To evaluate the effect of paramedic rapid sequence intubation (RSI) on outcome in patients with severe traumatic brain injury.
Adult major trauma victims were prospectively enrolled over two years using the following inclusion criteria: Glasgow Coma Scale (GCS) 3–8, suspected head injury by mechanism or physical examination, transport time > 10,“ and inability to intubate without RSI. Midazolam and succinylcholine were administered before laryngoscopy; rocuronium was given after tube placement was confirmed using physical examination, capnometry, syringe aspiration, and pulse oximetry. The Combitube was used as a salvage airway device. For this analysis, trial patients were excluded for absence of a head injury (Head/Neck AIS score < 2), failure to fulfill major trauma outcome study criteria, unsuccessful intubation or Combitube insertion, or death in the field or in the resuscitation suite within 30” of arrival. Each study patient was hand matched to three nonintubated historical controls from our trauma registry using the following parameters: age, sex, mechanism of injury, trauma center, and AIS score for each body system. Controls were excluded for Head/Neck AIS defined by a c-spine injury or death in the field or in the resuscitation suite within 30“ of arrival. χ2, odds ratios, and logistic regression were used to investigate the impact of RSI on the primary outcome measures of mortality and incidence of a ”good outcome,“ defined as discharge to home, rehabilitation, psychiatric facility, jail, or signing out against medical advice.
A total of 209 trial patients were hand matched to 627 controls. The groups were similar with regard to all matching parameters, admission vital signs, frequency of specific head injury diagnoses, and incidence of invasive procedures. Mortality was significantly increased in the trial cohort versus controls for all patients (33.0% versus 24.2%, p < 0.05) and in those with Head/Neck AIS scores of 3 or greater (41.1% versus 30.3%, p < 0.05). The incidence of a “good outcome” was lower in the trial cohort versus controls (45.5% versus 57.9%, p < 0.01). Factors that may have contributed to the increase in mortality include transient hypoxia, inadvertent hyperventilation, and longer scene times associated with the RSI procedure.
Paramedic RSI protocols to facilitate intubation of head-injured patients were associated with an increase in mortality and decrease in good outcomes versus matched historical controls.
From the Department of Emergency Medicine (D.D., P.R.), Department of Surgery, Division of Trauma, Department of Surgery (D.H., D.F., T.H.), and Division of Neurosurgery (L.M.), UC San Diego, and San Diego County Emergency Medical Services (M.O.), San Diego, California, U.S.A.
Submitted for publication October 13, 2002.
Accepted for publication December 5, 2002.
Presented at the 61st Annual Meeting of the American Association for the Surgery of Trauma, September 26–28, 2002, Orlando, Florida.
Support for this project was received from the Society for Academic Emergency Medicine.
Address for correspondence: Daniel Davis, MD, UCSD Emergency Medicine, 200 West Arbor Drive, #8676, San Diego, California, U.S.A. 92103-8676