Background: Hypoxemia following injury contributes significant morbidity and mortality following trauma. The identification of modifiable risk factors for the prevention of acute lung injury in trauma patients remains unclear. We hypothesize that acute crystalloid administration and tidal volume delivery from mechanical ventilation may represent novel modalities to mitigate negative pulmonary outcomes in critically injured patients.
Methods: A retrospective, case-control study was undertaken after merging of the institutional trauma registry, trauma ventilator registry and the electronic medical record data from 2/2011 to 8/2014. Patients with survival > 24 hours, at least one PaO2 to FiO2 (P/F) ratio recorded during the first 7 days of hospitalization, and a tidal volume recorded during the first 2 days were included in the final analyses. Multivariate logistic regression models were utilized to investigate the contributions of demographic and injury characteristics, as well as blood products, crystalloid, and tidal volume exposures to negative pulmonary outcomes. The primary outcome of interest was the development of moderate to severe hypoxemia (P/F ratio = 200 mm Hg) during days 1-7. A secondary composite pulmonary outcome was created that included the development of in-hospital pneumonia, tracheostomy, moderate to severe hypoxemia, acute respiratory distress syndrome (ARDS), or early death occurring during days 1-7.
Results: Of the 661 patients within the dataset, 531 met inclusion criteria. The median age was 42 years, ISS was 24, 77% were male, 26% suffered a penetrating injury, and 57% experienced a P/F = 200 mm Hg. The median tidal volume was 7.8 (7.0-8.7) mL/kg of predicted body weight and the median crystalloid exposure in the first 24 hours was 2.3 (1.4-3.4) half liters. Those with high tidal volume exposure (>8 ml/kg, n = 224) were significantly older, female, and received less platelets. Those with high crystalloid amounts (> 3 half liters, n = 175) were significantly younger, had more penetrating injuries, were more commonly hypotensive on emergency department evaluation, had a higher Glasgow coma scale on admission, and had more crystalloid, packed red blood cell, and plasma needs. High crystalloid users also had a greater intensive care unit and hospital lengths of stay, ventilator days, and developed ARDS more often. Age (odds ratio [OR] 1.02, 95% confidence interval [CI] 1.01-1.04, p < 0.01) and chest abbreviated injury score (AIS) (OR 1.43 [CI 1.25-1.62], p < 0.01) were risk factors for a P/F = 200 mm Hg. Age (OR 1.03 [1.02-1.04], p = 0.04), head AIS (OR 1.3 [CI 1.15-1.47], p < 0.01), chest AIS (OR 1.32 [OR 1.16-1.51], p < 0.01), and crystalloid during the first 24 hours (OR 1.17 [1.00-1.37], p = 0.05) were found to be risk factors for a negative composite pulmonary outcome.
Conclusions: Crystalloid, but not tidal volume, exposure appears to be a modifiable risk factor for pulmonary complications. Further work should focus on trauma sub-populations that may have the greatest benefit from these preventative acute interventions.