Nils Becker (Aachen / DE), Antonia Bernhardt (Aachen / DE), Felix Bläsius (Aachen / DE), Christian David Weber (Aachen / DE), Frank Hildebrand (Aachen / DE), Klemens Horst (Aachen / DE)
Abstract text (incl. references and figure legends)
Introduction In polytraumatized patients, factors of trauma-induced mortality are time-dependent. While early deaths are predominantly associated with injury severity, late deaths are often triggered by complications.1 In this retrospective analysis we assessed risk factors for a delayed mortality and the influence of different clusters of complications.
Materials & Methods Inclusion criteria: primary admission to a level I trauma center, Injury Severity Score (ISS) >16, age >18 years. We compared immediate (death within the first 24h after admission), early (death within 24-72h after admission) and delayed mortality (death later than 72h after admission). Statistical analysis was performed via SPSS software (IBM), using Mann-Whitney-U-Test and binominal logistic regression analysis.
Results A total of 382 patients (71.1% male, mean age 51.5 years, median ISS of 23) were included. A delayed death occurred in 31 patients (8%). Immediate and early death were significantly influenced by the ISS, but were not affected by the development of complications. Significant influence on a delayed mortality were observed by severe head injuries (AIS > 2; OR 3.4; 95%-CI 1.154; 10.036; p=0.026), age (OR 1.04; 95%-CI 1.017; 1.063; p<0.001) and the development of infectious complications (OR 2.665; 95%-CI 1.221; 5.814; p=0.014). Duration of mechanical ventilation (337h vs. 284h; p=0.039) and mean duration of ICU stay (15d vs. 13d; p=0.041) were significantly longer in patients with a delayed death.
Conclusion Infectious complications are a main factor of delayed mortality. Special attention should be applied to older patients and those with severe head injuries. Further studies are needed to understand the underlaying effects of trauma- and ventilation associated infections to reduce rates of posttraumatic infections.
References 1Pfeifer et al., PLoS One. 2016
Disclosure: Do you have a significant financial interest, consultancy or other relationship with products, manufacturer(s) of products or providers of services related to this abstract? (If not, please enter "No" in the text field.)
No