• COVID-19

Four years ago this month, something was brewing in Wuhan that would change not only the world as we knew it, but also the face of medical literature. A quick search of PubMed shows well over 1/4 of a million unique publications with COVID in the title have been published to date. So, with that in mind, is there still learning to be had from this disease for emergency clinicians? One of the most fascinatingly terrifying aspects of COVID emergency management remains the concept of “silent hypoxia.” It is therefore important that we publish an original research manuscript from Laredo and colleagues (our Editor’s Choice) this month that explores this concept further. In their cohort study, that included over 1000 patients with COVID and compared with a cohort admitted with lower respiratory tract infections, they appear to demonstrate that “silent hypoxia” in previously healthy individuals with COVID appears to be a different entity to those patients with comorbidities, namely COPD, although the prognostic implications remain unclear. In their commentary about this paper, Steve Goodacre and Ashleigh Trimble point out that even the definition of silent hypoxia in Covid remains unclear, eloquently asking: “Was it a clinically meaningful entity or just a curious feature of COVID-19?”

As many of us continue to work in EDs that are seemingly devoid of strict infection prevention control regulations applied rigorously to inpatient settings, the pre-hospital identification infected patients with COVID remains important. Spina et al explore this issue in a novel way by assessing the diagnostic accuracy of operator-based interviews and machine learning algorithms for the pre-hospital detection of COVID, as later confirmed by PCR. Using a very large training set of operator calls and robust algorithm development, which include both clinical and epidemiological variables, their results prove extremely promising. The highest performing machine learning algorithm achieved an Area Under the Curve of 0.85 for the detection of COVID in validation.

There are an awful lot of ultrasound probes being bandied around in EDs and limited evidence behind their use. How we optimise the use of point of care ultrasound remains uncertain, it is therefore great to see two original research articles exploring a relatively novel use in guiding the reduction of displaced distal radius fractures. The first of these, a multicentre randomised controlled trial (RCT) in four Dutch centres was unable to demonstrate a statistically significant reduction in the number of reduction attempts when POCUS was used, with reduction taking longer with ultrasound. Was it underpowered perhaps and was the right primary outcome chosen? You decide. The second took a step back from measuring definitive outcomes and examined the feasibility of a definitive trial. Recruiting 48 participants and using a clever control group that used a sham ultrasound, feasibility of recruitment and data collection was demonstrated, and the authors conclude a definitive RCT would be possible. Readers may question why we have published both a definitive and feasibility RCT with broadly similar interventions. One RCT is rarely enough to change practice and clearly there is some work to do to define optimum outcomes here. In both trials there was a significant training requirement which may have implications for generalisability of this approach were it to prove beneficial. Our third POCUS paper this month explores a more familiar use of ultrasound in trauma, and is our Reader’s Choice. Bouzid and colleagues from France examine the diagnostic accuracy of eFAST in victims of stabbings. For me, the findings of this observational study confirm what is clinically sensible, POCUS has benefits in the early detection of cardiac tamponade and expediting theatre. To evaluate the presence of other injuries, I’m still heading to CT.

Our last two original research articles this month come under the heading of “Practice Change” and cover tricky aspects of emergency medicine; one the patient with vertigo and the other, antibiotic stewardship, both focusing on clinician behaviours and attitudes. Herdman and colleagues use a survey and qualitative interviews to explore the implementation of evidence-based guidelines for the patient with acute vertigo. Their survey of clinicians demonstrates very few are using bedside tests in clinical practice and interviews confirmed that practice needed to change. Some useful implementation strategies arise. Their interview quotes from clinicians will be familiar to many readers with their references to Google and the difficulty in interpreting HINTS. In the second practice change paper, Vu et al use quality improvement methodology to evaluate barriers and facilitators to the implementation of an electronic clinical decision support system for antibiotic prescribing and demonstrate the importance of end-user feedback, together with improved adherence to guidelines.

From RCTs, through observational studies to mixed-methods and quality improvement, it is great to see researchers from across the world employing a huge variety of methods to tackle important clinically-relevant emergency medicine questions. Great work from all the research teams this month.