopy (NIRS) is used to monitor cerebral oxygenation during cardiac surgery. To date, reduced baseline cerebral NIRS values have been attributed to reduced cerebral blood flow primarily based on a significant positive correlation between left ventricular ejection fraction (LVEF) and baseline rSO 2 measured with the INVOS 5100C oximeter. Reportedly, however, rSO 2 , but not StO 2 measured with the FORESIGHT Elite oximeter, correlated with LVEF. We, thus, investigated associations among baseline NIRS values measured with three different oximeters before anesthesia for cardiac surgery and preoperative transthoracic echocardiography (TTE) variables, including LVEF, to examine whether there are inter-device differences in associations among baseline NIRS values and TTE variables. Using Spearman's correlation coefficient, we retrospectively investigated associations among 15 preoperative TTE variables, including LVEF, and baseline NIRS values, including rSO 2 , StO 2 , and TOI with the NIRO-200NX oximeter in 1346, 515, and 301 patients, respectively. Only rSO 2 (p < 0.00001), but not TOI or StO 2 (p > 0.05), positively correlated with LVEF. On the other hand, baseline rSO 2 , TOI, and StO 2 consistently, negatively correlated with the left atrial diameter index (LADI), early diastolic transmitral flow velocity (E), E-to-early diastolic mitral annular velocity ratio (E/e'), estimated right ventricular systolic pressure (eRVP), and inferior vena cava diameter index (IVCDI) (p < 0.0005 to p < 0.00001). Because all of these five TTE variables could be positively associated with right as well as left ventricular filling pressure, our results indicated that reduced baseline NIRS values were consistently associated not with reduced LVEF but with TTE findings indicative of elevated biventricular filling pressure. Our data suggest that regional venous congestion greatly contributes to reduced baseline NIRS values in patients undergoing cardiac surgery.