Prehospital Emergency Anaesthesia vs ED RSI

Background

Airway management in major trauma patients is a cornerstone of effective pre-hospital and in-hospital care. In the UK there is a national target that (where indicated) serious head injuries should receive prehospital emergency anaesthesia (PHEA) within 45 minutes of injury. That’s a really challenging target to hit and is a focus of PHEM services across the country. However, evidence comparing PHEA with emergency department RSI (EDRSI) remains limited. I am also frequently challenged about prehospital procedures by colleagues who may not understand what our capabilities are.

A couple of years ago, I attended a multiple stabbing incident where the patient was intubated at the scene, and a thoracotomy was performed. In the hospital debrief, concerns were raised about why we had not simply taken the patient to hospital for the surgical teams. These are usually easy questions to answer for issues such as thoracotomy, but for potentially less urgent interventions such as PHEA I still hear concerns. PHEA is at first glance more risky in the prehospital setting that in the emergency department. There is less equipment, more difficult environmental factors, fewer staff and no back up. So for many it appears more risky, although my opinion is it’s probably safer as a result of training, exposure, familiarity and good SOPs.

Doubts remain, though, and there is relatively little evidence out there. Does prehospital emergency anaesthesia really improve the times to intubation/ventilation or not? This week we have a paper from the UK that addresses just this question. The abstract is below, but as always please read the full paper and decide for yourself.

The Abstract

Background
Early rapid sequence induction of anaesthesia (RSI) and tracheal intubation for patients with airway or ventilatory compromise following major trauma is recommended, with guidance suggesting a 45-min timeframe. Whilst on-scene RSI is recommended, the potential time benefit offered by Helicopter Emergency Medical Services (HEMS) has not been studied. We compared the time from 999/112 emergency call to delivery of RSI between patients intubated either in the Emergency Department or pre-hospital by HEMS.
Methods
A retrospective observational cohort study of major trauma patients in South-East England who received a pre-hospital RSI (PHRSI) or Emergency Department RSI (EDRSI) between 2 January 2018 and 24 September 2019. Data were extracted from the UK Trauma Audit and Research Network database. The primary outcome was the time from emergency call to delivery of RSI. Secondary outcomes included mortality at 30-days or hospital discharge, time from arrival of service at hospital or scene to RSI, time from emergency call to Computerised Tomography scan, and conveyance interval. Linear regression was used to model time to RSI in both groups.
Results
Of 378 eligible patients, 209 patients met inclusion criteria. 103 received a PHRSI and 106 received an EDRSI. Most patients were male (n = 171, 82%) and the median age was 48 years (IQR 28-65). 94% sustained a blunt injury mechanism and head was the most injured body region for both cohorts (n = 134, 64% ) . 63% (n=67) of patients receiving a PHRSI were conveyed by helicopter. PHRSI was delivered significantly earlier with a median of 64 [IQR 51-75] minutes (95% CI, 60-68) compared with EDRSI with a median of 84 [IQR 68-113] minutes (95% CI, 76-94), p<0.001).
Conclusion
Major trauma patients who had a pre-hospital RSI received this time-critical intervention sooner after their injury than those who received an emergency anaesthetic after conveyance to a specialist hospital. Patient outcome benefit of HEMS delivered early RSI should be explored.
Heritage, D., Griggs, J., Barrett, J. et al. Helicopter emergency medical services demonstrate reduced time to emergency anaesthesia in an undifferentiated trauma population: a retrospective observational analysis across three major trauma networks. Scand J Trauma Resusc Emerg Med 32, 138 (2024).

This study set out to compare the time from emergency call to RSI delivery between PHEA and EDRSI, and to evaluate whether earlier RSI impacts patient outcomes.

What kind of study is this?

This was a retrospective observational cohort study conducted across three major trauma networks in Southeast England. Data were extracted from the UK Trauma Audit and Research Network (TARN) database for adult major trauma patients who received either prehospital emergency anaesthesia or ED RSI between January 2018 and September 2019.

The study analysed data from patients conveyed to three major trauma centres by either HEMS or ground ambulance services. Statistical modelling, including linear regression, was used to assess the primary and secondary outcomes.

Tell me about the patients

The study included 209 patients, with 103 receiving prehospital emergency anaesthesia and 106 receiving EDRSI. Most patients were male (82%), with a median age of 48 years. Blunt trauma accounted for 94% of cases, with head injuries being the most common (64%).

In the PHEA group, 65% were transported by helicopter, whereas all EDRSI patients arrived via ground ambulance. Injury Severity Scores (ISS) were higher in the PHEA group, reflecting a greater burden of injury.

What were the measured outcomes in this study?

Primary Outcome: Time from emergency call to RSI.
Secondary Outcomes:
- Time from arrival of service at the scene or ED to RSI.
- Total pre-hospital time.
- Time from emergency call to CT scan.
- Mortality at 30 days or hospital discharge.

Additional analyses examined predictors of time to RSI, including age, Glasgow Coma Score (GCS), and ISS.

What are the main results?

PHEA was delivered significantly earlier than EDRSI (median 64 minutes vs. 84 minutes from emergency call).
Time from arrival of an RSI-capable team to intubation was shorter in the ED group (18 minutes) compared to PHEA (25 minutes).
Total pre-hospital time was significantly longer in the PHEA group (115 minutes) than in the EDRSI group (64 minutes).
Mortality rates and hospital lengths of stay did not differ significantly between the two groups (63 and 65% indicating a very seriously injured cohort).
Average ISS was similar, but there were more patients with high ISS in the PHEA group, indicating greater injury severity.
Time from injury to CT scan: This was significantly less in the PHEA group 107 vs. 132 mins. For me this is arguably the most telling of the secondary outcomes as CT is a key decision point for the vast majority of our major trauma patients. Definitive care often follows the CT scan so an overall reduction of 25 mins feels clinically important to me.

How robust are the findings?

At first glance this appears to support the use of prehospital teams in PHEA, but as always there are caveats when we look at this. Although overall times were different we have to remember that as a retrospective observational cohort study the patient groups are likely to be different. We can see this as there are some obvious differences between them in the data.

Selection Bias: Patients in the PHEA group had higher ISS, indicating a potentially sicker cohort.
Data Accuracy: Retrospective studies are vulnerable to incomplete or imprecise data collection.
Unmeasured Confounders: Factors such as specific injury patterns, physiological parameters, or scene complexities were not fully accounted for.
Geographical Variability: Differences in rural vs. urban environments may have influenced response times, and be associated with different injury patterns.
Age: The data in the study is now roughly 6 years old. Many things may have changed since then and some more contemporaneous data would be useful.

So it’s not really a like vs. like study which we could only really achieve if we conducted an RCT which would not really happen in the modern era. That said the use of the TARN database for follow up and the linkage to the HEMS service data is a positive, although insufficient data was available for 68 patients which is a sizeable proportion (18%) of their initial cohort. Ideally we would want better data collection from this group of severely injured patients.

We also don’t really have any patient outcome data beyond mortality at 30 days to really determine whether the modest difference in times (the median of which is still outside of the 45 min national target) makes a difference to patients in terms of longer term survival or neurological outcome. Even before that we might be able to look at peri-intubation complications such as hypotension and hypoxia to get a better feel for whether there are similar or different rates around safety and stability.

Should we change practice based on this study?

This study supports the hypothesis that PHEA reduces the time to airway management compared to EDRSI. However, whether this time-saving translates into meaningful improvements in patient outcomes remains uncertain. The absence of a mortality difference suggests that factors beyond time to RSI—such as pre-hospital care quality and transport logistics—are also crucial.

PHEA appears to benefit patients who are far from major trauma centres, particularly those with severe head injuries. Further research, including prospective studies, is needed to clarify the clinical impact of earlier RSI delivery.

Summary

This retrospective observational study demonstrates that PHEA reduces the time to RSI in major trauma patients compared to EDRSI. While time savings are clear, no significant difference in mortality was observed. The findings highlight the operational value of HEMS in delivering RSI to severely injured patients, especially in geographically challenging areas.

Future research should focus on whether earlier RSI improves neurological outcomes, and on identifying the patients most likely to benefit from PHEA.

References and further reading

Heritage, D., Griggs, J., Barrett, J. et al. Helicopter emergency medical services demonstrate reduced time to emergency anaesthesia in an undifferentiated trauma population: a retrospective observational analysis across three major trauma networks. Scand J Trauma Resusc Emerg Med 32, 138 (2024). https://doi.org/10.1186/s13049-024-01313-y