BACKGROUND: We hypothesized that near-infrared spectroscopy (NIRS)-derived tissue oxygenation saturation (StO2) could assist in identifying shock in casualties arriving to a combat support hospital and predict the need for life-saving interventions (LSIs) and blood transfusions.
METHODS: We performed a prospective observational trial at a single US Army combat support hospital in Iraq from August to December 2007. Arriving casualties had NIRS-derived StO2 recorded in the emergency department. Minimum (StO2 min) and initial 2-minute averaged StO2 and tissue hemoglobin index readings were used as end points. Outcomes measured were requirement for LSIs, any blood transfusion, massive transfusion (>10 units in 24 hours), and early mortality. The data were subjected to univariate and multivariate logistic regression modeling.
RESULTS: Of the 147 combat casualties enrolled in the trial, 72 (49%) required an LSI, 42 (29%) required blood transfusion, and 10 (7%) required massive transfusion. On multivariate logistic regression analysis of the whole study group, systolic blood pressure (SBP), international normalized ratio, tissue hemoglobin index, and hematocrit predicted blood transfusion with an area under the curve of 0.90 (0.84-0.96), with a confidence interval of 95%. When just the group with an SBP >90 was analyzed, independent predictors of patients requiring blood transfusion on logistic regression analysis were StO2 min (odds ratio of 1.35) and hematocrit (odds ratio of 2.66) for an area under the curve of 0.84 (0.76-0.92).
CONCLUSIONS: NIRS-derived StO2 obtained on arrival predicts the need for blood transfusion in casualties who initially seem to be hemodynamically stable (SBP >90). Further study of this technology for use in the resuscitation of trauma patients is warranted.