Introduction: Balance between oxygen delivery and demand is typically assessed based on venous oxyhemoglobin saturation (SvO2), which requires repeated blood sampling from a central location. Alternative, non-invasive methods such as near-infrared spectroscopy (NIRS) demonstrate poor sensitivity to tissue hypoxia. Methods: We explored the possibility of measuring regional tissue oxyhemoglobin saturation in the esophagus (EsSO2) using a new, linear probe using resonance Raman spectroscopy (RRS, 420 nm excitatory light, A) to quantify the fraction of oxyhemoglobin. The RRS probe was passed from the mouth into the mid-esophagus in Yorkshire swine (n=7, weight 11.2-13.4 kg). Cerebral (cNIRS) and splanchnic NIRS (sNIRS, Somanetics) were continuously recorded. Swine underwent complete asphyxia by endotracheal tube occlusion for 12 minutes, during which time they were resuscitated using CPR and drugs. EsSO2, cNIRS, sNIRS were compared with SO2 based on arterial blood gas co-oximetry (SaO2). For the purpose of analysis, SaO2 was considered functionally equivalent to SvO2 due to the obligate intrapulmonary (veno-arterial) shunting in this physiology. Results: During asphyxia, SO2 rapidly declined to nearly 0%. This was reflected in EsSO2, but cNIRS and sNIRS decreased to a lesser extent (B). EsSO2 exhibited a strong correlation with SO2 (r 2 =0.63, p<0.001) compared to sNIRS (r 2 =0.53, p<0.001) and cNIRS (r 2 =0.06, p=0.106, C). Defining SO2 <30% as the presence of critical hypoxia, EsSO2 had a higher sensitivity than both the sNIRS and cNIRS (D). EsSO2 was <30 in 100% (33/33) of timepoints in which critical hypoxia was present, significantly more sensitive than rSO2 based on cNIRS (which was <30 in 58% (19/33) of timepoints) and sNIRS (<30 in 24% (8/33) of timepoints). Conclusion: EsSO2 accurately detects critical hypoxic states, more so than cNIRS or sNIRS, and may represent an alternative, continuous, less invasive approach to central venous monitoring.
Read full abstract