Authors: Celestine Weegenaar, Michael Connelly, Jessica Woods, Alex Dodd, Lakshmi Shenoy, Stephanie Milne, Daisy Carter, James Massey / Editors: Celestine Weegenaar, Michael Connelly, Rajesh Chatha, Robert Hirst / Codes: C3AP1a, C3AP2a, C3AP2b, CMP3, HAP19, HAP29, HMP3 / Published: 08/04/2021
Welcome to the second of TERN’s Top Papers. Each month, a different region review a selection of papers on a different theme each month. Participation will help EM trainees provide evidence for the research specialty learning outcome (SLO10) and will help aspiring EM trainees demonstrate commitment to specialty. If you are interested in getting involved email us with your name, grade, and deanery.
This month it was Severn’s turn. Last month, colleagues in South East Scotland (led by one of the TERN education reps Raj Chatha) looked at interesting papers from the past year. This month we have looked at papers with a trauma or orthopaedic focus published over the past year. April’s TERN Top Papers was led by Celestine Weegenaar (TERN Severn regional rep and the other TERN education rep) and Michael Connelly (TERN Severn regional rep).
Can the “important brain injury criteria” predict neurosurgical intervention in mild traumatic brain injury? A validation study. Lessard et al.
The aim of this paper was to validate criteria for important and clinically unimportant brain injuries in those with a minor traumatic brain injury (MTBI) and an acute abnormal finding on CT. These criteria were designed to predict which patients would go on to require neurosurgical intervention.
MTBI was defined as an acute brain injury with initial GCS of 13-15. An important brain injury (IBI) was defined as cerebral contusion ≥ 5mm in diameter, subarachnoid haemorrhage (SAH) ≥ 1mm thick, subdural haematoma (SDH) ≥ 4mm thick, epidural or intracerebral haematoma, intraventricular haemorrhage, depressed skill fracture, diffuse oedema, or pneumocephalus. A clinically unimportant brain injury (CUBI) was a solitary contusion <5mm in diameter, localised subarachnoid blood <1mm thick, SDH <4mm thick, isolated pneumocephaly, or closed depressed skull fracture not through inner table.
This was a single-centre study in a Canadian tertiary care trauma centre, with post-hoc analysis of patients over 18 presenting with a MTBI and an abnormality on initial CT between 2008-12. The study included patients referred to the trauma centre from regional hospitals. Unlike many similar studies, those on antiplatelet therapy and warfarin were included (unfortunately, likely due to the age of the data set, DOACs are not mentioned).
678 eligible patients were identified, of which only 11 were lost to follow up. The primary outcome was neurosurgical intervention. Those discharged on initial presentation without neurosurgical intervention were followed up at 4 weeks via a MTBI clinic and any subsequent attendances were reviewed to look for late adverse outcomes.
It was found that the IBI criteria identified those requiring neurosurgical intervention with 100% sensitivity [CI 95% 96.8–100], though only 34.8% specificity [CI 95% 30.8–38.8].
Many UK emergency departments will refer all patients with acute abnormalities on CT to neurosurgical teams for review and consideration for transfer. Due to their high sensitivity, these criteria could be used to rationalise referrals to tertiary neurosurgical teams from the emergency department by identifying patients who are extremely unlikely to require intervention, thereby reducing workload for both ED and neurosurgery clinicians, as well as reducing patient waiting times in ED. These criteria may also facilitate identification of patients who can be discharged from ED rather than admitted for prolonged observation.
 Lessard J, Cournoyer A, Chauny JM, Piette E, Paquet J and Daoust R. Can the ‘important brain injury criteria’ predict neurosurgical intervention in mild traumatic brain injury? A validation study. American Journal of Emergency Medicine. 2021. 38(3): 521-525. https://doi.org/10.1016/j.ajem.2019.05.043
Regional anesthesia for rib fractures: A pilot study of Serratus anterior plane block. Schnekenburger et al
This prospective observational pilot study aimed to assess the safety and efficacy of ultrasound guided serratus anterior plane block in adult patients with multiple rib fractures requiring opioid analgesia. It was conducted in a Level 1 major trauma center in a convenience sample of 20 patients. The primary outcome was feasibility of the procedure which included time to completion of the procedure, number of attempts and complications due to the procedure. Secondary outcomes included safety and efficacy of the procedure.
The median time to perform the procedure was 5.5 mins. 80% of procedures were performed at the first attempt. The median pain score was reduced from 6.5 to 3 at four hours post procedure. At least 40% of the patients achieved a 50% reduction in baseline pain within 30 mins of the procedure. Median length of stay in hospital was 5.9 days. No post-procedure complications, hospital acquired pneumonia or deaths were reported at the point of discharge.
Promising results are seen with this pilot study into serratus anterior plane blocks for pain control in rib fractures; if future larger studies mirror these findings, we could see an improvement in the analgesic management of these patients in the ED.
 Schnekenburger M, Mathew J, Fitzgerald M, Hendel S, Wais Sekandarzad and Mitra B. Regional anesthesia for rib fractures: A pilot study of Serratus anterior plane block. Emergency Medicine Australasia. 2021. https://doi.org/10.1111/1742-6723.13724
Effectiveness and safety of small-bore thoracostomy (≤ 20 Fr) for chest trauma patients: A retrospective observational study. Maezawa et al.
In 2018, the 10th edition of Advanced Trauma Life Support  recommended the use of smaller chest tubes (28-32 Fr) based on evidence demonstrating that smaller tube sizes did not reduce effectiveness of tube thoracostomy. The aim of this observational study was to investigate whether even smaller chest tubes could be safely used to treat traumatic chest injuries without compromising effective drainage.
This was a retrospective observational study performed in a single Japanese emergency department over a period of 5 years. It included all adult patients who required tube thoracostomy for traumatic pneumothorax, haemothorax or haemopneumothorax. All patients had either a 20 Fr chest tube or 8 Fr pigtail catheter placed using the blunt dissection method by an emergency physician or thoracic surgeon. The primary outcome was tube-related complications – tube obstruction, retained haemothorax or unresolved pneumothorax.
Out of 107 tube thoracostomies performed on 102 patients only 8 led to tube-related complications. There were 4 cases of retained haemothorax and 4 cases of unresolved pneumothorax, none of which were caused by tube obstruction. All 8 of these patients had effective drains but ultimately required further invasive intervention for definitive management of ongoing bleeding or air leak, such as embolization, thoracostomy or VATS.
Smaller chest drains could be used to provide effective drainage of traumatic pneumothorax and haemothorax while being less invasive and less painful for the patient. This was a small, observational, single-centre study without performing statistical analysis – but reported no incidences of tube obstruction out of the 107 tubes recorded. The optimal tube diameter for trauma chest drains remains unclear, but this paper suggests that further comparison between large and small-bore tubes is warranted.
 Maezawa T, Yanai M, Huh JY and Ariyoshi K. Effectiveness and safety of small-bore tube thoracostomy (<20Fr) for chest trauma patients: a retrospective observational study. American Journal of Emergency Medicine. 2020. 38(12) 2658-2660. https://doi.org/10.1016/j.ajem.2020.09.028
 Committee on Trauma. Advanced Trauma Life Support® Student Course Manual. 10th ed. Chicago: American College of Surgeons; 2018.
 Inaba K, Lustenberger T, Recinos G, Georgiou C, Velmahos GC, Brown C, et al. Does size matter? A prospective analysis of 28–32 versus 36–40 French chest tube size in trauma. J Trauma Acute Care Surg. 2012;72(2):422–7.
Propofol Target-Controlled Infusion in Emergency Department Sedation (ProTEDS): a multicentre, single-arm feasibility study. Burton et al.
This was a preliminary feasibility study which would go on to aid study design for a larger multicentre randomised control trial, comparing propofol target controlled infusion (TCI) with bolus administration for ED sedation.
The study was conducted across 4 EDs in Scotland and looked at 18-65 year-old patients presenting with acute traumatic anterior shoulder dislocation. They had a recruitment rate of around 20% with 25 patients recruited in total, of whom 2 were excluded.
Emergency consultants received training from an experienced anaesthetist via a 2-hour interactive online tutorial followed by practice setting up the TCI pump (BD Carefusion). Depth of sedation was recorded using Modified Observer’s Assessment of Alertness/Sedation Scale (OAA/S) with a target sedation level of 3. The study designers modified their dosing protocol after the first 5 patients as 2 out of the 5 did not achieve adequate sedation.
Mean time to achieve sedation of OAA/S 3 was 25 +/- 9 minutes, and mean time to reduction was 28+/- 10 minutes. This was compared to 10 +/- 6 minutes for standard bolus sedation in one of the participating centres. This is potentially due to the model used for TCI delivery – the study used Marsh (plasma concentration target) as opposed to Schneider (brain effector site target) and the authors propose that moving to the Schneider model may improve time to sedation in a further study.
In the 20 patients who achieved sedation of OAA/S 3, there was a 100% reduction rate of shoulder dislocation. Primary endpoints for the feasibility study were patient satisfaction, which was recorded using a visual analogue scale, and number of patients recruited vs number of patients screened. They have suggested that the future RCT will be designed to compare incidence of adverse events in bolus vs TCI administration, in order to establish whether there was a statistically significant improvement in safety associated with switching to TCI.
The study concludes that in this cohort, propofol TCI was acceptable to patients as a method of sedation in the ED. Propofol TCI is frequently discussed amongst ED practitioners as a potentially safe and useful tool to add to the skillset of ED clinicians. However, the timescale of sedation delivery alone, as found by this study, may prevent it becoming widely used in Emergency Departments in the UK for patients who are likely to tolerate procedures under entonox or who would be low risk for IV sedation in a bolus format.
For patient groups with a higher risk of adverse events during propofol sedation, such as elderly or co-morbid patients not included in this study, the question still remains whether these patients are appropriate for ED sedation regardless of the method of delivery. Whether we should be trying to increase the scope of ED practice to encompass sedation in those patients for whom TCI would potentially be shown to have a statistically significant improvement in safety in a larger RCT remains unclear.
 Burton F, Lowe D, Millar J, Corfield A, Watson M and Sim M. Propofol target-controlled infusion in emergency department sedation (ProTEDS): a multicentre, single-arm feasibility study. Emergency Medicine Journal. 2021. 38(3): 205-210. https://doi.org/10.1136/emermed-2020-209686
Musculoskeletal Ultrasonography to Diagnose Dislocated Shoulders: A Prospective Cohort. Secko et al.
This study aimed to determine the accuracy of a new point-of-care ultrasound (POCUS) technique for diagnosis of dislocation and fracture, determine the optimal glenohumeral distance, and the time to diagnosis compared to radiographic techniques.
This was a multicentre observational study using ultrasonography fellows and fellowship-trained physicians who used a novel posterior approach to the joint. Shoulder XR was used to confirm diagnosis. 65 patients were enrolled in the study. Specificity and sensitivity of POCUS to detect dislocations was 100% (95% CI 87% to 100%) and 100% (95% CI 87% to 100%) respectively. POCUS was 92% sensitive (95% CI 60% to 99.6%) and 100% specific (95% CI 92% to 100%) for non-Hill-Sachs/Bankart’s fractures of the humerus. Novel US technique for diagnosis was faster from triage than standard radiology with a median difference of 43 minutes.
Novel use of POCUS in diagnosing shoulder dislocations has a promising outlook with benefits of high accuracy, faster time to diagnosis and no radiation in comparison to routine radiographic diagnosis. This method would also allow for a check of successful relocation during the procedure.
Larger studies are required to determine the value of ultrasound in diagnosis of shoulder dislocation in the ED, however it seems that POCUS techniques have the possibility of becoming the primary means of diagnosis of dislocation in ED.
 Secko M, Reardon L, Gottlieb M, Morley E, Lohse M, Thode H and Singer A. Musculoskeletal ultrasonography to diagnose dislocated shoulders: a prospective cohort. Annals of Emergency Medicine. 2020. 76(2): 119-128. https://doi.org/10.1016/j.annemergmed.2020.01.008
Introduction of the low-risk ankle rule in a paediatric emergency department. Tormey et al. 
This paper looked at the introduction of the low risk ankle rule (LRAR) in a paediatric emergency department. The LRAR is a tool used in paediatrics which states if the child has only isolated tenderness and swelling over the distal fibula and/or adjacent lateral ligaments distal to the tibial anterior joint line an ankle x-ray is not necessary and a high-risk injury is very unlikely. This is in contrast to the widely used Ottawa ankle rules we use in UK emergency departments.
This was a single-site quality improvement project. In this paper, the authors initially conducted an audit of x-ray rates in ankle injuries and used the information gathered from that process to determine a baseline rate of ankle radiography and length of stay. They then introduced education sessions and created x-ray ordering prompts that encouraged clinicians to implement the LRAR. A re-audit was then performed.
Findings showed that of 969 patients who presented with ankle injuries in the pre-intervention period 90.7% went on to have an xray and the median LOS was 109 minutes. Post intervention, only 43.4% (40/92) of patients with ankle injuries went on to have an xray and median length of stay was 101 minutes. During the intervention arm the department saw 92 patients with ankle injuries, 9 patients were excluded and in 4 patients the clinician did not use the LRAR. When the LRAR was used (79 patients), 30 patients had a negative exam and went on to have x-ray (negative exam means that the LRAR cannot be applied). 49 patients had a positive LRAR exam and only one went on to have an x-ray due to parental concern. No clinically significant fractures were missed
The use of LRAR Criteria may safely and effectively reduce the rate of radiography in paediatric ankle injuries, without missing any clinically significant fractures.
 Tormey P, Calendar O, Fitzpatrick P, Okafor I, McNamara R and Kandamany N. Introduction of the low risk ankle rule to a paediatric emergency department. Injury. 2020. 51(3): 633-635. https://doi.org/10.1016/j.injury.2020.01.040
Risk of significant traumatic brain injury in adults with minor head injury taking direct oral anticoagulants: a cohort study and updated meta- analysis. Fuller et al.
This study looked at the outcomes of elderly patients who had suffered a minor head injury while taking a direct oral anticoagulant (DOAC). It was a single centre observational cohort study performed in a UK major trauma centre. It included 148 elderly patients (median age 82) who had suffered a mild head injury (GCS 14-15) while on a DOAC. The most common reason for injury was a ground level fall (96%) and the majority were asymptomatic (75%) at assessment. Almost all (90.5%) underwent a CT head. The primary endpoint was an adverse outcome within 30 days, this was defined as neurosurgical intervention, intracerebral haemorrhage (ICH) or death.
The overall risk of adverse outcome was 3.4% or 5/148 patients (95% CI 1.4%-8.0%). Of the five patients who suffered from an ICH one died within 30 days. All were admitted for observation and had their regular oral anticoagulation held. One was treated with prothrombin complex but none had a critical care admission or neurosurgical intervention. With the exception of taking a DOAC, the majority (75%) did not meet other NICE imaging criteria. No asymptomatic patient with a normal GCS and ground level fall had an adverse outcome (95% CI 0.0% – 4%). The authors also performed an updated meta-analysis using this new data which reported the weighted adverse outcome risk as 3.2% (n = 29/787 95% CI 2.0%-4.4%).
This study indicated that the risk of adverse outcome following mild head injury while taking a DOAC is low, especially in patients with a normal GCS, no symptoms and a non-dangerous mechanism of injury. It also reported that only one patient received a substantive change in management as result of CT head imaging. These findings would support shared patient – clinician decision making rather than routine imaging after minor head injury while taking DOACs.
 Fuller G, Sabir L, Evans R, et al. Risk of significant traumatic brain injury in adults with minor head injury taking direct oral anticoagulants: a cohort study and updated meta-analysis. Emergency Medicine Journal 2020;37:666-673.http://dx.doi.org/10.1136/emermed-2019-209307