Send letter to journal:
Re: Author's reply

Dear Editor,

We would like to thank the respondents for expressing their strong views regarding our review. Many thanks to Mason for highlighting the core issue at stake and to Dawes for informing us about the proceedings of the EMS physicians meeting in 2004.We made a sincere endeavour to present the current evidence in the few epidemiological studies which attempted to investigate use of pre-hospital airway management in trauma patients. As Doyle correctly mentions, it is inappropriate to derive ‘causality’ from studies which are of different designs and possess heterogeneity in their sample populations. And the review highlights the hypotheses which could have led to a perceptible increase in mortality in patients who underwent protection of airway by endotracheal intubation. An ‘association’ has been observed in the non-randomised trials and it may be worthwhile to test the association in prospective randomized controlled trials which may, however, be tad bit difficult. Saying that, we know of an Australian team which is investigating the above in an RCT and the results would surely be interesting.

‘Bets’, as we understand, profess to state what the evidence is. How that evidence is then used in practice is influenced by other factors, namely the cost, the experience of the clinician and the characteristics of the particular patient. To follow it blindly would be a recipe for danger and promote ‘cook-book medicine’ which evidence-based medicine is surely not. The studies under consideration were ones where anaesthetic drugs were used. We also attempted to use generic inverse variance to analyse the odds ratio of mortality in those studies where it was reported. Generic inverse variance is a statistical meta-analysis technique which can be used in non-randomised studies of different study designs as elucidated by the Cochrane Non-randomised methods group. The graphical representation is shown below.

However, due to the heterogeneity in studies and due to clinical considerations, our comments have elucidated the need for better trials and merely projected the association of increased mortality.

French et al question our understanding of pre-hospital critical care and need for airway management. We would certainly defer to their more informed experience but wish to add that we never intended to question the benefits of definitive airway management in pre-hospital care. That is an aspect, we believe, is well known to an expert audience. The question which all the studies have attempted to answer is in a ‘time-intensive, resource-poor’ setting with varying levels of airway skills, would it be better to attempt intubation or think of adjuvant means of airway protection like LMA which is a faster and quicker way of airway protection. We never suggested ‘bagging’ patients for long distance transfers. As we all realise, our first dictum is ‘primum non nocere’ (first do no harm). Would attempting intubation lead to increased transit times or worse increased risk of misplacement considering the absence of a good capnograph/meter in most paramedic ambulances? An ‘association’ however flawed needs to be tested and this is where research begins. Corticosteroids use in head injury (CRASH trial) is a notable example of this. As also, when John Snow ascribed the outbreak of cholera to the white flocculent particles in the water from the pump on Broad Street in London! Suffice to conclude that what is considered evidence today may not be evidence tomorrow. But we are in a state of flux and evolution as we strive to better our nebulous understanding of scientific research and methodology.

French et al have stated that ‘good doctors use clinical expertise with evidence.’ Our review never doubted the importance of clinical judgment. As someone said, ‘the stage is surely set for developing an educational method that draws productively on both traditions of evidence- based medicine and old-fashioned clinical intuition. Far from being a contradiction in terms, it is surely the epistemological marriage we have all been waiting for.’

Ayan Sen

Send letter to journal:
Re: Two wrongs don’t make a right

Dear Editors,

To “shoot the messenger” is to reply to an argument by attacking the person presenting the argument rather than the argument itself. It is a time-honoured way of dealing with unpleasant messages. The underlying sentiment is perhaps best expressed by Sophocles: “How dreadful knowledge of the truth can be when there is no help in the truth” (1).

Dr Mason suggests that the criticism of Sen and Nichani’s BET is motivated by a dislike of the message itself rather than a deep and genuine concern about the strength and reliability of the underpinning evidence (2). As a co-author of the letter by French et al.(3), I would like to emphasise that the criticism of the BET was, as is reflected in the title, on methodological and clinical grounds. It was not a personal reaction to the ‘distress generated by uncertainty and the realization of the limits of our knowledge’ (4).

In promoting the ILMA as an alternative to tracheal intubation in the pre-hospital setting, Dr Mason has, I fear, misunderstood the two main criticisms of the BET. The first was that the ‘clinical bottom line’ (prehospital endotracheal intubation is associated with increased mortality in patients with moderate to severe traumatic brain injury) could not be reliably concluded from the literature reviewed in the BET. It is, as Dr Mason acknowledges, “somewhat over-simplistic”. This seems a perfectly fair and reasonable criticism to make of an article published in a major emergency medicine journal.

The second was that the BETS process seems to ignore a key principle of evidence based practice: the combination of the best available evidence with clinical experience. Those of us with extensive pre-hospital experience do not question the clinical need for pre-hospital emergency anaesthesia, intubation, ventilation and retrieval to the most appropriate hospital for selected patients. We question how we might target this intervention more appropriately, how we might train paramedic practitioners to undertake it and how we might properly and thoroughly evaluate its safety and effectiveness compared to alternatives. The example given by French et al. was intended to highlight, to use Dr Mason’s words, another of the “ongoing absurdities in emergency medicine” – the historical acceptance of a standard of critical care in the pre- hospital phase which would be completely unacceptable in any hospital setting.

Neither the BET or the letter by French et al. concerned the role of the ILMA (or any other supraglottic airway device) or the role of professional paramedics in provision of pre-hospital critical care. The subject under discussion was the BET. Dr Mason's comments therefore seem a little unfair. Even if we can be accused of shooting the messenger, then two wrongs certainly don’t make a right. As a messenger who has been shot many times, I would ask Dr Mason to holster his gun, critically appraise the BET in question and re-read the correspondence related to it.

Roderick Mackenzie
PhD MRCP FFAEM
Clinical Fellow

Conflict of interest

Dr Mason and I have previously drawn pistols at dawn regarding the use of the ILMA in pre-hospital care (5,6).

References

1. Lloyd-Jones H (ed.) Sophocles. Ajax. Electra. Oedipus Tyrannus, Harvard University Press 1994.

2. Mason AM. Please don't shoot the messengers! EMJ Electronic Letter, 16 January 2006.

3. French J, Steel A, Clements R, et al., Best Bets. A call for scrutiny. EMJ Electronic Letter, 13 December 2005.

4. Choi PTL, Jadad AR. Systematic reviews in anesthesia: should we embrace them or shoot the messenger? Canadian Journal of Anesthesia 2000;47:486-493.

5. Mason AM, Use of the intubating laryngeal mask airway in pre- hospital care: a case report. Resuscitation, 2001;51:91-5.

Send letter to journal:
Re: Please don't shoot the messengers!

Dear Editors,

Sen and Nichani[1] should be congratulated for drawing our attention to one of the ongoing absurdities in emergency medicine; namely, that UK paramedics are provided with tracheal tubes, but are not given the drugs or monitoring equipment which enable their safe and effective use in salvageable trauma patients. The ability to intubate a trauma patient without the benefit of drugs is known to be an extremely grave sign[2], yet we don’t seem to be able to pluck up courage either to withdraw them from use, or to offer a viable alternative. Part of the problem comes from within the Ambulance Service itself, with many paramedics regarding the tracheal tube as the touchstone of their status.

The authors certainly don’t deserve to have their conclusions rubbished. Steel et al.[3] challenge them by asking if a potentially combative and physiologically compromised patient should preferentially undergo bag-valve-mask ventilation (BVMV) with an unsecured airway for a prolonged period, as if there were only two solutions to this problem; full-blown rapid-sequence intubation (RSI) or BVMV in the unsedated patient. Clearly, there is a third way that they omitted to mention, and that is the use of a supraglottic device in conjunction with appropriate sedation. They themselves had the opportunity to put forward evidence to convince us of the potential value and safety of endotracheal intubation in the hands of paramedics with or without drugs, but simply chose to take pot-shots at the messengers instead. Sen and Nichani’s bottom line may have been somewhat over-simplistic, but it has to be met with good evidence for the efficacy and safety of tracheal intubation in prehospital care before it can simply be dismissed.

References

1. Sen A, Nichani R. Prehospital endotracheal intubation in adult major trauma patients with head injury Emerg Med J 2005; 22.

2. Lockey D, Davies G, Coats T. Survival of trauma patients who have prehospital tracheal intubation without anaesthesia or muscle relaxants: observational study. Brit Med J 2001;323:141.

3. French J, Steel A, Clements R, Lewis S, Wilson M, Teasdale B, Mackenzie R, Black J. Best BETS. A call for scrutiny. EMJ Electronic Letter, 13 December 2005.

Send letter to journal:
Re: Prehospital Intubation – Delving deeper into the evidence

Dear Editors,

May I thank Ayan Sen and Raj Nichani for their recent “Best Bet” on prehospital intubation in head injury. It was a pity however, that they neglected to look deeper into the reasons why their conclusion, at least at this point in time, was that there is insufficient evidence to support its use. The very topic of prehospital rapid sequence induction (RSI), was the subject of a panel discussion and presentation at the National Association of Emergency Medical Service Physicians annual meeting in Arizona in 2004. They, fortunately, delved deeper into the issues surrounding RSI in head injured patients. One of the most important findings from this discussion was that most of the ambulance services involved in studies surrounding RSI / sedation assisted intubation, did so without the benefit of End-Tidal Carbon Dioxide (ETCo2) or even oxygen saturation monitoring. This, coupled with the widespread use of hyperventilation and inadequate preoxygenation went some way to explain the adverse findings found.

In one of the largest studies, the San Diego Paramedic RSI study, when one ambulance service introduced the use of ETCo2 monitoring, further analysis found hyperventilation (<30mmhg) occurred in 79% and severe hyperventilation (<25mmhg) occurred in 59% of intubated patients. Post introduction of ETCo2 monitoring, the incidence of inadvertent hyperventilation was significantly reduced. The only RSI subgroup without increased mortality were in those patients who underwent paramedic RSI but were then transported by air medical crews who had substantial experience using ETCo2 to guide ventilation.

The San Diego trial uncovered many adverse findings, but in a positive light, many important lessons were learned. First, advanced monitoring including pulse oximetry and ETCo2 should be mandatory when performing ETI with or without RSI. Second, adequate preoxygenation prior to RSI and close oxygen saturation monitoring during laryngoscopy should be routine. Third, hyperventilation should be avoided. In stark contrast to the San Diego study, the Whatcom Medic One program in Washington has experienced none of the desaturation/bradycardia issues and has an intubation success rate of 96.6%. All failed intubations were successfully managed. This successful RSI program is as a result of rigorous training, clinical governance, medical oversight, continuous quality assurance and of course the investment in adequate monitoring including ETCo2.

The most startling contrast between the USA and the UK, is that only physicians here undertake RSI. The monitoring described above is now mandatory in the emergency department (ED) and the anaesthetic room after a position statement by both the Royal College of Anaesthetists and our own faculty. In my scheme (Hampshire) and many others, we fully extend this to the prehospital theatre. In conclusion, if we are to accept that RSI in traumatic brain injury is a valid and meaningful intervention in the ED, then would it not follow that this is also true prehospital?

Dr Rob Dawes BM MFAEM DipIMC RCSed REMT-P

References

1. Ayan Sen and Raj Nichani: Prehospital endotracheal intubation in adult major trauma patients with head injury Emerg Med J 2005; 22.

2. Wang HE et al. Prehospital Rapid Sequence Intubation – What does the evidence show?: Proceedings from the 2004 national association of EMS physicians annual meeting: Prehospital Emergency Care Volume 8 No 4.

Send letter to journal:
Re: Association does not prove causality

Dear Editor,

I would like to briefly comment on the article entitled "Prehospital endotracheal intubation in adult major trauma patients with head injury" by Ayan Sen and Raj Nichani. In this excellent review, the authors point out that there are no prospective trials that have investigated the prehospital use of endotracheal intubation in adults. I believe it should be stressed that it is very difficult to account for all confounders using a retrospective design. It is extremely likely that the "sicker" patients were the ones who were intubated in the prehospital setting and therefore had worse outcomes. Until a prospective study is performed, I believe it is quite dangerous to jump to the conclusion that this association proves causality.

Send letter to journal:
Re: Best Bets. A call for scrutiny.

Dear Editors,

Best BETS are based on specific clinical scenarios and aim to provide a clinical bottom line which should indicate, in the light of the evidence, what the clinician would do if faced with the same scenario again.[1] The article by Sen and Nechani (EMJ 2005;22:887-889) serves to remind us that unless Best BETS are rigorously conducted their conclusions may be inappropriate.

Sen and Nechani wonder if pre-hospital intubation was of benefit to the major trauma patient they describe. They conclude that pre-hospital intubation is associated with increased mortality and imply that this intervention should not be undertaken.

There are two main problems with this. Firstly, evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.[2] Accumulating bad evidence does not make it good. Good evidence answers a highly specific question and the results are similarly specific to the circumstances. Sen and Nechani ask a poorly focused question and do not define the circumstances surrounding pre-hospital intubation in the studies they review – especially whether anaesthetic drugs were used. Even a cursory glance at these studies reveals major differences in quality, study design, patient populations, the experience and training of the operator, the use of anaesthetic drugs and the operational environment. The brief conclusion is therefore completely inappropriate.

Secondly, good doctors use individual clinical expertise together with the best available evidence: neither alone is enough.[2] Sen and Nechani question whether pre-hospital emergency anaesthesia is indicated in their patient. Such a question suggests that they do not appreciate the reality of pre-hospital critical care practice. The decision to anaesthetise and intubate an unconscious trauma patient is not controversial.[3] The controversy relates to whether this critical care intervention can be undertaken competently and safely. Are they really suggesting that their potentially combative and physiologically compromised patient should preferentially undergo bag-valve-mask ventilation with an unsecured airway for a prolonged period (often greater than half an hour) with no reliable measure of end tidal CO2? Would this be acceptable in the hospital critical care environment?

The EMJ has a responsibility to ensure that Best BETS are properly conducted and reviewed. This is not the first time that clinical bottom lines with major implications have been questionable – perhaps it is time to review the process again?

References

1. Mackway-Jones, K. Towards evidence based emergency medicine: Best BETs from the Manchester Royal Infirmary. EMJ 2005;22:887.

2. Sackett DL, Rosenberg WMC, Gray JAM, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn't. BMJ 1996;312:71-72.