Esophageal resonance Raman spectroscopy tracks mixed venous oxygenation in experimental shock and hypoxemia

Abstract Introduction Mixed venous oxygen saturation (SvO 2 ) reflects the balance between systemic oxygen delivery and consumption and is widely used as an indicator of tissue oxygenation in critically ill patients. However, its measurement requires invasive venous access. We evaluated whether a customized probe applied to the esophageal mucosa could provide a continuous measurement of regional capillary hemoglobin oxygen saturation (S Es O 2 ) that tracks changes in mixed venous oxygen saturation for early detection of experimental shock with minimal invasiveness. Methods A miniaturized 405 nm resonance Raman spectroscopy (RRS) probe was developed to quantify esophageal regional capillary hemoglobin saturation (S Es O 2 ) based on spectral decomposition of oxy- and deoxy-hemoglobin Raman peaks. Two rat models were investigated: progressive hemorrhage (n = 12) and graded hypoxia (n = 9). At each stage, SvO 2 , S Es O 2 , arterial oxygen saturation (SaO 2 ), and serum lactate were measured. Tracking ability and diagnostic accuracy compared to SvO 2 and transabdominal near-infrared spectroscopy (NIRS) were evaluated using linear mixed-effects models (animal random intercepts), (cluster-bootstrapped) receiver operating characteristic curve (ROC) analyses, and Pearson correlation. Results In both models, baseline S Es O 2 was ~20% higher than SvO 2 , consistent with a regional capillary measurement distinct from mixed venous blood. In hemorrhage, LME showed S Es O 2 was positively associated with SvO 2 (β = 0.53, 95% CI 0.43–0.62; p < 0.0001) and inversely associated with lactate (β = −0.15, 95% CI −0.18 to −0.12; p < 0.0001). In ROC analysis, S Es O 2 discriminated SvO 2 < 40% with AUC 0.98 (95% CI 0.95–1.00). For transabdominal NIRS, the AUC was 0.82 (95% CI 0.70–0.95). During graded hypoxia, S Es O 2 was also positively associated with SvO 2 (β = 0.51, 95% CI 0.43–0.60; p < 0.0001) and inversely associated with lactate (β = −0.11, 95% CI −0.13 to −0.09; p < 0.0001). Discrimination of S Es O 2 for SvO 2 < 40% was AUC 0.92 (95% CI 0.82–1.00). Transabdominal NIRS demonstrated discrimination with high uncertainty (AUC 0.56, 95% CI 0.23–1.00). Conclusions Esophageal RRS enables molecularly specific measurement of regional capillary hemoglobin oxygen saturation and thereby tracks SvO 2 across distinct experimental shock paradigms. These preclinical findings support further investigation of esophageal RRS as a potential minimally invasive monitoring approach for detection of systemic oxygen imbalance.