Fever in trauma patients: friend or foe?

BACKGROUND: In trauma patients, elevated body temperature is a common and noninfective procedure soon after injury. We hypothesized that the absence of this febrile response is associated with failure to meet metabolic demands and results in adverse outcomes.
METHODS: We collected retrospective data of 253 consecutive trauma patients admitted to the intensive care unit during a 3-year period. Patients were stratified according to their daily maximum body temperature from days 1 to 10 (Tmax 1-10): no fever (<37.5 degrees C), low fever (37.5-38.4 degrees C), moderate fever (38.5-39.0 degrees C), and high fever (>39.0 degrees C). The area under the curve (AUC) of core temperature during the first 24 hours after admission was calculated for each patient at a baseline of 36 degrees C. The infection and mortality rates were measured.
RESULTS: Sixty-three patients (24.9%) developed an infection, and the overall mortality was 7.5% (19 patients). Patients with no Tmax 1 and a low or high Tmax 4 to 10 had a significantly high infection rate; those with no fever on days 1 and 2 had a significantly high mortality rate. A low AUC was also associated with significantly higher infection and mortality rates. Multiple logistic regression analysis controlled for age, injury severity score, Tmax 1, AUC, initial temperature at admission, and time taken to reach 36 degrees C (if hypothermia was present) revealed that age, injury severity score, low AUC (odds ratio, 0.96; 95% confidence interval, 0.94-0.99; p = 0.002), and initial temperature were independent predictors of infection. Age and lower AUC (odds ratio, 0.87; 95% confidence interval, 0.81-0.92; p < 0.001) were both predictors of mortality.
CONCLUSIONS: A febrile response until day 4 after injury did not increase morbidity, and a low AUC is independently associated with adverse outcomes. These findings show that a nonfebrile response soon after injury results in poor prognosis.