The term anaphylaxis refers to a rapidly progressive, systemic allergic reaction that may be fatal. The definition of anaphylaxis in the literature is quite varied and its aetiology may include both IgE and non IgE mediated mechanisms.

The current definition as agreed upon by the Joint Council of Allergy, Asthma and Immunology; and the American College of Allergy, Asthma and Immunology is as follows [12]:

This definition can be difficult to remember and apply to patients in the setting of a busy Emergency Department. However, it is necessary to maintain such a definition to differentiate anaphylaxis from less severe allergic reactions.

Anaphylaxis is a common cause for presentation to the Emergency Department. A recent retrospective case based study of patients aged 13 years and above, over a period of 1 year was performed in a Brisbane hospital, revealing an incidence of 1 in 439 Emergency Department presentations. One conclusion drawn from this study was that a moderate sized Emergency Department could expect to see at least one case of anaphylaxis each week [3].

Given that this is a common and sometimes fatal condition we sought to audit our Emergency Department's management of anaphylaxis and determine if an educational intervention could improve outcomes.

We performed a retrospective review of the electronic records for all paediatric and adult cases that presented to our Emergency Department. We identified all triage labels or final diagnoses with the terms anaphylaxis, allergy, allergic reaction, rash, urticaria, and angioedema with an aim to identify cases of anaphylaxis and milder forms of immediate type allergic reactions presenting with a combination of urticaria and nonairway threatening angioedema. While the presence of cutaneous manifestations with nonairway threatening angioedema is not strictly defined as anaphylaxis, we included these events to determine how they were being managed.

We graded the severity of each case using a severity grade adapted from Brown 2006 with mild reactions characterised by urticaria or angioedema not threatening the airway; moderate reactions characterised by nausea, vomiting, abdominal pain, dyspnoea, wheeze, stridor, throat constriction, diaphoresis or chest tightness; and severe reactions characterised by cyanosis or SpO₂ <92%, hypotension (systolic blood pressure <90 mm Hg in adults), confusion and syncope [4].

This grading system was chosen in preference to others because of its practicality and ease of use in an Emergency Department setting. The use of adrenaline, antihistamines, and corticosteroids was reviewed as well as observation time and organisation of follow up with an immunologist/allergist.

After 5 months a targeted education program consisting of a 45-minute lecture discussing the recognition, grading and management of anaphylaxis was provided to the Emergency Department physicians. Our proposed treatment algorithm was also provided (Table 1) and kept in the Emergency Department for rapid access. Data was collected for a further 5 months following this intervention.

As this was a retrospective, deidentified audit for educational and quality control purposes, an ethics submission to our Institutional Ethics Committee was not required.

In the first 5 months of the audit 38 patients met the inclusion criteria (Table 2). When compared to our proposed management algorithm of anaphylaxis many deficiencies in management became apparent. There were no fatalities amongst the 38 patients reviewed.

Our audit reveals a number of obstacles and deficiencies which have been highlighted in multiple reviews examining the management of anaphylaxis [356]. The first major obstacle relates to the lack of a simple definition of anaphylaxis that can be easily recalled from memory by emergency staff.

As our audit revealed, recognition of the severest forms of anaphylaxis is often achieved leading to prompt and appropriate management in the majority of cases. Patients with moderate severity anaphylaxis are frequently inappropriately categorised and often receive inadequate or inappropriate treatment.

Another common problem identified was the inadequate use of adrenaline. Despite the well established role of adrenaline in the management of anaphylaxis it is not uncommon for this life saving medication to be withheld. A multicenter study looking at Emergency Department visits for food allergy revealed that among patients with severe reactions only 24% received adrenaline. Of the 97% of patients that were discharged from the Emergency Department, only 16% of patients were prescribed an adrenaline auto injector [7].

In our cohort, 54% of patients with moderate and severe anaphylaxis received adrenaline prior to the educational intervention. This increased to 75% following the intervention. Misconceptions about the safety profile of adrenaline lead to it being withheld in preference for drugs that are perceived to be safer alternatives i.e., antihistamines, salbutamol and corticosteroids. While these can supplement the use of adrenaline, none take its place in the treatment of bronchospasm and vascular collapse which are common features of severe anaphylaxis. In the context of moderate to severe anaphylaxis the benefits of using adrenaline outweigh the risks [8910].

Prospective controlled trials looking at the efficacy of adrenaline cannot be performed for ethical reasons as patients cannot be randomised to a placebo arm. However, retrospective studies have shown that fatalities, caused by anaphylaxis, generally occur as a result of delay in administration of adrenaline supporting the beneficial role of this drug [1112].

Antihistamines aid in treatment of urticaria and rhinorrhoea associated with anaphylaxis but do not prevent or relieve bronchoconstriction, airway angioedema or vascular collapse [13]. First and second generation H₁ antihistamines are very frequently prescribed for the management of anaphylaxis as they are often viewed as being harmless drugs and safer to give than adrenaline. This misconception can result in delay in the administration of adrenaline. Promethazine, the only H₁ antihistamine available in the intravenous form in Australia, is often given in preference to oral formulations of nonsedating H₁ antihistamines and can exacerbate hypotension by causing vasodilatation. Promethazine is capable of crossing the blood brain barrier occupying over 70% of the H₁ receptors within the brain [14]. This can lead to side effects such as drowsiness which can confuse the clinical presentation of severe anaphylaxis with central nervous system hypoperfusion and as such should be avoided in the setting of anaphylaxis [1314]. Second generation H₁ antihistamines do not readily cross the blood brain barrier and are generally free of these side effects. They are also less likely to cause the antimuscarinic, antiserotonin, and anti-α adrenergic side effects common to first generation H₁ antihistamines [14].

Insufficient observation time was another problem identified. The observation time appeared to be proportionate to the severity of the reaction in most cases. This could reflect a lack of knowledge about biphasic and protracted anaphylaxis and the need for appropriate observation time. Unfortunately there was an 11% reduction in the number of patients observed for at least 4 hours. This is likely to reflect a pressure to clear beds and meet bench marks around patient turnaround times in the Emergency Department.

The final issues related to discharge with an adrenaline auto-injector and follow up with an immunologist. Only 11 of the 38 patients reviewed prior to the education intervention had an appointment organised with an immunologist and none were prescribed an adrenaline auto-injector or given an anaphylaxis action plan unless they were reviewed by an immunologist while in hospital. Most of the patients were advised to return to the Emergency Department if their symptoms returned. Again this probably reflects a lack of appreciation of the life saving role of adrenaline in the management of anaphylaxis and a lack of knowledge on how to prescribe this drug from the Emergency Department. In the 5 months after the educational intervention 17 patients were identified that should have been prescribed an adrenaline auto-injector but only 6 received a prescription. While this is an improvement compared to the pre-educational intervention group, there is still much progress to be made in this area. Fortunately the number of referrals to an allergist/immunologist increased by 24% an indication that the educational intervention increased awareness regarding the need for appropriate follow-up.

We have shown that a single educational intervention can improve many outcomes including appropriate adrenaline use, sedating antihistamine avoidance, and referral to allergy/immunology services. However, there is much progress to be made and this will likely require regular education sessions with the Emergency Department staff particularly as there is usually a high turnover of staff in these departments.

In conclusion, our audit reveals a number of deficiencies in the management of anaphylaxis. Unfortunately these deficiencies are not unique to our Emergency Department. Failure to identify and grade anaphylaxis correctly often leads to delayed or inappropriate treatments. Inadequate Immunology/Allergy follow-up and failure to prescribe adrenaline auto injectors on discharge from the Emergency Department are all major deficiencies which need to be addressed. We have shown that targeted educational activities aimed at the Emergency Department hospital staff may improve outcomes.