A systematic review of human and veterinary applications of noninvasive tissue oxygen monitoring

Evaluation of tissue oxygen saturation in naturally occurring canine shock patients

OBJECTIVE

To measure tissue oxygen saturation (StO₂ ) in a population of dogs with naturally occurring shock and to evaluate the relationship of StO₂ with an established veterinary severity scoring system (Acute Patient Physiologic and Laboratory Evaluation) and patient survival.

DESIGN

Prospective observational study.

SETTING

University teaching hospital.

ANIMALS

Twenty-five adult dogs presenting in shock, as determined by the presence of hypotension, the calculated shock index, and hyperlactatemia.

INTERVENTIONS

StO₂ was measured prior to any therapeutic interventions. Blood samples were also collected for measurement of plasma lactate, complete blood count, and a serum biochemical profile. Abdominal and thoracic focused assessment with sonography was also performed.

MEASUREMENTS AND MAIN RESULTS

Dogs enrolled in this study had lower mean (±SD) StO₂ values (65.12 ± 17.7%) than previously reported in experimental models of canine hemorrhagic shock. There was a moderate correlation between lower StO₂ and increasing Acute Patient Physiologic and Laboratory Evaluation scores. A single StO₂ value, assessed prior to therapeutic intervention, was not a sensitive predictor of mortality in this population.

CONCLUSIONS

Dogs with naturally occurring shock have lower mean StO₂ values than those previously reported in dogs with experimentally induced shock. A lower initial StO₂ was associated with worse disease severity but was not a significant predictor of survival in this population.

Tissue oxygen saturation in dogs with acute hemorrhage

OBJECTIVE

To evaluate initial tissue hemoglobin oxygen saturation (StO₂ ) in dogs presenting to an emergency room (ER) for acute hemorrhage.

DESIGN

Prospective, observational study.

SETTING

University veterinary teaching hospital.

ANIMALS

Thirty-eight dogs with acute hemorrhage were enrolled between July 2009 and October 2010. Seventy-eight normal dogs from a previous observational study were included to represent healthy controls ("no shock").

INTERVENTIONS

Tissue oxygen saturation measurement was obtained at enrollment on dogs presented to the ER for acute hemorrhage. Baseline clinicopathologic (CBC, serum biochemical profile, prothrombin time, and activated partial thromboplastin time) and physiologic (plasma lactate concentration, venous blood gas, blood pressure, and hemoglobin oxygen saturation by pulse oximetry) data were recorded from all patients with hemorrhage. An ER clinician blinded to the StO₂ value guided patient management. Patient survival to discharge from the hospital in the study group was recorded. Once data collection was complete, 3 emergency and critical care clinicians blinded to the StO₂ data retrospectively classified patients into 1 of 4 shock categories (no shock, mild, moderate, or severe shock).

MEASUREMENTS AND MAIN RESULTS

The historical group of healthy dogs had higher StO₂ concentrations compared to the dogs classified with shock at all 3 levels (mild, moderate, and severe, P = 0.0006, <0.0001, and 0.0018, respectively); however, there was no statistical difference in StO₂ between the levels of shock. A cut-off StO₂ value of 87.6% identified a patient as having shock (area under the curve: 0.824, 95% confidence interval 0.749, 0.899).

CONCLUSIONS

Dogs with hemorrhagic shock have lower StO₂ than a population of healthy dogs.

Diffuse optical monitoring of peripheral tissues during uncontrolled internal hemorrhage in a porcine model

Reliable, continuous and noninvasive blood flow and hemoglobin monitoring in trauma patients remains a critical, but generally unachieved goal. Two optical sensing methods - diffuse correlation spectroscopy (DCS) and diffuse reflectance spectroscopy (DRS) - are used to monitor and detect internal hemorrhage. Specifically, we investigate if cutaneous perfusion measurements acquired using DCS and DRS in peripheral (thighs and ear-lobe) tissues could detect severe hemorrhagic shock in a porcine model. Four animals underwent high-grade hepato-portal injury in a closed abdomen, to induce uncontrolled hemorrhage and were subsequently allowed to bleed for 10 minutes before fluid resuscitation. DRS and DCS measurements of cutaneous blood flow were acquired using fiber optical probes placed on the thigh and earlobe of the animals and were obtained repeatedly starting from 1 to 5 minutes pre-injury, up to several minutes post shock. Clear changes were observed in measured optical spectra across all animals at both sites. DCS-derived cutaneous blood flow decreased sharply during hemorrhage, while DRS-derived vascular saturation and hemoglobin paralleled cardiac output. All derived optical parameters had the steepest changes during the rapid initial hemorrhage unambiguously. This suggests that a combined DCS and DRS based device might provide an easy-to-use, non-invasive, internal-hemorrhage detection system that can be used across a wide array of clinical settings.