Introduction

Emergent endotracheal intubation is a core skill for Emergency Medicine (EM) physicians, is performed frequently in Emergency Departments (EDs) globally, and is a complex and challenging task in the care of critically ill patients [1]. Complications include failed first attempts, the need to perform a surgical airway, and other life-threatening adverse events such as hypotension, hypoxemia, aspiration, dysrhythmias, and cardiac arrest [2,3,4,5,6,7]. Many EDs have implemented airway registries to capture endotracheal intubation practices and monitor the frequency of adverse events, with the goal of optimizing patient outcomes [1, 2, 8]. Airway registries have been implemented in many regions including Australia and New Zealand [9,10,11], North America [12,13,14,15,16,17,18], Europe [19, 20], Asia [21,22,23,24], and Africa [25].

Airway registries offer significant potential utility as sources of data for quality assurance (QA), quality improvement (QI), and clinical research [2, 23, 25,26,27]. Despite the growing body of research involving airway registries, there is no consensus of their methodology, how airway registry data is utilized, or their definitions of key performance indicators of intubation. One review exploring this topic identified this problem but did not assess the full scope of airway registries as critical grey literature sources were not included [28].

We performed a scoping review to answer three important questions regarding airway registries. The primary objective is to describe the current scope and prevalence of ED airway registries globally. Secondary objectives include describing utilization of airway registry data and determining how various airway registries define adverse events and key performance indicators during the peri-intubation period.

Methods

A scoping review was completed using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) framework [29].

Eligibility criteria

We included all published studies (full-text publications, conference abstracts, conference posters) which reported the implementation of an airway registry to monitor intubation practices among adult patients within an ED. We defined an airway registry as a formal process implemented within the ED to monitor intubation practices by recording information pertinent to intubation and adverse events independent from a finite intervention or research hypothesis. Non-English publications, studies that reported greater than 50% of their intubations were performed in paediatric patients, studies that did not utilize an airway registry for data collection, or those that included intubations performed primarily outside of the ED (intensive care unit [ICU], operating theatre, pre-hospital, in-patient wards) were excluded. No date limitations were applied.

Information sources and search

A search conducted in Ovid Medline (1946–2021), Embase (1947–2021), Scopus (2021), Cochrane Libraries (2021), Web of Science (2021), and Google Scholar (2021) was initially completed on November 3, 2021. A systematic search strategy was developed and translated into each of the databases’ syntax. The full search strategy for all databases is shown in Additional file 1. The intent of this search was to identify all peer-reviewed and grey literature pertaining to ED airway registries. The search of all databases followed by study screening was repeated on July 26, 2022, with a date limitation applied (November 3, 2021–July 26, 2022).

Study selection

Studies identified were imported to Covidence, an online platform for screening and organizing data for literature reviews, and automatically de-duplicated. Any subsequent duplicates noted were manually excluded by authors collaboratively throughout the review process. An initial review of abstracts was completed using the aforementioned eligibility criteria. Full texts (if available) were reviewed for eligibility by two reviewers, independently. Discrepancies were resolved by a third team member.

Data charting

Data was charted by two independent reviewers (SM and HJ) using a standardized data charting form created in Microsoft Excel (2018) [30] specifically for this review. The data charting form was piloted for usability by two reviewers for 5 studies each prior to data charting. Charted study information was informed by the Arksey and O’Malley Framework for Scoping Reviews [31]. Additional Files 2 through 5 depict the totality of all data charted for each included publication.

Synthesis of results

The total number of patients for each registry was reported as the largest number of patients available in identified publications within each registry. Years active was determined based on time periods of data collection for all included studies in each registry. To be as representative of current trends as possible, data was reported from the most recent, full-text publication (if available), within each registry. If the most recent publication was missing any of this information, it was reported from the next most recent publication, and so on.

To examine the utilization of airway registry data, studies were grouped into research or QA/QI. For the purposes of this review, QA was defined as studies that measure compliance against certain required standards [32]. QI was defined as studies implementing a proactive approach to improve specific processes or systems [33]. Utilization of airway registry data from all publications was synthesized.

Results

The combined searches yielded 2349 results, including 1084 duplicates. Following title/abstract and full-text screening, 124 studies were included in our review consisting of 29 conference abstracts, one conference poster, and 94 full-text, peer-reviewed publications, as illustrated in Fig. 1. Included studies were published between the years of 1999 and 2022.

Fig. 1
figure 1

PRISMA Flow diagram for study inclusion

Full size image

Characteristics of identified airway registries

A summary of characteristics of identified airway registries is provided in Table 1. A complete list of the characteristics for each included publication can be found in Additional File 2. A total of 22 airway registries were identified globally with seven registries in North America, seven registries in Asia, four registries in Australia and New Zealand, three registries in Europe, and one registry in Africa. Of the registries identified, nine of 22 have only a single full-text publication, conference abstract, or poster available. Nine of 22 registries provide data from multiple centres ranging from three centres in the British Columbia Airway Registry for Emergencies (BCARE) to 43 centres reported in the Australian and New Zealand Emergency Department Airway Registry (ANZEDAR) [1, 34]. The largest and longest running dataset is the National Emergency Airway Registry (NEAR) which exists as three iterations reporting on intubation practices across four countries (USA, Canada, Singapore, and Australia) from 1996 to 2018. NEAR includes a total of 29, 296 ED intubations between NEARI, NEARII, and NEARIII reporting 1288, 8937, and 19,071 intubations, respectively [12,13,14].

Table 1 Characteristics of airway registries
Full size table

Seventeen of the 22 registries identified began their data collection in 2005 or later. The earliest time periods of data collection are seen in NEARI, II, and III which began in 1996, 1997, and 2002 respectively [12, 35, 36], Emergency Department Intubation Registry (EDIR) which began in 1999 [37], and the Singapore General Hospital Emergency Airway Registry which began in 2000 [38]. Only two registries included data on intubations that took place outside of the ED setting. The Cleveland Clinic Emergency Airway Registry includes 30% prehospital intubations and BCARE includes 30% ICU intubations and 16.3% ward intubations [34, 39]. Six registries also specify the exact percentage of their data that reflects paediatric intubations, ranging from 0.2% of patients aged 13 or less in EDIR, to 13.9% patients aged zero to 18 in NEARI [12, 19]. This reflects variable definitions of the age which constitutes a paediatric patient across the identified airway registries depending on the region or country of origin.

Types of publications and utilization of airway registry data

The utilization of airway registry data varied; eight of 22 registries were used in QA studies, eight of 22 for QI studies, and 21 of 22 for research. Note that one publication was classified as both QI and research [10]. Table 2 provides a summary of QA and QI publications identified. Five of 10 QA studies focused on auditing the practices and outcomes of their single local centre by comparing its performance to benchmark data from other centres within the same registry, or from a larger, more established registry. Of the QI publications identified, all but three focused on dedicated QI programs implemented at a local centre. The remaining three studies assessed the impact of procedural changes in response to disease outbreaks (COVID-19 and SARS [10, 38]) or conducted a cross-sectional evaluation of simulation-based training programs between centres [40]. A list of all included QA and QI publications is provided in Additional File 3.

Table 2 Utilization of airway registry data—QI/QA studies
Full size table

The most common areas of focus amongst the research-based publications include 34 studies comparing between laryngoscope types, 15 regarding evaluations of medication choice, dosing, and outcomes, and 12 reporting summaries of captured airway management statistics as seen in Table 3. Of the 105 research publications identified, 73 were retrospective analyses and 29 were prospective observational studies. One study used a combination of retrospective analyses and qualitative surveys [41], while two abstracts did not specify their study methodology [34, 42]. A complete list of research publications included can be found in Additional file 4.

Table 3 Utilization of airway registry data—Research studies
Full size table

Intubation practices and adverse events of identified airway registries

Intubation practices of various airway registries are summarized in Table 4. A complete list of all publications’ intubation practices and adverse events can be found in Additional File 5. EM physicians are reported as the most common intubator in 19 of 22 airway registries. Of these registries, 12 reported that the most common intubator is EM residents. One registry, the Cipto Mangunkusumo General Hospital Airway Registry, reported Anesthetists as the most common intubator.

Table 4 Intubation practices in identified airway registries
Full size table

Fourteen registries reported the most common indication for intubation as medical conditions including intracranial hemorrhage/stroke and airway protection. Only one registry reported trauma as the most common indication, Defense Registry for Emergency Airway Management (DREAM), a registry at Brooke Army Medical Centre, which is a level 1 trauma centre.

First pass success (FPS) rates were reported by 20 of 22 airway registries identified, with the exception of NEARI and the Middlemore Hospital Airway Registry. FPS ranged from 74% in the Japanese Emergency Airway Network I and II (JEANI + II) and the Cleveland Clinic Emergency Airway Registry to 93.9% in The Royal North Shore Emergency Airway Registry.

Seventeen of 22 registries reported the most used intubation device, either the video laryngoscope (VL) or direct laryngoscope (DL) in all of these registries. Twelve of these registries reported the VL as the most used with rates ranging from 52% in the South African ED Registry to 94% in the Continuous QI Database.

Medication use for intubation and rates of rapid sequence intubation (RSI) varied widely across the identified registries. Among the 13 registries that reported rates of RSI, rates ranged from 9% in the Cipto Mangunkusumo General Hospital Airway Registry to 85.2% in the Continuous QI Database. The most common induction agent was etomidate (7 of 22 registries). Ketamine, Propofol, Midazolam, Fentanyl, and Thiopental were each reported as the most common induction medication in one registry. Ten registries did not report their most common induction agent. Paralytic agent use was reported in 11 of 22 registries, with five of these reporting Rocuronium and six reporting Succinylcholine as the most used paralytic.

Total rates of adverse events were reported by 17 of 22 airway registries. These rates ranged from 6.5% in the Korean Emergency Airway Management Registry (KEAMR) to 33% in the South African ED Registry. Seventeen of 22 identified registries reported the most common adverse events associated with intubation. The most common adverse events were hypoxia in eight of 17 registries (7.5% to 16.2% of patients), hypotension in four of 17 registries (4.1% to 10% of patients), and esophageal intubation in five of these 17 registries (2.9 to 6% of patients).

Definitions in identified airway registries

The identified airway registries provided various definitions of an intubation attempt, FPS, and adverse events, including hypoxia and hypotension. Table 5 provides a summary of these definitions in each registry, if available. Definitions provided in each included publication can be found in Additional File 5. Six of 22 registries defined an attempt at intubation as “a single passage of the laryngoscope blade into the mouth” while five registries made this definition more specific defining passage of the laryngoscope blade or endotracheal tube past various anatomical structures including past the lips (Royal North Shore), the teeth (JEANI + II and KEAMR), the vocal cords (Singapore General Hospital Emergency Airway Registry) or the alveolar ridge (NEARIII) as an intubation attempt. Eight registries did not define an intubation attempt.

Table 5 Definitions in identified airway registries
Full size table

FPS was most often defined as “successful placement of the endotracheal tube on the first laryngoscope insertion” in seven of 22 registries, while three defined FPS as “correct placement through the vocal cords on the first attempt” (National Emergency Resuscitation Airway Audit [NERAA], NEARII, KEAMR), and two stipulated that placement had to be confirmed by end-tidal capnography to be deemed successful (JEANI + II and Royal North Shore Emergency Airway Registry). Ten registries did not define FPS.

Thirteen of 22 registries defined specific parameters for hypoxia to be reported as an adverse event. Six registries reported hypoxia as peripheral oxygen saturation < 90%, six reported < 93%, and one reported < 80% within 30 min of intubation. NEARIII also specified that hypoxia would be reported as an adverse event if there was a desaturation of > 10% of absolute oxygen saturation.

Seven registries provided specific parameters to define hypotension as an adverse event. Hypotension was most often defined as a systolic blood pressure (SBP) of < 90 mmHg in six of these seven registries, whereas NEARIII defined hypotension as an SBP of < 100 mmHg. Among the registries that did not report a specific parameter, hypotension was defined as requiring treatment with either intravenous fluids or vasopressors in five registries. Both the South African ED Registry and the Samsung Medical Centre Emergency Airway Program also reported hypotension as an adverse event if a 20% change from the patients’ baseline SBP was recorded. The Samsung Medical Centre Emergency Airway Program included a third parameter to measure hypotension as a mean arterial pressure of < 65 mmHg. Ten registries did not define hypotension.

Discussion

Of the identified registries, commonalities of reported information include patient demographics (most often age, sex, and weight), indication for intubation, method of intubation and adjuncts used (stylet, bougie), device used for each attempt, intubator level of training and specialty, number of attempts and the outcome of each attempt, presence of any difficult airway characteristics and predicted difficulty of intubation, induction and/or paralytic medications used, patients’ pre- and post-intubation vital signs, and complications or adverse events. Other commonly reported information included patient positioning, intubation maneuvers or rescue techniques, and the use of a pre-RSI checklist. Commonly reported adverse events included hypoxia, hypotension, bradycardia, esophageal intubation, failed airway requiring surgical airway, dysrhythmia or cardiac arrest, vomiting or aspiration, endobronchial intubation, dental or airway trauma, and laryngospasm.

There is a suggestion that the context of the registry is also important when comparing intubation performance. For example, the Cleveland Clinic Emergency Airway Registry also included pre-hospital intubations that comprised 30% of their data in one study [39]. Interestingly, this registry also reported the lowest FPS of North American airway registries, possibly due to the inclusion of pre-hospital intubations, which may be more difficult than ED intubations due to the inability to obtain assistance, lack of proper equipment, variability of intubator experience and training, and difficult intubation environments [43]. Additionally, 11 of 22 identified registries include intubations performed on patients of all ages, which complicates comparisons of intubation performance between institutions because of the added complexity of managing paediatric airways due to their smaller size, and anatomical and physiological differences from adult airways [44,45,46]. These inconsistencies of information reported, location of intubation, and patient population among airway registries may limit comparisons of intubation practices.

Airway registries are a critical QA tool that allow some degree of comparison of adverse events rates and key performance indicators of intubation practice between centres, both nationally and internationally. For example, Powell et al., 2018 utilized data from ANZEDAR as a benchmark to monitor intubation practices in a small, rural ED in New Zealand [47]. Using the data pooled from 43 centres in ANZEDAR, this study demonstrated that FPS rates and complication rates were similar in this rural ED to data collected from across Australia and New Zealand [47]. On a larger scale, Park et al., 2017 utilized real-time clinical data collected from airway registries internationally to determine an international benchmark for FPS of 84.1%, and provided rates of the most common peri-intubation adverse events [48].

Several studies identified in this review demonstrate that airway registry data play a crucial role in identifying deficiencies in ED intubation practices, developing targeted solutions, and monitoring the efficacy of multi-faceted QI programs to improve FPS while minimizing rates of peri-intubation adverse events [2, 11, 13, 39, 40, 49,50,51,52,53,54,55]. For example, Hwang et al., 2018 standardized intubation practices by introducing an ED intubation protocol checklist that directed staff to optimize pre-oxygenation, utilize VL, use RSI as a standard method for intubation, and limit intubation attempts to two by a single provider or with the same device [50]. This centre also provided evidence-based lectures on airway management to EM physicians and nurses, as well as hands-on skills training with various airway devices in a simulated setting [50]. After implementation of these changes, airway registry data collected in the subsequent three-year study period demonstrated that FPS increased by 11%, and overall peri-intubation complication rates decreased by 8% [50]. Tracking the efficacy of these QI interventions using airway registry data is crucial as it is well-established that multiple intubation attempts lead to higher rates of life-threatening peri-intubation complications, an increased risk of intubation failure on subsequent attempts, and lower probability of return of spontaneous circulation [3, 21, 22, 56,57,58,59,60].

The airway registries reviewed provide insight into historical practices and current trends. Airway registries identified in this review revealed a trend towards higher rates of intubations performed by RSI in America [13, 61,62,63], Europe [19, 64], and Australia and New Zealand [1, 47]. While rates of RSI are increasing in more recent publications, rates of RSI remain low in Japan [65,66,67] and Indonesia [24]. Likewise, DL has historically been the standard of care for performing ED intubations; however, trends in the identified airway registries demonstrate the increasing prevalence of VL as this intubation technique allows for higher rates of FPS with lower rates of adverse events [2, 8, 13, 27, 61, 68,69,70,71,72,73]. Trends in medication selection are also described in airway registry data, and this review demonstrated a trend towards increasing use of etomidate, despite the persistence of geographic variation in medication selection. Airway registries can be a valuable tool to increase awareness of how local practices compare to global trends in clinical care and their impact, such as standardizing the use of RSI for intubation, to provide better patient outcomes [4, 8, 52, 71, 76,77,78].

FPS is often used as a marker of intubation proficiency because multiple intubation attempts are associated with increased rates of adverse events [3, 21, 56,57,58,59,60]. Several studies included in our review reported the definition of an intubation attempt and FPS utilized at their institution; however, these definitions had some heterogeneity. The most recent systematic review performed to determine an international benchmark of FPS recommended defining FPS as “the proportion of endotracheal tubes placed successfully after the first attempt” and defining intubation attempt as “any single insertion of the laryngoscope into the mouth” [48]. This aligns with the most commonly reported definitions of FPS and intubation attempt identified in this review; however, there is still heterogeneity and a lack of transparent reporting of the definitions used by many airway registries identified which limits international comparison. To allow for the creation of reliable international benchmarks and equivalent comparisons of airway management performance between centres, we recommend that these definitions are used at all institutions implementing an airway registry.

Similarly, several identified airway registries did not provide definitions of common adverse events such as hypoxia and hypotension [12, 13, 15, 23, 34, 39, 49, 55]. Among those that provided definitions of these adverse events, there was a great deal of variability. A lack of agreed upon definitions of peri-intubation adverse events may explain the variability in reported overall adverse event rates among studies identified in this review, ranging from 6.5 to 33% [25, 79]. While there is little consensus of adverse event definitions that should be used in ED airway registries, guidelines do exist for these definitions in pre-hospital airway management [80]. Intubation attempts at a peripheral oxygen saturation of 93% or less have a much higher risk of progressing to critical oxygen desaturations [81]. This is in-keeping with the rationale that ANZEDAR provided for reporting hypoxia as an oxygen saturation of less than 93%, as they preferred this higher cut-off to provide a margin of safety before critical desaturation occurs [11]. Additionally, the most commonly used value of hypotension identified in this review was a SBP of less than 90 mmHg, in-keeping with the value recognized in some guidelines provided in pre-hospital airway management that report hypotension as a SBP less than 90 mmHg or a decrease in SBP of greater than 10% from the patient’s baseline value [82]. Based on the available literature, airway registries should consider reporting peri-intubation adverse events as hypoxia if peripheral oxygen saturation is 93% or less and hypotension if SBP falls below 90 mmHg or greater than 10% from the patient’s baseline measurement.

Of note, few publications provided meaningful information regarding the logistics of the airway registry use, sources of funding, methods of data collection or dissemination of information which would be valuable for consideration of how to optimally implement an ED airway registry.

Limitations

This study reports only published data from airway registries or registry-like reporting procedures. We suspect that data from smaller or non-academic institutions, or those who did not provide evidence of an ongoing airway registry in their publication were likely not represented in this review which likely amounts to a degree of selection bias. Additionally, non-English publications were excluded from this review leading to an English-language bias.

Several of the identified registries include only a single publication or publications that are now dated. This may not be representative of the current state of intubation practices or the ongoing function of the airway registry within those institutions. The wide range of publication dates also means that data among various airway registries may no longer be representative of that airway registry. We have also included abstract only publications in this review to be as representative as possible of global airway registries; however, these publications do provide limited information when compared to full-text publications.

Lastly, we attempted to report the most recent and representative trends within each included airway registry. Data was reported from the most recent available publication. If the most recent publication was missing any of this data, it was reported from the next most recent publication, and so on. This method often led to intubation practices being reported from a single study within a registry and may not be representative of the nuances of emergency airway management among included institutions.

Conclusions

This scoping review identified 22 airway registries globally that monitor ED intubation. Airway registries appear to be a crucial tool to improve intubation procedures and patient outcomes by contributing meaningful longitudinal data to guide local intubation practices, ensure quality of care, and serve as a platform to innovate and test clinically relevant questions, while providing important information regarding current intubation procedure trends and guidance for quality improvement initiatives. However, standardized definitions and transparent reporting of adverse events and other key performance indicators, such as FPS, would allow airway management performance to be compared on a more equivalent basis and allow for the determination of more reliable international benchmarks. With some consensus among pre-hospital airway registries and identified ED airway registries, it would be reasonable to adopt reporting hypoxia as a peripheral oxygen saturation of 93% or less and hypotension as SBP of less than 90 mmHg or a decrease of greater than 10% from the patient’s baseline. Additionally, FPS should be defined as “the proportion of endotracheal tubes placed successfully after the first attempt” and intubation attempt as “any single insertion of the laryngoscope into the mouth.” Given the importance of airway registry data, the creation and implementation of an airway registry should be considered by any centre looking to investigate and improve their emergency airway practices.