Mortality of Adult Respiratory Distress Syndrome in Trauma Patients: A Systematic Review over a Period of Four Decades

Abstract

Background: Acute respiratory distress syndrome (ARDS) is associated with high morbidity and mortality. In addition, its aetiologies are heterogeneous, and the outcome depends on the underlying cause. In trauma care, changes in diagnostics, management and treatment may influence the outcome of posttraumatic ARDS over time, which are attributable to geographic distribution, relate to the definition used and depend on the injury severity.

Methods: We selected studies by searching PubMed using the MeSH terms "acute respiratory distress syndrome", "mortality" and "trauma" and all their combinations. Studies that reported mortality rates for polytrauma patients with ARDS were included. We excluded studies with selected population collectives (e.g. burn patients, isolated thoracic trauma and paediatric trauma) and non-trauma studies. Studies were stratified according to the recruitment year of patients into Decade 1 (prior 1990), Decade 2(1990-1999), Decade 3 (2000-2009) and Decade 4 (later 2009); geographic location (North America or Europe), the definition used in the different studies (American European Consensus Conference (AECC) definition or Berlin definition) and the mean injury severity of the patient populations, respectively.

Results: Twenty-three studies between 1 January 1980 and 30 June 2018 were included in the analysis (486,861 patients, 52,561 with posttraumatic ARDS). There was a wide variation in mortality rates among the studies (4-39%). The overall pooled mean mortality rate for all studies was 21.8%, SD ± 8.3%. The mean mortality over the last four decades was similar between decade 1 (25.8%), decade 2 (20.6%), decade 3 (20.7%) and decade 4 (22.5%). Geographical observations comparing Europe and North America revealed no difference [Europe (22.1%) and North America (21.7%)]. The ARDS mortality in publications using the Berlin definition (23%) was comparable to the mortality rate in publications using the AECC definition (22.9%). The mortality rate based on the injury severity showed no alteration (ISS ≤ 25 points, 22.6%, ISS > 25 points, 22.6%).

Conclusion: In this systematic review, there was no change in the mortality rate over the last four decades, no geographical difference within Western societies, no dependence on the ARDS definition used and no association with injury severity, respectively.