Machine learning for real-time aggregated prediction of hospital admission for emergency patients.

Machine learning for hospital operations is under-studied. We present a prediction pipeline that uses live electronic health-records for patients in a UK teaching hospital's emergency department (ED) to generate short-term, probabilistic forecasts of emergency admissions. A set of XGBoost classifiers applied to 109,465 ED visits yielded AUROCs from 0.82 to 0.90 depending on elapsed visit-time at the point of prediction. Patient-level probabilities of admission were aggregated to forecast the number of admissions among current ED patients and, incorporating patients yet to arrive, total emergency admissions within specified time-windows. The pipeline gave a mean absolute error (MAE) of 4.0 admissions (mean percentage error of 17%) versus 6.5 (32%) for a benchmark metric. Models developed with 104,504 later visits during the Covid-19 pandemic gave AUROCs of 0.68-0.90 and MAE of 4.2 (30%) versus a 4.9 (33%) benchmark. We discuss how we surmounted challenges of designing and implementing models for real-time use, including temporal framing, data preparation, and changing operational conditions.

Azithromycin versus standard care in patients with mild-to-moderate COVID-19 (ATOMIC2): an open-label, randomised trial.

BACKGROUND: The antibacterial, anti-inflammatory, and antiviral properties of azithromycin suggest therapeutic potential against COVID-19. Randomised data in mild-to-moderate disease are not available. We assessed whether azithromycin is effective in reducing hospital admission in patients with mild-to-moderate COVID-19. METHODS: This prospective, open-label, randomised superiority trial was done at 19 hospitals in the UK. We enrolled adults aged at least 18 years presenting to hospitals with clinically diagnosed, highly probable or confirmed COVID-19 infection, with fewer than 14 days of symptoms, who were considered suitable for initial ambulatory management. Patients were randomly assigned (1:1) to azithromycin (500 mg once daily orally for 14 days) plus standard care or to standard care alone. The primary outcome was death or hospital admission from any cause over the 28 days from randomisation. The primary and safety outcomes were assessed according to the intention-to-treat principle. This trial is registered at ClinicalTrials.gov (NCT04381962) and recruitment is closed. FINDINGS: 298 participants were enrolled from June 3, 2020, to Jan 29, 2021. Three participants withdrew consent and requested removal of all data, and three further participants withdrew consent after randomisation, thus, the primary outcome was assessed in 292 participants (145 in the azithromycin group and 147 in the standard care group). The mean age of the participants was 45·9 years (SD 14·9). 15 (10%) participants in the azithromycin group and 17 (12%) in the standard care group were admitted to hospital or died during the study (adjusted OR 0·91 [95% CI 0·43-1·92], p=0·80). No serious adverse events were reported. INTERPRETATION: In patients with mild-to-moderate COVID-19 managed without hospital admission, adding azithromycin to standard care treatment did not reduce the risk of subsequent hospital admission or death. Our findings do not support the use of azithromycin in patients with mild-to-moderate COVID-19. FUNDING: National Institute for Health Research Oxford Biomedical Research Centre, University of Oxford and Pfizer.

Emergency department procedural sedation: the London experience.

AIM: The aim of this study was to develop a procedural sedation guideline and an audit tool to identify the medications chosen, the incidence of predefined adverse events and the factors associated with their occurrence. METHODS: We performed a prospective observational study using a standardized proforma. We obtained data on patient demographics, the procedure performed, the drugs used and the dosages and predefined adverse events. Our target population was adult patients attending three urban UK teaching hospital Emergency Departments (EDs) over a 12-month period. We included all patients who were to undergo procedural sedation in the ED. RESULTS: Data were obtained on 414 patients. The majority of procedures were reductions of fractures and dislocations (89%). Midazolam was the sedative agent used most commonly (45%), and morphine was the most frequently used analgesic agent (36%). Overall 10% of sedations involved predefined complications (95% confidence interval: 7.1-12.9). There were no cases of airway obstruction requiring intervention; no patients showed clinical evidence of aspiration and no patient required intubation. Procedural success was 96%. Propofol was associated with fewer adverse events (95% confidence interval: 0.024-0.572), as well as higher procedural success. CONCLUSION: Procedural sedation and analgesia can be safely and effectively performed in the ED by appropriately trained emergency physicians.

Safe cervical spine clearance in adult obtunded blunt trauma patients on the basis of a normal multidetector CT scan--a meta-analysis and cohort study.

BACKGROUND: A true gold standard to rule out a significant cervical spine injury in subset of blunt trauma patients with altered sensorium is still to be agreed upon. The objective of this study is to determine whether in obtunded adult patients with blunt trauma, a clinically significant injury to the cervical spine be ruled out on the basis of a normal multidetector cervical spine computed tomography. METHODS: Comprehensive database search was conducted to include all the prospective and retrospective studies on blunt trauma patients with altered sensorium undergoing cervical spine multidetector CT scan as core imaging modality to "clear" the cervical spine. The studies used two main gold standards, magnetic resonance imaging of the cervical spine and/or prolonged clinical follow-up. The data was extracted to report true positive, true negatives, false positives and false negatives. Meta-analysis of sensitivity, specificity, negative and positive predictive values was performed using Meta Analyst Beta 3.13 software. We also performed a retrospective investigation comparing a robust clinical follow-up and/or cervical spine MR findings in 53 obtunded blunt trauma patients, who previously had undergone a normal multidetector CT scan of the cervical spine reported by a radiologist. RESULTS: A total of 10 studies involving 1850 obtunded blunt trauma patients with initial cervical spine CT scan reported as normal were included in the final meta-analysis. The cumulative negative predictive value and specificity of cervical spine CT of the ten studies was 99.7% (99.4-99.9%, 95% confidence interval). The positive predictive value and sensitivity was 93.7% (84.0-97.7%, 95% confidence interval). In the retrospective review of our obtunded blunt trauma patients, none was later diagnosed to have significant cervical spine injury that required a change in clinical management. CONCLUSION: In a blunt trauma patient with altered sensorium, a normal cervical spine CT scan is conclusive to safely rule out a clinically significant cervical spine injury. The results of this meta-analysis strongly support the removal of cervical precautions in obtunded blunt trauma patient after normal cervical spine computed tomography. Any further imaging like magnetic resonance imaging of the cervical spine should be performed on case-to-case basis.

Pneumomediastinum secondary to mephedrone inhalation.

A 23-year-old man presented to the emergency department with severe, central, pleuritic chest pain. The pain was present on waking and exacerbated by movement. On examination, supraclavicular and anterior chest wall surgical emphysema was noted, otherwise examination and initial observations were normal. A chest x-ray and subsequent chest CT demonstrated a pneumomediastium with significant subcutaneous emphysema. There was no history of vomiting or chest trauma, but the patient had inhaled mephedrone, a synthetic stimulant drug, 36 h prior. Pneumomediastinum is an uncommon complication of inhalational drug use.

Spontaneous upper limb monoplegia secondary to probable cerebral amyloid angiopathy.

Cerebral amyloid angiopathy is a clinicopathological disorder characterised by vascular amyloid deposition initially in leptomeningeal and neocortical vessels, and later affecting cortical and subcortical regions. The presence of amyloid within the walls of these vessels leads to a propensity for primary intracerebral haemorrhage. We report the unusual case of a 77-year-old female who presented to our emergency department with sudden onset isolated hypoaesthesia and right upper limb monoplegia. A CT scan demonstrated a peripheral acute haematoma involving the left perirolandic cortices. Subsequent magnetic resonance imaging demonstrated previous superficial haemorrhagic events. One week following discharge the patient re-attended with multiple short-lived episodes of aphasia and jerking of the right upper limb. Further imaging demonstrated oedematous changes around the previous haemorrhagic insult. Cerebral amyloid angiopathy is an overlooked cause of intracerebral haemorrhage; the isolated nature of the neurological deficit in this case illustrates the many guises in which it can present.

Single fascia iliaca compartment block for pain relief in patients with fractured neck of femur in the emergency department: a pilot study.

OBJECTIVES: Fractured neck of femur (NOF) is a common injury in the elderly, which causes severe pain and distress. Fascia iliaca compartment block using the 'Two Pop Technique' is safe and relatively easy to perform. It has been traditionally performed by anaesthetists in a controlled environment for perioperative pain relief with very good results. The purpose of the study was to investigate its applicability and success rate in a busy emergency department (ED) with changing junior staff and time pressures. MATERIALS AND METHODS: A prospective cohort study was carried out on 137 patients presenting to the ED with fractured NOF. The blocks were performed by emergency physicians (EPs) of different grades after a short training session. Pain scores (no pain=1, worst pain=10) were documented at presentation, 30 and 60 min after the block. RESULTS: A reduction in the pain score of three points or more was regarded as a successful block. This was achieved in 77.4% of all documented cases. No adverse events were reported and the participating EP found it very easy to master the technique. CONCLUSION: The fascia iliaca compartment block offers a very good, efficient and safe alternative to conventional analgesia in the treatment of pain in patients with fractured NOF. It should be used more by EPs as it is easy to learn and it requires inexpensive equipment.

Barbicidal overdose.

Acute severe methemoglobinaemia is an uncommon but life-threatening condition caused by a variety of oxidizing agents commonly used in both health care and industrial settings. Thus, recognition is important as it is readily treatable. The oxygen transport is compromised as a result of abnormal levels of oxidized haemoglobin, and this leads to skin discolouration and a variety of symptoms. Diagnostic confusion occurs as the oxygen saturations (SpO2) on the pulse oximeter are unreliable (Sharma V, Haber A. Acquired methaemoglobinaemia: a case report of benzocaine-induced methaemoglobinaemia and a review of the literature. Clin Pul Med. 2002;9(1):53-8). A case of severe methaemoglobinaemia due to self poisoning with barbicide is presented with a brief discussion of the patho-physiology and an overview of the treatment. A barbicidal overdose has never been reported before.