Non-invasive monitoring of microcirculation dynamics in hypovolemic shock: a novel application of diffuse correlation spectroscopy.

Background: Microcirculatory dysfunction is a poor prognostic indicator for the management of critically ill patients, highlighting the need for the development of appropriate assessment methods. Current microcirculatory parameters are often indirect, invasive, or lack immediacy and continuity, with no standardised markers for critical care. Diffuse correlation spectroscopy (DCS), a near-infrared optical technique, facilitates the non-invasive real-time monitoring of microvascular dynamics via the blood flow index (BFI). However, the relationship between BFI and conventional microcirculatory parameters in hypovolemic shock remains unclear. This study examined the utility of DCS in assessing the microcirculation during hypovolemic shock in a canine model.

Methods: Six male beagle dogs underwent controlled blood withdrawal to induce hypovolemic shock, defined as a ≥ 30% decrease in cardiac output (CO) and mean arterial pressure (MAP) < 60 mmHg or systolic arterial pressure (SAP) < 90 mmHg. BFI was measured using a DCS device attached to the skin of the forelimb. From baseline to blood withdrawals followed by transfusions, changes in BFI were compared with microcirculatory parameters, mixed venous oxygen saturation (S_vO₂), core-to-skin temperature gradient (ΔT), veno-arterial difference in partial pressure of carbon dioxide (PCO₂ gap), and serum lactate. Correlation and receiver operating characteristic (ROC) analyses were performed to determine the cut-off value of relative BFI for predicting whether lactate levels exceeded 22.5 mg/dL.

Results: Blood withdrawal resulted in significant reductions in BFI, CO, and radial artery blood flow, with the corresponding deteriorations in the ΔT, S_vO₂, and PCO₂ gap and lactate levels. BFI showed significant correlations with ΔT (correlation coefficient [CC] = - 0.48, 95% confidence interval [CI] - 0.69 to - 0.18, p < 0.01), SvO₂ (CC = 0.67, 95% CI 0.43 to 0.81, p < 0.01), and PCO2 gap (CC = - 0.63, 95% CI - 0.79 to - 0.39, p < 0.01). ROC analysis identified a relative BFI threshold of 35.5% of the baseline for predicting elevated lactate levels, with 62% sensitivity and 100% specificity (AUC = 0.75).

Conclusions: Blood flow index measured by DCS reflects peripheral perfusion changes and is significantly correlated with clinical parameters during blood withdrawal and transfusion, highlighting its potential for non-invasive, continuous microcirculation monitoring in hypovolemic shock.