A prospective, randomized, controlled trial compared paramedic rapid sequence intubation with hospital intubation in adults with severe traumatic brain injury in four cities in Victoria, Australia. The primary outcome was neurologic outcome at 6 months postinjury.
Training
Paramedics already experienced in ‘cold’ intubation (without drugs) undertook an additional 16-hour training program in the theory and practice of RSI, including class time (4 hours), practical intubating experience in the operating room under the supervision of an anesthesiologist (8 hours), and completion of a simulation-based examination (4 hours).
Methods
Patients included in the study were those assessed by paramedics on road ambulances as having all the following: evidence of head trauma, Glasgow Coma Score ≤9, age ≥15 years, and ‘intact airway reflexes’, although this is not defined or explained. Patients were excluded if any of the following applied: within 10 minutes of a designated trauma hospital, no intravenous access, allergy to any of the RSI drugs (as stated by relatives or a medical alert bracelet), or transport planned by medical helicopter. Drug therapy for intubation consisted of fentanyl (100μg), midazolam (0.1 mg/kg), and succinylcholine (1.5 mg/kg) administered in rapid succession. Atropine (1.2 mg) was administered for a heart rate <60/min. A minimum 500 mL fluid bolus (lactated Ringers Solution) was administered. A half dose of the sedative drugs was used in patients with hypotension (systolic blood pressure <100 mm Hg) or older age (>60 years).

Cricoid pressure was applied in all patients. After intubation and confirmation of the position of the endotracheal tube using the presence of the characteristic waveform on a capnograph, patients received a single dose of pancuronium (0.1 mg/kg), and an intravenous infusion of morphine and midazolam at 5 to 10 mg/h each. If intubation was not achieved at the first attempt, or the larynx was not visible, one further attempt at placement of the endotracheal tube over a plastic airway bougie was permitted. If this was unsuccessful, ventilation with oxygen using a bag/mask and an oral airway was commenced and continued until spontaneous respirations returned. Insertion of a laryngeal mask airway was indicated if bag/mask ventilation using an oral airway appeared to provide inadequate ventilation. Cricothyroidotomy was indicated if adequate ventilation could not be achieved with the above interventions. In all patients, a cervical collar was fitted, and hypotension (systolic blood pressure <100 mm Hg) was treated with a 20 mL/kg bolus of lactated Ringers Solution that could be repeated as indicated. Other injuries such as fractures were treated as required. In the hospital emergency department, patients who were not intubated underwent immediate RSI by a physician prior to chest x-ray and computed tomography head scan.
Follow up
At 6 months following injury, surviving patients or their next-of-kin were interviewed by telephone using a structured questionnaire and allocated a score from 1 (deceased) to 8 (normal) using the extended Glasgow Outcome Scale (GOSe). The interviewer was blinded to the treatment allocation.
Statistical power
A sample size of 312 patients was calculated to achieve 80% power at an alpha error of 0.05. Three hundred twenty-eight patients met the enrollment criteria. Three hundred twelve patients were randomly allocated to either paramedic intubation (160 patients) or hospital intubation (152 patients). A mean Injury Severity Score of 25 indicated that many patients had multiple injuries.
Success of intubation
Of the 157 patients administered RSI drugs, intubation was successful in 152 (97%) patients. The remaining 5 patients had esophageal placement of the endotracheal tube recognized immediately on capnography. The endotracheal tube was removed and the patients were managed with an oropharyngeal airway and bag/mask ventilation with oxygen and transported to hospital. There were no cases of unrecognised esophageal intubation on arrival at the emergency department during this study and no patient underwent cricothyroidotomy.
Outcome
After admission to hospital, both groups appeared to receive similar rates of neurosurgical interventions, including initial CT scan, urgent craniotomy (if indicated), and monitoring of intracranial pressure in the intensive care unit.
Favorable neurologic outcome was increased in the paramedic intubation patients (51%) compared with the hospital intubation patients (39%), just reaching statistical significance with P = 0.046. A limitation is that 13 of 312 patients were lost to follow-up and the majority of these were in the hospital intubation group. The authors do point out that the difference in outcomes would no longer be statistically significant whether one more patient had a positive outcome in the treatment group (P = 0.059) or one less in the control group (P = 0.061). The median GOSe was higher in the paramedic intubation group compared with hospital intubation (5 vs. 3), however, this did not reach statistical significance (P = 0.28).
More patients in the paramedic intubation group suffered prehospital cardiac arrest. There were 10 cardiac arrests prior to hospital arrival in the paramedic RSI group and 2 in the patients allocated to hospital intubation. Further detail on these patients is provided in the paper. The authors state that it is likely that the administration of sedative drugs followed by positive pressure ventilation had adverse hemodynamic consequences in patients with uncontrolled bleeding, and that it is possible that the doses of sedative drugs administered in this study to hemodynamically unstable patients were excessive and consideration should be given to a decreasing the dose of sedation.
Authors’ conclusions
The authors overall conclusion is that patients with severe TBI should undergo prehospital intubation using a rapid sequence approach to increase the proportion of patients with favorable neurologic outcome at 6 months postinjury. Further studies to determine the optimal protocol for paramedic rapid sequence intubation that minimize the risk of cardiac arrest should be undertaken.
Prehospital rapid sequence intubation improves functional outcome for patients with severe traumatic brain injury: a randomized controlled trial.
Ann Surg. 2010 Dec;252(6):959-65.
Victorian Ambulance Service protocols are available here, which include their current paramedic RSI protocol