Using Serial Hemoglobin Levels to Detect Occult Blood Loss in the Early Evaluation of Blunt Trauma Patients.

BackgroundSerial hemoglobin measurement (ΔHgb) is intended to aid in the early identification of blunt trauma patients who have significant blood loss requiring intervention. However, the utility of ΔHgb has yet to be rigorously studied.ObjectiveWe sought to determine if ΔHgb is a reliable diagnostic tool in assessing blood loss in blunt trauma patients.MethodsWe enrolled consecutive blunt trauma patients ≥18 years of age who presented to a level I trauma center. We measured 2 hemoglobin levels spaced 5 min apart and calculated the difference (ΔHgb) for each patient. We also recorded whether each patient required any of the following interventions to treat their injuries: 1) operation or procedure to control hemorrhage; 2) radiographic embolization; 3) administration of blood and blood products; 4) administration of ≥3 liters of intravenous fluids; and 5) exsanguination. Our primary outcome was the area under the receiver operating characteristic (ROC) curve.ResultsWe enrolled 251 patients, including 192 males and 59 females with a mean age of 40 years. Interventions occurred in 56 patients and were withheld in 195. The median ΔHgb was -0.1 gm/dL (interquartile range -0.5 to 0.1 gm/dL) for patients requiring intervention and 0.0 gm/dL (interquartile range -0.6 to 0.3 gm/dL) for patients not requiring intervention. We found the area under the ROC curve to be 0.53 (95% confidence interval 0.44-0.62).ConclusionsOur results indicated that ΔHgb does not reliably distinguish between blunt trauma patients who require intervention and those who do not