The decreasing incidence and mortality of acute respiratory distress syndrome after injury: A 5-year observational study

Background: Acute respiratory distress syndrome (ARDS) after major trauma has been associated with increased morbidity and mortality rates. Recently, there have been marked advances in defining etiologic factors and optimal management strategies for ARDS. We sought to examine whether there has been a corresponding change in the incidence and outcomes of ARDS after injury in recent years. Methods: A prospective observational study of all adult trauma intensive care unit (ICU) admissions over 5 years. Patients were evaluated daily for pre-defined ARDS criteria. Patient data, illness and injury severity, and ARDS incidence were compared by year of admission. Logistic regression analysis was used to identify independent predictors of ARDS and mortality. Results: There were 1,913 patients identified; the majority were male (79%) and suffered blunt trauma (62%). Two hundred seventy-four patients (14%) met criteria for ARDS. The incidence of ARDS showed a significant decrease from 23% in 2000 to rates of 8.4% and 9% for 2003 and 2004 (p < 0.01), respectively. There was no significant difference by year for trauma mechanism, age, sex, Injury Severity Score, Acute Physiology and Chronic Health Evaluation, ICU length of stay, or mortality. The strongest independent predictor of ARDS was year of ICU admission, with an odds ratio of 2.9 (95% confidence interval, 1.7-5.0) for admission in 2000 versus subsequent years (p < 0.001). After adjusting for age and injury severity, patients with ARDS had more days on mechanical ventilation and longer hospital and ICU stays (all p < 0.01), but there was no significant difference in mortality with or without ARDS (p = 0.57). Conclusion: There has been a more than 50% reduction in the incidence of ARDS after injury during the past 5 years in our institution despite similar patient demographics and injury severities. Development of ARDS increased hospital and ICU stays but not hospital mortality.