Prehospital dynamic tissue oxygen saturation response predicts in-hospital lifesaving interventions in trauma patients

Abstract

Background: Tissue oximetry (StO2) plus a vascular occlusion test is a noninvasive technology that targets indices of oxygen uptake and delivery. We hypothesize that prehospital tissue oximetric values and vascular occlusion test response can predict the need for in-hospital lifesaving interventions (LSI).

Methods: We conducted a prospective, blinded observational study to evaluate StO2 slopes to predict the need for LSI. We calculated the DeO2 slope using Pearson's coefficients of regression (r2) for the first 25% of descent and the ReO2 slope using the entire recovery interval. The primary outcome was LSI defined as the need for emergent operation or transfusion in the first 24 hours of hospitalization. We created multivariable logistic regression models using covariates of age, sex, vital signs, lactate, and mental status.

Results: We assessed StO2 in a convenience sample of 150 trauma patients from April to November of 2009. In-hospital mortality was 3% (95% confidence interval [CI], 1.1-7.6); 31% (95% CI, 24-39) were admitted to the intensive care unit, 6% (95% CI, 2.8-11.1) had an emergent operation, and 10% (95% CI, 5.7-15.9) required transfusion. Decreasing DeO2 was associated with a higher proportion of patients requiring LSI. In the multivariate model, the association between the need for LSI and DeO2, Glasgow Coma Scale, and age persists.

Conclusion: Prehospital DeO2 is associated with need for LSI in our trauma population. Further study of DeO2 is warranted to determine whether it can be used as an adjunct triage criterion or an endpoint for resuscitation.

Figure 1

(A) Scatter plot showing the relationship of DeO₂ slope and lowest SBP. (B) Scatter plot showing the relationship of ReO₂ slope and lowest SBP.

Figure 2

Receiver operating curve for DeO₂ as a predictor of LSI.