Article of the Month

June 2019

Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Annals of Internal Medicine 2001; 135(2): 98-107. 

An older paper this month, but one of significant importance in impacting the way we practice medicine. This paper is the foundation for the Wells score used to predict pulmonary embolism (PE) and risk stratifying patients into different groups. This paper recognised the low specificity of the D-dimer assay and developed a scoring system reduce the number of D-dimer tests being requested.

The study aimed to determine the safety of combining the pre-test probability ( using Well’s score) and D-dimer assay in detecting PE, recruting 930 patients from 4 centres. The large sample size from several centres makes this study more robust, and it describes selection of the cases (consecutive). In the study population, patients with a low Well’s score and a negative D-dimer result received no further investigation, whilst all other patients underwent V/Q scan. Further testing was based on pre-test probability and V/Q scan result. Primary endpoint was determined to be a diagnosis of thromboembolism within 3 months of testing.

PE diagnosis was met in 81 patients on initial assessment. 849 patients were diagnosed as not having PE initially but 90 excluded due to incorrect application of the diagnostic criteria – suggesting the study construct was adhered to closely. 437 patients had a low risk Well’s score and a negative D-dimer and only 1 developed PE during follow-up, providing a negative predictive value of 99.5%.

This study was externally validated in 2004 by Wolf, splitting Well’s score into two or three tiers of risk i.e. unlikely vs. likely, or low vs. moderate vs. high probability of PE. Pre-test probabilities for PE in the unlikely or low-risk groups were found to be 3% & 2% respectively and when combined with a negative D-dimer the incidence of PE is less than 0.5% within 3 months.

External validation and large population samples provide a strong evidence-base for Well’s scoring and D-dimer assay in assessing PE risk, and these studies underpin our practice in the emergency department. Numerous further studies have contributed to the body of evidence to support Well’s score, however there remains the significant issue of over-investigation due to poor specificity of D-dimer assay testing.


May 2019

Impact of Emergency Medicine Consultants and Clinical Advisors on a NHS 111 Clinical Assessment Service. Sen B et al. Emergency Medicine Journal 2019; 36: 208-212.

The 111 NHS advice service faces significant criticism from acute healthcare workers as a result of the large number of resources it demands in assessing patients in a hospital setting. Studies found that it contributed to a 3% rise in attendances to A&E by ambulance and no reduction in overall attendances. Although clinical advisors are available to 111 call operators, only a small minority of calls are escalated and there is still insufficient evidence to support the skill mix needed to operate an effective 111 service. This study introduced emergency medicine consultants into the NHS 111 assessment centres to assess their impact on the service.

Previously, callers were triaged by call operators and disposition decided. Those where correct disposition was unclear or patients refused disposition were escalated to the ‘clinical advisors’, however the new pathway identified patients who should ‘attend ED within 1 or 4 hours’ and these were then passed to emergency physicians or non-physician clinical advisors (NPCAs) and was employed during peak call times. The trial excluded patients unlikely to benefit from additional telephone triage i.e. penetrating injuries, assaults or foreign bodies. Documented were the amount of calls where disposition was changed away from an A&E attendance and those upgraded to 999 calls (i.e. stroke, sepsis).

The results demonstrated a large shift away from walk-in centres and emergency departments (96.7% pre-intervention vs. 40.1% physician-assessed call vs. 28.1% NPCA-assessed calls) with a shift towards self-care and GP in- or out-of-hours. This change could reduce the load experienced by A&Es as a result of attendances, however it requires significant resources to implement through recruiting NPCAs and clinicians to confidently provide advice and support to the 111 service. There was also a rise in the time-to-call-back from 10 minutes to 24 minutes and this additional wait time would need to be balanced against the benefits it provides. This short increase in wait will be acceptable for most complaints, however for a minority this could result in increased morbidity and mortality.


April 2019

Optimising antiplatelet utilisation in the acute care settin: a novel threshold for medical intervention in suspected acute coronary syndromes. Reynard C. Emergency Medicine Journal 2019; 36: 163-170. 

Emergency medicine doctors work with significant uncertainty and when presented with chest pain suspicious of cardiac origin they must weigh up the benefits of full treatment of acute coronary syndrome (ACS) against the risks of the drugs being administered. Often patients without ACS are given treatment that ultimately turns out to not be required. Positive predictive value for a single troponin may only be 50%, suggesting up to half of results are not due to ACS. Confirmed ACS is managed with aspirin, ticagrelor and, if coronary intervention is unlikely acutely, fondaparinux. However, ACS may not be confirmed until 6 hours after initial pain, leaving the patient exposed to greater risk of morbidity and mortality. This is where risk stratification and decision-making is paramount and this article investigates the point at which full ACS treatment could be considered.

The article first compares the risk of anti-platelet therapies in excess of aspirin alone (i.e. with clopidogrel or ticagrelor) and used two large trials to evaluate ticagrelor (PLATO) and clopidogrel (CURE trial), as well as generating utility data for each management strategy based on clinical outcomes from another large trial. As expected, more anti-platelet therapy improved outcomes but held a higher risk of side-effects (i.e. bleeding). This data was plotted to determine the point at which the benefits of each strategy outweigh the associated risks and the resultant threshold points were then applied in real-world situations at Manchester Royal Infirmary recruiting patients presenting within 12 hour of pain suspicious of ACS.

Patients were divided into 5 treatment groups: 1) as per the new threshold model, 2) standard practice, 3) no antiplatelets, 4) aspirin & ticagrelor for all suspicious of ACS, 5) ticagrelor only added for patients with troponin >52ng/L. The new model was analysed using a validated tool to extrapolate results to produce number of endpoints at 12 months and the utility of each pathway calculated. The Troponin-only Manchester ACS (T-MACS) decision aid was used to calculate risk of ACS. Overall it was found that patients with <8% likelihood of ACS were likely to have better utility from aspirin alone, whilst patients with higher ACS risk would benefit from ticagrelor in addition to aspirin when followed up at 30 days.

This study provides understanding of the risks and benefits of prescribing ticagrelor in addition to aspirin in suspected ACS and highlights the importance of applying decision aids to identify patients who would likely benefit from ticagrelor. In this study, these patients were those with an 8% likelihood of ACS using T-MACS. Many trusts still rely on TIMI scores and HEART scores to risk stratify patients, however with increasingly novel decision aids and additional research, there may be a shift towards alternative methods of risk stratification.


March 2019 

Indoor accidental hypothermia in the elderly: an emerging lethal entity in the 21st century. Paal P. Emergency Medicine Journal 2018; 35(11): 667-668

Hypothermia has become an increasingly important problem amongst an ageing population due to a number of factors influencing thermoregulation. Paal examines the results of a recent Japanese study that explored the outcomes of hypothermia by age, co-morbidities and other clinical and biochemical markers. Accidental hypothermia triggered by the cold only is classified as primary, whilst secondary hypothermia has a host of underlying aetiology related predominantly to neurological, endocrine or dermatological conditions.

The outcomes of hypothermia seem to be closely related to age, where ventricular fibrillation occurs more commonly in elderly patients during rewarming, a process that is aimed at slowing the rate of myocardial cooling. Underlying co-morbidities also results in more serious outcomes, linked to a four-fold rise in mortality (12% vs. 48%). Other studies found a higher mortality in patients following indoor hypothermia where living alone leads to unwitnessed falls in light clothes, leading to long lies whilst comorbidities prevent suitable compensation.

The summary of these studies from Japan and Europe has led to the development of several risk factors for indoor hypothermia and in-hospital death. These include being age over 75 years, having major co-morbidities, and having cardiac arrest or significant hypotension on admission. A variety of metabolic derangements such as hyperlactataemia and metabolic acidosis were also associated with more significant hypothermia and poorer outcomes.

They concluded that accidental hypothermia should be considered all year round in an increasingly elderly population, and any comorbidity predisposing to hypothermia should be treated aggressively to minimise its impact on the patient’s overall condition. There are currently significant variations in the quality of care provided to hypothermic patients depending on the frequency with which it is encountered; however robust methods should be developed to improve treatment of accidental hypothermia.


February 2019

Improvised first aid techniques for terrorist attacks. Loftus A. Emergency Medicine Journal 2018; 35:516-521.

This article explores terrorist attacks as a medium for multiple trauma events and management in the field. The article employs first aid using regular items with a ‘CABCDE’ approach. Most interventions taken go some way to providing analgesia (i.e. wound/burn dressing, splinting fractures).

In 2016 there were 37 terrorist attacks daily across the world using a variety of weapons and rescuers may have to rely on virtually nothing. ‘Cover’ is preferred over ‘concealment’, the former providing protection from high-velocity weapons while the latter provides protection only until discovered (i.e. hedges or car doors).

Prioritisation of catastrophic haemorrhage control before airway comes from avoiding the ‘lethal triad’ of coagulopathy, hypothermia and acidosis that exacerbates haemorrhage and cardiovascular instability. Dealing with major limb haemorrhage requires tourniquet application above the injury, followed by another proximal to this if bleeding continues. In the field this can be achieved with any fabric strip (i.e. cut trouser leg) and an improvised windlass between knots, turned to exceed arterial pressure. Failure rate for fabric alone is 99% but adding a windlass reduces this to 32%.

Cervical spine immobilisation is defensively overused to prevent neurological sequalae however no trial has proven benefit. Unstable cervical spine injuries occur in 1.7% of injuries, and only 0.1% show neurological injury. In mass casualties immobilisation is time consuming and patients immobilise themselves readily through neck muscle spasm. A Montana neck brace is improvised with a sheet to provide immobilisation. The recovery position is as a simple airway manoeuvre when working through multiple casualties.

Pressure bandages are useful for non-catastrophic haemorrhage and are improvised with nappies and sanitary pads to provide bulk, pressure and absorbance. Multiple tampons can pack wounds and enlarge as they absorb blood to decrease free volume and increase pressure.

Fractures from high power vehicles and blasts cause visceral damage and major haemorrhage, particularly pelvic fractures where reduction is critical. Pelvic binders improvised with sheets or trouser legs are placed beneath greater trochanters with legs together and ankles bound, and the sheet tightened to reduce the pelvis and its internal volume. Upper limb fractures can be splinted using sheets or cling-film slings, and mouldable metal sheets (i.e. tin cans) used for splinting joints.

Copious bottled water irrigates chemical injuries or burns, covering burns with cling-film to prevent fluid loss, but not circumferentially around a limb as delayed swelling may result in distal ischaemia by a tourniquet effect. Hypothermia is avoided using blankets and clothing, as well as sharing body heat with other survivors.


January 2019

Postural modification to the standard Valsalva manoeuvre for emergency treatment of supraventricular tachycardias (REVERT): a randomised controlled trial. Appleboam. The Lancet 2015.

This original trial investigated the effect of adding a passive leg lift to Valsalva manoeuvres in attempting to restore sinus rhythm in supraventricular tachycardia (SVT) in emergency departments, using a multi-centred RCT across 10 departments over 2 years. Patients not included were those requiring immediate electrical cardioversion, contraindications to Valsalva manoeuvre, and those previously included in the study.

428 cases were randomly allocated to standard Valsalva sitting up (control) or Modified Valsalva where they performed Valsalva followed by being made supine and legs elevated to 45 degrees (intervention). Both groups had Valsalva standardised to 40mmHg using a manometer to measure strain force. The procedure was repeated once if it failed the first time.

The primary outcome was sinus rhythm on ECG at one minute, reported as 17% in the control and 43% in the intervention groups (odds ratio 3.7, p<0.0001). Secondary outcomes favoured the intervention group in terms of: need for adenosine or other anti-arrhythmic treatment. There was no statistically significant difference in discharge or side-effects experienced.

This trial demonstrated a more effective, cost-free Valsalva manoeuvre that could be taught to patients on discharge to reduce re-attendance rates. A number needed to Treat (NNT) of 4 makes this intervention both effective and efficient without risk of harm.

Further Reading:




December 2018

Falling down a flight of stairs: The impact of age and intoxication on injury pattern and severity. Chatha, H. Trauma 2018 Vol. 20(3) 169-174

As the number of elderly patients living independently steadily increases, the major cause of trauma in the ED has changed from young high velocity accidents to elderly falls where low impact appears to result in major injury. This retrospective observational study of one major trauma centre looks at the impact of age on injury pattern and severity using falls down stairs, which are often difficult to assess as they are unwitnessed.

Previous studies have shown that this mechanism is important as it accounts for over half of all traumatic brain injuries in older people. Injuries after falls include traumatic brain injury (5.6%), limb fracture (22.4%), thoracic injury (7.4%). For this reason falls down stairs are now considered ‘fall from height’ to activate the trauma team.

481 patients were included, 208 over 65 years (median 78) and compared to 273 below 65 years (median 49). It was found that younger patients were more commonly intoxicated, whilst elderly patients were more likely to sustain polytrauma, particularly involving the head and face, chest and spine. Elderly also had non-statistically significant higher injury severity scores (16 vs. 13, p=0.130). Spinal injury severity was most markedly more severe (p<0.001).

The findings are supportive of a careful and diligent approach to falls in elderly patients as the risk of polytrauma is much higher than in younger patients and occult injuries could be easily missed. Spinal and head injuries are particularly important to be vigilant for as they can be subtle or delayed. Although the study does not support the use of whole body CT scanning in all patients it is likely to become more frequently used as a means to exclude significant polytrauma in these patients, particularly as observations and neurological status may only be subtly altered.

Key Points:

  1. Elderly trauma has become the most common form of trauma encountered in emergency departments, and a high index of suspicion for polytrauma is needed.
  2. Trauma from falling down stairs most commonly involves head, face, chest and spinal injuries, and in elderly patients the severity is more marked.
  3. As trauma teams are more frequently activated in what is now considered a ‘fall from height’, whole body CT imaging in these patients is becoming more common to exclude occult injury.


November 2018

Bilateral blood pressure differential as a clinical marker of acute aortic dissection in the emergency department. Um, SW et al. Emergency Medicine Journal 2018; 35:556-558

Identifying aortic dissection early is important as, untreated mortality is up to 50% within 48 hours and 90% in 1 year, but with intervention 30-day survival is close to 90%. Traditionally pulse deficit or systolic pressure difference of >20mmHg was suggestive of acute aortic dissection, the result of an intimal flap causing limb hypo-perfusion and pseudo-hypotension. Most research has focused on the combined predictive value of arm blood pressure differences and pulse deficit in aortic dissection, so this case-control study was the first focusing purely on blood pressure difference.

Data was extracted from notes using several techniques to ensure rigor and 111 cases and controls over 12 years analysed, aiming for a power of 80% and confidence interval of 95% to detect an odds ratio of 3. The data revealed that systolic BP difference of >10mmHg (odds ratio 1.8, p=0.020) and >20mmHg (OR 2.7, p=0.003) were associated with aortic dissection. Pulse deficit was much more accurate (OR 28.9, p<0.001) at detecting aortic dissection and is very specific (99.1%) but poorly sensitive (21.1%) i.e. if present, aortic dissection is highly likely, but if absent aortic dissection was still possible. This is supported by another similar study looking at pulse deficit. It seemed combining the two variables did not increase the diagnostic yield significantly.

Other authors also showed that 53% of patients attending ED have systolic BP difference of >10mmHg and 19% >20mmHg incidentally without aortic dissection after thorough investigations. The study suggests that although bilateral systolic BP differences >20mmHg are associated with aortic dissection, variability in measuring technique limits the interpretation of results and usefulness of bilateral blood pressure measurement.

Key Points:

  1. Blood pressure differentials can be associated with aortic dissection, however the specificity of this finding is low.
  2. Pulse deficit is very specific for aortic dissection, however it’s absence does not exclude this serious pathology.
  3. Using both variables does not increase the sensitivity significantly.
  4. Assessment of these factors by different staff will influence interpretation of findings and thus detection rate.


October 2018

Fluid Therapy in the Emergency Department: an Expert Practice Review

Harris, T et al. Emergency Medicine Journal Aug 2018 Vol. 35 Issue 8

This article explores the evidence for fluid prescribing in medically unwell patients in the emergency department. Previous studies have focused mainly on sepsis, so the work of Harris builds on a less thoroughly researched area by addressing common questions faced daily by clinicians assessing patients who are acutely unwell but haemodynamically normal. Normal saline is the most often prescribed formulation and it is not physiologically equal to plasma, an error in calculating its composition being responsible for this.

The use of colloids in critical illness is now a well-recognised as a cause of increased mortality and requirement for renal replacement therapy after a number of large randomised-control trials in intensive care units.

Similarly, 0.9% sodium chloride has a higher association with hyperchloraemic acidosis and renal injury compared to balanced solutions such as Hartmanns compound lactate which cause minimal acid-base upset.

The recommendation is currently to choose 0.9% saline in patients with intracerebral pathology (where the hyperosmolality may improve symptoms), and balanced solutions i.e. Hartmanns in patients with renal impairment or acidosis. 

The question of ‘how much’ remains unanswered, certainly in shock the aim is to increase venous return thus stroke volume and ultimately cardiac output to restore perfusion. Excess fluid will distend the heart beyond its optimum and risk pulmonary oedema. Volume status has long been used as a way to guide fluid therapy, and this appears to be as good as fluid resuscitation with central venous pressure monitoring or inferior vena cava collapsibility index (amount of vena cava collapsibility during positive pressure ventilation in intensive care).

Lactate is commonly used as a marker of resuscitation in A&E, and clearance is associated with adequate resuscitation, however relief of other stimulants of the sympathetic nervous system (i.e. pain, stress response) can improve lactate clearance.

A fluid challenge can be used to assess volume responsiveness – 500ml bolus in less than 30 minutes will detect blood pressure response and subsequent maintenance or fall in blood pressure. However harm can be caused as: giving fluid to a non-responsive patient risks causing fluid overload, whilst fluid boluses where organ hypoperfusion is not a problem (i.e. normal lactate) are quickly redistributed into extravascular compartments leading blood pressure to subsequently return to pre-fluid challenge levels.

The article concludes the following:

  1. Intravascular fluid deficit in sepsis is one part of a complex of cardiac and vascular dysfunction – fluid resuscitation is useful where tissue hypoxia is present.
  2. Use balanced solutions in renal injury or acidosis, and 0.9% sodium chloride in cerebral injuries.
  3. In the absence of need for fluid resuscitation, oral fluids may be a suitable alternative.
  4. Over-resuscitation is linked to poor outcomes as much as under-resuscitation is.
  5. Trials are still required to decide whether to target improved organ perfusion or cardiovascular endpoints.


September 2018 Part 1

When consulting patients in secondary care with acutely sore throat it is often obvious when intravenous antibiotics and referral to Ear, Nose & Throat specialist is required for suspected peritonsillar abscess, but there is a significant grey area when considering oral antibiotics for patients being discharged. Fortunately NICE have produced a set of guidelines on managing this group of patients, with an overarching message of antibiotic avoidance for most sore throats as most will recover within a week.

NICE recommends initial assessment using either the FeverPAIN score (1 each for Fever, Purulent tonsils, Attendance within 3 days of onset, severely Inflamed tonsils, No cough), or Centor score (1 each for Tonsil exudate, cervical lymphadenopathy, Fever, No cough) to stratify patients into 3 groups based on the likelihood of bacterial Group A beta-haemolytic streptococcal (GABHS) infection.

FeverPAIN 0 or 1, Centor 0-2 (15% chance of GABHS) – No antibiotic with safety netting

FeverPAIN 2-3 (35-40% chance of GABHS) – No antibiotic / delayed 3-5 days

FeverPAIN 4-5, Centor 3-4 (35-60% chance of GABHS) – Immediate antibiotic with safety netting

There is increasing evidence that antibiotics provide little benefit and do not shorten the course of the illness, but may cause adverse effects such as diarrhoea & nausea. A delayed prescription, although may increasing overall prescribing of antibiotics, may help prevent collection through education and empowerment.

Regarding lozengers and anaesthetic sprays there remains little evidence for their use even for analgesia.

NICE recommends phenoxymethylpenicillin 500mg four times daily for 5-10 days (Clarythromycin 250-500mg twice daily for 5 days if penicillin allergic) in adults, and age-adjusted doses of the same antibiotics for children.

Self care advice should be given to all patients being discharged, including use of regular analgesia such as paracetamol and ibuprofen, plenty of oral fluids, and that they may consider the use of medicated lozenges and mouthwashes though evidence for these is lacking).

Full guideline

Brief summary



September 2018 Part 2

A similar guideline to NICE sore throat (acute) exists for the management of otitis media (acute). Ear pain, discharge and bleeding are common presenting complaints to the emergency department in adults, but more commony children where coryza and fever are the presenting symptoms.

As before, severely systemically unwell patients or those at high risk of complications (such as mastoiditis) should be managed as in-patients.

All patients should be offered education and self-care advice including duration of illness and use of regular analgesia.

Other patients require assessment and division into several categories:

Age <2yrs with bilateral infection – delayed (3 days) / immediate antibiotic prescription

Discharge after perforation – delayed (3 days) / immediate antibiotic prescription

Age over 2yrs, unilateral and no discharge – delayed / no antibiotic prescription with safety netting

Antibiotics have not shown improvement to recurrence, peroforation or short-term hearing loss rates and may increase risk of side-effects such as diarrhoea and nausea.

NICE recommends Amoxicillin 500mg 3 times daily in adults (clarithromycin 500mg twice daily if penicillin allergic / erythromcin 500mg 4 times daily if penicillin allergic and pregnant) and age-adjusted doses of these in children. Patients that are worsening despite treatment can be changed to Co-amoxiclav 500/125mg 3 times daily if not penicillin allergic.

Brief Summary



August 2018

The original article can be found here.

Much research has been focused on the use of troponin measurement to reliably exclude acute coronary syndrome (ACS) in patients presenting with chest pain in efforts to avoid admission to hospital for serial troponin measurement. With the advent of high-sensitivity troponin (hs-Tn) this strategy is becoming a more realistic possibility as patients can potentially be assessed and investigated in the emergency department setting with a single negative rule-out troponin result, though trials still suggest that the sensitivity of a single troponin result is not sufficient in adequately ruling out ACS, particularly in patients who attend early (within 2 hours of chest pain) where troponin may not have started to rise.

Further studies have assessed whether this strategy can be used in ‘low-risk’ patients identified with a risk stratification tool, and a strategy has been published by the National Institute of Clinical Excellence (NICE) in identifying these patients and applying the use of a single negative troponin result to these patients and can be found here.

This article, published in the Emergency Medicine Journal (EMJ), looks further at the use of single hs-Tn assays in patients presenting with chest pain and compares the risk stratification tools available to clinicians in the emergency department including TIMI, HEART & GRACE scoring methods.

The article examined patients with suspected cardiac chest pain without significant ECG changes and each patient had their TIMI, GRACE & HEART scores calculated and a single hs-Tn taken. The primary end-point was 30 day major adverse cardiac event (MACE). The study cohort of 1,000 patients was followed and 189 found to develop MACE by 30 days. When combined with a negative single troponin result, HEART score 3 or less was found to be most effective in identifying low risk patients (sensitivity 99.5%, negative predictive value 99.6%). TIMI score of zero performed less well (sensitivity 97.4%, NPV 97.8%), while GRACE 0-55 performed least well (sensitivity 95.2%, NPV 95.8%). These findings are mirrored in previous similar studies.

These results are not sufficient to adequately exclude ACS in most patient groups, and hs-Tn assays are still not available in many centres, therefore standard practice of serial troponin results continues to be the norm until further work can further confirm the safety of a single troponin assay in ruling out cardiac causes of chest pain.


July 2018

Petkar S, Bullock I, Davis S, et al. Transient Loss of Consciousness: summary of NICE guidance 

The National Institute of Clinical Excellence (NICE) published their first guideline for transient loss of consciousness in 2010 as this condition became a far more common presentation to both general practitioners and the emergency department. This summary of the guidelines presented in the Heart Journal in 2013 appraises the NICE guidance and presents it in a readily digestible format.

The NICE guidelines help to divide ‘collapses’ and ‘blackouts’ into those patients suffering simple faints or situational syncope and those with more significant underlying pathology warranting further investigation by specialists, be that neurology input after a suspected first fit, or cardiology assessment for suspected cardiac syncope. It employs a simple initial assessment including history, examination and simple investigations to stratify patients into these groups, followed by further specialist assessment as needed.

This article summarises the features NICE recommend identifying in the history after transient loss of consciousness to help diagnose simple faints, cardiac syncope, and epilepsy. It also covers significant examination findings, and features in the 12 lead ECG that are of concern and warrant further assessment by a specialist as well as exploring the different causes of cardiac syncope (cardiomyopathy, structural/valvular disease, and arrhythmia) and the approach a cardiologist might take when investigating these further.



June 2018

Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801–810. doi:10.1001/jama.2016.0287

In 2016, 15 years after the previous definitions of sepsis were coined, a new consensus was reached about what ‘sepsis’ actually means. These changes were developed based on a broader understanding of the continuum (or lack thereof) that from sepsis through severe sepsis to culminate in septic shock.

These new definitions set the stage for the National Institute for Clinical Excellence (NICE) and other healthcare bodies to develop new guidelines in the recognition and management of sepsis. The new definitions of sepsis discard the old belief that sepsis could be anything but severe. By definition sepsis is defined as ‘a dysregulated host response to an infectious trigger resulting in life-threatening organ dysfunction’ and therefore should always be treated aggressively if suspected.

We previously relied on recognition of sepsis by a Systemic Inflammatory Response Syndrome (SIRS) with evidence of infection, however it is now understood that many patients that ultimately have ‘sepsis’ do not initially display evidence of SIRS due to a number of factors that may blunt the inflammatory response to infection. NICE have now suggested recognising the possibility of infection initially, followed by using a series of red and amber flags (presenting symptoms and serum biochemistry/lactate) to risk stratify patients and direct early therapy.

The use of scoring systems for organ dysfunction such as SOFA in intensive care and quick SOFA (respiratory rate >22/min, altered mental state, or systolic blood pressure <100mmHg) at the bedside have been validated for use in early recognition of sepsis, however the qSOFA specificity is poor due to its variables being seen in a vast number of diseases.

This study helps to improve the definition of sepsis and how to decide its early management and these, along with NICE guidelines, should be absorbed, understood and applied. The article still fails to adequately describe sepsis which remains a nebulous entity with multiple presentations, a variety of triggers and a host of unconnected risk factors.