INFLAMMATORY MARKERS - SUPERIOR PREDICTORS OF ADVERSE OUTCOME IN BLUNT TRAUMA PATIENTS

Objective: To assess whether variables reflective of early metabolic r esponses to injury are predictors of outcome in critically ill trauma patients. Design: Clinical inception cohort study comparing convention al measures of injury severity with early host response markers for th e correlation of each with outcome. These data are prospectively colle cted in a group of patients being evaluated in a nutritional support i nvestigation. Setting: Intensive care unit (ICU) of a major Level I tr auma center. Patients: Seventeen blunt trauma patients, aged 18 to 60 yrs with an Injury Severity Score of greater than or equal to 15, requ iring early mechanical ventilation. Interventions: Blood and urine sam ples were routinely obtained from patients undergoing nutritional supp ort by one of three routes. Measurements and Main Results: Conventiona l assessment was consistent with moderate severity and variation: Inju ry Severity Score, 41 +/- 15; Glasgow Coma Score, 11 +/- 4; admission circulating lactate concentration, 4.8 +/- 2.2 mmol/L; and first 24-hr transfusion requirement, 3.1 +/- 2.9 L. The mean concentrations of in flammatory marker during the first week were: cholesterol, 2.67 +/- 0. 80 mmol/L (103.2 +/- 31 mgd/dL); C-reactive protein, 23 +/- 11 mg/dL; transferrin, 1.44 +/- 0.47 gl/L; glucose, 9.21 +/- 2.27 mmol/L (166 +/ - 41 mg/dL); albumin, 26 +/- 5 g/L; and nitrogen loss, 24 +/- 9 g/d. H ospital outcome variables were: ventilator days, 17 +/- 7; ICU days, 2 6 +/- 10; hospital days, 38 +/- 15; occurrence rate of adult respirato ry distress syndrome (ARDS), 35%; infections, 82%; multiple organ fail ure, 71%; and total of hospital plus professional charges, $125,000 +/ - $56,000. A significant (p < .05), but weak, correlation existed betw een all seven outcome variables and the inflammatory markers: ventilat or days with cholesterol and C-reactive protein; ICU days with transfe rrin; total stay with cholesterol; ARDS with C-reactive protein; infec tions with glucose, cholesterol, and nitrogen loss; multiple organ fai lure with albumin and C-reactive protein; and financial charges with g lucose. However, a significant correlation existed between only two of seven outcome variables and conventional measures of severity: multip le organ failure with lactate and financial charges with transfusion r equirement. Conclusion: Readily obtainable inflammatory marker measure ments may better reflect the summation effects of the early perfusion deficit and tissue injury in the blunt trauma patient compared with co nventional measures of injury severity.