Background: Rapid sequence intubation (RSI) has been instituted in some prehospital settings to improve the success of endotracheal intubation (ETI); whether RSI improves outcomes is unclear.
Objectives: We sought to determine if trauma patients intubated with RSI in the prehospital setting had better survival compared to those intubated without RSI.
Methods: Retrospective cohort analysis. We analyzed all injured adults (aged >or= 15 years) meeting state trauma system criteria, having a prehospital ETI attempt, and transported from the scene (19 counties) to one of the two state Level 1 trauma centers from 2000-2005. To adjust for the non-random selection of patients for field RSI, we built a propensity score from 15 important confounders, including demographics, injury severity, blood transfusion, surgical procedures, comorbidities, alcohol use, transport mode, injury mechanism, and initial field physiologic values. A propensity-adjusted multivariable logistic regression model (outcome = in-hospital mortality), with a time-based variable for system-wide implementation of changes in airway management, was used to test the association between RSI-ETI and mortality.
Results: There were 877 consecutive trauma patients who had prehospital ETI during this period and were included in the analysis. Of these, 496 (57%) had RSI-ETI. In univariate analyses, those with RSI-ETI had less severe injuries, better prehospital physiology (i.e., higher Glasgow Coma Scale score and blood pressure), fewer operations, fewer blood transfusions, and lower unadjusted mortality than those intubated without RSI. However, in the propensity-adjusted model, there was no statistical difference in mortality between the two groups (odds ratio 0.74, 95% confidence interval 0.52-1.06).
Conclusions: Patients selected for RSI-ETI were less seriously injured, with better prognostic factors than intubated patients for whom RSI was not used. After adjusting for these differences, use of prehospital RSI-ETI was not associated with improved survival.
Copyright (c) 2010 Elsevier Inc. All rights reserved.