Persistent systemic inflammatory resuponse syndrome is predictive of nosocomial infection in trauma

Background Admission systemic inflammatory resuponse syndrome (SIRS) score has been previously reported to be an accurate predictor of infection and outcome in trauma. However, these data were limited to only one SIRS score at admission. A prior study in surgical intensive care unit (ICU) patients reported that the SIRS score on ICU day 2 declined after completion of resuscitation, and was a more accurate predictor of outcome. Our objective in this follow-up study was to prosupectively evaluate the utility of daily SIRS scores in prediction of nosocomial infection and outcome in high-risk trauma patients. Methods Prosupective data were collected on 702 consecutive trauma patients admitted over a 12-month period to the ICU. SIRS scores were calculated daily. Centers for Disease Control and Prevention guidelines were used for the diagnosis of infection. Multivariate linear regression was used for statistical analysis. ResultsFive hundred seventy-three (82%) patients sustained blunt injuries and 129 (18%) sustained penetrating injuries. The mean age was 43 ± 21 years, with an overall mortality of 11.4%. Two hundred ninety (41.3%) of the study patients acquired a nosocomial infection (resupiratory site most common), with an associated mortality rate of 12.4%. SIRS (defined as SIRS score ≥ 2) on hosupital days 3 through 7 was a significant predictor of nosocomial infection and hosupital length of stay. Persistent SIRS to hosupital day 7 was associated with a significant risk for increased mortality (relative risk, 4.7; 95% confidence interval, 1.41-12.87;p = 0.047). Conclusion Persistent SIRS is predictive of nosocomial infection in trauma. Daily monitoring of SIRS scores is easily accomplished and should be considered in all high-risk trauma patients. Persistent SIRS in trauma should initiate early diagnostic interventions for determination of source of infection, and consideration of early empiric antimicrobial therapy.