Reviews and summaries of the latest Emergency Medicine research papers. We try to cover 10 of the best articles each month. Please visit our website for written summaries and to search past editions:

May 21st, 2019    

Boxer’s fracture: does it really need a plaster cast?

Background: One fifth of hand injuries are fifth metacarpal neck fractures ('boxer's fractures'). If stable, these are traditionally treated with an ulnar gutter back slab in the first instance. Several trials have looked at buddy strapping as an alternative, but in 2005 a Cochrane review​1​ concluded that existing studies were underpowered to detect a difference in functional outcomes. Since that time, however, at least 3 trials​2–4​ have been published, each coming to the same conclusion. Here we take a look at the most recent one, which is currently only available online (it hasn't been published in written form yet).


 Pellatt R, Fomin I, Pienaar C, Bindra R, Thomas M, Tan E, et al. Is Buddy Taping as Effective as Plaster Immobilization for Adults With an Uncomplicated Neck of Fifth Metacarpal Fracture? A Randomized Controlled Trial. Annals of Emergency Medicine. 2019 March​4​


The paper: An Australian RCT that recruited 97 people with uncomplicated boxer's fractures. The fracture was classed as 'complicated' if there was a rotational deformity, an associated tendon injury, it was intra-articular or angulated more than 70 degrees (in the UK we would class this as 30 degrees angulation from the axis of the shaft on lateral view). The control arm was treated with immobilisation in an ulnar gutter plaster cast, while the intervention arm received buddy strapping of ring and little fingers. No reduction was attempted before immobilisation.


In terms of the primary outcome, there was no difference in reported hand function at 12 weeks (assessed using the quickDASH questionnaire). Secondary outcome measures were also identical: pain, satisfaction, days off work, return to sports and overall quality of life. The only difference found was in terms of cost, with buddy strapping costing an estimated $2 compared to $22 with plaster casting.


There are several limitations to this study. No strictly objective assessments were included as outcome measures: in particular, grip strength, range of motion and radiographic angulation. However, one could argue that the outcomes chosen were more patient-centred than these surrogate markers. Also, cosmetic appearance was not directly assessed and some studies​2​ suggest that loss of the knuckle is commoner with conservative management.


The bottom line: Buddy strapping is a viable alternative to cast immobilisation for uncomplicated boxer's fractures. This study found no difference in functional outcome, and there is a considerable saving in terms ED resources.


Note: Why is there a cut off for the degree of acceptable angulation? Simply that as the angle increases the distance between the origin and insertion of flexor digiti minimi decreases, which creates slack and reduces grip strength. A study from 1999​5​ provided the evidence that led to the 30 degrees cut off.


Expert commentary:
"I've always put them in a POP as I thought it was more comfortable for the patient. If there's no difference in outcome then I think this could change practice. The only thing to be wary of is managing 'complicated' fractures in this way. We need to keep the above features in mind to ensure we are selecting the right group of patients for this intervention."
Dr Dwynwen Roberts, ED Consultant


Other FOAMed on this topic:
Life in the Fast Lane



  1. Poolman RW, Goslings JC, Lee J, Statius Muller M, Steller EP, Struijs PA. Conservative treatment for closed fifth (small finger) metacarpal neck fractures. Cochrane Database of Systematic Reviews [Internet]. 2005 Jul 20; Available from:
  2. van Aaken J, Fusetti C, Luchina S, Brunetti S, Beaulieu J-Y, Gayet-Ageron A, et al. Fifth metacarpal neck fractures treated with soft wrap/buddy taping compared to reduction and casting: results of a prospective, multicenter, randomized trial. Arch Orthop Trauma Surg [Internet]. 2015 Nov 11;135–42. Available from:
  3. Dunn JC, Kusnezov N, Orr JD, Pallis M, Mitchell JS. The Boxer’s Fracture: Splint Immobilization Is Not Necessary. Orthopedics [Internet]. 2016 Mar 29;188–92. Available from:
  4. Pellatt R, Fomin I, Pienaar C, Bindra R, Thomas M, Tan E, et al. Is Buddy Taping as Effective as Plaster Immobilization for Adults With an Uncomplicated Neck of Fifth Metacarpal Fracture? A Randomized Controlled Trial. Annals of Emergency Medicine [Internet]. 2019 Mar; Available from:
  5. Ali A, Hamman J, Mass D. The biomechanical effects of angulated boxer’s fractures. J Hand Surg Am [Internet]. 1999 Jul 1;24(4):835–44. Available from:

May 16th, 2019    

Introducing the modified Sgarbossa criteria


Background: I saw a patient last month who had chest pain and an ECG showing an acute MI in the context of LBBB. The ECG did not meet the original Sgarbossa criteria but did when the modified rule was applied. I was suitably thrilled by this. No one else was. In fact, most of my colleagues were not familiar with the original Sgarbossa, let alone the modified version...



Warning: you can probably tell already that this post is going to be a little more nerdy and detailed than you're used to. Bear with me - Sgarbossa is not really difficult and it can really help when confronted with ECGs showing LBBB or ventricular pacing. The in-built ECG computer usually reports paced ECGs as: "no further analysis attempted". Be better than the machine! You can analyse them and, in particular, you can spot a STEMI with 80% sensitivity if you use the rule described below.



Dr Elena Sgarbossa proposed a clinical prediction rule for diagnosing acute myocardial infarction in the presence of LBBB way back in 1996 (in this paper). It consisted of 3 simple ECG criteria, and was later found to be equally helpful for diagnosing MI in the presence of a paced rhythm (see this post from The BREACH and the ESC guidelines on the management of STEMI here). What you may not know is that Dr Stephen Smith modified the Sgarbossa criteria in 2012 to make them more sensitive (this paper). Here we consider the proposed changes, and review a validation study that was published 3 years later.



Myers HP, Limkakeng T, Jaffa EJ, et al. Validation of the modified Sgarbossa criteria for acute coronary occlusion in the setting of left bundle branch block: a retrospective case-control study. Am Heart J. 2015 Dec;170(6):1255-64



The paper: A retrospective case-control study performed by chart review. Over 3 years every patient with LBBB and cardiac-sounding chest pain was included (294 in total - 45 STEMIs and 249 controls). The modified Sgarbossa criteria performed significantly better than the original weighted Sgarbossa criteria: 80% vs 49% sensitivity, and 99% vs 100% specificity. Against the unweighted criteria, modified Sgarbossa was again significantly better: 80% vs 56% sensitivity, and 99% vs 94% specificity.



'Weighted', by the way, refers to the system where each criterion receives a certain number of points, which are then totalled to give a final result. When Sgarbossa is used 'unweighted' (any one criterion gives a positive result), the sensitivity increases while the specificity reduces. With the modified criteria however, the sensitivity is pretty good already, so there is no need to bother with a points system. This makes it much easier to remember and use in practice.



The bottom line: The modified Sgarbossa criteria are superior to the original criteria for identifying acute STEMI in the context of LBBB




Original criteria:
1. Concordant ST elevation >1mm in leads with a positive QRS
2. Concordant ST depression >1mm in leads with a negative QRS (V1-3)
3. Excessively discordant ST elevation >5mm in leads with a negative QRS (V1-3)




Modified criteria:
1. Unchanged
2. Unchanged
3. ST elevation >25% of the depth of the S wave in leads with a negative QRS (V1-3)
i.e. the new criteria make the degree of ST elevation proportional: dependent on the amplitude of the preceding QRS rather than being an absolute figure



Note: Dr Smith's modified criteria were referenced and accepted by Dr Sgarbossa herself in this 2013 paper.



Expert commentary:
"This is interesting, especially if it was validated by Dr Sgarbossa. I think there needs to be more education on this for our department (and for EM trainees in general) as it is not widely known."
Dr Dwynwen Roberts, ED Consultant


May 11th, 2019    

New guidelines on lower GI bleeding


Background: Lower GI bleeding (LGIB) is a fairly common ED presentation, but it may surprise you to learn that there are no UK national guidelines on the acute management of this condition. Who can go home? Who needs admission? What is the best initial management? Thankfully, the British Society of Gastroenterology has recently published a paper covering all of this, and it's a big one: stakeholders include the Association of Coloproctology of Great Britain and Ireland, the British Society of Interventional Radiology, the Royal College of Radiologists and NHS Blood and Transplant.



Oakland K, Chadwick G, East J, et al. Diagnosis and management of acute lower gastrointestinal bleeding: guidelines from the British Society of Gastroenterology. Gut. 2019 Jan;0:1-14



Summary of recommendations





  • The Oakland score is recommended to risk-stratify patients
  • Patients with a minor self-limiting bleed (Oakland 8 points or less) can be discharged for urgent outpatient investigation
  • Patients with a major bleed (Oakland >8 points) should be admitted for colonoscopy on the next available list
  • Patients who are haemodynamically unstable (shock index >1 after initial resuscitation) should be admitted for urgent CT angiography (CTA)





  • Colonoscopy is the preferred initial investigation in most cases of major LGIB as it has the potential for diagnosis and treatment
  • CTA is preferred over colonoscopy in unstable patients, as it can localise a bleeding source in the upper and lower GI tract and has no requirement for bowel preparation
  • If no source is identified on CTA, patients should undergo an upper endoscopy (OGD)
  • Some unstable patients should proceed directly to OGD following senior discussion, as shock is more commonly found in UGIB than LGIB
  • No patient should proceed to emergency laparotomy unless every effort has been made to localise bleeding by radiological or endoscopic modalities



Blood transfusion:


  • Patients should be transfused if haemoglobin level is less than 70g/L
  • Patients with a history of cardiovascular disease should be transfused if haemoglobin is less than 80g/L



Expert commentary:
"This article is really interesting: a useful summary for ourselves and the admitting speciality. The Oakland score looks promising, as it seems to have fairly standard criteria on it."
(Dr Robert Tan, ED Consultant)



More on the Oakland score:
The score can be found on MDCalc. It predicts the probability of safe discharge, specifically the absence of rebleeding, transfusion, interventions, readmission or death at 28 days. It is based on 7 variables: age, sex, history of LGIB, blood on PR, heart rate, blood pressure and haemoglobin level. A score of 8 or less corresponds to a 95% probability of safe discharge. The score was derived in a large UK-based multi-centre study (The Lancet, 2017) involving around 2,300 patients. It has not been internationally validated yet.


May 6th, 2019    

Should we cardiovert everyone with recent-onset fast AF?


Background: Cardioverting a patient who presents with recent-onset fast AF is great. You consent them, sedate them, shock them, and they wake up feeling better and (often) go home. You feel awesome - quick action, visible result, everyone happy. But how necessary is it to get these patients back into sinus rhythm on their initial visit? And how many will cardiovert without any intervention? The RACE-7 trial can help us answer these questions.



Pluymaekers NAHA, Dudink EAMP, Luermans JGLM, et al. Early or delayed cardioversion in recent-onset atrial fibrillation. N Engl J Med 2019 Apr;380(16):1499-1508



The paper: A total of 437 patients were randomised over a 4-year period across 15 EDs in the Netherlands. Each presented with recent-onset (within the last 36 hours) stable AF and were randomised to immediate or delayed cardioversion. Those in the delayed group received rate-controlling medication in the ED and were followed up at 48 hours. If they were still in AF, they were cardioverted at this time. In the immediate cardioversion group, sinus rhythm was restored in 94% of cases (16% spontaneous, 38% chemical, 40% electrical). In the delayed group, 69% had spontaneously cardioverted by the time of follow up, and the remaining 28% were successfully cardioverted at this time (4% chemical, 24% electrical). There was no significant difference in the authors' primary endpoint (the presence of sinus rhythm after 4 weeks), but for me the interesting part of this study is the rate of spontaneous cardioversion in both groups.



The bottom line: These data suggest that the majority of patients presenting with recent-onset stable AF will spontaneously cardiovert by 48 hours. Maybe we should give them a chance before reaching for the defib pads.



Expert commentary:
"I think this backs up our usual practice: most get rate control or IV fluids and they often slow down/convert. I'm not always convinced that the change in rate/rhythm is anything to do with what we've done to the patient but that they just slow down/convert on their own! This study lends support to this practice."
(Dr Dwynwen Roberts, ED Consultant)


May 1st, 2019    

Upside-down position for SVT in children

Background: Supraventricular tachycardia (SVT) is the commonest arrhythmia in children, having two peaks – one in the first year of life and another around age 6. We generally attempt vagal manoeuvres before medication or electricity to restore sinus rhythm. The commonest techniques are the ‘mammalian dive reflex’ (submerge the face of an infant in freezing water) and the Valsalva method (forced expiration against a closed glottis). There is, however, a new kid on the block – the imaginatively-named upside-down position. This is the first pilot study of this technique. The authors are currently conducting a larger RCT.


Bronzetti G, Brighenti M, Mariucci E, et al. Upside-down position for the out of hospital management of children with supraventricular tachycardia. Int J Cardiol 2018;252:106-109


The paper: A single-centre study of 24 children with known paroxysmal SVT. The parents were taught to identify SVT using palpation of the brachial or carotid pulse and with a stethoscope. Each family was randomly assigned to standard vagal manoeuvre (VM – blowing into a 10ml syringe for 15 seconds) or upside-down manoeuvre (UD – manually flipping children weighing less than 30kg or helping heavier children to do a handstand for 30 seconds). If this was unsuccessful, the other manoeuvre was attempted. The rate of cardioversion at the first attempt was 33% (VM), compared to 67% (UD). Of those in whom the attempted manoeuvre was unsuccessful, 50% subsequently cardioverted with UD, whereas none cardioverted with VM. In addition to these results, parents reported that their children enjoyed the upside-down position more than syringe-blowing.


The bottom line: This is a tiny pilot study, but the results are very much in favour of the upside-down position for the treatment of SVT in children.


Note: It is well-known that vagal manoeuvres work better if they are carried out before adrenergic tone rises, which is why they work better when performed soon after the onset of SVT, and at home rather than in the ED


Expert commentary:
“Interesting… to be honest I have never tried this technique. However, it seems very promising and I’m glad that further research is happening. The British Heart Foundation leaflet for parents on SVT actually recommends the upside-down method so it’s certainly becoming well-established.”
(Mohsin Jafri, Paediatric Consultant)

April 27th, 2019    

New tetanus guidelines


Background: Have you ever seen a case of tetanus? Don't feel bad if you haven't - there were only four in the whole of England during 2016. The disease is admittedly terrible: the muscle spasms can be so severe they cause fractures, and serious complications loss of airway and respiratory arrest. Thankfully, vaccination has made the condition extremely rare in the developed world. But it often feels as though confusion reigns over the best approach to vaccination in the ED... Who needs a booster? Who needs immunoglobulin? What kind of patient? What kind of wound? The Department of Health recently updated their guidance on tetanus immunisation, replacing the 2013 recommendations. Here is the bite-sized BREACH summary of the document...



Department of Health (2018). Tetanus: The Green Book, chapter 30


See for summary tables


April 17th, 2019    

Right ventricle myocardial infarction


Background: A case report in 1974 was the first to mention right ventricle (RV) infarction as a distinct entity, noting that patients with this pathology exhibit "a unique clinical and haemodynamic syndrome". This is because the resulting weakness of the RV makes the patient's cardiac output extremely preload sensitive. They often present in shock and even if they are normotensive initially, giving GTN can precipitate a severe drop in blood pressure or even trigger cardiac arrest. These are an important group of patients to identify!



Thygesen K, Alpert J, Jaffe A, et al. Fourth universal definition of myocardial infarction. European Heart Journal 2019;40(3):237-269



The paper: This is another nugget from the 4th Universal Definition of Myocardial Infarction, published last year. It highlights the necessity of obtaining recordings from supplemental leads in certain patients. They note that an RV infarct is very rarely isolated but usually occurs as part of an inferior STEMI (Up To Date suggests 30-50% of inferior STEMIs have this complication). Supplemental leads V3R and V4R are placed on the right side of the chest, opposite the standard V3 and V4. ST elevation >0.5mm in either of these supplementary leads is considered diagnostic of RV infarction. Other ECG changes suggestive of RV infarction include ST elevation in aVR or V1, but several experts recommend deploying the supplemental leads in every case of inferior MI, because missing an RV infarct can be devastating for the patient.



The bottom line: Be aware of RV infarction as a potential complication of inferior STEMI and actively look for it by deploying supplemental leads. Don't give GTN to these patients! They need IV fluids in the first instance.



Note: Amal Mattu (Cardiology EM consultant at University of Maryland, USA and producer of the excellent ECG Weekly) suggests looking for the following triad to identify patients at risk of RV infarction...

- Chest pain
- Hypotension
- No crackles on auscultation

You need all three because the other main differential for MI + shock is flash pulmonary oedema.



Expert commentary:
"This is really great learning, and the triad is easily memorable."
(Dr Robert Tan, ED Consultant)



More FOAMed on this topic:
Life in the fast lane
Dr Smith's ECG blog (to take things to the next level)
ECG weekly - not quite FOAMed but (I think) the best ECG resource around and it's pretty inexpensive


April 10th, 2019    

Needle-less anaesthesia for mucosal lacerations


Background: It is difficult to treat oral mucosal lacerations in children. They are often terrified, in pain, and won't remain still enough to allow safe delivery of local anaesthetic or fine suturing. Lidocaine can either be injected locally or used in a nerve block, but both of these options are tricky in the paediatric population. The authors of the current paper present a novel approach to achieving satisfactory local anaesthesia.



Nickerson J, Tay ET. Dripped lidocaine: a novel approach to needleless anesthesia for mucosal lacerations. J Emerg Med 2018;55(3):405-7



The paper: A case series of 3 children (aged 2, 8 and 15) who presented with lip lacerations. Each received local anaesthesia by dripping 1% lidocaine into the wound at a rate of 1 drop per second until the wound was saturated with solution. Adequate anaesthesia was achieved after around 5 minutes, and each wound was sutured closed with minimum discomfort. The authors hypothesised that the increased blood supply of the oral mucosa meant that the lidocaine was readily absorbed by the wound and surrounding tissue.



The bottom line: If you have a child with an oral mucosal laceration that needs closing, try dripping lidocaine onto the wound to achieve local anaesthesia.



Note: I have personally tried this 3 times (bringing our 'collective case series' to 6!). Twice it worked like a charm. In the last case, there was some discomfort, but I was able to inject some more lidocaine to eliminate this - the drips certainly helped reduce the pain of injection.



Expert commentary:
"This technique is useful as LAT gel cannot be used on mucus membranes. The only potential risks are swallowing the solution and reaching the maximum dose for weight. Incidentally, there is no hard-and-fast rule about who can close lacerations across the vermilion border - it's purely about the competence of the practitioner and the agreement of the patient / parent."
(Dr Dwynwen Roberts, ED Consultant)



"This is a potentially amazing idea, especially in needle phobic children, ADHD, ASD, etc."
(Dr Mohsin Jafri, Paediatric Consultant)


April 4th, 2019    

Ultrasound for diagnosing retinal detachment


Background: Retinal detachment is one of several sight-threatening conditions seen in the ED. Flashes and floaters are often the first sign that something is amiss, and the standard ED workup involves slit lamp examination and fundoscopy. However, subtle abnormalities are difficult to pick up on fundoscopy, especially in the non-dilated eye. In the FOTO-ED study (2011), ED physicians only performed fundoscopy in 33 out of 350 appropriate cases, and they missed the diagnosis in every one (confirmed by photos taken of the fundus at the time of examination).

Is there a better way to check the back of the eye? The first observational trial of ocular ultrasound in the ED (17 years ago in 2002) found it to be very accurate and since then many more trials have appeared. The average 'bench-to-bedside' period is often quoted as 17 years - has the time come for ocular ultrasound to become common practice?



Gottlieb M, Holladay D, Peksa GD. Point-of-care ocular ultrasound for the diagnosis of retinal detachment: a systematic review and meta-analysis. Acad Emerg Med 2019;00:1-9



The paper: A systematic review that found 11 observational studies with a total of 844 patients. Ocular ultrasound was performed by ED physicians on patients with suspected retinal detachment, and their findings were compared to final ophthalmology diagnosis. They found that ultrasound was 94.2% sensitive and 96.3% specific for the diagnosis of retinal detachment.



The bottom line: Ocular ultrasound is both sensitive and specific for the diagnosis of retinal detachment and, with training, is probably a better examination tool than fundoscopy for this condition.



Note: The Royal College of Ophthalmologists (2010) recommend immediate referral if the patient has "visual field loss or fundoscopic signs of detachment or vitreous haemorrhage". They recommend indirect ophthalmoscopy within 24 hours if there is "no visual field loss, normal visual acuity and normal fundoscopic exam" (i.e. reported flashes and floaters only)...

So where does ocular ultrasound fit into this? If your patient has normal visual acuity and no field deficit, perform ultrasound. If there are signs of detachment, refer them immediately; if not, they can be seen the following day.


March 30th, 2019    

Can we suture dog bite wounds closed?


Background: The traditional teaching is to leave dog bite wounds open, allowing them to heal by secondary intention or having them closed in theatre after a thorough wash-out. The idea was that this reduced the risk of infection. The studies that led to this approach, however, have many methodological problems and are at least 30 years old. Recent research has looked at primary closure.



Paschos NK, Makris EA, Gantsos A, Georgoulis AD. Primary closure versus non-closure of dog bite wounds. A randomised controlled trial. Injury 2014;45(1):237-40



The paper: An RCT involving 168 consecutive patients with dog bite injuries. Each had their wounds cleaned using high-pressure irrigation and iodine and each received a course of prophylactic antibiotics. They were randomised to have their wounds sutured closed or left open, and were followed up at weeks 1 and 4 post-injury to determine the rate of infection and the cosmetic appearance of the wound. The overall infection rate was 8.3%, and there was no significant difference between the two groups. The cosmetic appearance at 4 weeks (independently assessed by a surgeon using the Vancouver Scar Scale) was significantly better in the sutured group.



The bottom line: This study is probably the best evidence to date (it's hard to see how they could have improved this type of trial - you can't blind the clinicians!). If dog bite wounds are carefully cleaned and debrided, and if they are not too large or complicated, it seems reasonable to close them in the ED.


- Older Posts »