Context: Red blood cell distribution width (RDW) has been used to predict mortality during infection and inflammatory diseases. It also been purported to be predictive of mortality following traumatic injury.
Objectives: To identify the role of RDW in predicting mortality in trauma patients. We also sought to identify the role of RDW in predicting the development of sepsis in trauma patients.
Methods: A retrospective observational study was performed of the medical records for all adult trauma patients admitted to Loyola University Medical Center from 2007 to 2014. Patients admitted for fewer than four days were excluded. Admission, peak, and change from admission to peak (Δ) RDW were recorded to determine the relationship with in-hospital mortality. Patient age, development of sepsis during the hospitalization, admission to the intensive care unit (ICU), and discharge disposition were also examined.
Results: A total of 9,845 patients were admitted to the trauma service between 2007 and 2014, and a total of 2,512 (25.5%) patients fit the inclusion criteria and had both admission and peak values available. One-hundred twenty (4.6%) died while in the hospital. RDW values for all patients were (mean [standard deviation, SD]): admission 14.09 (1.88), peak 15.09 (2.34), and Δ RDW 1.00 (1.44). Admission, peak, and Δ RDW were not significant predictors of mortality (all p>0.50; hazard ratio [HR], 1.01–1.03). However, trauma patients who eventually developed sepsis had significantly higher RDW values (admission RDW: 14.27 (2.02) sepsis vs. 13.98 (1.73) no sepsis, p<0.001; peak RDW: 15.95 (2.55) vs. 14.51 (1.97), p<0.001; Δ RDW: 1.68 (1.77) vs. 0.53 (0.91), p<0.001).
Conclusions: Admission, peak, and Δ RDW were not associated with in-hospital mortality in adult trauma patients with a length of stay (LOS) ≥four days. However, the development of sepsis in trauma patients is closely linked to increased RDW values and in-hospital mortality.