• syncope
• chest pain
• risk assessment
• Covid-19
• psychiatric emergencies

• Download figure
• Open in new tab
• Download powerpoint

Much of what emergency physicians do is assess risk. Sometimes it’s the likelihood of a diagnosis – does this chest pain represent ischaemic heart disease? And sometimes we know the diagnosis but not how likely the patient is to suffer adverse consequences if discharged home. Syncope, of course, is one where we usually know what happened but not the underlying aetiology, and thus, whether the episode can be evaluated on an outpatient basis. This has led to several decision rules which have had varying success in validation studies. In this issue, Thiruganasambandamoorthy and colleagues pooled data from two large cohort studies to come up with Canadian Syncope Risk Scores that suggest high, medium and low risk for 30 day serious outcomes after a syncopal episode. The authors created an on-line calculator and visual aids that are available to readers to us in patient discussions to allow shared decision making.

Another knotty problem is judging whether the patient with chest pain has ACS and will suffer MACE in the next ³⁰ days. A variety of approaches have been proposed and validated, most of which require one or more troponin tests. But these tests add time, expense and pain to the patient’s visit. Could you risk stratify without a troponin? The article by Todd et al looks at the accuracy of ‘HEAR’ and ‘He-MACS,’ two proposed risk stratification scores that do not use a troponin, both adapted from scores that do use this test.

Now let’s make the experiment even harder – no troponin, no ECG, no vital signs, and no ability to eyeball the patient. Yes, you are on the receiving end of telephone triage call. How well will you do? In the study of NHS24 in Scotland assessing young patients with chest pain, Hodgins et al found that triage to the ED or an immediate home visit was associated with increased odds of patients being admitted to hospital with ‘a serious diagnosis’ while only 0.1% of patients directed to self-care were admitted to hospital in the next 7 days However, only 8.2% of patients were directed to self-care suggesting that although safe, there was substantial over-triage. The study is also quite fascinating in that it determines that among 102 822 calls, there were 1251 different pathways to receiving care.

Ultimately, underlying these studies is the question: what proportion of cases is it ‘acceptable’ to miss? For the missed MI, 2% is a number often tossed about. Several papers suggest that patients are willing to accept more risk than physicians, and certainly in my own practice, most patients have turned down my recommendation of a lumbar puncture after a negative CT scan to definitively rule out a subarachnoid haemorrhage. So, I was surprised by the findings from the study by Greenslade et al. These authors presented a scenario to real ED patients with complaint of chest pain in which the ‘doctor’ explained that their risk of an adverse event in the next 30 days was low, but could be made lower if they underwent some tests today. However, the risk of an adverse event from the diagnostic work up was 2%. The unexpected (IMHO) result suggests that much more work needs to be done in understanding what patients want, and the influence of how we present risks to them.

While psychiatric ‘boarding’ is now a major aspect of long stays in our departments, documentation of its impact is sparse. Our Editor’s Choice is a study of the association of waiting times of psychiatric patients in the ED with the length of their inpatient stay. The study found only a small increase in length of stay, and our editorial postulates why that may be the case. Another interesting aspect of this study is that it uses a different, but potentially more relevant approach, to the analysis than whether the difference is ‘statistically significant’. Rather this Bayesian approach tells us that, whatever the size of the impact, it is highly likely that there is an association of psychiatric boarding, and hospital LOS, and so solutions are needed.

A previous study by Marsden et al (EMJ 36 (7) 395–400, 2019) found half of trauma patients who received TXA received it in hospital, rather than in the prehospital setting. This month’s Reader’s Choice is a qualitative study of interviews with 21 paramedics about barriers and facilitators to giving TXA in the field, which should provide direction for increasing its pre-hospital use. We cover several other trauma topics this month, including needs assessment for REBOA, and the association of haemodynamic profiles with neurologic outcomes in children with traumatic brain injury. Further, while many studies about prehospital intubation focus on patients in cardiac arrest, Strucker et al explore risk factors and outcomes of unrecognised endobronchial intubation in trauma patients.

Our Concepts paper describes the development and implementation of a unique international cardiac arrest registry joining data from France and Canada. And our COVID-19 paper this month details the clinical outcomes of a nurse-led alternate care centre for assessing patients with COVID-19, which might prove an otion for some EDs should we see more surges in the current pandemic or as we plan for the inevitable next one. Do also read the correspondence regarding two prior EMJ papers; reader letters (and author replies) are an important contribution to scientific discourse and peer review.