Background: Multiple-organ failure (MOF) is common among the most seriously injured trauma patients. The ability to easily and accurately identify trauma patients in the emergency department at risk for MOF would be valuable. The aim of this study was to derive and internally validate an instrument to predict the development of MOF in adult trauma patients using clinical and laboratory data available in the emergency department.
Methods: We enrolled consecutive adult trauma patients from 2005 to 2008 from the Denver Health Trauma Registry, a prospectively collected database from an urban Level 1 trauma center. Multivariable logistic regression was used to develop a clinical prediction instrument. The outcome was the development of MOF within 7 days of admission as defined by the Sequential Organ Failure Assessment (SOFA) score. A risk score was created from the final regression model by rounding the regression β coefficients to the nearest integer. Calibration and discrimination were assessed using 10-fold cross-validation.
Results: A total of 4,355 patients were included in this study. The median age was 37 years (interquartile range [IQR], 26-51 years), and 72% were male. The median Injury Severity Score (ISS) was 9 (IQR, 4-16), and 78% of the patients had blunt injury mechanisms. MOF occurred in 216 patients (5%; 95% confidence interval, 4-6%). The final risk score included patient age, intubation, systolic blood pressure, hematocrit, blood urea nitrogen, and white blood cell count and ranged from 0 to 9. The prevalence of MOF increased in an approximate exponential fashion as the score increased. The model demonstrated excellent calibration and discrimination (calibration slope, 1.0; c statistic, 0.92).
Conclusion: We derived a simple, internally valid instrument to predict MOF in adults following trauma. The use of this score may allow early identification of patients at risk for MOF and result in more aggressive targeted resuscitation and improved resource allocation.
Level of evidence: Prognostic and epidemiologic study, level III.