Time to Sedation and Analgesia After Rapid Sequence Intubation in the Emergency Department

Authors: Callum Barnden, Biswadev Mitra, Amit Maini, Christopher Groombridge

Abstract

Objectives

Rapid sequence intubation (RSI) is a common procedure within the Emergency Department. “Sedation gap”—the time between induction and post intubation sedation—has been associated with adverse outcomes. The aim of this study was to measure the median time to sedation or analgesia post intubation in adult patients within an Australian Emergency Department (ED). The secondary aim was to assess the difference in “sedation gap” among medical and trauma patient cohorts.

Methods

The retrospective study was conducted using data within a 12‐month period at an Australian adult tertiary referral hospital emergency department. Data were sourced from an online airway registry and electronic medical records. Time to sedation or analgesia was summarised as median with interquartile range.

Results

There were 158 patients included in the study. Overall median time to sedation was 7.5 min (IQR 4.0–14.0). This did not differ significantly between medical and trauma patient groups (8.0 min [IQR, 4–18 min] vs. 7.0 min [IQR, 4–13 min]). For induction of anaesthesia the main sedative agent was ketamine (n = 120, 75.9%) and paralytic agent rocuronium (n = 151, 95.6%). First attempt success rate was 95.6% (n = 151).

Conclusions

We observed a median time of 7.5 min between paralysis and administration of post‐intubation sedation or analgesia within an Australian ED. Ongoing surveillance with strategies to minimise “sedation gap” is indicated.

Introduction

The most common method of emergency airway management in Australian emergency departments (ED) is Rapid Sequence Intubation (RSI). This involves induction of anaesthesia with an intravenous anaesthetic agent followed by administration of a fast‐acting paralytic agent to facilitate endotracheal intubation whilst reducing the risk of pulmonary aspiration.

The concept of accidental awareness occurs when there is failure to deliver sufficient general anaesthesia or patient resistance to anaesthesia resulting in a degree of consciousness [1]. Awareness within the ED can occur due to insufficient anaesthetic or sedative agent being delivered to a patient being intubated and mechanically ventilated. Induction agents are usually short acting and inadequate or delayed administration of sedation and analgesia can expose patients to physiological and psychological risks [2, 3]. This delay has been referred to as a ‘sedation gap’ [2, 3, 4].

Rates of awareness in the ED population have not been rigorously studied. A prospective observational study of 383 mechanically ventilated patients reported a 2.6% prevalence of awareness [3]. Others found 7.4% of patients who received paralytic agents to facilitate emergent intubation recalled awareness in the ED [5]. Established risk factors for awareness with paralysis within the operative room setting include underdosing or delayed anaesthetic, paralytic agent use with longer half‐life than the sedative, lack of protocolised sedation depth monitoring and high level of consciousness prior to intubation [5, 6].

Awareness in mechanically ventilated patients has been shown to have poor adverse outcomes, including psychological distress and patient discomfort [2, 3]. Single‐centre retrospective cohort studies in emergency departments report a wide range of delays from RSI to administration of analgesia or sedation, with median ‘sedation gap’ reported up to 40 min [4, 7, 8, 9]. The association with ‘sedation gap’ and use of rocuronium for RSI in the ED has yielded conflicting results [8, 10]. Some studies have suggested mean time to sedation or analgesia is longer in patients that receive rocuronium when compared to suxamethonium [9, 11]. These findings highlight the need for ongoing assessment into the practice of prompt sedation and/or analgesia delivery in the setting of RSI.

The aim of this study was to quantify this ‘sedation gap’ in an adult tertiary referral hospital. These findings may provide a benchmark for standard of care and inform future auditing or quality improvement initiatives across other Australian emergency departments. We hypothesised that time to sedation or analgesia will be different among the medical and trauma patient cohorts.

Methods

Setting

The study was conducted at The Alfred Hospital in Melbourne, Australia, an adult tertiary referral hospital. The emergency department (ED) receives over 70,000 patients per year, of which over 1600 are classified as major trauma patients.

Data Sources

The data for the intubated patients in the study period were sourced from The Alfred Airway Registry that recorded information as part of an ongoing clinical improvement and education program. The registry data is submitted by the airway clinician contemporaneously through a secure online form (Jotform). Data collected has previously been described [12] and includes patient demographics and details of both the preparation for, performance of, and complications related to the airway management episode. For each patient, timing data were calculated from medication timestamps recorded in the Electronic Medical Records by a single investigator (CB). In our centre, all invasive airway procedures are performed by resuscitation teams. Scribe nurses record timestamps of medication administration in real time and enter these data into the Electronic Medical Record.

Design

Data from all adult patients (≥ 18 years) included in the Airway registry over a 12‐month period of Jan‐Dec 2024 were extracted. Patients were excluded if administration of a muscle relaxant, sedation, or analgesia was not documented in the electronic medical record. Patients intubated during cardiac arrest were also excluded.

A planned subgroup analysis was undertaken focusing on the indication for intubation, categorised into medical versus trauma. Medical cases included patients intubated for indications such as hypoxemic or hypercapnic respiratory failure and decreased level of consciousness. Trauma cases were intubated for reduced level of consciousness due to head injury, chest trauma, or multisystem trauma and shock.

Co‐variates assessed included clinical and procedural factors. Clinical variables were age, sex, and Glasgow Coma Scale (GCS) prior to intubation. Procedural factors included First Attempt Success, time of day, predicted airway difficulty, use of Emergency Intubation Checklist (see Supporting Information), preoxygenation and apnoeic oxygenation methods, sedative and paralytic agents, laryngoscope type, and first laryngoscopist.

Key time points measured were the administration of paralytic agent, sedation, and analgesia. Due to the negligible difference in time of administration of sedative and analgesic agents being not clinically relevant, we chose to use a single primary outcome measure defined as interval in minutes from administration of the paralytic agent at intubation to administration of the earliest of post intubation sedation or analgesia.

Analysis

Descriptive statistics are reported for baseline characteristics and outcomes. Continuous data were summarised with medians with interquartile ranges (IQR) given the skewed distribution. Categorial variables are presented as frequencies (n) and proportions (%). All analyses were performed using Stata V.17 (Stata Corp., College Station, Texas, USA).

Ethics and Consent

Ethics approval was obtained from the Alfred Hospital Human Research Ethics Committee (Approval number 300/25). The requirement to seek informed consent from patients or person responsible was waived.

Results

There were 183 patients identified and 158 were included in the final analysis. Twenty five patients were excluded from the 21 due to missing medication documentation, 2 were cardiac arrest intubations without administration of a paralytic agent and 2 in which Alfred Airway Registry data was unable to be linked to Electronic Medical Records. Table 1 presents baseline demographic and procedural characteristics of patients undergoing rapid sequence intubation (RSI) and Table 2 is stratified into medical and trauma indications. Almost two thirds of patients were male (n = 102, 64.4%) and the most common indication for intubation was medical (n = 86, 54.4%). Intubation First Attempt Success rate recorded in 151 cases (95.6%), consistent with prior reported studies at this centre [12, 13]. An intubation preparation checklist covering team roles, patient preparation (monitoring, access, positioning, preoxygenation), drug preparation (RSI and rescue drugs, sedation plan), and equipment readiness (airway adjuncts, laryngoscopes, tubes, suction, surgical airway kit) was used 86.1% of the time. The main induction agents were ketamine as the sedative (n = 120, 75.9%) and rocuronium as the paralytic agent (n = 151, 95.6%). For the primary outcome, the overall median time to sedation was 7.5 min (interquartile range 4.0 to 14.0). Median time to sedation between medical and trauma subgroups was 8.0 min (IQR, 4–18 min) and 7.0 min (IQR, 4–13 min), respectively.

Trauma patients were more frequently male (78% vs. 53%) and more frequently were GCS 14 or greater on presentation (36% vs. 9%) than medical patients. A predicted difficult airway was more prevalent in trauma than medical cases (42% vs. 17%). Laryngoscope selection differed between groups, with video laryngoscopy (CMAC or HAVL) used more commonly in trauma cases. Bougie was used to facilitate intubation in most cases across both groups (94% vs. 85%). Medical RSIs were predominantly performed by emergency registrars (n = 59, 69%), whereas trauma RSIs more frequently involved anaesthetic registrars or consultants (n = 24, 33% combined). Other variables including age, time of day, checklist use, preoxygenation and apnoeic oxygenation method, sedative and paralytic agent were similar between groups.

Table 3 presents the mean and median induction dosages of sedation and paralytic agents used in this study. The mean dose of sedative given in the trauma subgroup was marginally greater than that of medical patients. Mean dose of ketamine in trauma and medical subgroups was 95.6 (SD 48.6) vs. 88.8 mg (SD 40.7), respectively. Mean dose of propofol in trauma and medical subgroups was 107.3 (SD 47.6) vs. 93.8 mg (SD 39.8), respectively. Dose of the predominant paralytic agent rocuronium was very similar between trauma and medical patients (117.3 [SD 29.0] vs. 117.2 mg [SD 28.6], respectively).

Figure 1 presents Kaplan–Meier estimates of time to sedation or analgesia following paralysis administration, stratified by indication for intubation (medical vs. trauma). Both groups demonstrated a rapid increase in the cumulative probability of sedation within the first 10 min.

Discussion

Currently there are no Australian guidelines focused on timely post intubation sedation or analgesia. Given contemporary literature demonstrating long times to administration of sedation following intubation in the ED, the aim of this study was to assess our own performance and identify opportunities for improvement.

Our primary outcome demonstrated the median time from paralysis administration to initiation of analgesia or sedation was 7.5 min. Intubation First Attempt Success was recorded in 151 cases (95.6%). These outcomes may be attributable to the overall high adherence (n = 136, 86%) to an intubation preparation checklist incorporating a post‐intubation sedation plan (see Supporting Information). This checklist may serve as a cognitive aid, promoting consistency in post intubation care, particularly for unconscious patients. For example, the majority of medical patients had a GCS < 8 (n = 57, 66%) and the use of a checklist may help reduce clinician‐to‐clinician variability in initiating sedation in this group.

Improvements in ‘sedation gap’ associated with the use of paralytic agent may be attributable to protocol‐based approaches, pharmacy and nursing advocacy, education intervention and regular audit [7, 9, 14]. This centre continues to report a high First Attempt Success rate (n = 151, 96%) [12, 13, 15] which may reflect the rigorous quality improvement and governance processes undertaken at this site. This culture may in part explain the minimised ‘sedation gap’ in this ED.

The association between rocuronium use and the occurrence of a post‐intubation ‘sedation gap’ in the emergency setting remains uncertain, with existing studies reporting conflicting results [8, 10, 11, 16]. Our ‘sedation gap’, occurred in a setting where rocuronium was the predominant paralytic agent (n = 151, 95.6%).

The UK‐Ireland national audit NAP5 identified induction and the early maintenance phase as the highest‐risk periods for accidental awareness during general anaesthesia, with risk further increased when neuromuscular blockade is used [1]. The ED “sedation gap” period after RSI is a comparable vulnerable window. Structured intubation processes, including checklist use (see Supporting Information), may be attributable to the ‘sedation gap’ observed in our study compared with delays spanning 8.5–47 min in previous literature [4, 7, 8, 9]. Further research is needed to measure accidental awareness as an outcome and whether it is associated with this ‘sedation gap’.

Limitations

The single centre nature of this study restricts the generalisability to other Emergency Departments across Australia. The retrospective nature of this study may also impact data accuracy. For example, medication time stamps are often retrospectively entered into the Electronic Medical Record and may not accurately reflect exact times of medication administration. Our primary outcome of time to sedation does not take into account the adequacy of the dose of sedation or an assessment of the depth of sedation achieved. Given the self‐reported nature of the airway registry, there is an inherent risk of underreporting, particularly for failed intubation attempts or challenging intubations.

Conclusion

A median time of 7.5 min between paralysis and administration of post‐intubation sedation or analgesia was observed in this single centre sample from an adult tertiary referral in the ED. This was despite the predominant use of long acting paralytic at this site. Ongoing quality improvement processes such as an intubation preparation checklist and regular audit‐based education may have contributed to this outcome and require ongoing surveillance and implementation. Further prospective study is required to determine the rate of awareness within the ED population and if the current ‘sedation gap’ results in optimal patient outcomes.

Ethics Statement

Ethics approval was obtained from the Alfred Hospital Human Research Ethics Committee (Approval number 300/25).

Conflicts of Interest

B.M. is a Section Editor for Emergency Medicine Australasia.

Supporting information