Send letter to journal:
Re: Assessment and risk stratification of patients with acute chest pain

We read with interest Manini et al’s ¹ recent paper suggesting ischaemia-modified albumin assays could assist in decreasing the rate of inappropriate discharges from the ED, and that further studies into diagnostic tools for use in chest pain are warranted.

We conducted an audit at Ealing Hospital that also supports the need for developing new diagnostic algorithms for chest pain. Medical histories of 147 patients admitted to the Emergency Department’s Chest Pain Unit for monitoring were studied. All patients gave a history consistent with chest pain of cardiac origin. We studied whether positive troponin-T results correlated with the presence of Framlingham cardiovascular risk factors, or with a past medical history of cardiac or vascular disease (i.e. a history of angina, coronary artery bypass grafting, previous MI, stroke, TIA, heart failure, previous angiography or history of claudication).

In our patient group the percentage of positive troponin results did not increase in line with the number of Framlingham risk factors present, however these patients were highly likely to receive further in-patient investigation. Conversely, while past medical history of cardiac or vascular disease did not appear to have a bearing on the emergency physician’s decision to admit a patient for investigation, those patients with 1-3 conditions appeared to form the majority of patients with positive troponin results (this group made up 55.9% of patients with positive troponin results, but only contributed 36.9% of those with negative troponins).

This data supports a number of other studies on this topic ^2,3, and we believe that physicians should consider treating chest pain patients with significant cardiovascular history with a higher index of suspicion that is currently the case. We also believe that further research into the use of past medical history in the assessment and risk stratification of acute chest pain is required.

References:

1. Manini AF, Ilgen J, Noble VE et al.. Derivation and validation of a sensitive IMA cutpoint to predict cardiac events in patients with chest pain. Emerg Med J 2009;26:791-796.
2. Body R, McDowell G, Carley S et al.. Do risk factors for chronic coronary heart disease help diagnose acute myocardial infarction in the Emergency Department? Resuscitation 2008;79:41-5.
3. Cakir B, Blue K. How to Improve the Management of Chest Pain: Hospitalists and Use of Prediction Rules. South Med J 2007;100:242-7.

Send letter to journal:
Re: Basic Life Support should be kept basic.

Although the definition of Basic Life Support (BLS) does vary between sources, it is probably best regarded as, "a level of medical care that can be used to treat patients with life-threatening illness or injury without the use of any advanced or invasive medical procedures or intravenous access". It should be possible for any rescuer in any situation to render BLS simply by using hands and lungs, although simple improvised items (such as a handkerchief plus necktie or tights to apply local pressure to a bleeding wound) should also be allowed. Some would extend the list of permitted adjuncts to include a face shield/pocket mask or even an oropharyngeal airway. However, to stretch the concept of BLS to include the use of a bag-valve-mask device (BVMD), laryngeal mask airway (LMA) or laryngeal tube (LT) probably strays too far into the realms of advanced life support. If this is the case, then the title of the paper by Dixon et al.[1] was inaccurately worded.

In reality, the BVMD is probably one of the trickiest items to use correctly in prehospital care, and some would say its use requires two persons – one to hold the mask securely against the patient’s face and the other to squeeze the bag. Certainly, to ventilate the lungs correctly with a BVMD without inflating the stomach requires considerable skill.

The authors failed to state which versions of the selected supraglottic airway devices (SADs) were used in the trial. The Laryngeal Tube is available with both a single lumen (LT & LT-D) and dual lumens (LTSII & LTS-D), the dual-lumen versions featuring a gastric drainage channel in addition to the airway tube. With its distal balloon inflated within the manikin’s upper oesophagus, it is hardly surprising that the authors found a low incidence of gastric insufflation with the LT, particularly if the device also featured a gastric drainage channel opening beyond the distal balloon. Also, the authors failed to state which type of LMA was used by way of comparison. If this was a basic LMA (e.g., the LMA Classic, LMA Unique, Ambu LM or Softseal LM) then a direct comparison with any LT device was probably unfair with respect to the rates of gastric insufflation. It would have been better to have compared the LT with either the LMA Proseal or the disposable LMA Supreme or i-Gel airways - all of which have gastric drainage channels like the LT-D. It should also be noted that the correct sizing of a supraglottic airway device is particularly important when attempting to ventilate manikins, and the same size of airway does not always provide the optimum fit with a particular manikin across the entire range of SADs from different manufacturers. An ill-fitting basic LMA will never be a match for a correctly-sized LT on any of its performance characteristics.

Bearing in mind all these points, the authors’ conclusions need to be viewed with a degree of caution.

Send letter to journal:
Re: Importance of log roll and careful spine evaluation

Send letter to journal:
Re: Pellegrini-Stieda

This case appears to be describing the Pellegrini-Stieda syndrome/lesion. This is, in fact, a well-known and fairly common finding, generally felt to reflect a post-traumatic ectopic ossification, either acute or chronic. MR is a useful test, as the signal void from the calcification can be seen in relation to the MCL. It may also uncover bone bruising from an associated avulsion fracture (the so-called Stieda fracture). As an academic exercise, MR can also subclassify the lesion into 4 different types, dependent on the exact pattern of ossification (although this has limited, if any, clinical importance).

Send letter to journal:
Re: A response from the International Academies of Emergency Dispatch

Send letter to journal:
Re: Primum, non nocere.

Dear Editor,

I was disturbed to read the article by Body and Foex [1] advocating the embrace of Utilitarian values in medicine. I hope it was merely a misuse of words. All penguins are birds, but not all birds are penguins. Utilitarianism is a form of consequentialism, but not all ethical thinking that considers the consequences of one’s actions is Utilitarian. The authors of the article correctly make a clear case for efficiency and risk -benefit “consequential” thinking in medicine, and in particular in decision analysis. However, this is not Utilitarianism.

Utilitarianism is an ethical theory in which the “right” action is that which maximises the aggregate “good” outcome across a population. The “good” can variously be defined as pleasure, or preference satisfaction, or as in this article, health benefit. There are practical problems with the theory which are common to any consequential thinking: the difficulties of making predictions and calculating relative risk-benefits for people with different perceptions of what is good for them. However, the big problem with utilitarianism itself is that it justifies any action, so long as there is an aggregate net increase in what is considered good. Plagiarism, falsified research data, dishonest job applications, unfair discrimination - could all in some situations be justified. And so could bribery, theft, or murder. Of course, doctors would never be involved in inhuman human experiments, forced sterilisations, “eugenic” murder, or torture for the “greater good”… but they would, and they were, and it didn’t stop with Nazi Germany or Tuskegee. And why should it, if you accept Utilitarianism? Why not kill an unwilling patient in order to harvest his healthy organs to save five others who would otherwise die?

If we seek to do net good in medicine, the command to “do no harm” is a helpful warning against the evil of imposing a centrally defined, collective “good”, on vulnerable individuals. As such, it is good that Hippocrates is said to have made it a priority. Bentham might have considered rights to be “nonsense upon stilts”,[2] but it is because consequence-based ethical thinking is so inherently dangerous, that we need human rights, and aphorisms such as “first, do no harm”.

No-one outside the extreme wing of the health-and –safety lobby would suggest that “do no harm” means that one must never perform an action that might be painful. Foex and Body are setting up a “straw man” by using examples such as venous cannulation to argue against the importance of non -maleficence. Patients consent to undergo discomfort or risk in order to achieve a later benefit: this expresses their right to choose for themselves (autonomy), and is a balance of good and harm (beneficence and non-maleficence). If the patient cannot consent, then we have to weigh up very carefully what is in the patient’s best interests (risk-benefit) – not primarily the interests of the State.

Utilitarianism can impose deliberate involuntary harm on an individual for others’ benefit. By reminding ourselves of the importance of autonomy and doing no harm, we would remember that the good we seek is that of the patient we are treating, that the harm that may ensue is accidental, and the risks agreed by the one facing them.

Consequence-based reasoning and good medicine are of course “inescapably intertwined”.[1] But for goodness’ sake, do not call this “Utilitarianism”.

Sincerely, Giles N Cattermole.

[1]Body R, Foex B. On the philosophy of diagnosis: is doing more good than harm better than “primum non nocere”? Emerg Med J 2009;26:238-40.

Send letter to journal:
Re: Sensitivity or specificity?

With great interest we read the recent article by Bevan and colleagues reporting how predictive ALT is for liver injury in children with blunt abdominal trauma. In the result section the authors describe the following: "The presence or absence of liver injury can be predicted with a sensitivity of 96% and a specificity of 80%". In their conclusion they reported: "a threshold of >104 IU/L gave a 96% specificity for the detection of liver injury". We would like to emphasize the difference and think the authors mean 96% sensitivity in their conclusion section.

Yours sincerely,

Send letter to journal:
Re: ToPDoG is in progress

Dear Sirs

I commend the work of Geelhoed and MacDonald in their sentinel dose- finding studies regarding the minimum effective dose of dexamethasone for croup, it does seem to suggest a 'ceiling' effect. This work has been recently followed by a descriptive paper (accepted for publication, not yet published) outlining the experience over 27 years at their institution, clearly demonstrating the real-world effectiveness of the lower dose (0.15mg/kg), which became local policy more than two decades ago. Unfortunately, every Systematic Review by the Cochrane Collaboration has found the numbers of patients enrolled in the initial RCTs of Geelhoed and MacDonald too small to rigorously prove the efficacy of the lower dose of dexamethasone in croup. It is with this background that we have recently started recruiting subjects for a large RCT, the ToPDoG Study: Trial of Prednisolone / Dexamethasone oral Glucocorticoid. This trial aims to settle the question of dexamethasone dose (0.6 vs 0.15mg/kg), and compare the efficacy of prednisolone (1mg/kg)in a three-armed equivalence study. Details can be found at: http://www.anzctr.org.au/trial_view.aspx?ID=83722

References:

Oral dexamethasone in the treatment of croup: 0.15 mg/kg versus 0.3 mg/kg versus 0.6 mg/kg. Geelhoed GC, Macdonald WB. Pediatr Pulmonol. 1995 Dec;20(6):362-8. PMID: 8649915

Sixteen years of croup in a Western Australian teaching hospital: effects of routine steroid treatment. Geelhoed GC. Ann Emerg Med. 1996 Dec;28(6):621-6. PMID: 8953950

Send letter to journal:
Re: Propofol is not safe for sedation for hip relocation

University Department of Anaesthetics Level 2, University Block Glasgow Royal Infirmary 10 Alexandra Parade Glasgow G31 2ER

19 February 2009

Dear Sir,

We read with interest the clinically based study, on the use of propofol to sedate patients for relocation of hip prostheses in the emergency department.[1] The authors rightly point out that there are problems with the safety and efficacy of using midazolam, and conclude that the described technique is both effective and safe. In another paper by the same authors they demonstrate this technique of “sedation” has a better success than midazolam, reduces the delay in these patients going to theatre, and therefore the patients discomfort (although there is no mention of pain scores of these patients).[2] However we disagree strongly with the conclusions that adverse effects were acceptably uncommon, and argue that the authors have not demonstrated the safety of this technique.

First, we would like to comment on the sedation protocol. Disappointingly there is no attempt to describe the depth of sedation provided. The report of the Academy of Royal Colleges on Safe Sedation Practice states clearly that “verbal contact with the patient is maintained throughout the period of sedation”.[3] To us, 1mg.kg-1 of propofol in this age group is a dose close to that required for induction of anaesthesia [4], and without documentation of the maintenance of verbal contact it cannot be termed sedation. By your own admission, many of the patients in this study were, in fact, anaesthetised. The Academy of Royal Colleges document (to which the Faculty of Accident and Emergency Medicine were party) again is quite clear that “provision of sedation deeper than this (verbal contact)… is bordering on anaesthesia. As such, this depth of sedation must be supervised by those with the same level of training and skills necessary to provide general anaesthesia”.[4] Given that many of these patients may have been anaesthetised, we have several concerns regarding this protocol pertaining to training, monitoring, and fasting:

Training: the staff responsible for this procedure had only undergone one hour of in-house training. The Royal College of Anaesthetists mandate to its own trainees that they should undergo an initial assessment of competency before being allowed to give any anaesthetic not directly supervised. This assessment is usually after a full three months.[5] Monitoring: the level of monitoring recommended for patients undergoing general anaesthesia should include capnography.[6] Fasting: we find it incredible that in emergency patients suffering pain (and consequently at higher risk of pulmonary aspiration) that you stated that fasting guidelines were used “as a guide and not a rule”. Evidence on the necessity of fasting for elective procedures are clear after almost 40 years of evidence.[7] Guidelines are less clear for emergency cases as normal fasting times may be insufficient, necessitating protection of the patients’ airway.

More worryingly we refute the interpretation of these data as evidencing safety. It would have been useful to present the incidence of adverse events with confidence intervals (CI). This allows one to estimate the true population incidence of a rare event, which could be as much as the upper level of the 95% CI.[8] We have taken the liberty of doing this for you: 8% (95% CI 2.6 to 13.4) of patients suffered arterial oxygen desaturation, 4% (95%CI 0.2-8) required bag-valve-mask ventilation and 4% (95%CI 0.2 to 2.8) required vasopressors. Therefore your actual population rate may be anywhere between 2.6% and 13.4%. This rate of airway/respiratory events equates to 80/1000 (but could be anywhere between 26 and 134/1000 patients). This compares very unfavourably with those of other non anaesthetic groups (Australian GPs) of 4.1 (95%CI 3.3 to 4.9) /1000 and even less favourably with anaesthetists of 2.6 (95%CI 1.6 to 4.2) /1000. 9 Our department has trained non-medical sedationists to provide true conscious sedation for a different painful procedure (oocyte retrieval for assisted conception), and have audited experience of 3000 patients with an adverse incidence rate of 0.3 (95%CI -0.3 to +0.9) /1000patients. In this context your described results cannot be remotely construed as demonstrating safety. We would also point out that a sample size of thousands not hundreds would be necessary to convince us of the safety of this non-standard technique.[10]

In conclusion we are not surprised that the hip relocation rate is higher with your technique as you have compared propofol anaesthesia with midazolam sedation. We can entirely understand the desire to reduce delays for your patients waiting in pain for hip relocation in theatre. However, our answer to the title of your article “Is propofol a safe and effective sedative for relocating hip protheses?” is a resounding no. It is our interpretation that this technique has not been demonstrated as safe, and would be difficult to justify in the event of a permanent serious complication.

Yours sincerely,

Dr A Puxty, Dr M Sim, Dr KJ Anderson, Professor J Kinsella