Pre- and Apnoeic high flow oxygenation for RApid sequence intubation in The Emergency department (Pre-AeRATE): study protocol for a multicentre, randomised controlled trial

Rapid sequence intubation (RSI) is the most common method of intubation used in the Emergency Department (ED) [1]. RSI involves the immediate administration of a paralytic agent following delivery of an induction agent that brings about unresponsiveness; it is a procedure which aims for successful endotracheal intubation with no ventilation via bag-valve-mask while awaiting the onset of paralysis. Under these circumstances, the maintenance of oxygen saturation during the apnoeic phase depends on the patient’s underlying oxygen reserve, and the adequacy of pre- and apnoeic oxygenation.

Critically ill patients have a shorter safe apnoeic time due to physiological stressors that accompany critical illnesses, such as decreased cardiac output, increased shunting, and reduced pulmonary reserves [2]. Oxygen saturation in these patients can rapidly drop to a critical hypoxic state of less than 70% within seconds [3]. Furthermore, with increasing prevalence of obesity both locally and worldwide [4, 5] from increasing affluence, a large proportion of our critically ill patients are also obese, which further shortens safe apnoeic times. It is estimated that a moderately ill normal-sized adult would reach an oxygen saturation (SpO₂) of 90% at slightly under 5 min, while a 127-kg obese adult would reach such a level within 2.5 to 3 min [6].

The risk of adverse outcomes escalates with an increasing number of attempts at endotracheal intubation and with the development of hypoxia [7‐9]. Prolongation of safe apnoea time and adequate pre-oxygenation would ameliorate some of these negative outcomes by preventing hypoxia and allowing more time to achieve successful intubation. As such, the current usual practice of non-rebreather mask for pre-oxygenation and standard nasal cannula of 15 L/min for apnoeic oxygenation may not be adequate in maintaining oxygenation during RSI of critically ill ED patients.

The purpose of this multi-centre randomised controlled trial is to determine the best practices for maintaining oxygenation during rapid sequence intubation of critically ill patients in the ED. Our primary aim is to test the hypothesis that the use of humidified high-flow oxygenation via nasal cannula (HFNC) at 60 L/min for pre-oxygenation and apnoeic oxygenation will maintain a higher SpO₂ when compared with current usual care of non-rebreather mask (NRM) for pre-oxygenation and standard nasal cannula of 15 L/min of oxygen flow for apnoeic oxygenation.

This will be a randomised controlled study (Fig. 1), enrolling adult patients aged 21 years and older who require RSI due to medical, surgical, and traumatic conditions in the EDs of the National University Hospital (NUH) and the Ng Teng Fong General Hospital (NTFGH), Singapore. The planned Consolidated Standards of Reporting Trials (CONSORT) flow diagram is presented in Fig. 1. The protocol is reported according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT; Fig. 2 and Additional file 1).

The NUH is a 1225-bed tertiary academic medical centre in Singapore and its ED receives over 110,000 attendances annually, of which about 47% of the cases require urgent (42.5%) or immediate (4.5%) care. The NTFGH is a 700-bed acute care general hospital located within the same healthcare cluster with annual ED attendances that approximate over 120,000. Both institutions perform an average of 12 to 20 RSIs per month.

Adequate pre-oxygenation is the cornerstone of successful RSI and an essential step in achieving prolonged safe apnoeic times during intubation [2]. Pre-oxygenation with a high FiO₂ de-nitrogenates the lungs and maximises oxygen reserve and oxygen saturation in the bloodstream [3]. The standard of care is to apply a non-rebreather mask (NRM) with tidal volume breathing for 3 min for pre-oxygenation. However, depending on the patient’s respiratory rate and depth of breathing which constitutes their minute volume, the actual FiO₂ delivered may vary from 0.8 to 0.6 [12], sometimes even lower. Other more efficacious methods such as applying positive airway pressure and pre-oxygenating in a head-up position of 20° have been described, but are not always possible in emergency situations or with obtunded patients [18, 19].

HFNC oxygenation has been gaining interest as an alternative method of pre-oxygenation and apnoeic oxygenation [20]. It can deliver up to 60 L/min of oxygen and an FiO₂ of 1.0. As it is humidified and heated, the high flow is well tolerated by awake patients [21]. Moreover, the high flow has been shown to wash out carbon dioxide in anatomical dead space, create positive airway pressure adequate enough for alveolar recruitment, and maintain a constant FiO₂, regardless of the patient’s minute volume [20]. Additionally, it has the convenience of continuing oxygenation during the apnoeic phase after induction and paralysis, since the high flow maintains a positive end-expiratory pressure and can insufflate the lungs, resulting in apnoeic oxygenation. Apnoeic oxygenation is possible due to the differential solubility of oxygen and carbon dioxide in blood. Carbon dioxide is more soluble and therefore moves less readily down its concentration gradient from the bloodstream into the alveoli during apnoea. As a result, more oxygen moves from the alveoli into the bloodstream. This creates a sub-atmospheric pressure in the alveoli, allowing oxygen to flow from the pharynx to the alveoli during apnoea.

Despite this sensible physiological rationale, a recent large randomised controlled trial reported no difference in median lowest arterial oxygen saturation in ICU patients with apnoeic oxygenation using 15 L/min of nasal cannula compared with no apnoeic oxygenation at all [22]. However, limitations in this study prevent definitive conclusions to be drawn. Firstly, the oxygen flow of 15 L/min was delivered through routine nasal cannulae, which are not designed for such flows. Secondly, a flow rate of 15 L/min may not be high enough to provide apnoeic oxygenation; it has been shown that flows of 30 L/min are necessary to generate positive airway pressures [23]. Thirdly, the cohort in the study consisted of mostly medical ICU patients with advanced respiratory failure, in whom 15 L/min of oxygenation may not generate enough positive airway pressure to overcome pulmonary shunting. Furthermore, the results cannot be generalised to patients who do not typically present to medical ICUs, for instance patients with severe trauma or a surgical diagnosis. There have been conflicting results from previous studies evaluating pre-oxygenation and apnoeic oxygenation using true HFNC. The PREOXYFLOW trial is the largest randomised controlled trial to date (n = 124) and found no difference in the lowest SpO₂ achieved when HFNC oxygenation was compared with a high fraction-inspired oxygen facial mask for pre-oxygenation. These results differ from another quasi-experimental before-after study (n = 101), which showed that patients who were pre-oxygenated with HFNC had significantly higher SpO₂ during the apnoeic phase [17]. Both studies were conducted in ICU patients and continued the HFNC during the apnoeic phase for apnoeic oxygenation.

Until now, there have been no studies conducted in the ED investigating the use of HFNC for pre- and apnoeic oxygenation. The incidence of failed intubations in the emergency setting is estimated to be 20 times that compared with intubation done electively [24], and ED patients differ from ICU patients because they are less likely to have been fasted thus increasing the risk of aspiration if re-oxygenation with bag valve mask is required during the intubation attempts. Aspiration of gastric contents contributes to increased risk of death during intubation [25]. Furthermore, hypoxia from failure to intubate is associated with increased ICU length of stay, irreversible brain damage, increased number of ventilated days, and higher incidence of tracheostomy [26]. This translates to longer hospital stays, adversely affecting overall bed occupancy, as well as higher healthcare costs. Given the existing high occupancy in Singapore’s restructured hospitals, which is expected to worsen as the population ages, it is vital that we prevent such adverse events from occurring by maintaining the lowest SpO₂ above 90% and prolonging safe apnoea times for intubation without the need for bag valve mask ventilation for re-oxygenation during RSI. If our hypothesis is confirmed, the results from this study could potentially change current clinical practice and reduce demands on an already stressed healthcare system.

Recruitment for the Pre-AeRATE trial is currently ongoing and expected to be completed by November 2019. This study protocol reports protocol version 2.0, dated 29 December 2017.