Evaluation of tissue hemoglobin saturation (StO2) using near-infrared spectroscopy during hypoxemia and hyperoxemia in Beagle dogs.

OBJECTIVE: To determine the relationship between tissue oxygen saturation (StO2) and oxygen delivery (D˙O2) during hypoxemia and hyperoxemia.

STUDY DESIGN: Prospective, randomized study.

ANIMALS: Eight purpose-bred Beagle dogs.

METHODS: Dogs were anesthetized with isoflurane, ventilated to eucapnia, and instrumented for thermodilution cardiac output, invasive mean arterial pressure (MAP), sartorius muscle StO2 and airway gas monitoring. Dogs were administered rocuronium to facilitate mechanical ventilation and esmolol to minimize anesthetic effects on cardiac output. Instrumentation and baseline data collection were at 0.21 fractional inspired oxygen (FIO2). Dogs were evaluated at high (0.40 then 0.95) and low (0.15 then 0.10) FIO2 sequences in random order with a 60 minute rest period at FIO2 0.21 between sequences. Target FIO2 was achieved by manipulating nitrogen and oxygen flow rates. Data collected at each FIO2, after a 10 minute period of stabilization, included heart rate (HR), MAP, cardiac index (CI) and StO2. Arterial oxygen content (CaO2) and oxygen delivery index (D˙O2I) were calculated at each FIO2. Data analysis included Pearson's correlation analysis and mixed-model anova (p < 0.05).

RESULTS: There were no significant differences in HR, MAP or CI across all FIO2 values. Significant decreases occurred in mean ± standard deviation StO2 (90 ± 4% to 69 ± 18%; p = 0.0001), D˙O2I (458 ± 70 to 281 ± 100 mL minute(-1) m(-2); p = 0.0008) and CaO2 (13.2 ± 1.53 to 8.4 ± 2.05 mL dL(-1); p = 0.0001) from FIO2 0.21 to 0.10, but not at remaining FIO2 values. The correlation between StO2 and D˙O2I across all FIO2 values was strong (r = 0.97; p = 0.0013) and linear.

CONCLUSIONS AND CLINICAL RELEVANCE: In this model of hypoxemia and hyperoxemia, the strong correlation between StO2 and D˙O2I suggests that StO2 can be used to estimate D˙O2.