Background: Out-of-hospital cardiac arrests continue to be a challenge in emergency medicine. Transport to a cardiac arrest centre with ongoing resuscitation can be reasonable in selected patients. As time is crucial, choosing a suitable mode of transport can be vital. This study was set out to investigate the association of transport mode (ground vs helicopter) with the 30-day outcome in patients with ongoing resuscitation on admission. Methods: We included patients ≥18 years with a non-traumatic out-of-hospital cardiac arrest between January 2013 and May 2021 from the prospective Vienna Clinical Cardiac Arrest Registry. Only ongoing resuscitations on admission without a sustained return of spontaneous circulation (ROSC) were included. Transport was performed either by ground emergency medical service (GEMS) or helicopter (HEMS). The Cerebral Performance Category (CPC) scale on day 30 was defined as primary outcome. We used a logistic regression to investigate the association of transport mode with the dichotomised CPC score (good: 1-2 vs poor: 3-5). Age (years), gender (male/female), witnessed cardiac arrest (yes/no), basic life support (BLS, yes/no), first analysed rhythm (shockable yes/no) and cumulative epinephrine dose (milligram) served as covariates. Results: 725 Patients were included in this study (611 [84.3%] GEMS and 114 [15,7%] HEMS). Age (57.5 ± 14.5 years vs 54.3 ± 15.6 years), the rate of female patients (22.9 % vs 18.4 %), the rate of witnessed cardiac arrests (80.9% vs 75.4%) as well as the cumulative epinephrine dosage (8.5 mg ± 3.9 vs 8.1 mg ± 3.9) did not differ between the GEMS and HEMS group. The BLS (70% vs and 84.2 %, p = .002) and the rate of shockable first rhythm (57% vs 43.5%, p=.008) were higher in the HEMS group. A CPC score of 1-2 was achieved in 3.3% in the GEMS group and 7.0% in the HEMS group (p = .057). The mean time from collapse to arrival at the hospital was 53.6 min (± 21.4) in the GEMS and 60.27 min (± 20.11) in the HEMS group (p = .07). In logistic regression, age, witnessed cardiac arrest, first analysed rhythm, and cumulative epinephrine dose were significantly associated with good neurological outcome after 30 days (p ≤ .03). However, gender (OR 0.64, 95%CI 0.62 - 1.91, p = .50), BLS (OR 0.49, 95%CI -0.65 - 1.63, p = .22) and the mode of transport (OR 0.53, 95% CI -0.45 - 1.51, p = .21) did not show a significant association. Discussion & Conclusions: In the present study, no significant association between the mode of transport and good neurologic outcome after 30 days was found in patients with ongoing resuscitation on admission. This is especially interesting as some characteristics of the HEMS patients would suggest a better outcome (higher BLS and shockable first rhythm rate). Interestingly, our data showed no benefit in transport time. This might be due to logistical reasons (e.g. the greater distance travelled or difficulties stemming from landing in urban areas). In summary, our data do not show a ubiquitous benefit of airborne transport in patients with ongoing resuscitation.

Funding: This study did not receive any specific funding. Ethical approval and informed consent: The study was approved by the Ethics Committee of the Medical University of Vienna (2376/2020).

Apnoeic oxygenation during endotracheal intubation. A tertiary Emergency Department’s airway registry Authors Names: Izak Petrus Scholtz, Etimbuk Umana, Francis O’Keeffe Authors Affiliation: Emergency Department, Mater Misericoridiae University Hospital, Dublin, Ireland Background Endotracheal intubation during emergency airway management is an integral part of the management of the critically unwell patient that presents to an emergency department (ED). Intubations, performed by rapid sequence intubation, is a potentially lifesaving intervention in this cohort of patients. However, they present with altered physiological parameters and elevated metabolic demands which predispose them to hypoxia and post intubation complications. The use of apnoeic oxygenation in combination with standard care during intubation has been described as a potential buffer against hypoxaemia during and after endotracheal intubation. Apnoeic oxygenation refers to the use of nasal cannula to provide passive flow of oxygen against a gradient to oxygenate alveoli and to delay hypoxaemia in apnoeic patients. The aim this study was to evaluate the trend in the use of apnoeic oxygenation during endotracheal intubation in our institution and the role of apnoeic oxygenation on post intubation desaturation. Method A retrospective audit was undertaken of the Mater Misercoridea University Hospital Emergency Department’s Airway registry which evaluated data collected prospectively for the period of August 2018 until June 2022. The data used was recorded contemporaneously by clinician performing the endotracheal intubation. Intubations were excluded from the audit if it was performed during cardiac arrest. Data recorded in the registry included the following: age, patient gender, weight, indication for intubation, speciality performing intubation, anticipated difficult airway, formal airway assessment, use of standard care or apnoeic oxygenation and first pass success rate. Post intubation desaturation was defined as SpO2<93%. Results Over a 45-month period, 326 endotracheal intubations were recorded in MMUH. When cardiac arrest related aetiology excluded, only 220 meet inclusion criteria. Of the 220, majority were male (n=149, 68%) and the indication for intubation was mostly medical (n=174, 79%). Median age was 51,5 years (37,75 - 65). Majority of intubations were performed by EM clinicians (n=192, 87%) Of the 220 patients, 99 (45%) patients had apnoeic oxygenation, while 121 (55%) received standard care during intubation. The trend of apnoeic oxygenation use increased over the last 4 years (Year 1-(41%), Year 2-(34%), Year 3-(55%) and Year 4-(55%)). In terms of post intubation desaturation, the apnoeic oxygenation group had a 5% rate of post-intubation desaturation compared to 8% in the standard care group. Discussion In our study intubations were more commonly performed by EM physicians. There was also an increased trend in the use of apnoeic oxygenation and a lower rate of post intubation desaturation. Our findings compare to other international studies. Apnoeic oxygenation should be used in daily practice in all disciplines performing intubations in critically ill patients in the ED.

Background: Persistent microperfusion alterations after return of spontaneous circulation (ROSC) are associated with poor survival. During cardiac arrest, sufficient microperfusion is associated with survival and highly correlates with coronary perfusion pressure in animal models. Capillary refill time (CRT) is a simple method to assess microperfusion of the skin. Thus, we hypothesized that CRT during cardiopulmonary resuscitation (CPR) might be a predictor for ROSC. Methods: We performed a pilot study with prospective inclusion from 3/3/2021 to 3/31/2022 with 30 day follow up. During the study period, a special research car was dispatched to all out of hospital cardiac arrests in the city of Vienna during weekdays additionally to regular EMS units. This car was staffed with a physician and a paramedic. Patients ≥18 years with witnessed cardiac arrest and ongoing CPR were included to the study. Exclusion criteria were peripheral arterial disease, Raynaud’s disease, hypovolemia, severe hypo- or hyperthermia and logistic factors (e.g. limited space, study team had to perform CPR). The primary endpoint was ROSC. CRT was measured by applying firm pressure to the ventral surface of a finger’s distal phalanx and the earlobe with a glass microscope slide. Pressure was increased until the skin went blank and was then maintained for 10 seconds. The time for return of the initial skin color was registered with a chronometer. CRT >10 seconds was defined as “no refill”. Measurements were repeated every two minutes until ROSC, termination of resuscitation efforts or decision to transport to hospital with ongoing CPR. The first and last measured CRT on finger and earlobe are depicted with median and interquartile range. For statistical analyses, values were classified in quartiles. The association between ROSC and CRT was compared using Fisher’s exact test and estimated with exact logistic regression. Cochran Armitage (CA) was used for trend analyses. Results: 50 patients (mean age 73.1 years, 26% female, 30% ROSC) could be included. Neither CRT on the earlobe (first CRT: noROSC 3 [3-4] vs. ROSC 4 [3-4] sec., p=0.431, CA: 0.211; last CRT: noROSC 3 [3-5] vs. ROSC 3 [3-4] sec., p=1.0, CA: 0.845) nor on the finger at first measure (noROSC 7.5 [6- >10] vs. ROSC 6 [5-9] sec., p=0.306, CA: 0.265) were statistically significant different. CRT at the last measurement on the finger was shorter in patients with ROSC compared to noROSC (6.5 [4-9] vs. >10 [9- >10], p=0.007, CA: 0.129). At last measurement on the finger, the odds ratio for achieving ROSC with CRT >10sec was 0.099. Discussion & Conclusions: In this study, only the last measured CRT on the finger was shorter in patients with ROSC. Upper venous congestion may influence the findings on the earlobe, whereas microperfusion alterations on the finger might only evolve over time. However, CRT may be of use in the multifactorial decision to terminate a resuscitation. Acknowledgements: We want to thank all Field Supervisors of the Emergency Medical Service of Vienna for their great support! We further want to thank C. Kienbacher, I.A.M. Magnet, G. Gelbenegger and A. Nuernberger for patient inclusion!

Trial registration: clinicaltrials.gov (NCT04791995). Funding: This study was funded by the Medical Scientific Fund of the Mayor of the city of Vienna (AP21171) Ethical approval: The study was approved by the Ethics Committee of the Medical University of Vienna (EK 2427/2020), written informed consent was obtained after patients regained consciousness.

Background: Cardiopulmonary resuscitation (CPR) skills are of the great importance to healthcare providers. CPR skill decay is a well-known phenomenon and occurs when skills are not used regularly. Recently a new law requiring Lithuanian healthcare providers to regularly take part in mandatory CPR training sessions was passed. CPR refresher courses were conducted for healthcare teams from hospitals and outpatient clinics in Vilnius district. Our team surveyed CPR course attendees to collect various demographic, CPR experience data and measured self-assessed CPR skills score. Goal of our research was to look for factors associated with self-assessed CPR skills score. Methods: Prospective observational study was conducted from 2022 January until May. A questionnaire was constructed evaluating gender, age, specialty, type of work settings, years in clinical practice, previous clinical experiences in CPR, previous training in CPR. Participants had to evaluate their CPR skills and likelihood of taking part in CPR outside of clinical environment using 5 item score. Data was processed using IBM SPSS statistical package. Difference between the variables was reliable if p<0,05. Results: A total number of 201 cases were analysed. Mean age was 53.3 (±9.64) with 96.5% female (n=194), 3,5% male (n=7). 24,4% physicians, 62,2% nurses and 14% non-clinical staff were taking part in survey. Mean working experience in years 28.1 (±10.33). 64% working in outpatient setting, 36% in hospitals – 7 % of them in intensive care units. 142 participants had no previous clinical CPR experience. Best self-evaluated CPR skill was chest compressions (mean=2,67) and worst was defibrillation (mean=1,97). We found that factors associated with better self-assessed CPR skills were: working in hospital, time form last CPR course, times of CPR performed in clinical setting, time from last CPR case (p<0,05). Higher self-assessed CPR skills score was associated with higher self-assessed likelihood to perform CPR outside of clinical environment (p<0,05). Discussion & Conclusions: Educational effort directed to healthcare personnel working outside of hospital environment could improve self-confidence in resuscitation skills thus improving likelihood of participation in CPR outside of clinical environment. CPR course instructors could put more emphasis on defibrillation skills to improve overall self-confidence in ability to resuscitate.

-

Background: This study uses machine learning and multi-center registry data for analyzing the determinants of favorable neurological outcome in the out-of-hospital cardiac arrest (OHCA) patient and developing its decision support systems for various subgroups. Methods: Data came from Korean Cardiac Arrest Research Consortium registry with 2679 OHCA patients aged 18 or more with the return of spontaneous circulation (ROSC). The dependent variable was favorable neurological outcome (Cerebral Performance Category scores 1-2) and 68 independent variables were included, e.g., first monitored rhythm, in-hospital cardiopulmonary resuscitation (CPR) duration and post-ROSC pH. The random forest was used for identifying major determinants of favorable neurological outcome and developing its decision support systems for various subgroups stratified by major variables. Results: Based on random forest variable importance, major determinants of OHCA outcome were in-hospital CPR duration (0.0824), in-hospital electrocardiogram on emergency room arrival (0.0692), post-ROSC pH (0.0579), prehospital ROSC before emergency room arrival (0.0565), coronary angiography (0.0527), age (0.0415), first monitored rhythm (EMS) (0.0402), first monitored rhythm (community) (0.0401), early coronary angiography within 24 hours (0.0304) and scene arrival to CPR stop (0.0301). It was also found that patients can be divided to 6 subgroups in terms of prehospital ROSC and first monitored rhythm (EMS) and that a decision tree can be developed as a decision support system for each subgroup to find its effective cut-off points regarding in-hospital CPR duration, post-ROSC pH, age and hemoglobin. Conclusions: We identified the major determinants of favorable neurological outcome in successfully resuscitated OHCA patients using machine learning. This study demonstrated the strengths of the random forest as an effective decision support system for each stratified subgroup (prehospital ROSC and first monitored rhythm by EMS) to find its own optimal cut-off points for major in-hospital variables (in-hospital CPR duration, post-ROSC pH, age and hemoglobin).

This work was supported by the National Research Foundation of Korea grant (No.2019R1A2C1007110)

Background: SARS-CoV-2 related pneumonia is characterized by moderate-to severe hypoxemia often associated with hypocapnia. Currently few studies investigated the possible role of hypocapnia as a severity predictor in SARS-CoV-2 related pneumonia. Aims: evaluate if hypocapnia can predict the need for non-invasive mechanical ventilation (NIV) in subjects with SARS-CoV-2 related pneumonia. Matherial and methods: we prospectively included in the study 52 subjects with moderate-severe SARS-CoV-2 related pneumonia. The end point of the study was the need for non-invasive mechanical ventilation. All the data were collected at admission to the Emergency Department and we included: clinical and laboratory data, blood gas analysis and lung ultrasound. All blood gas analysis were performed in room air. Lung ultrasound (LUS) was performed using a convex probe while patients were in sitting position and we used the 12-zone protocol, assigning a score from 0 to 3 for each zone for calculating the total LUS score. Results: of the 52 subjects included 30 were males and 22 were females; mean age was 61 ± 12. 33 subjects (63,4%) reached the end point. Mean time between admission and NIV start was 20,3 ± 13 hours. At univariate analysis we did not find any significant association among age, gender, number of comorbidities, systolic and diastolic blood pressure, heart rate, respiratory rate, white blood cells count, platelets, blood urea, creatinin, glomerular filtration rate, d-dimer, aspartate aminotransferase (AST), alanine aminotransferase (ALT), serum lactic dehydrogenase (LDH), and C-reactive protein (CRP), arterial oxygen partial pressure (PaO2), LUS SCORE and the need for NIV, but we observed a statistically significant inverse association between carbon dioxide partial pressure (PaCO2) and the need for NIV (OR 0,82, CI 95% 0,689-0,976, p .025). At multivariate analysis we observed that PaCO2 predicted the need for NIV independently from age, gender, number of comorbidities, d-dimer, CRP, PaO2 and LUS SCORE (OR 0,838, CI 95% 0,710-0,988, p .035) Conclusions: our data suggest that hypocapnia could be a good predictor of rapid respiratory failure worsening in subjects affected by SARS-CoV-2 related pneumonia independently from other already known predictors of unfavourable outcome. We speculate that hypocapnia could reflect the occurrence of a spontaneous respiratory pattern characterized by deep and frequent breathing, typical of the initial phase of the disease, which can generate an excessive swing of transpulmonary pressure inducing a real risk of producing a self-induced lung injury (P-SILI). We observed that the majority of subjects need NIV support after few hours from admission suggesting that hypocapnia allows the early recognition of the presence of the respiratory pattern which can cause the P-SILI, thus allowing the early recognition of the subjects at higher risk of rapid deterioration. Further multicentric studies are needed for validating these data on greater populations.

Introduction: The COVID-19 pandemic has imposed pressure for changes in the care provided in emergency departments (ED). We sought to explore the perceptions of different types of ED workers involved in resuscitation teams regarding these practice changes, aiming to understand the barriers and facilitators affecting their adoption. Methods: We conducted this exploratory qualitative study using a narrative analysis approach. Participants were members of a multidisciplinary resuscitation team in an urban tertiary care academic ED (nurses, orderlies, respiratory therapists, emergency physicians, and case managers). They were recruited using a purposive sampling technique to maximize variation and aim for even representation of occupation and shift type. Focus groups with workers in the same occupation were interviewed in a semi-structured manner. The data were analyzed thematically using an inductive approach, combined with a structuring framework (Theoretical Domains Framework). Two members of the research team corroborated the coding and analysis of the data. The selected themes and quotes were submitted to the participants for validation. Results: Thirty-three participants took part in six focus groups in March and April 2021. The changes described were related to protective measures, site organization, team functioning and care protocols. Participants identified barriers to their adaptation such as: (1) information overload due to multiple sources of information and frequent changes, (2) protocols established without fully considering the reality of the field, (3) communication challenges due to fragmentation of teams in isolation spaces and use of protective equipment, (4) perceived negative impact of new protocols and measures on patient care. Facilitating factors were: (1) identification of trusted peers to filter information, (2) sense of professional identity as an ED worker, (3) collaboration in resuscitation situations, as well as in the reorganization of work, and (4) peer support. Conclusion: This project incorporated input from participants from often underrepresented groups of ED professionals. Many barriers were encountered by resuscitation team members in coping with the changes brought about by the pandemic. Potential solutions to consider include: (a) strategies to streamline the transmission of information from decision makers to teams, (b) intermediaries to bridge the gap between the field and decision-making bodies, (c) improvement of communication in the resuscitation room, and (d) the implementation of formal peer support measures.

Chaire Docteur Sadok Besrour grant Association des spécialistes en médecine d'urgence du Québec grant This study was supported by the Fonds de recherche des Urgentistes de l’hôpital Sacré-Cœur de Montréal.

Background: Coma is a disorder of consciousness, a true medical emergency determined by various pathologies whose identification and treatment in a timely manner is critical for the patient's outcome. The management begins with stabilization of the vital functions, adequate breathing and circulation, followed by a thorough physical examination to identify the possible causes. Definitive treatment and prognosis depends on the treatable disorders and reversible causes. Methods: We conducted a retrospective observational study on a total of 129,028 patients presented at the Emergency Unit of Sibiu in 2019 and 2021, a year before and one during the pandemic, using our means and frequencies to characterize and describe the susceptible population of risk most liable to this pathology. Results: The total number of presentations at the Emergency Unit remained similar in 2019 and 2021. Out of a total of 361 patients with comas, the number of cases decreased from 226 cases in 2019 to 135 cases in 2021. The age distribution during the study was the following: Age category 0-20 years old: 4 cases (1,10%); Age category 21-40 years old: 26 cases (7,20%); Age category 41-60 years old: 80 cases (22,16%); Age category over 60 years old: 251 cases (69,52%). The predominant pathologies that determined this condition were represented by: Strokes: 33 cases in 2019 and 22 cases in 2021; Traumas: 18 cases in 2019 and 6 cases in 2021; Cardio-respiratory arrests: 103 cases in 2019 and 62 cases in 2021; Metabolic acidosis: 8 cases in 2019 and 3 cases in 2021; Hypercapnia: 10 cases in 2019 and 8 cases in 2021. Discussion & Conclusions: The number of presentations in the Emergency Unit remained similar in both years, even though we are speaking of a year before the pandemic hit and one in the midst of it. The number of comas caused by strokes decreased from 2019 to 2021 although we would expect the number to increase due to the pro-coagulant effect of Covid-19. Patients do not have as many routine investigations as during the time before the pandemic and they no longer had access to preventative methods and updated medication. Hospitals are clogged and access is becoming increasingly difficult, which is why some patients, who had a stroke at home, had neglected it due to mild onset symptoms. The second stroke, or maybe even the first because of the lack of preventive methods, that took place, were much more severe and not leading into a coma but directly into cardio-respiratory arrest. As we expected, it is still the prerogative of the elderly to have a stroke. Chronic respiratory pathologies, such as chronic obstructive pulmonary disease, that have evolved into hypercapnia, remained constant. Traumas, especially in young people, are still lower in number in 2021, thanks to the pandemic traffic restrictions and social distancing, which significantly lowered the risk of becoming a victim in road accidents.

Background: The impact of myosteatosis on septic patients had not been fully revealed. The aim of study was to evaluate the impact of myosteatosis area and percentage on the 28-day mortality in patients with septic shock. Methods: We conducted a single center, retrospective study from prospectively collected registry of adult patients with septic shock who presented to emergency department and performed abdominal computed tomography (CT) from May 2016 to May 2020. Myosteatosis area defined as the sum of low attenuation muscle area and intramuscular adipose tissue at the level of the third lumbar vertebra was measured by CT. Myosteatosis percentages were calculated by divide myosteatosis area into total abdominal muscle area. Odds ratios (ORs) and 95% confidence intervals (CIs) for 28-day mortality were estimated using a multivariate logistic regression model. Results: Of the 896 patients, 28-day mortality was 16.3%, and abnormal myosteatosis area was commonly detected (81.7%). Among variables of body compositions, non-survivors had relatively lower normal attenuation muscle area, higher low attenuation muscle area, higher myosteatosis area and percentage than that of survivors. Trends of myosteatosis according to age group were different between male and female group. In subgroup analysis with male patients, the multivariate model showed that the myosteatosis percentage (adjusted OR 1.02 [95% CI 1.01 – 1.03]) was an independent risk factor for 28-day mortality. However, this association was not evident in female group. Conclusions: Myosteatosis was common and high myosteatosis percentage was associated with short-term mortality in patients with septic shock. Our results implied that abnormal fatty disposition in muscle could impact on increased mortality. And this effect was more prominent in male patients.

This study was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI18C1216).

Introduction: Transesophageal echocardiography (TEE) is an emerging tool that can aid emergency physicians in treating patients in cardiac arrest and undifferentiated shock. TEE can aid in diagnosis, resuscitation, identify cardiac rhythms, guide chest compression vectors, and shorten sonographic pulse checks. The objective of this study is to evaluate proportion of patients who underwent a change in their resuscitation management as a result of emergency department resuscitative TEE. Methods: This was a retrospective cohort study that took place at an academic hospital in Toronto, Canada of all patients who underwent ED resuscitative TEE from 2015-2019. The primary outcome was the proportion of patients who underwent a change in their resuscitation management as a result of resuscitative TEE. Secondary outcomes were change in working diagnosis, complications, patient disposition, and survival to hospital discharge. Results: 25 patients (median age 71, 40% female) underwent ED resuscitative TEE. All patients were intubated prior to probe insertion. The most common indication for resuscitative TEE was cardiac arrest (16/25) followed by undifferentiated shock (7/25) and post-cardiac arrest (2/25). Resuscitative TEE was performed by senior emergency medicine residents or ultrasound fellows under direct supervision in 10 cases. Probe insertion was successful in all 25 examinations (100%) with difficult insertions occurring in 9/25. Adequate TEE views were obtained for every patient. The most commonly obtained TEE views were the mid-esophageal four chamber (100%), mid-esophageal long axis (100%), mid-esophageal descending aorta (100%), trans-gastric short axis (96%), and mid-esophageal bicaval (68%). After resuscitative TEE, the management changed in 76% (N=19) and information was diagnostically influential in 76% (N=19) of patients. Therapeutic recommendations included guidance of hemodynamic support with volume (8/25) or vasoactive medications (6/25), decision to transfer the patient to the cardiac catheterization lab (3/25), and decision to terminate resuscitation (3/25). The most common diagnostic contributions included hypovolemic shock (5/25), cardiogenic shock (4/25), pulmonary embolism (4/25), cardiac standstill (3/25), and acute coronary syndrome (2/25). Ten patients died in the ED, 15 were admitted to hospital, and eight survived to hospital discharge. There were no immediate complications (0/15) and two delayed complications (2/15), both of which were minor gastrointestinal bleeding. Conclusions: Resuscitative TEE is emerging as a valuable diagnostic and therapeutic tool for patients with cardiac arrest and undifferentiated shock in the ED. It is a relatively new emergency medicine modality with the first use described in the ED in 20085. It has been shown that it can be relatively easily taught to operators for resuscitations in the ED2,10. There is limited published evidence on the use of ED TEE and this study contributes important data to the literature. In this study we found that the use of ED resuscitative TEE was associated with significant therapeutic changes in critically ill patients and resulted in a higher rate of adequate cardiac visualization than TTE alone. There was a low complication rate.

Background: Current decision rules for minor head trauma (MHT) are reliable when applied to a population consistent with original trials, but their effectiveness decreases if more heterogeneous series of patients were considered. Although infrequent, clinical relevance of complications after MHT makes indispensable effective and appliable to all patients guidelines . The aim of this study was to develop a statistical model able to predict the individual risk of intracranial bleeding in patients suffering from MHT. Methods: In this prospective observational study all patients consecutively observed for MHT from January 2018 to June 2019 in the Emergency Department of University Hospital of Verona were considered. Inclusion criteria were: MHT defined as Glasgow Coma Scale 14 or 15; age > 18 years; CT scan performed within 48 hours from trauma. All clinical features and common risk factors were reported at admission. Main outcome was CT report of any intracranial lesion. Secondary outcomes were need for hospitlization, neurosurgery or neurological intensive care support. Statistical analysis was aimed to create and validate two prediction tools: risk score and nomogram. All risk factor associated with outcome in the univariate analysis (signficance level p<0.1) were submitted to multivariate analysis by Logistic Regression. Those variables confirmed as significant predictors (accuracy level at least 0.5%) were included in the final model. We further validated the risk score by means of discrimination and calibration models. Discrimination was calculated through area under receiver operating characteristics (ROC) curve and calibration by calibration plot. The nomogram was based on multivariate logistic model and translated the probability of intracranial lesion into a risk rate per cent. Nomogram was developed using a statistical software package and also validated by discrimination and calibration models. Results: In this study we enrolled 3722 patients (1826 males, 1896 females, mean age 74 years). CT scan positive for intracranial bleeding was reported in 316 patients (8.5%), but only 28 (0.75%) needed for neurosurgery. Independent variables proved significant after multivariate analysis were: dangerous mechanism, anticoagulant therapy, loss of consciousness, amnesia, repeated vomit, suspected fracture, GCS<15 (all p<0.001), neurological deficit (p=0.001), signs of trauma (p=0.003). The final score (range 0-26) was the total of single variable score derived from its statistical weight. Mean outcome was significantly related to increasing score, with concordance index in ROC curve equal to 0.809 (p<0.05). In 82% out of 11 patients with score > 12 CT scan was positive (specificity 0.99, positive predictive value 0.82). The 95.4% of 769 score < 3 patients and all of 71 completely risk-free ones presented a negative CT scan (sensibility 1.0, negative predictive value 1.0). Nomogram reached high accuracy level (concordance index 0.802, p<0.05). In ROC curve probability scores correlated with defined specificity and sensibility values. The 76.5% of patients with probability > 0.8% suffered from intracranial bleeding (specificity 0.99, positive predictive value 0.76), while in patients with probability < 0.2% (87.9%) specificity rate increased to 92.4%. Conclusions: Our combined statistical model appears a reliable tool to assess the individual risk of intracranial bleeding after minor head trauma.

Trial Registration of the study is TCM-NOMO_RET (889 CESC) The study received no fund

Background: Right upper quadrant abdominal pain is a common presentation for a general surgery on-call. Differentials include choledocholithiasis, for which ASGE guidelines utilize a risk scoring system to determine appropriate ongoing investigations and treatment modalities. Their scoring system categorizes patients with suspected common bile duct (CBD) stones into low, intermediate, and high risk of choledocholithiasis, and based off risk stratification, patients are either advised to proceed straight to ERCP followed by cholecystectomy or for MRI/Endoscopy prior to any therapeutic intervention. Our audit reviewed the ASGE 2019 scoring system and management guidelines against current practices at a tertiary hospital to assess how well common practice matches guideline recommendations for the management of choledocholithiasis based on risk scoring. Method: An independent audit across a three-month period was conducted of all patients presenting to a tertiary general surgical on-call with symptoms suggestive of choledocholithiasis (pain, jaundice, fevers). Primary quantitative and qualitative data was collected from electronic medical records for each patient, and this was used to retrospectively risk stratify each patient into low, intermediate, or high risk. The proceeding management that each patient received was analyzed against the recommended guidance, to assess compliance of current clinical practice. Results: In total 83 patients were included. 23% categorized as, 58% as intermediate and 19% as high risk. The low-risk group had the greatest compliance to guidance; 89% of patients either had or were listed for a cholecystectomy. 75% of patients categorized as high risk went onto have an ERCP prior to consideration for cholecystectomy, without any further imaging (MRI or EUS). However, 50% had an MRI first (no one had an EUS) and for 75%, the results corroborated the initial clinical suspicion. The same proportion of patients, with or without further imaging, had an ERCP followed by plan for cholecystectomy. Of this group, only 1/3 of the patients had either complications or required repeat procedures. In the intermediate-risk group, only 38% of the group had further imaging and 4 confirmed to have stone disease. In those patients that did not have further imaging, 53% did not have a cholecystectomy either. Just under half of these were because intervention was considered too high risk. Age >55 scores immediate risk regardless of other factors, and a sub-group analysis of those only scoring immediate risk based on age was conducted. All patients had either an USS, CT or MRCP. There were no findings suggestive of choledocholithiasis on any imaging in this cohort and only 11% had an elective laparoscopic cholecystectomy either done or listed. Conclusions: Overall, our audit showed that risk stratification is relatively safe and inclusive of appropriate patients with suspected choledocholithiasis presentation and current management does match a risk-scoring approach in certain aspects. However, some aspects of the criterion for each risk-strata may need revising and most often patients do not always fit into one risk group throughout their presentation and nor does solely basing on specific criteria acknowledge for comorbidities and frailty in an ever-aging population, which will impact management plans, regardless of perceived risk.

An Assessment of the Management of Choledocholithiasis Based on Risk Scoring Guidelines at a Tertiary General Surgery Referral Centre

Introduction: Falls are a major problem in the older emergency department population. Falls occur frequently after the age of 65 and account for 73% of cases of major trauma in that age group. They are associated with reduced mobility, functional decline, and death. Prognostic markers to distinguish between patients at high risk of death or adverse events and those who would benefit from rapid discharge would facilitate the time-consuming work-up in the emergency department (ED) and is therefore of utmost interest. D-dimer levels are commonly used to rule out thromboembolic disease and pulmonary embolism, while troponin T and I are used as a sensitive marker for myocardial infarction. It has been suggested that D-dimer might be helpful as a non-specific prognostic marker. This study aimed to determine the rate of elevated troponin levels in older patients presenting to the ED with falls and their prognostic value. Additionally, we aimed to evaluate whether D-dimer levels are predictive of 30d-mortality in the same population. Methods: This trial was conducted as a prospective international multicentre, cross-sectional observational study. Data collection was performed from November 2014 until January 2018. Study centres were University Hospital Basel, Charité Berlin (tertiary care hospitals), Hospital Bruderholz, and Hospital of Liestal (regional hospitals). Patients older than 65 years presenting to the ED within 24 hours after a fall and giving informed consent, were enrolled in the study. Upon presentation, Demographic baseline data, vital signs, D-dimer, and troponin T and I levels were determined. Follow-up analyses were performed after 30 days, 90 days, 180 days, and 1 year. The primary outcome of this study was 30-day mortality; the secondary outcomes were mortality after 30 days, 90 days, 180 days, and 1-year mortality. Results: Of 825 screened patients, 587 patients could be included in the study. 226 had missing D-dimer levels and 5 were lost to follow-up resulting in a final study population of 356. The median age was 83 years [IQR 78, 89], and 236 (66.3%) were female. Mortality rates were: 3.1% for 30 days, 5.1% for 90-days, 7% for 180 days and 12.1% for 1 year. 321 (90.2%) patients had elevated D-dimer levels with the regular cutoff (<500 ng/mL) applied while the age-adjusted cutoff (patient's age * 10 ng/L) resulted in 281 (78.9%) patients. None of the 11 Non-survivors after 30 days and the 18 Non-survivors after 90 days had a D-dimer below the regular or adjusted cutoff. Troponin data will be presented at the conference. Discussion: To our knowledge, this trial is the first prospective study investigating the mortality of older patients presenting to the ED after a recent fall - or the prognostic value of biomarkers regarding mortality in this population. We could confirm the generally high rate of elevated D-dimer levels in older patients questioning the value of the defined cut-offs for this age group. On the other hand, a low D-dimer level seems to be a strong predictor of survival in the next 30 and 90 days.

This study did not receive any specific funding.

Introduction: The introduction of high-sensitivity cardiac troponin (hs-cTn) assays has enabled earlier rule-out of acute coronary syndromes (ACS) in the Emergency Department (ED), often with a single blood test at the time of arrival. However, all current decision aids and pathways still leave some patients with uncertain diagnoses in the ‘observation zone’. Computed tomography coronary angiography (CTCA) could be used to resolve uncertainty for patients in the ‘observation zone’. In January 2018, we implemented a novel ambulatory CTCA service for such patients. With this pathway, patients in the ‘observation zone’ after applying the T-MACS decision aid were potentially eligible for CTCA via an Ambulatory Care Unit (ACU), as an alternative to hospital admission. Eligible patients were identified in the T-MACS electronic application and the referral procedure was ‘pushed’ directly to clinicians. We aimed to investigate the uptake of this service and its effects on patient outcomes including length of stay and the use of percutaneous coronary intervention (PCI). Methods: We conducted a retrospective, single-centre cohort study. From January 2018, patients were eligible for CTCA if they were in the ‘observation zone’ (moderate risk with T-MACS; troponin <99th percentile; no acute ECG ischaemia). The data for all patients with suspected ACS were captured prospectively using the T-MACS electronic application. We extracted data for consecutive patients in the ‘observation zone’ both prior to implementing ambulatory CTCA (June 2016 – December 2018) and post-implementation (January 2018 – January 2020). The primary outcome was adherence, defined as the number of eligible patients who received an ambulatory CTCA. Secondary outcomes include location of stay, length of stay, time to CTCA, and number of PCI. We summarised data using descriptive statistics. Results: We identified 1,341 patients in the ‘observation zone’ (524 pre-implementation of ambulatory CTCA; 817 post-implementation; 820 [61%] were male; mean age 57 [SD 14] years; 743 [55%] white; 91 [7%] black; 410 [31%] Asian; 97 [7%] other). Pre-implementation, 53% (n=280) of these patients were admitted to inpatient beds. Post-implementation, 68% (n=559) were admitted to inpatient beds, while 32% (n=258) were admitted to the Ambulatory Care Unit. However, only 7 (1.3%) eligible patients received ambulatory CTCA after the service was implemented. A further 25 patients in the ‘observation zone’ received CTCA from an inpatient area. For patients who received an ambulatory CTCA in adherence with local guidance, the median time to CTCA from admission was 1 (IQR 0.75-3) day. A total of 5 (71.4%) of the ambulatory CTCAs were abnormal, identifying significant coronary artery disease. PCI was undertaken in 83 (6.2%) patients overall. Three patients (4.3% of all those undergoing CTCA) with abnormal CTCA underwent PCI, although only one had followed the ambulatory CTCA pathway. Conclusions: Despite providing a novel ambulatory CTCA pathway for ‘observation zone’ patients with electronic decision support to identify eligible patients, our two-year evaluation found that clinicians continued to admit patients to inpatient areas. This indicates suboptimal utilisation, potentially because of a lack of confidence from clinicians. Future work should focus on identifying barriers to adherence.

Trial Registration: Not applicable as this is a service evaluation and there is no appropriate register. Funding information: This study did not receive any specific funding.

Background Acute upper gastrointestinal bleeding (AUGIB) is one of the major causes of morbidity and mortality in the geriatric population in emergency departments (EDs). Therefore, it is important to rapidly identify high risk group in this vulnerable population and commence aggressive resuscitation and proper medical intervention. The aim of this study is to compare the ability of three simple physiological-based scoring systems including Modified Early Warning Score (MEWS), Rapid Emergency Medicine Score (REMS), and Rapid Acute Physiology Score (RAPS) in predicting adverse clinical outcome in geriatric patients with AUGIB in EDs. Methods This is a s retrospective cohort study conducted in a single suburban regional hospital from July 01,2016 to July 31, 2021 in Chiayi, Taiwan. Patients aged older than 65-year-old visited the ED with the diagnosis of AUGIB confirmed by endoscopy were recruited. Data including vital signs in triage, demographics, underlying comorbidities, presenting signs and symptoms, laboratory findings and endoscopic findings were recorded. Three physiological-based scoring systems including RAPS, REMS and MEWS were calculated by triage vital signs. Outcome was defined as in-hospital mortality or intensive care unit (ICU) admission during the hospitalization. Univariate logistic regression analyses were used to identify the variables that were significantly associated with in-hospital mortality or ICU admission. A two-tailed p value of less than 0.05 was considered statistically significant. The diagnostic performance with area under curve (AUC) was further evaluated by receiver operating characteristic curve analysis. Results A total of 317 geriatric AUGIB patients were recruited, 16 of which have poor clinical outcomes including ICU admission or in-hospital mortality. Patients with adverse clinical outcomes present with lower SBP (103.50 v.s. 121.84 mmHg) and more rapid respiratory rate (21.19 v.s. 19.08 time/min). Lower Hemoglobin (7.67 v.s. 9.26 g/dL) and albumin level (2.90 v.s. 3.47 g/dL) are in the group with adverse outcomes. The area under the receiver operating characteristic curve values for MEWS, RAPS and REMS were 0.739 [95% confidence interval (C.I.): 0.603-0.875, p = 0.001], 0.666 (95% C.I.: 0.523-0.808, p = 0.025), and 0.647 (95% C.I.: 0.471-0.802, p = 0.048), respectively. Conclusion Among the 3 physiological-based scores, MEWS has the best AUROC in predicting adverse clinical outcomes in geriatric patients with AUGIB in EDs. This simple, rapid, obtainable-by-the-bed parameters that can be calculated immediately could assist emergency physicians for initial management for this vulnerable group. However, to evaluate the real clinical impact, further prospective study is warranted.

Trial Registration: Not applicable for observational study. Funding: This study did not receive any specific funding. Ethical approval and informed consent: This study was approved by the Chang Gung Medical Foundation Institutional Review Board (IRB:201900828B0C102), waiving the need for obtaining the informed consent of the study participants.

Introduction The Royal College of Emergency Medicine defines Frequent Attenders (FA) as anyone who attends the Emergency Department(ED) five or more times per year. This group has a high mortality rate and is a significant burden on services. The Bristol Royal infirmary (BRI) is a city centre adult only ED, 51828 individual patients are seen per year 1.8% of which attend 5 times or more per year. Mortality rate of our adult FAs is 20%. The BRI High Impact User (HIU) team was established with the aim of supporting this population. Our aim is to further develop a triage tool used by the service, by using predicted mortality. The first step, outlined here, was to determine which factors increase mortality in this specific local population of FAs. Methods Data was collected retrospectively on a cohort of 250 patients, randomly selected from1780 FAs attending BRI ED between January 2016 and January 2017. Six variables were chosen for analysis: current mental health problems, homelessness, injecting drug use, alcohol misuse, chronic medical problems and number of attendances during the 12-month period. Data was collected from electronic patient notes. Data on age, gender and 5-year mortality were also recorded. Logistic regression modelling was performed to determine which factors best predicted 5-year mortality. This wasn’t a clinical research study therefore did not require NHS research ethical approval. Local Service development and information governance guidelines were followed. Results In univariate analysis, age and chronic medical problems were the only two variables were independently associated with 5-year mortality. All except one of the 15 patients with >15 attendances per year had mental health problems(MHP). Patients who attended 10-20 times per year with MHP(n=22) had a higher mortality rate 31.3% than those in the >20 attendances group (n=6) where 5-year mortality was 0%. Multivariate analysis suggested different predictors of mortality depending on the presence or absence of MHP. Therefore two different algorithms were derived; one for those without MHP and one for those with MHP, both of which had AUROC of over 0.7. Discussion This was a small-scale service development using a specific cohort of frequent attenders in inner-city Bristol, and therefore our findings are not generalisable. However, there are some interesting results. In particular the complex relationship between mortality and frequency of attendance in patients with mental health problems. We must recognise that in most FA programmes there is a dual aim of reducing mortality, along-side reducing the burden of frequent use on emergency services. Our finding prompts the question of whether this second aim risks inadvertently discouraging high risk patients from accessing our services where their interactions may be protective. This is an under-researched and under-resourced population and we hope to encourage other local EDs to develop similar tools for their populations of FAs. Our service plans to now prospectively collect data using these scoring systems to further validate our tool in our specific population.

This wasn’t a clinical research study therefore did not require NHS research ethical approval. Local Service development and information governance guidelines were followed. There was no specific funding for this work which was carried out by members of the HIU team and doctors employed by the Emergency Departmentin Bristol.

Introduction Despite poor predictive performance, with overestimation of risk in younger and underestimation of risk in older Emergency Department (ED) patients, the National Early Warning Score (NEWS) has been widely implemented and adopted. To improve predictive performance for all ages, the aim of this study was to develop and externally validate an International Early Warning Score (IEWS) based on a recalibrated NEWS model including age and sex and evaluate its performance independently in three age categories (18-65y;66-80y;>80years).       
Methods An international multicenter cohort study including all ED patients ≥18years using the Netherlands Emergency department Evaluation Database (NEED) including three Dutch EDs. External validation was performed using a Danish Multicenter Cohort (DMC). The primary outcome was in-hospital mortality.

Results In total, 95,553 patients were included for development of the model and 14,809 were included for external validation. Mortality rate was 2.4%. Overall, the IEWS performed significantly better than NEWS with an Area Under the Curve (AUC) of 0.89 (Confidence Intervals (CI) 0.89-0.90) versus 0.82 (CI 0.82-0.83) in the NEED and 0.87 (CI 0.85-0.88) versus 0.82 (CI 0.80-0.91) in external validation. For each age category discrimination improved significantly with IEWS. AUC was 0.92 (CI 0.90-0.93) for the age category 18-65, 0.85 (CI 0.84-0.86) for 66-80, 0.83 (CI 0.82-0.85) for >80, compared to NEWS that showed AUC of 0.87 (CI 0.85-0.88), 0.82 (CI 0.82-0.83), and 0.80 (CI 0.79-0.81) in the same age categories, respectively. NEWS underestimated risk in older patients and overestimated risk in the youngest, while IEWS improved calibration with a substantial reclassification of patients from low to high risk. With IEWS, 47.5% more patients were justifiably considered low risk, whereas 27.6% more patients who died were considered as medium or high risk.

Conclusions The IEWS, a recalibrated NEWS including age and sex, substantially improves mortality prediction for all ages. With better classification into low and high risk, IEWS may potentially contribute to better decision policies and should therefore replace NEWS in the ED.        

None

Background: The qSOFA, the CRB65 and the Fine score are simple tools allowing respectively identification of sepsis and prediction of mortality in acute community-acquired pneumonia . As SARS-CoV2 infection is an emerging disease, several scores have been proposed for the evaluation of its severity in the emergency department. The objective of this work was to evaluate the performance of the qSOFA, the CRB 65 and the fine score in the prediction of mortality due to SARS-CoV2 infection. Methods: Prospective, observational and prognostic study including patients aged over 15 years , treated in the emergency room for SARS-COV2 infection (October 2020-July 2021). Calculation at admission of qSOFA, CRB65 and Fine score. Study of early mortality (D07) and in-hospital mortality: Comparison of deceased groups versus survivor groups univariate study with elaboration of ROC curves and precision of their characteristics Results: We included 709 patients. Median age=65 years; IQR (56,74). Sex ratio (M/F) =1.27. Forty-three patients (6.1%) had qSOFA 2 and 28 patients (4%) had qSOFA at 3. A CRB65 2 score was found in 20.5% of cases (145) and a Fine score >90 was noted in 40.3% of patients (145). The rates of mortality at D7 and in-hospital mortality were respectively N (%): 141(19.9) and 301(42.5). All 3 scores were statistically higher in the deceased (at D07 and in-hospital) versus survivor groups. In the univariate study, qSOFA, CRB65 and Fine score were predictive of mortality at D07: qSOFA 2 (OR=3.469, CI95%: [1.820 - 6.612], p<0.001); CRB65 2 (OR=2.984, CI95%: [1.984 - 4.487], p<0.001); Fine score > 90 (OR=3.086, CI95%: [2.106 - 4.522], p<0.001) and in-hospital mortality: qSOFA 2 (OR=3.789, CI95%:[1.912 - 7.509], p<0.001); CRB65 2 (OR=4.533, CI95%:[3.044 - 6.751], p<0.001); Fine score > 90 (OR=2.585, CI95%:[1.897 - 3.522], p<0.001). The Fine score presented the best ROC curve characteristics either for mortality at D07: AUC 0.690; 95% CI [0.639 - 0.741]; Sensitivity: 62.4%; NPV: 87.4% and for in-hospital mortality: AUC: 0.677; 95% CI [0.639 - 0.716]; Sensitivity: 69.3%; NPV: 66.7% Conclusion: In our study, qSOFA, CRB65 and Fine score were predictive of mortality from SARS-CoV2 infection. These are simple and rapid prognostic scores commonly used in emergency departments to identify patients at risk of worsening. Their performance in the context of SARS-CoV2 infection would be a good tool for triage and management of patients during a pandemic.

Background Frailty, an aging-associated decline in function across multiple physiologic systems, is a common clinical syndrome in older adults and carries an increased risk for poor health outcomes. The proportion of older people in Swedish Emergency Departments accounts for approximately 40%. Proper identification and stratified management based on frailty is needed to direct the use of limited ED resources to patients most susceptible to adverse events. EDs in Sweden are implementing clinical judgment of frailty as a compulsory part in the care of elderly. There are many frailty tools of different precision and time demand. Loss of Independence (LOI) defined as inability to rise unaided, has been shown to be prognostic regarding risk for readmission as well as mortality. Adding LOI to triage increased the prognostic precision for mortality within 24-72 hours compared to Modified Early Warning Score (MEWS). The aim of this study was to investigate whether LOI, can be a simple alternative frailty assessment tool to predict adverse outcomes such as mortality, hospital admission, Length of Stay (LOS) in ED and Length of Hospitalisation (LOH). Methods The study was a multi-centre, prospective, observational cohort study of a consecutive sample of geriatric patients (≥65) visiting one of three EDs within a regional healthcare system in Sweden serving 450 000 inhabitants. LOI was assessed by either the physician, registered nurse or the assistant nurse. The primary outcome variable was mortality at 30 days after index visit. The secondary outcome variables were: Mortality at 7 respectively 90 days, LOS, admission rate, LOH and return visits to the ED. Descriptive statistics have been used for outcomes of admission rate, mortality at 7, 30 and 90 days, LOS and LOH. Variables regarding LOI was set to LOI 1 = independent, LOI 2 = frail and LOI 3 = very frail. LOI 1 was set to non-frail and LOI 2 + 3 = frail for dichotomisation in statistical processing using Chi-square test and Fisher’s exact test (Graphpad prism). Results 2169 patients were included and 1929 (52% male) remained for analysis after excluding planned revisits to the ED and patients with missing data concerning LOI and regular frailty assessment scores (Clinical Frailty Scale). 25% (n=482/1929) were deemed frail at index visit. Frail patients had a significantly higher admittance rate (66% (319/482) compared to 37% (n=529/1447) of the non-frail (p<0.0001). Preliminary data suggests a doubled relative risk of death 30 days after index visit comparing frail (44/482) to non-frail (25/1447). The LOS in the ED as well hospitalisation was less influenced by frailty but showed a wide variation. Discussion and conclusions Adverse outcomes associated with frailty has been shown in many studies with different frailty scoring systems. LOI is a simple assessment that seems associated with increased admission rates and higher mortality. The data may lead to relocation of resources in the ED to the patients in most need as well as add to future re-dimensioning of hospital resources.

The study was registered on ClinicalTrials.gov identifier: NCT04877028 This study did not receive any specific funding. Ethical approval and informed consent: The study was approved by the Swedish Ethical Review Authority (permit 2021-00875). Written informed consent was obtained.

Background: Biomarkers of brain injury, such as neuron specific enolase (NSE), can be a good component of the multimodal prognostication model predicting post-cardiac arrest outcomes. Adding more ideal parameters will improve prognostic accuracy. Recently, spectrin breakdown products (SBDP) have been studied as biomarkers of traumatic brain injury. However, SBDP have not been investigated as biomarkers of post-cardiac arrest neurological injury. Therefore, we evaluated the prognostic accuracy of the serum SBDP 145 kDa for the long-term outcome after cardiac arrest. Methods: This prognostic accuracy study was prospectively planned before testing. A sample size of 56 was calculated with an expected area under curve (AUC) of 0.85 (null hypothesis 0.65, power 0.95, and beta 0.2). Unconscious, adult survivors regardless of the cardiac arrest rhythm were consecutively recruited in two university-affiliated hospitals. NSE was selected as the reference test. Blood samples were obtained as soon as possible (0h), 24 hours, and 48 hours after the return of spontaneous circulation (ROSC). Clinical information was unavailable to testers. Modified Rankin Scale greater than 4 at six months after cardiac arrest was defined as a poor outcome. There were no preliminary data for the cut-off value of SBDP predicting a poor outcome. Therefore, we used the median value of SBDP145 as the cut-off value. According to the cut-off value, we calculated AUC, sensitivity, specificity, and likelihood ratio (LR). We compared AUCs of biomarkers using the DeLong test. Median values of SBDP in each outcome group were compared using the Mann-Whitney test. Results: Between November 2019 and February 2021, 56 patients were recruited. General characteristics were as follows: Mean age 57 years; 38 males; 54 out-of-hospital cardiac arrest; 22 shockable arrest rhythm. Within 24 hours after ROSC, seven patients expired and 49 samples were obtained at 24h. At 48 hours after ROSC, eight more patients expired and 41 samples were available. Forty-three patients were found to have poor outcomes six months after ROSC. AUC of SBDP145 was 0.58 (95% CI 0.44-0.71) at 0h, 0.56 (95% CI 0.41-0.7) at 24h, and 0.5 (95% CI 0.34-0.66) at 48h, respectively. AUC of NSE was 0.51 (95% CI 0.37-0.65) at 0h, 0.91 (95% CI 0.78-0.98) at 24h, and 0.98 (95% CI 0.86-1) at 48h, respectively. AUCs of NSE at 24h and 48h were significantly larger than those of SBDP145 (p<0.05). Median value of SBDP145 was 6.6 ng/mL at 0h, 6.5 ng/mL at 24h, and 6.7 ng/mL at 48h, respectively. Prognostic accuracy of SBDP145 predicting poor outcome was poor at all time points (0h: sensitivity 53.49%, specificity 61.54%, positive LR 1.39, negative LR 0.76; 24h: sensitivity 50%, specificity 61.54%, positive LR 1.3, negative LR 0.81; 48h: sensitivity 51.72%, specificity 41.67%, positive LR 0.89, negative LR 1.16). Median values of SBDP145 were not significantly different between the poor outcome group and good outcome group at all time points (p>0.05). Discussion & Conclusions: SBDP145 did not discriminate between poor and good outcomes six months after cardiac arrest at all time points, while NSE predicted poor outcomes at 24 and 48h after ROSC.

This work was supported by the Catholic University of Korea, Uijeongbu St. Mary's Hospital Clinical Research Laboratory Foundation program (UJBCR201918).

Title A mixed-methods study examining student perceived educational benefit of using students as simulated patients in place of mannequins in an undergraduate multidisciplinary simulation event. Authors Lennie I., Wallace R., Moulds Y., Bell L. Background Previous literature has demonstrated significant educational benefit when healthcare professionals role-play as patients during simulation training. We set out to establish student perceived educational benefit of role-playing patients in place of simulator mannequins during undergraduate multidisciplinary simulation. Methods Undergraduate final and penultimate year medical and nursing students undertaking their emergency medicine placements were included in the study. Study period extended between August 2021 and April 2022. Existing University of Glasgow ‘5EM’ simulation scenarios were adapted, and new scenarios designed, to facilitate students role-playing patients in place of simulator mannequins. All students were briefed on the format of the day and familiarised with the simulation environment. Chest and abdomen mannequins and a venepuncture arm were used to prevent unsolicited exposure/examination of actors and elicit clinical signs. Participants were invited to volunteer to play the role of a patient and were briefed and given an instruction card with the details of the patient scenario. The volunteer students were also provided with an earpiece to allow researchers to guide them to scenario template during the scenario. During the debrief and at the end of the simulation event, students were asked to provide feedback on their patient experiences. Feedback was collected both quantitatively, via a feedback form, and qualitatively via focus groups. Feedback forms were statistically analysed. Analysis of the qualitative data is ongoing. Results 52 undergraduate medical (28) and nursing (24) students participated in the research project. 36 participants were female. Age of participants ranged from 19-46 (mean =23). 90% of participants agreed that, as a result of being a simulated patient, they developed new insights to incorporate into their working role. 90% of participants agreed that playing the role of a patient in a simulation session was a valuable learning experience. 95% of participants agreed or strongly agreed that they gained a new perspective on how patients experience resuscitation. 91% of participants agreed they feel more empathy towards patients undergoing resuscitation. Early analysis of focus group feedback has shown improved empathy, compassion and better understanding of the workings of the multi-disciplinary team. Students have commented on how this will positively influence their future practice through enhanced communication and empathy. They described enhanced simulation fidelity compared to previous experience of simulation using mannequins. Discussion To our knowledge, this is the first research project examining the potential educational benefits of using the patient experience in both a multidisciplinary simulation setting and undergraduate medical curriculum. The project reinforces previous literature findings and demonstrates student-perceived learning, particularly increased patient empathy. Future work will involve further quantitative feedback forms, 6 months following simulation, to examine whether there has been sustained learning and if changes to practice have been implemented. Cohort studies comparing scenarios with mannequin patients vs student actor patients would provide a more robust data set to further establish educational benefit.

Background: Delirium is a highly prevalent condition in elderly patients, characterized by fluctuating disturbances in attention, cognition, and consciousness level. Its diagnosis is based on clinical scores since no blood test is helpful as a marker. Therefore, this study addresses whether plasma levels of brain cell damage markers may be used as diagnostic and prognostic tools for delirium identification in the emergency department. Methods: We conducted a prospective cohort study of consecutive patients (>65 years old) admitted to the Emergency Department in a tertiary academic hospital between April 2019 and November 2020. We collected baseline sociodemographic/clinical data and performed the confusion assessment method (CAM), a standardized and evidence-based instrument to recognize the occurrence of delirium at admission. In addition, blood samples were collected at admission, and cytokines and brain cell injury markers (S100beta, Neuron Specific Enolase, Tau protein) were measured by Luminex or ELISA. Statistical analyses were performed using the nonparametric Kruskal-Wallis chi-squared command in R Studio (version 4.0.3 for Windows) for categorical variables and the results expressed in percentage. Mann-Withney test was used for continuous variables and results expressed in median and interquartile range. A p-value < 0.05 was considered significant. Results: Two hundred patients were included and divided according to the presence of delirium (CAM+) or not(CAM-). In addition, each group was also divided regarding the presence of infection (INF+) or not (INF-). Patients in CAM+ were older than CAM- (median age 77 [73-85] vs. 72 [68-77.5] years, respectively), and the previous diagnosis of dementia and cerebrovascular events were more prevalent. Other demographic characteristics and previous conditions were equally distributed. Moreover, patients in the CAM+ group present higher mortality rates and ICU stays (31.92% and 42.55%, respectively) when compared to CAM- group (7.89% and 24.34%) and more extended hospital stay. There was no significant difference in circulating proteins that are markers for neuronal death between the CAM+ and CAM- groups. However, when only the CAM+ patients were evaluated, higher levels of NSE (2.68 [1.63-3.40] ng/ml) and Tau (75.09 [43.8-102.1]) were measured in the INF+ patients when compared to non-infected patients (NSE 1.51 [0.85-2.05]; Tau 55.56 [40.09-84.78] ng/ml) Discussion and conclusion: Circulating proteins used to detect neuronal death (S100Beta, NSE and Tau) do not help diagnose and prognosis of delirium. However, they may be useful for identifying the infectious etiology of delirium at Emergency Services. Ethical approval and informed consent: The study protocol was approved by the local Ethics Committee (opinion number 3.990.817; CAAE: 30417520.0.0000.0068), which also waived the need for written informed consent. We adhere to STROBE guidelines.

Funding: FAPESP and HCFMUSP

Background: Rural citizens are an underserved community in terms of health care with access to trained emergency physicians (EPs) even less available for this population. Emergency medicine (EM) residencies in the United States have historically been located in urban and suburban areas. Upon completion of training, newly minted emergentologists typically enter practice in larger communities where access to other specialists and diagnostic and therapeutic options generally mirrors the environment in which they were trained. Rural hospitals are much more resource limited in terms of specialists, procedures and available diagnostic tests. EPs in such an environment are required to perform tasks that are unnecessary for them in tertiary care centers including the management of patients longer while effecting transfers to a higher level of care. Methods: We were recruited by the leadership of the hospital (Magnolia Regional Health Center in Corinth, Mississippi) to establish a rural EM residency. The facility had a preexisting Office of Graduate Medical Education with an internal medicine residency and cardiology fellowship. Upon recruitment of adequate faculty and establishing a partnership with the University of Mississippi Medical Center for a required Pediatric Intensive Care Unit rotation, the residency application was submitted and approved. Our first residency class began July 1, 2020. Results: Quality EM metrics such as patient satisfaction scores improved, presumably, from increased facetime with physicians by both patients and their families. Our EM residents are easily meeting metrics on required procedures because there is no competition with residents in other programs. Initial concerns over lack of interaction with residents and fellows from other disciplines were allayed by communication with members of the medical staff in those specialties and with accepting physicians when transfers were arranged. The medical staff welcomed the interaction and integration of EM residents into the facility. Conclusions: A quality EM residency can and must exist in a rural setting to facilitate recruitment and retention of sufficient numbers of trained EPs into such a practice. Collaboration with larger educational institutions is essential to procure required or needed services and rotations that are unavailable locally. Additionally, residents at larger institutions would benefit from rotations in rural programs by (1) acquiring experience working closely with the medical staff; (2) learning quality measures that are tracked and reported; (3) having more responsibility in patient care due to reduced competition from other residency programs for procedures, etc. and (4) honing skills rarely needed in a tertiary care hospital such as transferring patients to another facility and caring for them until the transfer can be effected. A symbiotic relationship should exist between rural and traditional residency programs. Additionally, our model should be exportable in developing rural residencies.

None

Introduction: The rapid evolution and growing popularity of extended reality systems (virtual, mixed and augmented reality) have expanded their use as a learning tool, especially in healthcare. This novel technology brings significant benefits, such as the possibility of training in safe environments, the capacity for repetition to perfect skills, or the opportunity to better internalise the knowledge and skills acquired, thanks to greater realism and immersion in the simulation. Considering the essential role of healthcare first responders in mass casualty incidents (MCI), it is crucial to enhance the training they receive to strengthen their preparedness and response to this type of situation. Consequently, the European project MED1STMR will develop a new mixed reality training to train first responders for disasters. Objectives: In the initial phase of the project, a systematic review is being conducted to: First, synthesise the existing scientific literature on extended reality to train emergency medical teams in MCI. Determine the effectiveness of these training systems. Finally, explore the perception and acceptability of healthcare first responders. Method: For transparency and reproducibility, the recommendations of the PRISMA guidelines were followed, and this systematic review was registered in PROSPERO. To identify relevant studies, a search strategy was developed with the support of a librarian. This strategy is based on inclusion/exclusion criteria. It comprises terms related to the four main aspects: out-of-hospital emergency medical personnel (participants), education/training (intervention), MCI (setting/environment), and extended reality. The search was conducted in four databases: Medline, EMBASE, CINAHL, LILACS, and grey literature and reference lists of included studies. To facilitate the article selection processes and duplicate data extraction, the Covidence platform was used, in which META-QAT was also integrated as a tool to assess study quality. The information collected is being analysed and summarised in narrative and semi-quantitative form. Results: The search strategy identified 2799 studies, of which finally included 18. Preliminary results reveal that it is not possible to conduct a meta-analysis, given the heterogeneity of the articles concerning participants, types and duration of the training, and ways of assessing effectiveness. Most studies were conducted in North America (n=12), and publication date ranges from 2007 to 2020. Participants include combinations of physicians, nurses and EMTs, paramedics and students from the professions mentioned earlier. The MCI scenarios are varied, although explosions are the most commonly reported (n=8). The most frequently used type of extended reality is virtual reality (n=15), and the most commonly learned skills during training are triage (n=15). Regarding the effectiveness of the interventions, there is much diversity in measurement methods, ranging from subjective Likert scales to validated questionnaires. However, in all studies where participants' perception or experience was explored (n=12), satisfaction and acceptance were high and positive. Conclusions: This systematic review will serve as a basis for the MED1STMR to measure the effectiveness of extended reality training. It will also provide insight into the previous experiences of first responders. Particular emphasis should be placed on this innovative training methodology to improve the preparedness and performance of out-of-hospital emergency medical teams for future MCI.

This project is funded by the European projects Horizon 2020 with Grant Agreement number: 101021775 — MED1stMR — H2020-SU-SEC-2018-2019-2020 / H2020-SU-SEC-2020

Background: Ethnical and racial disparities are widespread in health care. Racial bias has been found in physicians and medical students, mainly in the US. To our knowledge it is however yet unknown whether medical students in Europe show a bias towards immigrant patients. Methods: This study was conducted in a 2 x 2 randomised controlled design. All medical students at the Medical University of Vienna enrolled in the mandatory doctor-patient communication course during their second year of studies were included. Two professional actors were instructed to depict patients suffering from abdominal pain of a strength of 4 out of 10. Four different patients were portrayed: Two males, two females, each with and without immigration status. Immigration status was depicted by stereotypical clothing and name. Besides gender and immigrant status, the patients reported the same complaints and identical medical history. Students were randomised to be shown one of the four videos. Afterwards, they were asked whether they would administer pain medication (yes/no, primary outcome), as well as their assessment of strength of pain (secondary outcome). For the primary outcome, the proportion of patients receiving pain medication by the students, a logistic regression model was used. Migration status of the patient, gender of the patient, age of the student and gender of the student served as covariates. The study was performed using an online teaching platform (Moodle®). Results: A total of 607 medical students (56% female [n=337], 42% male [n=255], 2% other [n=15]) were recruited between January and February 2022. The majority was in the age-group of 18-22 years (64%, n=387), followed by 23-27 years (27%, n=163). For the patients without immigrant status, 95.2% of students stated they would administer pain medication, and for patients with migration status 94.6% of students stated they would do so (absolute difference 0.6%, 95% CI -2.9% - 4.1%). The logistic regression model showed no significant association between migration status of the patient and the proportion of suggested pain medication (OR 0.80, 95% CI 0.38 - 1.67, p = 0.56). Furthermore, the gender of the patient and the age of the student did not show a significant association for pain medication rate. Discussion & Conclusions: Our data showed no association between migration background and the proportion of pain medication administered by second year medical students. It is important to note that these results do not rule out bias towards patients with immigration status. However, medical students appear to show no explicit bias in our theoretical example.

Funding: This study did not receive any specific funding. Ethical approval and informed consent: This study was approved by the independent Ethics Committee of the Medical University of Vienna and complies with the Declaration of Helsinki.

Introduction: All residency programs, especially emergency medicine (EM), incorporate simulation-based education (SBE) in their curriculum as an integral part. This is in addition to traditional teaching methods like didactics. Simulation-based education is generally considered to be better than traditional methods in developing not only clinical but also non-clinical skills according to many studies. This is attributed to the educational theories which underpin the concept of simulation. It is essential that any successful emergency medicine residency program should have an effective simulation-based learning experience for its residents. Objective: We aimed to study the effect of simulation-based education in improving the emergency medicine residents’ clinical and non-technical skills along with their overall confidence, learning experience and patient care. Method: We designed a prospective observational study at our Emergency Medicine Institute, Cleveland Clinic Abu Dhabi. All the current emergency medicine residents (Year 1, R1 = 3 and Year 2, R2 = 3) were included. We started an Emergency Medicine Simulation Series based on simulation-based curriculum which was specifically designed for EM residents. Each month, a 4-hour simulation session was conducted involving 3 simulation scenarios separately for both R1 and R2. From August 2020 to April 2022, total of 20 such simulation sessions were conducted (20 x 3 = 60 scenarios). Expert faculty used, our own designed, objective Emergency Medicine Simulation Assessment Tool (EMSAT) to record performance against clinical and non-technical skills. Total 15 sub-tasks were divided into 60% clinical (9 sub-tasks) and 40 % non-clinical/non-technical (6 sub-tasks). Overall competency assessment score (CAS) was calculated at the end. Residents were also given a feedback form at the end of each session and evaluations were sought against a 5-point rating scale. We divided the data into two groups, first 10 sessions (30 scenarios) as phase-I and last 10 sessions (30 scenarios) as phase-II. We used the two-tailed paired T test to compare both phases. Primary outcome was considered to be the improvement in clinical and non-technical skills using EMSAT as percentage scores. The secondary outcome was considered to be improvement in confidence, learning and patient care using resident’s feedback against 5-point rating scale. Results: Results of the paired T test indicated that there was a significant large difference for clinical skills between phase-I (M=65.7, SD=9.4) and phase-II (M=84.9, SD=4.4), t(29) =10.1, p < .001. Similarly, for non-technical skills the difference was noted between phase-I (M=59, SD=7.7) and phase-II (M=79.1, SD=7.3), t(29) =9.1, p < .001. The secondary outcomes showed similar trend for confidence (phase-I (M=2.2, SD=0.8) and phase-II (M=3.9, SD=0.8), t(29) =8.9, p < .001), learning experience (phase-I (M=1.9, SD=0.8) and phase-II (M=4, SD=0.8), t(29) = 9.1, p < .001) and patient care (phase-I (M=2, SD=0.9) and phase-II (M=4.1, SD=0.8), t(29) =9.2, p < .001). Conclusion: A significant improvement is noted in emergency medicine resident’s clinical and non-technical skills as well as their confidence, learning experience and patient care with simulation-based education. In addition to the standard clinical training for residents in the emergency medicine, simulation-based education is a valuable adjunct.

Trial registration was not required. No funding was required. Study was done as a part of regular simulation program for EM residents.

Introduction: Structured clinical debriefs in the emergency department (ED) have been shown to improve team performance and departmental processes yet debriefs are infrequently performed. One reported barrier is staff discomfort in leading debriefs stemming from the lack of formal training in clinical debriefing. The purpose of our study was to develop and evaluate the impact of a simulation-based workshop designed to train ED staff in clinical debriefing. Our primary outcome was measurement of participant comfort in leading debriefs post-workshop. Additional outcome measures included frequency of debriefs and debrief leadership skill-level following the workshop. Methods: We developed a two-hour simulation-based workshop with clinical experts that was refined through an iterative process. The workshop was administered virtually by two physicians experienced with clinical debriefing through a video-conferencing platform. Recruitment of ED staff included physicians, nurses and allied health staff at a large single-center Urban hospital in Toronto, Canada via convenience sampling. Participants were provided with a structured debriefing tool that was already in use in the ED at the institution. Simulation sessions in the workshop required participants to lead observed clinical debriefs with feedback provided from facilitators. For data collection, we created surveys with clinical and research experts modeled after the Kirkpatrick model of training evaluation. Participants completed structured surveys at three time points i) pre-workshop, ii) post-workshop, and iii) two-months post-workshop. Survey questions consisted of Likert-Scale items that were later coded for statistical analysis. Data analysis was done using IBM SPSS analytical software and included student’s paired t-test and Wilcoxon rank-sum test for ordinal data. Our study was approved by the hospital research ethics board as a department education project. Results: A total of 36 participants were recruited in the study. 72% of participants completed all three surveys. Significantly more participants rated an increase in level of comfort leading debriefs at the two-month follow-up (mean Likert score 4.15, p 0.001) compared to pre-workshop (mean Likert score 3.23). Similarly, the debrief leadership skill-level at two-month follow-up was significantly higher (mean Likert score 3.22, p 0.01) compared to pre-workshop (mean 2.66). Regarding the frequency of debriefs, there was no significant change in ED clinical debriefs performed by participants before and after the workshop. Sources of bias that limited our study were selection bias and participant re-call bias. Discussion and Conclusion: Our results indicate that a case-based virtual simulation workshop is an effective educational tool to improve staff learning and performance in leading clinical debriefs in the ED. Furthermore, the workshop displayed versatility as an effective educational tool for nurses and allied health professions in addition to physicians. The workshop did not impact the number of debriefs completed. A possible explanation is that other barriers to performing debriefs in the ED, such as time constraints on staff, need to be addressed in addition to formalized training of staff. Additionally, our follow-up time may have been too short at the two-month period to assess frequency of debriefs.

none

Background: Seniors have a high risk of re-visits to emergency departments (ED) varying between 12-20% within the first month and 19-33% within three months after ED discharge. The rate of ED re-visits within 72 hours is a key indicator for quality of care in emergency medicine and varies between 1-15%. Reasons for ED re-visits are the natural course of illness, misdiagnosis, lack of homecare and self-discharge against medical advice. However, risk factors for ED re-visits have not been fully investigated. Therefore, the aim was to analyse the incidence of ED re-visits and to identify risk factors for ED re-visits within 72 hours one and three months after ED discharge. Methods: In this retrospective study, seniors (≥ 70 yrs.) were consecutively enrolled if they presented with an Emergency Severity Index of 2 or 3 in a tertiary care ED in 2019, and were discharged after the ED visit. The frequencies of unplanned ED re-visits within 72 hours, one and three months after ED discharge as well as reasons and potential risk factors for ED re-visits were investigated by univariate and multivariable logistic regression models. Results: 592 seniors were enrolled of whom 30 patients (5.1%) re-visited the ED significantly more often for gastrointestinal complaints (OR 2.9, 95% CI 1.04 – 8.2, p=0.043) such as vomiting and abdominal pain within 72 hours after ED discharge. Furthermore, discharge against medical advice (OR 5.6, 95% CI 1.7 – 18.1, p=0.004) and ED presentation during night shifts (OR 2.7, 95% CI 1.2 – 6.1, p=0.014) were significant risk factors for ED re-visits within 72 hours after discharge. Sixty-six seniors (11.2%) re-visited the ED within one month after discharge and 79 patients (13.3%) within three months. Hospital stay in the preceding six months before the index ED presentation was a risk factor for ED re-visits within one (OR 2.0, 95% CI 1.2 – 3.4, p=0.009) and three months (OR 2.0, 95% CI 1.2 – 3.3, p=0.005) after ED discharge. Seniors who presented at the index ED visit with uro-genital symptoms (OR 2.1, 95% CI 1.0 – 4.5, p=0.050), during night shifts (OR 1.9, 95% CI 1.1 – 3.4, p=0.028) or living alone (OR 1.9, 95% CI 1.04. – 3.5, p=0.036) had an increased risk for ED re-visits within three months after discharge. Conclusion: Although the frequencies of ED re-visits among seniors at 72 hours, one and three months after discharge tends to be low and most re-visits were illness-related, seniors need to be assessed for risk factors for ED re-visits. Discharge should be carefully evaluated in order to improve patient safety and in balance to ED overcrowding and health care costs.

It is not a trial therefore, no trial registration. No funding.

Background: The Royal College of Emergency Medicine provides standards for inducting junior doctors in training who are rotating into emergency medicine, including how to gain senior support, access guidelines and workplace policies as well as how to raise concerns. Despite these guidelines junior doctors often feel underprepared and anxious about starting their new role. There is no national consensus on how to best deliver an induction program, but there is increasing evidence to support online induction in combination with face to face near pear teaching and simulation. These methods have shown improvement in junior doctor knowledge retention, reduction in anxiety and improved confidence. Aim: This evaluation study assessed the effectiveness of online videos followed by a near-peer simulation as methods for delivering junior doctor induction in a paediatric emergency department. Methods: Participants were purposefully selected as junior doctors rotating into a tertiary paediatric emergency department in the UK between February and March 2022. They were a mix of both paediatric and emergency medicine trainees. They received a 2-day induction comprising of both online videos and simulation. The online induction covered a wide range of topics from introductions to senior team members, accessing guidelines and how to make referrals. As part of the face-to-face induction the participants were orientated to the department and then underwent a near-peer simulation. The simulation was tailored to include scenarios that would be most useful to the participants based on their training background. A questionnaire was distributed to collect qualitative and quantitative data on the induction. Results: A total number of 9 participants rotated into the ED between February and March 2022. 90% rated the induction as either good or excellent. Comments included: ‘the best induction I’ve had during training’ and ‘a mixture of online and face to face is the best way forward.’ Online induction: There was an 89% response rate to the questionnaire. 80% of participants either strongly agreed or agreed that the induction had prepared them to start working in the department. 91% felt welcomed and well orientated to the departments. Comments included: ‘excellent online videos, can go at own pace’ and ‘videos introducing the consultants were friendly and welcoming’. Near-peer simulation: There was a 78% response rate to the questionnaire. 100% of participants found the simulation sessions extremely or very useful and 100% of participants either strongly agreed or agreed that the induction simulation had improved their confidence when dealing with unwell patients in the department. Conclusion: This evaluation study highlights the benefits of using both online and near-peer simulation to orientate and prepare junior doctors to start work in a paediatric ED. Online induction video allows user flexibility, whereas the simulation provides hands-on skills and training related to the job and the specific department. These simple methods can be incorporated by other trusts to help junior doctors feel orientated and prepared to start work in their new role.

Background: Child maltreatment is known for its detrimental impact on children, families and society, but remains challenging to identify. To improve detection of child maltreatment at the emergency department (ED), screening tools combined with training have shown to be effective. Multidisciplinary child abuse teams and policy officers support healthcare professionals and initiate appropriate follow-up. The presence of these strategies should be embedded in a hospital policy to ensure healthcare professionals are supported by the hospital for adequate management of child maltreatment cases. We investigated whether having a hospital policy regarding child maltreatment is associated with the use of different strategies for recognition of child maltreatment in the ED. Methods: A survey on child maltreatment recognition in European EDs was conducted focusing on screening methods, training and hospital policies. The survey was distributed through key members from collaborative emergency medicine societies and research networks (European Society for Emergency Medicine (EUSEM), Research in Pediatric Emergency Medicine (REPEM) and European Society for Emergency Nursing (EuSEN)) from September 2018 to January 2019. The following strategies for recognition of child maltreatment were investigated: use of a screening tool, training of hospital staff, availability of a child abuse team and presence of a child abuse policy officer. Descriptive analyses were performed with additional chi-squared tests to assess any associations between the current hospital policy (‘Does your hospital have a standardised policy or guideline for the detection of child abuse?) and the aforementioned strategies. Results: 185 surveys were analysed, representing 148 hospitals from 29 European countries. The estimated response rate ranged from 33.8% to 54.6%. Half of the respondents (51%; n=94/185) had a standardised policy regarding child maltreatment in place, while 24% (44/185) stated they did not have such a policy. 25% (47/185) had a missing or ‘unknown’ response to the question. The presence of a hospital policy was significantly associated with the use of screening strategies in comparison with the absence of a hospital policy: screening tool (p<0.001) (resp. 52%; n=49/94 vs. 5%; n=2/44), training (p<0.001) (resp. 63%; n=59/94 vs. 30%; n=13/44), child abuse team (p<0.001) (resp. 73%; n=69/94 vs. 27%; n=12/44) and policy officer (p<0.001) (resp. 51%; n=48/94 vs. 20%; n=9/44). Respondents with missing or ‘unknown’ responses on the question regarding a current hospital policy showed comparable results to respondents with no policy. Correlating respondents (n=185) to hospitals (n=148), analyses showed that out of the hospitals with a child maltreatment policy (55%; n=81/148), 28% (23/81) used all four strategies for recognition of child maltreatment. Discussion & conclusions: Our study shows that presence of a hospital policy on child maltreatment facilitates the use of different strategies for recognition of child maltreatment. Only half of the EDs in Europe have such a policy in place, and even in the presence of a policy, the different available strategies are not widely implemented on the ED floor. Development of a child maltreatment hospital policy and investigating barriers and facilitators for the implementation of different strategies are the next step to improve recognition and management of child maltreatment in Europe.

Trial registration: This study was not registered since this was a descriptive study amongst healthcare professionals and no appropriate register was available. Funding: This study did not receive any specific funding. Ethical approval: This study was approved by the Ethics Committee of the European Society of Emergency Medicine (EUSEM, April 16th 2018). The completion of the survey by the participants was taken as consent and data was analyzed anonymized.

Background: End tidal carbon dioxide measurement is mainly used for confirmation of intubation and continuous measurement of carbon dioxide levels in intubated pediatric patients. However it is possible to use it in non intubated patients as well as a non invasive method to measure carbon dioxide levels. Our aim in this study is to search for a correlation between end tidal carbon dioxide levels measured by a capnograph with carbon dioxide levels in blood gas analysis in non intubated pediatric patients. Methods: Our study was designed as a prospective cohort study. Patients that applied to Pediatric Emergency Unit of a 3rd level University Hospital with a complaint of acute gastroenteritis between July 1st 2018 and September 30th 2021 were included. Patients that had chronic diseases that would interfere with acid base levels (metabolic diseases, chronic diarrhea, renal tubular acidosis etc), patients that needed emergency interventions and the patients that did not want to participate were excluded. End tidal carbon dioxide was measured with capnograph with an attached mask. Correct sized mask is used for each patient. Statistical analysis was performed by SPSS v25.0 program. Descriptive statistics were presented with frequencies and percentages for discrete variables and with mean and standard deviation for continuous variables. Continuous variables were compared using the Mann-Whitney U test and correlation analysis was performed with Spearman's rank correlation coefficient. For all tests, p < 0.05 was considered statistically significant. Results: 60 patients were eligible for the study. Mean age of the patients was 65,3251,52 months and %63,6 was boys. Mean end tidal carbon dioxide level was 28,275,32 mm Hg and mean carbon dioxide level in blood gas analysis was 30,035,83 mm Hg with mean pH of 7,390,05. Correlation between mean carbon dioxide levels in capnography and blood gas analysis was significant correlated (r=0.410, p=0.0019). Discussion and Conclusion: End tidal carbon dioxide levels was shown to be valuable predictive tool in non intubated patients in literature1,2. However these studies mainly included adult patients and used side stream capnography. Our study showed that a simple capnograph with an attached face mask suitable for patient’s face can also be used safely or non invasive measurement of carbon dioxide levels in non intubated pediatric patients as well. 1Kartal M, Eray O, Rinnert S, Goksu E, Bektas F, Eken C ETCO2: a predictive tool for excluding metabolic disturbances in non intubated patients. American Journal of Emergency Medicine (2011) 29, 65–69. 2 Chebl RB, Madden B, Belsky J, Harmouche E, Yessayan L. Diagnostic value of end tidal capnography in patients with hyperglycemia in the emergency department. BMC Emergency Medicine (2016) 16:7 .

Background Evaluation of suspected child maltreatment remains a challenge for clinicians. Adequate physical examination, diagnostic work-up and follow-up with respect to somatic and psychological symptoms require specialized skills. In the Erasmus Medical Center in the Netherlands these children are evaluated by a child abuse expert team consisting of pediatricians, psychologists and social workers. We provide an overview of the characteristics and outcomes of children assessed at the emergency department by this expert team. Methods Prospective data were collected on children aged 0-16 years who were assessed for maltreatment from 2019 through 2021 in the Erasmus Medical Center (Rotterdam, the Netherlands). Patients were evaluated for signs of child maltreatment by physical examination and STD testing. All children were psychologically evaluated for trauma-related symptoms and they and their families were screened for risk factors of child maltreatment. The data were analyzed descriptively. Results 414 children (median age 7.0 years, 75% girls) were investigated. Children were referred to the expert team by the hospital’s own emergency department or specialists (10%, n=41), general practitioner (41%, n=169), other health care providers (16%, n=64), child protection services (16%, n=67), police (16%, n=68) or social services (1%, n=5). Children were referred with suspicions of (a combination of) the following types of maltreatment: sexual abuse (81%, n=337), physical abuse (21%, n=86), physical neglect (8%, n=35), psychological abuse (2%, n=9), psychological neglect (1%, n=6), domestic violence (2%, n=9) and pediatric falsification syndrome (0.2%, n=1). Of the children investigated for sexual abuse (n=337), 31% (n=105) were acute cases and 81% (n=274) were routinely tested for sexually transmitted diseases (STDs). 15 children aged 3-16 years were diagnosed with a STD: chlamydia (n=6), gonorrhea (n=5), both chlamydia and gonorrhea (n=3) and trichomonas (n=1). Of the children investigated for physical abuse or neglect (n=104), 64% had one or more physical injuries such as skin lesions (n=54), fractures (n=18) or traumatic brain injury (n=17). Two or more child-, parental- and/or family-related risk factors for child abuse were identified in 60% (n=250) of cases. In 46% (n=190) concerns for child abuse were confirmed, in 14% (n=60) refuted and in 40% (n=164) inconclusive by expert opinion of the team. In 64% (n=265) of cases a need for help was identified which corresponded to 84% (n=159) of cases with confirmed and 60% (n=99) of cases with inconclusive concerns for child maltreatment. Because of trauma-related symptoms 31% (n=130) of all children were referred to specialized mental health services. In 12% (n=48) of all children suspected child maltreatment was reported to the Safe at Home centre (national child safety organization). Conclusion Our study shows that a thorough evaluation of children suspected for maltreatment yields results, especially when performed by a child abuse expert team. Suspected sexual child abuse was the main reason for consultation and routine STD testing showed positive results including in the young. Psychosocial evaluation identified psychological trauma-related symptoms in a third of children suspected for abuse and a need for help in a large percentage of cases, even when the diagnosis of child maltreatment remained inconclusive.

N.a.

Background: Ketamine is one of the sedatives most used in North America for procedural sedation outside the operating room in children. Vomiting and recovery agitation are the adverse events most frequently reported after ketamine administration in children. The aim of this study was to identify genetic variants associated with the development of vomiting or recovery agitation after intravenous ketamine administration. Material and methods: This was an explorative single-center prospective pharmacogenetic study performed at the tertiary-level children’s hospital Institute for Maternal and Child Health IRCCS Burlo Garofolo of Trieste, Italy. Children, between 1 and 17yrs of age, performing procedural sedation with intravenous ketamine as the sole sedative agent were enrolled. Children with intellectual disability or who received other sedatives during the procedure or who received antiemetics before the procedure were excluded from the analysis. Pharmacogenetic and pharmacokinetic analyses were predisposed. The genotyping was performed through Illumina Global screening array. The search for biomarkers was performed through the analysis of the variants of 10 candidate genes coding for proteins responsible for the metabolism (CYP3A4, CYP2B6, CYP2C9, CYP3A5, CYP2A6) and pharmacodynamics (GRIN1, GRIN2A-D) of ketamine. A genome-wide analysis was carried out to identify the presence of genetic variants associated with the development of adverse events but not related to the candidate genes. The plasma concentration of ketamine and norketamine were determined through liquid chromatography-tandem mass spectrometry and pharmacokinetic parameters were determined with R software. Results: From September 2019 to October 2021, 106 patients were enrolled in the study. Among them, 87 patients were analyzed. The median age was 8yrs (5-11). Forty-seven (55%) were females, 32 (37%) were affected by chronic diseases. Arthrocentesis and bone fracture reduction were the procedures most frequently performed. A median dose of 1.7mg/kg of intravenous ketamine was employed to perform the procedures. Thirty-one patients (35%) experienced vomiting or recovery agitation and were compared to the 56 patients (65%) without these events. No statistically significant differences were found between patients with vomiting, recovery agitation and without these events regarding the age, gender, weight, presence of chronic disease and median dose of ketamine received during the procedure. The pharmacogenetic analysis showed that the polymorphisms more associated with the development of vomiting and recovery agitation were GSA-rs71379063 on the GRIN2A gene on chromosome 16 (p 0.0038, OR 7.16) and GSA-rs28399442 on the CYP2A6 on chromosome 19 (p 0.0065, OR 4.97). The polymorphism GSA-rs71379063 on the GRIN2A gene was present in 7 patients (22.5%) with vomiting or recovery agitation, compared to 2 patients (3.5%) without these events. The polymorphism GSA-rs28399442 on the CYP2A6 gene was present in 3 patients (9.7%) with vomiting or recovery agitation compared to 1 patient (1.8%) without these events. We found a statistically significant different pharmacokinetic profile in patients with recovery agitation compared to the others regarding ketamine clearance (p<0.001). No significant differences were found considering Ke, T1/2, Vd, and AUC. Conclusion: The development of vomiting and recovery agitation after intravenous ketamine administration seems associated with specific genetic variants. These findings should be confirmed in a greater population.

Background: Paramedics are frequently called to attend seizures in children. High-quality evidence on second-line treatment of benzodiazepine (BZD)-refractory convulsions with parenteral long-acting antiepileptic drugs in children has become available from the ED.1,2 Aims: To determine the association of BZD use pre-hospital and the need for respiratory support; to address the potential need for an alternative pre-hospital agent. Methods: Retrospective observational study of state-wide ambulance service data (Ambulance Victoria (AV) in Victoria, Australia, population: 6.5 million). Children aged 0-17 years assessed for seizures by paramedics were analyzed for demographics, process factors, treatment and airway management. We calculated adjusted odds ratios (AOR) of the requirement for respiratory support in relation to the number of BZD doses administered. Results: Paramedics attended 5,112 children with suspected seizures over 1 year (July 2018 to June 2019). Overall, need for respiratory support was low (n=166; 3.2%). Before ambulance arrival, 509 (10.0%) had already received a BZD and 420 (8.2%) were treated with midazolam by paramedics. Of the 846 (16.5%) patients treated with BZD, 597 (70.6%) received 1 BZD dose, 156 (18.4%) 2 doses and 93 (11.0%) >2 doses of BZD (Table 1,2). Patients who were administered 1, 2 and >2 doses of BZD received respiratory support in 8.9%, 32.1% (AOR 4.6 vs. 1 dose, 95% CI 2.9-7.4) and 49.5% (AOR 10.3 vs. 1 dose, 95% CI 6.0-17.9) respectively. Conclusions: Increasing administration of BZD doses was associated with higher use of respiratory support. Alternative pre-hospital anti-epileptic drugs to minimize respiratory depression should be investigated in future research.

Background: The emergency department (ED) can be an extremely stressful environment with factors quoted to contribute to stress being that of heavy workload, unwell patients, psychosocial stressors. These are especially so in a paediatric setting.3 Challenges of those who work in the emergency have been studied but few studies have looked at the paediatric emergency setting. In addition, junior doctors who may be rotating into the environment for the first time are a subgroup who may have additional challenges of sharp learning curves and adjusting to the environment. Therefore, this study aimed to explore the challenges that junior doctors experienced in the paediatric emergency department. Methods: We conducted a qualitative study using interpretive phenomenological analysis to study the challenges of junior doctors working in a tertiary paediatric ED setting in Singapore. Prior to the commencement of the study, we obtained approval from the Centralised Institutional Review Board. A thorough literature review was conducted and an interview guide was formulated. Sixteen junior doctors were working in the Paediatric ED during the time of the study were recruited. Semi-structured interviews were then conducted within a one-month period in December 2019. Data was collected until saturation point. All interviews were recorded and transcribed verbatim manually and subsequently analysed. Results: The main challenges that the junior doctors studied faced were broadly categorised into subject matter, communication difficulties, emotional stressors and uncertainty in patient care. In terms of subject matter, they cited difficulties in terms of becoming familiar with the breadth of the information required to work in the emergency in view of the extensive number of conditions the emergency department deals with ranging from minor trauma cases and procedures to high acuity resuscitation situations. Communication difficulties were especially noted in communication with parents of children who may have been anxious or grieving. In terms of psychological stressors these were seen in exposure to traumatic or unfortunate cases involving children or adolescents which was an added emotional stress to the junior doctors. The junior doctors also reported stress relating to uncertainty in patient care in which they worried about the decisions they had made had impacted on the child negatively especially if the child was discharged from the care of the hospital. Conclusion: The challenges of the junior doctors in the emergency were explored. These surrounded areas over subject matter, communication difficulties, clinical uncertainty, and emotional stressors from traumatic scenarios. Knowing these challenges can help the individuals interacting with these junior doctors in the emergency setting understand them better and find ways to work together more efficiently. More research can be done in evaluating how the doctors had dealt with these challenges and how these junior doctors would like to be assisted.

Background: Systematic screening for child maltreatment at the emergency department (ED) has shown to increase the detection rate of child maltreatment. Our previous survey showed the need for a screening checklist, training of hospital staff and availability of a hospital policy regarding child maltreatment. Implementation of a child maltreatment toolkit containing these elements could fill this need, and improve recognition and management of child maltreatment. This study investigated healthcare professionals’ barriers and facilitators for the implementation of this toolkit at European EDs. Methods: A survey for healthcare professionals was developed based on the ‘Barriers and Facilitators Assessment Instrument’ and consists of statements (5-point Likert scale) on barriers and facilitators regarding implementation of our child maltreatment toolkit. In the survey, detailed information on the toolkit was given with illustrative videos and figures concerning the screening checklist, training of hospital staff and availability of a hospital policy. Collaborative emergency medicine societies and research networks (European Society for Emergency Medicine (EUSEM), Research in Pediatric Emergency Medicine (REPEM) and European Society for Emergency Nursing (EuSEN)), previous respondents (n=96) and additional contacts at European EDs were approached starting from September 2021 for distribution of the survey. Descriptive analyses were performed and missing answers on statements were excluded. Based on the statements of the majority of the respondents, barriers (>50% (fully) disagreed) and facilitators (>50% (fully) agreed) were identified. Results: 131 surveys were analysed, representing 87 hospitals from 24 European countries. The estimated response rate ranged from 34 to 52%. The following most important facilitators for implementation of the screening checklist, training and hospital policy were identified: - Not too time consuming (screening checklist 90%, training 71%, hospital policy 80%). - Fits into way of working at the ED (screening checklist 87%, training 77%, hospital policy 90%). - Leaves enough room to make own conclusions (screening checklist 90%, hospital policy 66%). - Sufficient to improve the recognition of child maltreatment (training 84%). - Mandatory (national) implementation (hospital policy 96%). - Trust in the implementation and use (hospital policy 79%). - Allocation of a part of the hospital budget to the recognition of child maltreatment (hospital policy 94%). - No resistance to working according to protocols (screening checklist 92%). With respect to the complete toolkit, respondents were willing to use the toolkit in the future (82%) and think other doctors/assistants (82%) and managers/directors (57%) will cooperate in applying the toolkit. Although above results reflect facilitators for implementation, as only barrier respondents stated the need for more information on the toolkit before implementation (>50%). Discussion & conclusions: The identification of the barriers and facilitators for a child maltreatment toolkit will enhance successful implementation. This survey amongst healthcare professionals at European EDs mostly provided facilitators for the implementation of a child maltreatment toolkit. These included that the child maltreatment toolkit fits into the dynamic working environment at the ED. Furthermore, our results showed that professionals are motivated to implement a toolkit for better recognition of child maltreatment.

Trial registration: This study was not registered since this was a descriptive study amongst health care professionals and no appropriate register was available. Funding: This study did not receive any specific funding. Ethical approval: This study was approved by the Ethics Committee of the Erasmus Medical Center (MEC-2021-0246) and does not fall under the scope of the WMO. Data was analyzed anonymized.

Background: The American Academy of Pediatrics recently published guidelines addressing the evaluation and management of infants 8 to 60 days of age with fever ≥38.0ºC. Evidence is lacking regarding the best approach to young febrile infants who have received immunizations within the previous 48 hours. For this reason, these infants were excluded from these guidelines. The objective of our study was to compare the prevalence of invasive bacterial infection (IBI) and urinary tract infection (UTI) in infants 42 to 90 days of age with fever without a source (FWS) who have received immunizations in the preceding 48 hours and those who have not. Methods: We carried out a secondary analysis of an observational prospective registry that includes all the infants ≤90 days of age with FWS attended in a pediatric ED. We analyzed those 42 to 90 days of age infants attended over an 11-year period (2010–2021). Preterm infants (<37 weeks’ gestation) were excluded. We classified the infants either as having received immunizations (RI group) within the 48 hours preceding the ED visit or as not having received immunizations (non-RI group) during that period. We compared the prevalence of IBI (isolation of a bacterial pathogen in blood or cerebrospinal fluid) and UTI (urine culture with growth of ≥10 000 CFU/mL in combination with leukocyturia) in both groups. Results: We registered 1490 episodes corresponding to infants 42 to 90 days of age born at term with FWS. Information about recent immunization was available in 1448 (97.1%): 174 (12.0%) included in the RI group and 1274 (88.0%) in the non-RI group. Among the infants included in the RI-group, 77.6% of them had received the immunization in the same day the fever began. Overall, 272 (18.7%) were diagnosed with an UTI and 18 (1.2%) with an IBI (10 UTIs with associated bacteremia, 7 occult bacteremias and 1 sepsis). The prevalence of UTI was significantly lower in the RI group [6.9% vs 20.4% in the non-RI group, p<0.01; Odds Ratio: 0.28 (95% CI: 0.15-0.52)]. None of the recently immunized infants were diagnosed with an IBI [vs 18 of the non-RI group (1.4%), p=0.15] Discussion & Conclusions: Although the prevalence of UTI in infants 42 to 90 days of age with FWS who have received immunizations within the previous 48 hours is lower, recommending screening for UTI seems appropriate. Prevalence of IBI was also lower among infants with a recent immunization. If further studies confirm this finding, the recommendation of performing blood tests systematically in these infants should be reconsidered.

The study was approved by Ethical Committee of our hospital (CEIC EI4/55). Given that all the data were extracted from a database in which patients were anonymous and inclusion in the registry did not imply any additional interventions, informed consent was waived. This study did not receive any specific funding.

Background: Deviations from international guidelines during the management of paediatric cardiac arrest (PCA) are still frequent and can affect clinical outcomes. Different cognitive aids have been developed to support health care professionals during the management of cardiac arrest, however very limited data are available on their effectiveness in the management of PCA. We aimed to assess the effectiveness of the use of a novel cognitive aid, in the format of a tablet app, named PediAppRREST, in improving the management of a simulated in-hospital PCA scenario. Methods: We conducted a multicentre three-parallel-arm simulation-based randomized controlled trial, between September 2020 and December 2021, at four Italian University Hospitals. Residents in Paediatrics, Emergency Medicine, and Anaesthesiology were recruited and randomized into teams of three members each. For each team, one resident needed to be a Paediatric Advanced Life Support (PALS) certified provider and had the role of team leader, while the two other team members had to be Basic Life Support providers. Teams were randomized by an external statistician to three study arms, in a 1:1:1 ratio, using a computer-generated sequence and sealed envelopes. In the intervention arm teams used the PediAppRREST app (PediAppRREST+), while in the first control arm teams used a paper based cognitive aid, the PALS pocket reference card (CtrlPALS+), and in the second control arm teams did not use any cognitive aid (CtrlPALS-). All the teams managed the same standardized simulated scenario of non-shockable PCA. The primary outcome was the number of deviations from guideline recommendations, measured by a novel checklist named c-DEV15plus. Secondary outcomes were team performance, measured by the Clinical Performance Tool (CPT) score, and perceived team leaders’ workload, measured by the Raw NASA-Task Load Index. Scenarios were video recorded and analysed by two independent reviewers. Sample size was calculated as 29 scenarios per arm. Primary and secondary outcome scores were compared between groups with one-way ANOVA model, followed by the Tukey-Kramer multiple comparisons adjustment procedure in case of statistical significance. Results: A total of 324 residents were randomized into 108 teams. Eight teams were excluded due to sick leave or withdrawal of consent. Overall, 100 teams were analysed (PediAppRREST+=32, CtrlPALS+=35, CtrlPALS-=33). Participants’ demographics and previous experience in simulation and resuscitation were not significantly different among the three study arms. c-DEV15plus scores were significantly lower in the PediAppRREST arm (mean 3.4, standard deviation [SD] 2.0) than control arms (CtrlPALS+ mean 6.4, SD 1.8; CtrlPALS- 6.0, SD 1.6, p<0.0001). CPT scores were significantly higher in the PediAppRREST arm than control arms (mean 8.9 [SD1.6] vs 7.5 [1.5] and 7.7 [1.6] respectively; p 0.0028). Team leaders’ perceived workload did not significantly differ among the three study arms. Conclusion: The use of the PediAppRREST app was associated with a lower number of deviations from international guidelines and a better team performance, and it was not associated with an increase in the team leaders’ perceived workload. Further studies and systematic training are necessary before considering the use of the PediAppRREST app during real events.

Trial Registration: This trial was registered at ClinicalTrials.gov (NCT04619498). Funding: This research has received funding support from internal grants from the University of Padova (BIRD 191291/2019, 203279/2020). Ethical approval and informed consent: The Human Ethics Committee of the University Hospital of Padova, coordinating centre of the trial, deemed the trial to be a negligible risk study and approved it through an expedited review process. All participants signed an informed consent form.

Background: Accurate weighing of pediatric patient is essential because drug dosage is based on body weight. It is difficult to determine the exact weight of the children in emergency situations, so current American Heart Association guidelines recommend using the length-based weight estimation tool such as Broselow tape (BT) or Pediatric Advanced Weight Prediction in the Emergency Room (PAWPER) XL tape. However, these tape-based weight-estimating methods have problems such as the need to carry the tape in emergency situations, spending time to find it, and lack of knowledge about the proper use of the tape. We developed the application for length-based weight estimation that uses augmented reality by combining the strength of the mobile phone’s accessibility and convenience. Our application was programmed to provide medication dosage, defibrillation energy, and equipment size as well. The aim of our study was to evaluate the performance of the application compared to BT and PAWPER XL. Methods: This prospective cross-sectional study was conducted on pediatric patients aged from 1 month to 12 years who had visited the emergency department of a tertiary university hospital in Gyeonggi-do, South Korea from August 2021 to March 2022. Weights were estimated using the BT, PAWPER XL, and application. Our application used length-based weight estimation adjusted by 7 stages of body habitus based on 2017 Korean National Growth charts and was developed by Unity engine. The performance was analyzed by calculating the intraclass correlation coefficient (ICC), median performance error, median absolute performance error (MAPE), and root mean squared error (RMSPE). In addition, the percentage of estimations within 10% of the actual weight (PW10) was calculated as index of the overall accuracy of the measurements. Results: In total, 1090 pediatric patients were enrolled in our study. Compared with BT and PAWPER XL, our application had the smallest MAPE (9.49%) and RMSPE (2.89%). Also, the PW10 of the application was 63.39%, indicating better accuracy than the BT (57.25%), and PAWPER XL (62.47%). The ICC of the BT, PAWPER XL and application between actual weight and estimated weight were 0.9516, 0.9694 and 0.9755, respectively (p<0.001). Discussion & Conclusions: The performance and accuracy of the application were slightly superior to the BT and PAWPER XL. The estimated body weight by our application was sufficiently correlated with the actual body weight. Our application may be useful in pediatric emergencies.

No appropriate register./National Research Foundation.