Investigating Changes in Cerebral Blood Flow and Oxygenation Using Optical Imaging Sensors and Swine Models of Hemorrhagic Shock

Introduction Direct assessment of organ perfusion during hemorrhage is essential for effective interventions. Initial blood loss triggers compensatory mechanisms to preserve vital organ perfusion, but failure leads to decreased perfusion and injury. Vital signs provide indirect measures of cerebral health, while diffuse correlation spectroscopy (DCS) and near-infrared spectroscopy (NIRS) enable continuous, noninvasive monitoring of cerebral microvascular blood flow (CBF), cerebral blood volume (CBV), and oxygenation, offering real-time insights for detection and intervention. Materials and Methods Adult swine models of controlled (30% blood volume reduction via syringe) and uncontrolled (partial hepatectomy) hemorrhage were studied. Systemic measures, i.e., blood pressure (BP), end-tidal CO₂ (EtCO₂), and heart rate, were compared with cerebral measures (CBF, CBV, and oxygenation) assessed by DCS-NIRS. Monitoring spanned 3 hours post-intervention, with periodic laboratory tests for injury confirmation. Results Systemic and cerebral hemodynamics were assessed pre-intervention, post-intervention, and during follow-up, revealing significant differences across timepoints for BP and cerebral measures (P <. 05). BP, CBF, and CBV decreased post-intervention in both cohorts, with recovery during follow-up only in the controlled cohort, where relative blood flow index (rBFI), total hemoglobin (HbT), and oxygenated hemoglobin (HbO) showed large reductions post-blood loss (large positive effects) and partial recovery during follow-up (large negative effects); however, rBFI and HbO remained below baseline, indicating incomplete recovery of cerebral perfusion and oxygenation. In the uncontrolled cohort, rBFI, HbT, and HbO decreased significantly post-injury, with reductions persisting below baseline during follow-up, reflecting impaired cerebral perfusion and oxygenation. EtCO₂ remained stable in the controlled cohort (χ²(2) = 3.8, P =. 15) but varied significantly in the uncontrolled cohort (χ²(2) = 9.25, P <. 01). Lactic acid levels differed significantly in both cohorts, while pH changes were significant only in uncontrolled hemorrhage. Conclusions DCS-NIRS-derived biomarkers for cerebral hemodynamics reflect changes observed in BP in controlled and uncontrolled hemorrhage, thus supporting our previous findings of DCS-NIRS's ability to detect hemorrhagic shock.